Solute transport in Sphagnum dominated bogs

Transcription

1 Wouter Patberg Solute transport in Sphagnum dominated bogs The ecophysiological effects of mixing by convective flow Solute transport in Sphagnum dominated bogs The ecophysiological effects of mixing by convective flow Wouter Patberg

2 Wouter Patberg Solute transport in Sphagnum dominated bogs The ecophysiological effects of mixing by convective flow

3 Colophon Graphic design Mirjam Patberg Pictures Wouter Patberg Cover: Diepveen, Dwingelderveld page 116: Veerles Veen, Dwingelderveld Printing Grafimedia, Facilitair bedrijf RuG The research reported in this thesis was carried out at the Laboratory of Plant Physiology, which is part of the Centre of Ecological and Evolutionary Studies (CEES) of the University of Groningen, P.O. Box 11103, 9700 CC Groningen, The Netherlands. This research was financially supported by ALW grant ALW (Earth and life sciences) is part of NWO, the Netherlands Organization for Scientific Research. This thesis was printed with financial support from the University of Groningen. ISBN book: ISBN print:

4 RIJKSUNIVERSITEIT GRONINGEN Solute transport in Sphagnum dominated bogs The ecophysiological effects of mixing by convective flow Proefschrift ter verkrijging van het doctoraat in de Wiskunde en Natuurwetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. E. Sterken, in het openbaar te verdedigen op vrijdag 16 december 2011 om 11:00 uur door Wouter Patberg geboren op 12 april 1976 te Hoorn

5 Promotores Prof. dr. J.T.M. Elzenga Prof. dr. A.P. Grootjans Copromotor Dr. A. J. P. Smolders Beoordelingscommissie Prof. dr. R. van Diggelen Prof. dr. J.G.M. Roelofs Prof. dr. H. Joosten

6 Contents Chapter 1 General introduction 7 Chapter 2 The transport of solutes by buoyancy-driven water flow 17 in a water-saturated Sphagnum layer; laboratory and field evidence Wouter Patberg, Gert Jan Baaijens, Christian Fritz, Ab Grootjans, Rodolpho Iturraspe, Alfons Smolders and Theo Elzenga Chapter 3 Field characteristics of buoyancy-driven water flow 33 and its global occurrence Wouter Patberg, Erwin Adema, Myra Boers, Gert Jan Baaijens, Christian Fritz, Ab Grootjans, Rodolfo Iturraspe, Alfons Smolders and Theo Elzenga Chapter 4 Physiological evidence for internal acropetal transport of nitrogen 45 in Sphagnum cuspidatum and S. fallax Wouter Patberg, Bikila Warkineh Dullo, Alfons Smolders, Ab Grootjans and Theo Elzenga Chapter 5 The importance of groundwater carbon dioxide 57 in the restoration of Sphagnum bogs Wouter Patberg, Gert Jan Baaijens, Alfons Smolders, Ab Grootjans and Theo Elzenga Chapter 6 Photosynthesis of three Sphagnum species after acclimatization 75 to high and low carbon dioxide availability Wouter Patberg, Jan Erik van der Heide and Theo Elzenga Chapter 7 Summary and synthesis 91 References 105 Samenvatting 119 Dankwoord 127

7 Chapter 1

8 General introduction

9

10 Background Bog ecosystems Bogs are wet, acidic, peat forming ecosystems which generally have a low cover of vascular plants and are dominated by mosses of the genus Sphagnum (peat moss or veenmos in Dutch; Rydin & Jeglum, 2006). Sphagnum mosses play an important role in creating their own environment, thereby gaining competitive advantage over other plant species (Kilham, 1982; Malmer et al., 1994; Van Breemen, 1995). Sphagnum mosses consist of densely clustered developing and expanding branches at the top of the plant, the so-called capitulum (plural: capitula), and the stem with fully developed branches. During growth, the stem elongates from the capitulum and the branches become distributed along the new stem. The lower portion of Sphagnum plants gradually die and will form peat. The capitulum is the part of the plant where the main production of biomass takes place. Also, the highest metabolic activity and nutrient uptake in Sphagnum mosses was measured in the capitulum (Aldous, 2002a; Johansson & Linder, 1980; Malmer, 1988; Malmer et al., 1994; Robroek et al., 2009; Rydin & Jeglum, 2006). For example, the contribution of photosynthetic activity of the capitulum in dense Sphagnum hummocks and lawns has been estimated to be 98% in Sphagnum fuscum and 60% in S. balticum (Johansson & Linder, 1980). Looking at a cross section of a Sphagnum bog, it can be divided into two layers (Clymo, 1984; Ingram, 1978). The upper 10 to 40 cm is called the acrotelm. This layer contains the living part of the Sphagnum mosses and is a highly permeable layer where the groundwater table fluctuates. The spongy acrotelm has a high hydraulic conductivity and the ability to retain water in dryer periods, thus having a strong self-regulating effect on the depth of the water table (Ingram, 1978). The layer below is called the catotelm, a slowly permeable, permanently water-saturated anaerobic layer which contains most of the peat (Ingram, 1978). Bog ecosystems are ombrotrophic, which means that they receive their nutrients solely by atmospheric deposition and are therefore characterized by a low nutrient availability (Rydin & Jeglum, 2006). The ability of Sphagnum mosses to deal with low nutrient availability is often attributed to their high nutrient retention capacity. The bog surface appears as a layer of densely packed Sphagnum capitula that efficiently intercept nutrients coming from the atmosphere (Aldous, 2002a; Woodin & Lee, 1987). Sphagnum mosses lack vascular tissue for water and nutrient uptake, but they can take up water and nutrients over the entire surface of the plant because they lack a cuticle (Brown, 1982; Brown & Bates, 1990). Sphagnum mosses are able to survive under low nitrogen conditions due to their very efficient nitrogen utilization (Bridgham, 2002; Li & Vitt, 1997) ranging from 50 to 90% (Aldous, 2002a; Li & Vitt, 1997). Woodin & Lee (1987) even measured a retention of 100% of inorganic nitrogen at an unpolluted site, whereas chloride and sulphate were passing freely through the moss mat. That the efficiency of nitrogen retention by Sphagnum results in a competitive advantage of Sphagnum over vascular plants was shown by Aldous (2002a): vascular plants received <1% of N recently added by wet deposition. Moreover, the ability of Sphagnum species to grow in an environment with very low nutrient concentrations is often attributed to their pronounced capacity to exchange hydrogen ions for mineral cations (Daniels & Eddy, 1985). By doing so, releasing H + ions in exchange for dissolved cations, Sphagnum species have the ability to acidify their environment (Clymo, 1963; Clymo & Hayward, 1982). Another important feature General introduction 9

11 of Sphagnum to survive in nutrient poor habitats is the ability to recycle nutrients from senescent Sphagnum tissue very efficiently (Aldous 2002a, 2002b; Malmer, 1988; Van Breemen, 1995). In contrast to vascular plants, Sphagnum mosses lack stomata. Consequently, they cannot control their water loss actively. Water lost by evaporation must be replaced by rain or by the water from the peat below. Since there are no specialized cells for water transport in the stem, the upwards water transport takes place externally by capillary movement facilitated by a network of spaces between leaves, stems and branches (Rydin & Jeglum, 2006). Sphagnum mosses are characterized by their high water-holding capacity (Clymo & Hayward, 1982). Much water can be retained in the hyaline cells, which are large, dead cells, which make up about 80% of the plant s volume. Hyaline cells can rapidly absorb water through their pores (with have diameters from 5-20 μm), and water can be retained against suction pressures of kpa (Van Breemen, 1995). The hyaline cells are enclosed in a network of narrower chlorophyllose cells, the cells that contain chlorophyll and enable the mosses to photosynthesize. The microtopography of a Sphagnum bog is characterized by a diversity of wet depressions (pools and hollows), relatively dry lawns and dry hummocks (Andrus et al., 1983). Each microhabitat is occupied by a different group of Sphagnum mosses, broadly defined by its water retention capacity (Andrus et al., 1983; Hayward & Clymo, 1982). Sphagnum species observed at increasing height above the water table have an increasing capacity to conduct water by capillary action (Clymo & Hayward, 1982). Due to the wet, anoxic and acidic conditions in the catotelm, the production of Sphagnum exceeds the decomposition of organic material, resulting in the accumulation of organic material or peat (Clymo & Hayward, 1982; Clymo et al., 1998). Peatlands store more carbon than any other terrestrial ecosystem. It has been estimated that the accumulation of peat has led to a carbon pool that is about one-third of the global soil carbon pool. This is quite remarkable since peatlands occupy only about 3% of the world s total land area (Rydin & Jeglum, 2006). Peatlands, including Sphagnum bogs, function as a net sink for CO 2 (Clymo et al., 1998; Gorham, 1991) and as a consequence, Sphagnum bogs play an important role in global carbon cycling (Bridgham et al., 2001a; Clymo et al., 1998; Gorham, 1991). Due to the extensive exploitation for fuel, agriculture and forestry over many centuries, and the ongoing global warming, living (peat forming) peatlands have become endangered ecosystems throughout the world (Rochefort & Price, 2003). Even nowadays (extensive) peat extraction activities take place for commercial use in, for example, Canada, Scandinavia, Ireland and the Baltic states (Joosten, 2009). Due to the important role of peatlands in the global carbon cycle, and because of their unique ecological values, conservation and restoration of these ecosystems is necessary, preventing stored CO 2 being released into the atmosphere, which will lead to accelerated global warming. Globally much effort is dedicated to the restoration of damaged peatlands. However, the restoration of bog remnants in particular, has proven to be fairly complicated and not always successful (Grootjans et al., 2006; Money & Wheeler, 1999; Money et al., 2009). For the successful conservation and restoration of Sphagnum-dominated bogs, knowledge about environmental constrains for Sphagnum growth is necessary. 10 Solute transport in Sphagnum dominated bogs

12 Nutrient supply in Sphagnum bogs For their nutrient supply, Sphagnum bogs mainly depend on wet and dry atmospheric deposition. However, it has been shown that under non-polluted conditions the annual input of nutrients from atmospheric deposition is often insufficient to sustain the observed primary production in these systems (Aerts et al., 1999; Aldous, 2002a, 2002b; Bowden, 1987; Bridgham, 2002; Damman, 1978, 1986; Morris, 1991; Pakarinen, 1978; Rosswall & Granhall, 1980; Urban & Eisenreich, 1988). Therefore, other nutrients sources must be involved. Under natural conditions Sphagnum bogs are often nitrogen-deficient (Aerts et al., 1992; Bragazza et al., 2004; Bridgham et al., 2001b; Gunnarsson & Rydin, 2000; Li & Vitt, 1997) and therefore the availability of nitrogen is of special interest. However, Sphagnum growth has also been shown to be limited by phosphorus (Aerts et al., 1992; Bridgham et al., 1996), potassium (Damman, 1978; Pakarinen, 1978) and carbon dioxide (Rice & Giles, 1996; Smolders et al., 2001). Sources of nitrogen include N-fixation by cyanobacteria associated with Sphagnum and other plants (Gerdol et al. 2006), the internal reallocation of nitrogen from older senescent tissues to the metabolically active capitula (Malmer, 1988) and the mineralization of senescent Sphagnum plants at the border of the acrotelm and catotelm. The mineralization of N has been shown to be the most important nitrogen source for Sphagnum (Aldous, 2002a; Bowden, 1987; Bridgham, 2002; Urban & Eisenreich, 1988). Gerdol et al. (2006) showed that direct retention of N from precipitation is less important than recycling of mineralized N to support Sphagnum growth (Aldous, 2002b; Bowden, 1987; Bridgham, 2002; Urban & Eisenreich, 1988). The importance of re-mineralization of nitrogen for Sphagnum growth has been demonstrated in situ by Urban & Eisenreich (1988). They calculated the assimilation of nitrogen by plants (primarily Sphagnum) to be 66 kg ha -1 yr -1, whereas only 14.6 kg N ha -1 yr -1 was supplied by total inputs. The remainder was supplied by mineralization of the peat. In Sphagnum bogs the highest mineralization rates are found in the aerobic zone at the transition zone from acrotelm to the anaerobic catotelm (Bridgham et al., 1998; Malmer, 1993; Waddington et al., 2001). In contrast, the highest metabolic activity and nutrient uptake in Sphagna takes place in the upper part of the plant, the capitulum (Aldous, 2002a; Johansson & Linder, 1980; Malmer, 1988; Malmer et al., 1994; Robroek et al., 2009; Rydin & Jeglum, 2006). The spatial separation between the actively growing photosynthesizing capitula and the mineralization of nutrients requires an efficient nutrient transport system. Several nutrient transport mechanisms have been described for Sphagnum bogs. Throughout the water layer nutrients are passively distributed by diffusion. Above the water layer solutes might be transported upwards to the capitula through the extracellular capillary spaces between pendant branches and stems (Hayward & Clymo, 1982). Rydin & Clymo (1989) demonstrated the internal acropetal transport of carbon and phosphorus. Complementary to the abovementioned transport mechanisms, Baaijens (1982) and Rappoldt et al. (2003) reported on a phenomenon called buoyancy-driven water flow as a possible mechanism for the external transport of nutrients in a water-saturated Sphagnum layer. Buoyancy-driven water flow Buoyancy-driven water flow is the vertical convective flow of water in a water-saturated peat moss layer, driven by the temperature difference between day and night. Due to the temperature drop during the night the surface of the water layer will cool down, resulting in a relative cold layer on General introduction 11

13 top of a warmer layer. Because of the difference in density between these two layers, the cold water will sink and the warm water will rise. Evidence for the occurrence of buoyancy-driven water flow in a water-saturated Sphagnum layer, based on theoretical and experimental grounds, was provided by Rappoldt et al. (2003). The development of buoyancy flow in a water-saturated Sphagnum layer is determined by the Rayleigh (Ra) number of that layer. Rappoldt et al. (2003) calculated that buoyancy flow occurs if the system s Ra number exceeds 25. For a typical peat moss layer, a temperature difference of 10 degrees between day and night will result in a Ra number of 80 which is suitable for the quick development of buoyancy flow (Rappoldt et al., 2003). Adema et al. (2006) provided evidence for the occurrence of buoyancy flow in the field; based on the hydraulic conductivity (k) of the Sphagnum layer and a temperature difference between day and night of 8 C, in a small Sphagnum dominated peat bog in the Netherlands, the calculated Ra number was sufficiently high to induce buoyancy flow. The convective flow of water will result in the mixing of solutes as well. However, direct evidence for nutrient transport is lacking. It is hypothesized that nutrients originating from decomposition in the lower acrotelm, will be transported upwards by buoyancy flow and may become available for the growing Sphagnum capitula, thereby contributing to the nutrient supply of the Sphagnum plants. Moreover, oxygen produced by photosynthesis in the upper Sphagnum layer will be transported downwards resulting in increased decomposition rates. In turn, nutrients will become available to the growing Sphagnum when transported upwards by buoyancy flow. Consequently, buoyancy flow might be an essential mechanism in the efficient recycling of nutrients in Sphagnum bogs. Aim and outline of this thesis Part I: Buoyancy-driven water flow as a transport mechanism The main objective of the first part of this thesis is to determine the importance of buoyancy-driven water flow in the nutrient distribution in Sphagnum bogs. In Chapter 2 (The transport of solutes by buoyancy-driven water flow in a water-saturated Sphagnum layer; laboratory and field evidence), we asked ourselves the basic question whether nutrients are indeed transported by buoyancy flow. To answer this question a straightforward, but effective mesocosm experiment was performed in a temperature-regulated climate chamber. Buoyancy flow was generated in a water-saturated Sphagnum matrix and the transport of solutes by buoyancy flow was visualized by the addition and subsequent monitoring of a coloring dye. It became evident that buoyancy flow can act as a fast and efficient transport mechanism (Chapter 2). In accordance with Rappoldt et al. (2003), a reversal of the nutrient gradient due to the occurrence of buoyancy flow was possible and thereby induce a stepwise increase in the nutrient concentration near the capitula. Consequently, the importance of buoyancy flow as a transport mechanism in supplying the capitula is also determined by the ability of Sphagnum capitula to enhance the uptake and assimilation by (and thus benefit from) this increased nutrient availability. The amount of nutrients taken up by Sphagnum depends on the nutrient concentration and the affinity of the uptake mechanism for the substrate. In the case of, for example, temporary high ammonium concentrations in the upper Sphagnum layer due 12 Solute transport in Sphagnum dominated bogs

14 to buoyancy-driven water flow, Sphagnum must have a suitable uptake mechanism to benefit optimally from the situation. Therefore, we determined the uptake kinetics of ammonium by the capitula of S. cuspidatum and S. fallax. The possible role of the cation-binding sites in the uptake of nutrients is taken into consideration as well. In Chapter 2 also the findings of a field experiment, in which the transport of labeled nitrogen ( 15 N) in a Sphagnum layer by buoyancy-driven water flow and the subsequent uptake by the Sphagnum capitula are reported. Whereas the laboratory experiments (Chapter 2; Rappoldt et al., 2003) are all conducted under controlled conditions and with homogeneous samples, in the field temporal and spatial variation in temperature and hydraulic conductivity might occur and influence the occurrence and size of the buoyancy cells. Therefore, in Chapter 3 (Field characteristics of buoyancy-driven water flow and its global occurrence), a series of vertical temperature profiles were recorded in a pristine Sphagnum bog to validate the theoretical predictions in a natural situation. Based on the measured hydraulic conductivities and ambient day and night air temperatures, the Ra numbers of the Sphagnum sites were calculated. Based on these Ra numbers, the occurrence of buoyancy flow in the field could be predicted using the model described by Rappoldt et al. (2003). Additionally in Chapter 3, the possible occurrence of buoyancy-driven water flow in Sphagnum bogs throughout the world was determined. Geographical Information System software was used to analyze worldwide daily temperature data and model the occurrence of buoyancy flow in peatlands throughout the world. The importance of buoyancy-driven water flow in the nutrient supply of Sphagnum and nutrient cycling in bogs depends on the transport rate relative to other transport mechanism. To date, diffusion and internal transport were the known mechanisms by which nutrients were transported throughout a water-saturated Sphagnum layer. Note that capillary transport is often mentioned as a nutrient transport mechanism (Hayward & Clymo, 1982), but this type of transport is only possible above the water table and therefore not taken into account here. Diffusion and internal transport are both slow processes. For example, the diffusion coefficients for oxygen and ammonium are respectively and cm 2 s -1 at 20 C (Boudreau, 1997). Internal transport is estimated to distribute solutes throughout the plant with a half time of about 11 days, an estimation based on the symplasmic apical transport of 14 C (Rydin & Clymo, 1989). Moreover, in a review on internal transport in non-vascular plants (Raven, 2003) it was stated that there is no evidence for symplastic transport in Sphagna faster than can be accounted for by diffusion. As a consequence, we expect buoyancy-driven water flow to play an important role in the nutrient distribution in Sphagnum bogs. In earlier studies (Aldous, 2002b; Bridgham, 2002) the contribution of translocation to the nitrogen supply of the capitula was shown to be significant. Although, the ability of Sphagnum to transport nitrogen internally was widely assumed (Bonnett et al., 2010; Bragazza et al., 2005; Gerdol et al., 2006; Limpens & Berendse, 2003; Limpens & Heijmans, 2008; Malmer, 1988), internal transport of nitrogen had never been demonstrated. The assumption that N is transported internally is mainly based on the observations of internally transported carbon and phosphorus (Rydin & Clymo, 1989). This idea was supported by the often observed higher C:N ratios in stems than in capitula (e.g. Malmer, 1988) and are taken as an indication for the internal reallocation of N from the stem to capitula. However, experimental evidence for the internal reallocation of N was lacking. Chapter 4 (Physiological evidence for internal acropetal transport of nitrogen in Sphagnum cuspidatum and S. fallax) deals with the contribution of internal transport in the upward translocation of mineralized nitrogen in Sphagnum bogs. Two Sphagnum species, General introduction 13

15 Sphagnum cuspidatum and S. fallax, were used in experiments in which diffusion and capillary transport were excluded and the internal transport of nitrogen was monitored. For both Sphagnum cuspidatum and S. fallax, a slow but significant acropetal transport of nitrogen through an internal mechanism was observed. Moreover, the rate at which nitrogen was transported internally was estimated and its importance relative to buoyancy-driven water flow, is discussed. Part II: The importance of carbon dioxide for the growth of Sphagnum The second part of this thesis focuses on the importance of carbon dioxide for the growth of Sphagnum. In contrast to vascular plants, Sphagnum mosses lack a cuticle and stomates to regulate photosynthesis (Rydin & Jeglum, 2006), but are surrounded by an external water film through which gas exchange for photosynthesis is taking place. The photosynthetic rate of Sphagnum mosses has been shown to be a compromise between external water content and the availability of CO 2 (Schipperges & Rydin, 1998; Silvola, 1990; Titus et al., 1983). At low water contents, dehydration inhibits photosynthesis whereas at very high water contents Sphagnum species may suffer from carbon limitation due to very thick boundary layers (Jauhiainen & Silvola, 1999; Rice & Giles, 1996; Silvola, 1990; Titus et al., 1983; Williams & Flanagan, 1996). Since the diffusion of CO 2 is about 10 4 times lower in water than in air, external water films can form large barriers for gas exchange, reducing the supply of CO 2 towards the carbon assimilating cells resulting in a reduced photosynthetic rate (Bowes & Salvucci, 1989; Rice & Giles, 1996; Silvola, 1990; Williams & Flanagan, 1996). To overcome this problem many aquatic plant species make use of carbon concentrating mechanisms (CCM), which enhances the accumulation of carbon under water (Maberly & Madsen, 2002). The most frequently used mechanism is the use of bicarbonate (HCO 3- ) as a carbon source in photosynthesis (Prins & Elzenga, 1989). Sphagnum mosses lack such a CCM. Like most aquatic bryophytes (Raven et al., 1985), Sphagnum mosses are known to be obligate CO 2 users (Bain & Proctor, 1980) and are therefore solely dependent on the diffusive supply of CO 2 to the site of carbon fixation. In obligate CO 2 users high rates of underwater photosynthesis can only be sustained when the leaves are exposed to high concentrations of CO 2 (Jauhiainen & Silvola, 1999; Silvola, 1990). Because CO 2 is continuously produced by aerobic and anaerobic decomposition processes dissolved CO 2 concentrations are normally much higher in upper peat layers than atmospheric ones (16 μmol L -1 vs μmol L -1 ; Bridgham & Richardson, 1992; Lamers et al., 1999; Silvola, 1990; Yavitt et al., 1997; Glatzel et al., 2004; Smolders et al., 2001; Waddington et al., 2001). High CO 2 concentrations can compensate for low diffusion rates and ensure the substrate delivery for photosynthetic carbon fixation to be sufficient (Maberly & Madsen, 2002; Silvola, 1990). The refixation of CO 2 from decomposition processes has been unambiguously demonstrated by Rydin & Clymo (1989) and Turetsky & Wieder (1999). This so-called substrate-derived CO 2 has been shown to be an important carbon source for aquatic and emergent Sphagnum mosses (Baker & Boatman, 1990; Paffen & Roelofs, 1991; Riis & SandJensen, 1997; Roelofs, 1983; Smolders et al., 2001; Smolders et al., 2003). As a consequence, increased ambient atmospheric CO 2 concentrations (up to twice ambient) appeared to have limited effect on the growth of Sphagnum as outlined by Smolders et al. (2001). Chapter 5 (The importance of groundwater carbon dioxide in the restoration of Sphagnum bogs) focuses on the importance of CO 2 for Sphagnum in a field situation. The study was performed in the Dwingelderveld, a nature reserve in the Netherlands characterized by numerous small, 14 Solute transport in Sphagnum dominated bogs

16 damaged Sphagnum bogs, distributed throughout the area. Since the start of restoration measures, the developmental success between bogs has varied significantly; some bogs developed well, whereas others did not. Peat extraction has removed the bulk of organic material and the highly decomposed, humified peat which is left behind, has only limited CO 2 production rates (Bridgham & Richardson, 1992; Glatzel et al., 2004; Tomassen et al., 2004; Waddington et al., 2001). Therefore, for the successful restoration of cut-over Sphagnum bogs an additional carbon source might be essential for the re-establishment of Sphagnum mosses. It is hypothesized that in these hydrologically degraded bog remnants the restoration of Sphagnum growth is limited by the availability of CO 2. Bog waters analysis showed that the well-developed bogs received C-rich water from outside the bogs. It was concluded that high CO 2 availability is a pre-requisite for the successful re-establishment of Sphagnum mosses and subsequent bog development. Despite the obvious importance of a high CO 2 availability for Sphagnum, the physiological background of this apparent high CO 2 requirement of Sphagnum has never been established. Therefore, the physiological background of carbon uptake by Sphagnum was investigated as well. In Chapter 5 the plants used in the experiments were grown under ambient CO 2 conditions. However, For Sphagnum fuscum, a hummock forming species, acclimation to CO 2 levels has been shown (Jauhiainen & Silvola, 1999). Culturing plant under high CO 2 availability resulted in lower photosynthetic rates compared to plants that were grown under CO 2 -limiting conditions (Jauhiainen & Silvola, 1999). Therefore, we expected physiological adaptations in carbon dioxide uptake in response to the CO 2 concentration during the culturing period. Therefore in Chapter 6 (Photosynthesis of three Sphagnum species after acclimatization to high and low carbon dioxide availability) the physiological characteristics of carbon uptake by three different Sphagnum species was investigated for plants grown for a long period at high and low CO 2 availability. Chapter 7 (Summary and synthesis) is a combined summary and synthesis of this thesis. The main focus in this chapter lies on the ecological importance of buoyancy flow in Sphagnum-dominated bogs. For different Sphagnum habitats (hollows, lawns and hummocks) the contribution of buoyancy flow in the nutrient supply of Sphagnum will be compared to other nutrient transport mechanisms. Finally, some findings in this thesis will be discussed in relation to the restoration and conservation of Sphagnum bogs. General introduction 15

17 Chapter 2

18 The transport of solutes by buoyancy-driven water flow in a water-saturated Sphagnum layer; laboratory and field evidence Wouter Patberg Gert Jan Baaijens Christian Fritz Ab Grootjans Rodolpho Iturraspe Alfons Smolders Theo Elzenga

19 Abstract Sphagnum bogs depend for their nutrients mainly on atmospheric deposition. Yet, the main nutrient source for Sphagnum growth has been shown to be the mineralization of organic material. The highest mineralization rates are found in lower peat layers at the border between acrotelm and catotelm. The highest metabolic activity and nutrient uptake, however, takes place in the capitula, found at the top of the living Sphagnum layer. This separation between the actively growing capitula and the site of nutrient mineralization requires an efficient nutrient transport system. Several nutrient transport mechanisms in bogs have been described; diffusion, internal transport and capillary transport. Complementary to these mechanisms, buoyancy-driven water flow was proposed as an external nutrient transport mechanism in a water-saturated peat layer. Buoyancy flow is the vertical movement of water driven by the difference in density between water layers, which is the result of nocturnal cooling. This chapter shows, by means of a mesocosm experiment, that solutes are rapidly and in large quantities transported by buoyancy-driven water flow in a Sphagnum matrix. As a consequence, nutrient concentrations are stepwise increased around the capitula. Consequently, the importance of buoyancy flow in the nutrient supply of Sphagnum is also determined by the nutrient uptake capacity of Sphagnum. The uptake kinetics of ammonium by Sphagnum indicates that Sphagnum is able to benefit efficiently from a stepwise increase in ammonium availability. Therefore, compared to diffusion and internal transport, buoyancy flow seems to be a quantitatively important transport mechanism for nitrogen in a water-saturated bog. Additionally, the transport of N by buoyancy-driven water flow and subsequent uptake by the capitula was shown in a field situation. Other solutes like carbon dioxide and oxygen will be redistributed as well by buoyancy flow and this presumably also plays an important role in ecosystem functioning. 18 Solute transport in Sphagnum dominated bogs

20 Introduction One of the main characteristics of Sphagnum bogs is the accumulation of peat (Clymo et al., 1998). Due to the wet and acidic environment in bogs the production of Sphagnum exceeds the decomposition of organic material, resulting in the accumulation of organic material or peat (Clymo & Hayward, 1982). Sphagnum is functioning as an ecosystem engineer by contributing to the creation of this environment (Van Breemen, 1995). Consequently, peatlands function as a net sink for CO 2 (Clymo et al., 1998; Gorham, 1991). Sphagnum dominated bogs receive nutrients mainly by atmospheric deposition (e.g. Van Breemen, 1995). Sphagnum mosses are able to very efficiently intercept nutrients from precipitation and recycle them (Li & Vitt, 1997; Malmer, 1988; Rudolph et al., 1993; Woodin & Lee, 1987). Consequently, they are able to dominate these low nutrient ecosystems (Aldous, 2002a; Li & Vitt, 1997; Malmer, 1988; Van Breemen, 1995). However, it has been shown that under non-polluted conditions the annual input of nutrients by precipitation is often insufficient to sustain the observed primary production in these systems (Aerts et al., 1999; Aldous, 2002a, b; Bowden, 1987; Bridgham, 2002; Damman, 1978, 1986; Morris, 1991; Pakarinen, 1978; Rosswall & Granhall, 1980; Urban & Eisenreich, 1988). The release of minerals by decomposition of organic material has been shown to be an important nutrient source to support Sphagnum growth (Aldous, 2002a; Bowden, 1987; Bridgham, 2002; Gerdol et al., 2006; Urban & Eisenreich, 1988). The highest mineralization rates are found in the aerobic zone of the acrotelm at the border of the anaerobic catotelm (Bridgham et al., 1998; Malmer, 1993; Waddington et al., 2001). In contrast, the highest metabolic activity and nutrient uptake in Sphagna takes place in the upper part of the plant, the capitulum (Aldous, 2002a; Johansson & Linder, 1980; Malmer, 1988; Malmer et al., 1994; Robroek et al., 2009; Rydin & Jeglum, 2006). Together, these processes result in a gradient in nutrient concentration, from low in the upper layer of a Sphagnum bog to high in the lower parts were decomposition takes place. In addition, the oxygen produced by the photosynthesizing capitula in the top layer and the consumption of oxygen by decomposition processes in deeper acrotelm layer will result in decreasing oxygen levels with increasing depths (Lloyd et al., 1998; Redinger, 1934). The spatial separation between the actively growing, photosynthesizing capitula and the mineralization of nutrients requires an efficient nutrient transport system. Several nutrient transport mechanisms have been described for Sphagnum bogs. Nutrients are passively distributed throughout the water layer by diffusion. Above the water layer solutes might be transported upwards to the capitula through the extracellular capillary spaces between pendant branches and stems (Hayward & Clymo, 1982). Nutrients are also transported internally by Sphagnum. Rydin & Clymo (1989) demonstrated the internal acropetal transport of carbon and phosphorus. In Chapter 4 of this thesis evidence for internal transport of nitrogen is provided. Complementary to these mechanisms, Baaijens (1982) and Rappoldt et al. (2003) reported on a phenomenon called buoyancy-driven water flow, as a possible external nutrient transport mechanism in a watersaturated peat moss layer. Buoyancy flow is the vertical redistribution of water, generated by nocturnal cooling. During the night the upper water layer cools down, leading to relatively dense cold water on top of warmer water. If the temperature drop is sufficiently large, the cold water sinks leading to mixing of the water column. Evidence for the occurrence of buoyancy-driven water flow in a water-saturated The transport of solutes 19

21 Sphagnum layer was provided, based on theoretical and experimental grounds, by Rappoldt et al. (2003). They showed that for a typical peat moss layer a temperature difference of about 10 degrees between day and night will result in a Rayleigh number (Ra) suitable for the development of buoyancy flow. Adema et al. (2006) provided field evidence for buoyancy-driven water flow in a Sphagnum dominated peat bog. Hydraulic conductivity and temperature measurements in a pristine bog in Tierra del Fuego, Argentina, indicated that buoyancy flow events occur regularly. Moreover, daily air temperature data collected around the world indicate that buoyancy-driven water flow is very likely to occur on a regular basis in peat lands throughout the world (Chapter 3). Flow of water will result in the mixing of solutes and this has been proposed to be the most important ecological consequence of buoyancy-driven water flow (Adema et al., 2006; Rappoldt et al., 2003). However, direct evidence for nutrient transport is lacking. This study focuses on buoyancydriven water flow as a nutrient transport mechanism in a water-saturated Sphagnum layer. It is hypothesized that nutrients originating from decomposition will be transported upwards and may become available for the growing capitula of Sphagnum and thereby contributing to the nutrient supply of the Sphagnum plants. A mesocosm experiment was set up to trace the transport of solutes during the occurrence of buoyancy-driven water flow in a Sphagnum matrix. The contribution of buoyancy flow in the nutrient supply of Sphagnum also depends on the nutrient uptake capacity of Sphagnum. Therefore, the uptake kinetics of ammonium by the capitula of S. cuspidatum and S. fallax were determined. Additionally, in a pristine Argentinean bog a 15 N source was placed in the deeper acrotelm and the uptake of labelled nitrogen by the capitula was measured with and without the obstruction of convective flow. The importance of buoyancy-driven water flow for several nutrients and its importance with respect to other nutrient transport mechanisms in a Sphagnum bog will be discussed. Materials and methods Mesocosm experiment A container (h=11.5 cm d=20 cm) was completely filled with demineralized water and Sphagnum magellanicum mosses to create a water-saturated Sphagnum matrix. The container was insulated with a ten centimetre thick layer of foam to prevent radial heat loss and placed on a vibration free foundation in a climate controlled room. Day temperature was set at 20 C (14 hrs) and night temperature at 8 C (10 hrs). Light conditions were 50 µmol m -2 s -1, continuously (Hansatech Quantitherm Light meter). The occurrence of buoyancy flow in the Sphagnum matrix was monitored by continuously measuring the vertical temperature profile using an array of five chromel-alumel thermocouples placed in the centre of the bucket at 5, 20, 40, 70 and 110 mm depth connected to a Graphtec GL200 midi logger (Graphtec Corp., Yokohama, Japan) on which data were logged and stored with one minute intervals. A vertical temperature profile was also monitored continuously in a similar container filled with wet potting soil (providing a matrix in which no convective flow was expected to occur). A blue dye (Coomassie Brilliant Blue G, No. B0770, Sigma Aldrich) was used to mimic the transport of dissolved compounds by buoyancy flow. A volume of 500 ml of water was extracted from the 20 Solute transport in Sphagnum dominated bogs

22 Sphagnum matrix by a siphon followed by the resettlement of the matrix for an hour. Subsequently, at the onset of night (t=0), 200 ml of Brilliant Blue solution (66 mg/l) was layered through a tube (3*5mm*40 cm) on the bottom of the Sphagnum matrix from a 500 ml Erlenmeyer flask with a flow rate of approximately 1.3 ml s -1. The temperature of the Brilliant Blue solution was 4 C to assist the positioning of the layer at the bottom of the container. The homogeneous Sphagnum matrix had a density of approximately 4 g DW L -1, which is comparable to a natural, green and growing peat moss layer (Clymo, 1970). After 15 minutes (t=15) the first water samples were taken from the mesocosm at 5, 55 and 105 mm depth, using an array of three black norprene tubes (l = 400 mm, 4.8 mm outer and 1.6 mm inner diameter; Saint-Gobain Performance Plastics, Verneret, France) in combination with a peristaltic pump (Masterflex L/S model , Cole Parmer Instrument company). Water samples were 1 ml each and collected in disposable polystyrene cuvettes ( mm; Sarstedt, Nümbrecht, Germany). Water samples were taken at 15 minutes intervals during the first 90 minutes of the experiment and at increasingly longer intervals during the remainder of the experiment. The experiment lasted for 25 hours. Care was taken that the dead volume in the tubes was excluded from the samples taken. After 2.5 hours additional samples (1 ml) were taken with a pipette at the periphery of the Sphagnum matrix at 5 mm depth. Immediately after sampling the extinction of the water samples was measured on a double-beam spectrophotometer (Uvikon 940, Kontron Instruments, Germany) at 580nm with demineralized water as a reference. Potential effects of temperature on the extinction values of the water samples were determined by simultaneously sampling a solution with a known concentration of Brilliant Blue. No temperature effects were observed. The abovementioned experiment was repeated eight times. Data were plotted and fitted using graphing software (Prism version 4.03, 2005; GraphPad Software, Inc., San Diego, CA, USA). The Sphagnum magellanicum plants used in the mesocosm experiment were collected in August 2009 in a small bog located in the Dwingelderveld (N52 50, E6 26 ), a nature reserve in the north of the Netherlands. The mosses used in this experiment were solely used to create a Sphagnum matrix with acrotelmic characteristics. To minimize the adsorption and/or uptake of Brilliant Blue during the experiment by the Sphagnum mosses they were incubated for at least 5 days in a Brilliant Blue solution (66 mg L -1 ) prior to the experiment. Before the plants were used to set up the matrix they were rinsed twice with demineralized water. The Sphagnum matrix was set up at least 12 hours before the start of the experiment to allow equilibration, thereby avoiding possible leakage of Brilliant Blue from the Sphagnum plants into the matrix solution during the experiment. Uptake kinetics Experimental design and analysis To determine the uptake kinetics of ammonium by the capitula of Sphagnum cuspidatum and S. fallax, the upper 2 cm of the plants were incubated for two hours in solutions containing 0, 5, 10, 25, 50 and 100 µmol 15 NH 4 Cl L -1 and 20mM MES at ph = 4. Per concentration and species, three capitula were incubated in 85 ml solution in a square Petri-dish (120*120mm; Greiner bio-one GmbH), in triplicate. The capitula were rinsed three times with demineralized water before and after the treatment. The experiment was performed in a climate chamber (light 185 µmol m -2 s -1 ; 18 C). To distinguish between the amount of ammonium taken up internally (assimilated) and bound to the cell wall, an additional experiment was performed. The experimental set up was The transport of solutes 21

23 exactly the same as the experiment described above except, to rinse of the 15 NH 4 + from the cell wall, in this experiment the plants were rinsed for two minutes with 100 ml 1mM KCl mm CaCl 2 solution (at 150 rpm) after incubation. Subsequently, the plants were dried for at least 48 hours at 80 C. From each Petri-dish the capitula were pooled and grinded to a fine powder using a ball miller (Retsch MM2, Haan, Germany). Per sample the %N and % 15 N were measured at the University of California Davis Stable Isotope Facility, Davis, California, USA, using a PDZ Europa ANCA- GSL elemental analyzer interfaced to a PDZ Europa isotope ratio mass spectrometer (Sercon Ltd., Cheshire, UK). The data are expressed as the amount of 15 N taken up (assimilated and bound to the cell wall) by the plants, in µmol g DW -1 calculated by the formula (% 15 N * %N / 100)*10000/15, where %N is the percentage of total N in the sample, % 15 N is the percentage of 15 N of total N and 15 is the molecular weight of the stable N isotope. Plant material Sphagnum cuspidatum Ehrh. Ex Hoffm. plants were collected in a pool in a small bog in the forestry Gasselterveld, the Netherlands (N , E ). Sphagnum fallax (klinggr.) Klinggr. plants were collected in a small bog in the Dwingelderveld (N , E ). Before being used in the experiment, the plants acclimated for two weeks in a greenhouse. In the greenhouse, natural light was supplemented with high pressure sodium lamps to obtain a 14 hour photoperiod. During this period the plants were kept wet with demineralized water. No nutrients were supplied to the plants. Field experiment Rancho Hambre, Tierra del Fuego, Argentina In February 2009 the transport of labeled nitrogen ( 15 N) by buoyancy-driven water flow in a field situation was examined. The location was a pristine Sphagnum magellanicum bog complex, named Rancho Hambre (54 45 S W), located in the Argentinean province of Tierra del Fuego and characterized by the presence of numerous water hollows of different size (Mataloni & Tell, 1996). See Grootjans et al. (2010) for a more detailed site description. The experiment was performed in two contrasting Sphagnum habitats; a pool dominated by the aquatic moss S. fimbriatum, growing with their capitula at the level of the water table, and a lawn completely dominated by S. magellanicum with the capitula growing approximately 20 cm above the water table. See figure 2 for the experimental design. Per site, two treatments were applied and compared to a control situation. In both treatments a 15 N source was introduced by placing an agar cube (50*50*50 mm) containing 10 mg 15 NH 4 Cl g -1 agar at 10 cm depth below the capitula. In the first treatment, convective transport in the Sphagnum layer was excluded by the placement of a two centimeter thick agar disc (d=30 cm) just below the capitula. Before positioning the agar disc, the upper two centimeter of the Sphagnum plants were removed and placed back on top of the disc. In the second treatment convective flow was not blocked. The upper two cm of the Sphagnum mosses were cut off and immediately placed back, to rule out any effect of cutting. Cutting of the top 2 cm also will block possible internal transport of 15 N. Control values were obtained from a third treatment in which the Sphagnum plants were kept intact and no 15 N source was applied. All treatments were performed in triplicate. The distance between the experimental plots was at least one meter. All plots were sampled twice: just before and 10 days after application of the labeled nitrogen (10 February 2009). Per sample ten capitula were collected and transported in polyethylene bags to the 22 Solute transport in Sphagnum dominated bogs

24 laboratory where they were dried for at least 24 hours at 80 C. The samples were ground to a fine powder using a ball miller (Retsch MM2, Haan, Germany). The stable nitrogen isotope composition was measured for each sample with a Carlo Erba NA 1500 elemental analyzer (Thermo Fisher Scientific Inc.) coupled online via a Finnigan Conflo III interface with a Thermo-Finnigan Delta- Plus mass-spectrometer. Values of 15 N are expresses as percentage of total nitrogen concentration. Statistical differences in 15 N concentrations between treatments and species were tested using a two-way ANOVA with species and treatment as fixed factors (SPSS for Windows version , 2007; SPSS Inc., Chicago, IL, USA). The assumption of homogeneity of variance was not met, not even after transformation of the data. According to Heath (1995), the analysis of variance appears not to be greatly affected by heterogeneity in variance if sample sizes are more or less equal. Therefore, we decided to continue our analysis using non-transformed data. Air temperature data of the months January and February were obtained from a climate station located adjacent to the Rancho Hambre bog complex. Results Mesocosm experiment A clear difference between the vertical temperature profiles in the Sphagnum matrix and the container with potting soil was observed (figure 1a and b). Due to absence of convective flow in the potting soil, heat transfer takes place solely by conduction, resulting in a stratified temperature pattern and gradually decreasing temperatures during the nocturnal cooling period. The cooling of the Sphagnum matrix proceeds differently. Most remarkable is the small temperature increase during the nocturnal cooling in the upper Sphagnum layer after approximately 95 minutes. According to Rappoldt et al. (2003) and Adema et al. (2006) a telltale sign of the upwards transport of warmer water by buoyancy flow. Moreover, in the presence of convective flow the layers will mix and the appearance of stratification will be less, which is clearly shown by the smaller temperature differences between the layers in the Sphagnum matrix when compared to the potting soil temperature profile. Buoyancy flow has also been shown to result in a more efficient nocturnal cooling than the diurnal heating, indicated by a decreasing average temperature with increasing depth (cf. Rappoldt et al., 2003; Adema et al., 2006). This effect was also observed in the present study (figure 1d). In the Sphagnum matrix, the average temperature at 110 mm depth is about 0.75 C lower than at the surface of the matrix. In contrast, the average daily temperature in the container filled with potting soil is slightly increasing with depth. The addition of the Brilliant Blue solution at the beginning of the night, has led to a steep Brilliant Blue gradient in the Sphagnum matrix, with extinction values of 0.948, and at 105, 55 and 5 mm depth, respectively (figure 1c). The concentration of Brilliant Blue is reaching equilibrium at the end of the experiment (t=1500). The course to equilibrium, without the occurrence of buoyancy flow, can be fitted by a single exponential function, as plotted in figure 1c using the data from the first 95 minutes (before the occurrence of buoyancy flow) and the equilibrium value at t=1500. The transport of solutes 23

25 However, between 90 and 120 minutes the Brilliant Blue concentration at 105 mm depth suddenly drops from where it slowly increases to the equilibrium after 25 hours. The sudden drop at t=120 follows the small temperature increase in the upper Sphagnum matrix indicative for the occurrence of buoyancy flow. Up to that moment the levels of Brilliant Blue at 5 and 55 mm depth were at a constant low level but reach equilibrium relatively fast after that moment. Striking is the high concentration of Brilliant Blue at the edge of the upper matrix layer following the occurrence of buoyancy flow, reaching a concentration much higher than the fitted line based on exponential decay. If no mass flow mixing occurs and Brilliant Blue is solely redistributed by diffusion, the equilibrium value is the maximum concentration that will be reached in the upper Sphagnum layer and the minimum concentration for the layer at 105 mm. However, in the Sphagnum matrix, after the occurrence of buoyancy flow, the Brilliant Blue gradient is reversed, indicating the transport of Brilliant Blue by mixing of the different layers. Figure 1. Temperature profiles and Brilliant Blue course during the nocturnal period of the mesocosm experiment. (a) and (b) show the course of the vertical temperature profile in the container with wet potting soil (control) and in the Sphagnum matrix, respectively. The numbers in the temperature courses in (a) indicate the depths at which the temperature was measured. In (b) the occurrence of buoyancy flow is clearly visible by the small temperature increase after approximately 95 minutes (indicated by the vertical grey line). (c) shows the course of the Brilliant Blue solution throughout the Sphagnum matrix. The solution was added at the bottom of the Sphagnum matrix at t=0. The amount of Brilliant Blue at depths of 5, 55 and 110 mm is shown by respectively closed, grey and open circles. Additional samples taken with a pipette in the upper layer of the Sphagnum matrix are indicated by an open square. The initial concentration of Brilliant Blue at 110 mm depth is indicated by an open diamond. The 24 Solute transport in Sphagnum dominated bogs

26 black curve indicates the mixing of the Brilliant Blue throughout the Sphagnum matrix following a single exponential decay using the data from the first 95 minutes of the experiment (before buoyancy flow starts) and the equilibrium value reached at t=1500 (r 2 =0.9970). At the moment of buoyancy flow development (vertical grey line), a sudden decrease of Brilliant Blue at 110 mm depth was observed. (d) shows the average 24-hrs temperature with depth relative to the temperature at 5 mm depth for the Sphagnum matrix (closed circles) and the potting soils container (open circles) The points refer to the measurements in (a) and (b). Buoyancy flow results in a more efficient nocturnal cooling relative to the diurnal heating, indicated by a decreasing average temperature with depth. This is not the case in the container filled with potting soil. + Uptake kinetics of NH 4 The uptake of 15 N as function of the concentration 15 NH 4 Cl, for Sphagnum cuspidatum and S. fallax, are shown in figure 2. A distinction is made between the total amount of ammonium taken up (the fraction bound to the cell wall and the fraction taken up internally), the fraction 15 N taken up internally only (which are the 15 N concentrations after rinsing) and the fraction bound to the cell wall only (adsorption) which is derived from the difference between the total amount and adsorption. Hyperbolic curves according to the formula A=V max *[ 15 N]/(K m +[ 15 N]) + c were fitted to the data. Except for the internal assimilated 15 N which shows a linear response to the 15 N concentration. The uptake kinetic parameters for total uptake and adsorption are shown in table 1. Table 1. Kinetic uptake parameters of ammonium (±SD) for Sphagnum cuspidatum and S. fallax S. cuspidatum S. fallax Total uptake Adsorption Total uptake Adsorption V max (µmol 15 N g DW -1 hr -1 ) 31.3 (±4.1) 13.7 (±1.8) 20.3 (±2.4) 8.1 (±0.6) K m (µmol 15 N L -1 ) (±28.8) 60.0 (±18.4) (±22.8) 49.5 (±10.3) c 0.02 (±0.3) (±0.3) -0.1 (±0.2) -0.2 (±0.1) Figure 2. The amount of 15 N (µmol g DW -1 ) taken up by Sphagnum cuspidatum and S. fallax when incubated for two hours at different concentrations of 15 NH 4 Cl at 18 C. A distinction is made between overall uptake of 15 N (assimilated and adsorption solid circles) and assimilated only (open circles) and 15 N bound to the cell wall (adsorption (grey circles). Shown are means and standard deviations of three replicates. Data of overall uptake and adsorption are fitted to the hyperbolic curve A=V max *[ 15 N]/ (K m +[ 15 N]) + c and assimilation only data to the curve y=ax+b. The transport of solutes 25

27 Figure 3. The Rancho Hambre experiment. (a) Temperature difference between day and night from January 1 st to February , measured adjacent to the Rancho Hambre bog complex. The experimental period is indicated by the grey box. A temperature difference of 8 C is indicated by the dotted line. During the experimental period at least four times a difference of 8 C between day and night was measured, which has been shown to be sufficient for the formation of buoyancy-driven water flow in Sphagnum fimbriatum sites. (b) The relative increase in 15 N concentration in capitula of Sphagnum magellanicum (open symbols) and S. fimbriatum (closed symbols) 10 days after placing the 15 N source in the acrotelm. The solid line gives the average value of all non-buoyancy flow treatments (all treatments except the 15 N BF treatment at the S. fimbriatum site) and the dashed lines indicate the 95% interval of these samples. (c) A cross section of the upper part of a Sphagnum bog showing the schematic view of the experimental set-up. The experiment consisted of three treatments; control, 15 N-no BF and 15 N BF. Each treatment was performed in two different habitats; a site dominated by S. magellanicum and a site dominated by S. fimbriatum. The capitula (asterisks) of these mosses growing respectively 10 and 0 cm above the water level as indicated by the grey area. In the control treatment the plants were left untreated. In the other treatments a 15 N source was placed in the acrotelm at 10 cm below the capitula as indicated by the small grey boxes. In the second treatment convective flow was obstructed by placing an agar disc below the capitula (thick black horizontal line in the 15 N no BF treatment). In the third treatment, convective flow was not obstructed ( 15 N BF). 26 Solute transport in Sphagnum dominated bogs

28 15 N experiment Rancho Hambre The temperature difference between day and night adjacent to the Rancho Hambre bog complex in February 2009 are presented in figure 3a. During the experimental period a difference between day and night temperature of at least 8 C occurred several times in the Rancho Hambre bog complex. In Chapter 3 it is demonstrated that buoyancy flow develops in the Sphagnum fimbriatum sites in the Rancho Hambre bog complex at temperature differences of 8 C between day and night. Figure 3b shows the relative increase of 15 N in the capitula of Sphagnum magellanicum and S. fimbriatum capitula in all treatments after 10 days. Individual values per plot are shown. Main effects of species and treatment on 15 N concentration in the capitula were non-significant, (F(1,10)= 0.635, F(2,10) = 1.246, respectively; p>0.05). Also no significant interaction effect of species * treatment was found (F(2,10) = 0.177, p>0.05). However, in two of the non-obstructed Sphagnum fimbriatum plots ( 15 N - BF) a much higher increase of 15 N in the capitula was observed (2.28 and 2.72%,) compared to the average increase (0.97%) in the other treatments in which transport by buoyancy flow was not to be expected due to either too low water tables (all S. magellanicum sites) or obstruction of transport of 15 N by convective flow (S. fimbriatum, no BF). Discussion The mesocosm experiment clearly demonstrates that solutes are transported by buoyancy-driven water flow in a Sphagnum matrix. These findings indicate that buoyancy-driven water flow acts as an external nutrient transport mechanism in water-saturated Sphagnum habitats and thereby contributes to the supply of nutrients to the Sphagnum capitula in the upper bog layer and thereby to the recycling of nutrients. To date, diffusion and internal transport were the known mechanisms by which nutrients were transported throughout a water-saturated Sphagnum layer. Note that capillary transport is often mentioned as an (important) nutrient transport mechanism (Hayward & Clymo, 1982), but this type of transport is only possible above the water table and therefore not taken into account here. The mesocosm experiment presented here clearly shows buoyancy flow to distribute solutes more rapidly and in larger amounts than is possible by diffusion (figure 1) with the reversal of the Brilliant Blue gradient indicates mixing of the layers. It was assumed that diffusion was the only mechanism by which the Brilliant Blue was redistributed throughout the Sphagnum matrix before buoyancy flow started to occur. However, based on the concentration Brilliant Blue at t=0 at a depth of 105 mm, and the diffusion coefficient for Brilliant Blue, the decaying values in the first 95 minutes of the experiment and the equilibrium values at t=1500 are too high to be explained by diffusion alone. Other mechanisms play a role in the redistribution of Brilliant Blue in this phase of the experiment. This could involve small scale convective flow, adsorption by the Sphagnum plants and/or diffusion into the hyaline cells. However, the importance of buoyancy flow as a transport mechanism in supplying the capitula is co-determined by the nutrient uptake capacity of the capitula. Due to buoyancy flow, nitrogen gradients can be reversed in a water-saturated Sphagnum layer and thereby induce a stepwise increase in the nitrogen concentration near the capitula. In case of a full reversal of layers the maximum ammonium concentration in the surroundings of the capitula reflects the ammonium concentration in the deeper acrotelm. Literature values for bog water ammonium concentration are about The transport of solutes 27

29 3 µmol L -1 for a pristine bog in Ireland and 105 µmol L -1 for a Dutch bog suffering high nitrogen loads (Lamers et al., 2000). In the observed uptake kinetics for ammonium by Sphagnum cuspidatum and S. fallax, the V max of ammonium uptake when exposed to ammonium concentrations up to 100 µmol L -1 does not seem to be reached. In a separate experiment in which the time dependence of uptake was determined for 100 µmol NH 4 + L -1 equilibration was reached after about 24 hours for both S. cuspidatum and S. fallax (see Chapter 7, figure 1). These uptake characteristics allow Sphagnum to make full use of the stepwise increase in ammonium supplied by buoyancy flow, considering that in every 24-hour period only one buoyancy flow event can be expected. Interestingly, the observed uptake kinetics of ammonium by Sphagnum cuspidatum and S. fallax indicate the existence of two different processes: a high affinity adsorption and a low affinity + internal uptake mechanism. The high affinity of the adsorption of NH 4 to the cell wall implies that the fixation of ammonium at low concentrations is mainly realized by adsorption (table 1). With increasing concentrations the relative importance of adsorption to total uptake decreases; the cell wall will saturate and increased uptake will take place by intracellular uptake. These observations support the general assumption of the cell wall functioning as a temporal extension of nutrient availability for intracellular uptake (e.g. Buscher et al., 1990; Clymo, 1963; Hajek & Adamec, 2009; + Jauhiainen et al., 1998). Sphagnum fallax shows a higher affinity for NH 4 uptake than S. cuspidatum, a difference that could be reflected by the different habitat of these two species. Sphagnum species growing higher above the water level (hummocks) have higher cation exchange capacities (CEC) than species growing in wetter habitats like lawns and hollows (Jauhiainen et al., 1998; Hajek, 2009; Clymo, 1963). + Earlier studies report various values for NH 4 uptake rates by Sphagnum ranging from about + 20 to 130 µmol NH 4 g DW -1 hr -1 (Jauhiainen et al., 1998; Rudolph et al., 1993; Twenhoven, 1992; Wiedermann et al., 2009). However, these studies are difficult to compare since the amount of + NH 4 applied differed. + As in our study, Jauhiainen et al. (1998) applied 50 µm NH 4 to the capitula of seven Sphagnum + species in order to measure NH 4 uptake rates. For S. cuspidatum and S. fallax they found uptake rates of 67 and 78 µmol g DW -1 hr -1, respectively, which are high compared to the values found in this study; 8.7 and 6.3 µmol g DW -1 hr -1 for S. cuspidatum and S. fallax, respectively. Nitrogen uptake rates are suggested to be negatively correlated with nitrogen deposition rates and internal N concentration (Limpens & Berendse, 2003). If so, the relatively low uptake rates might be explained by the high N deposition at the sites were the plants were collected (28 kg ha -1 yr -1 ; RIVM, 2009), which was reflected by the internal nitrogen concentration of 11.1 ± 0.6 and 10.1 ± 2.0 mg g DW -1 for S. cuspidatum and S. fallax, respectively. Another explanation for these low uptake rakes might be the difference in ph between our experimental solution (4.0) and the one used by Jauhiainen et al. (1998): between 5.0 and 5.5. Since our data show the importance of the cation exchange capacity (CEC) of the cell wall in the ammonium uptake, a low ph, correlating with a low CEC (Richter & Dainty, 1989; Rudolph et al., 1993), might have resulted in lower uptake rates. Internal transport is estimated to distribute solutes throughout the plant with a half time of about 11 days, an estimation based on the symplasmic apical transport of 14 C (in NaHCO 3 ; Rydin & Clymo, 1989). Internal, apical transport of N showed even lower rates, a half time of 17 days (see Chapter 4). According to a review by Raven (2003) on long distance transport in non-vascular 28 Solute transport in Sphagnum dominated bogs

30 plants, there seems to be no evidence for symplasmic transport in Sphagna at speeds faster than can be accounted for by diffusion. Thus, in comparison with diffusion and internal transport, buoyancy flow seems to be a quantitative important nutrient transport mechanism in a watersaturated Sphagnum bog. Nitrogen Since Sphagnum growth in peatlands is often limited by nitrogen availability (Aerts et al., 1992; Bridgham et al., 1996; Gunnarsson & Rydin, 2000) the supply of nitrogen to the capitula is of special interest. The main N source for Sphagnum has been shown to be re-mineralized N (Aerts et al., 1999; Aldous, 2002b; Bridgham, 2002; Gerdol et al., 2006; Morris, 1991; Urban & Eisenreich, 1988). The importance of re-mineralization of nitrogen for Sphagnum growth has been demonstrated in situ by Urban & Eisenreich (1988). They calculated the assimilation of nitrogen by plants (primarily Sphagnum) to be 66 kg ha -1 yr -1, whereas only 14.6 kg N ha -1 yr -1 was supplied by total inputs. The remainder was supplied by mineralization of the peat. These findings are contradictory to the idea that Sphagnum and vascular plants utilize spatially distinct nutrient pools, with Sphagnum relying on N from precipitation and vascular plants on mineralization of senescing organic matter in the deeper acrotelm (Malmer et al., 1994; Pastor et al., 2002). Partly due to buoyancy flow, Sphagnum can also rely on mineralized nitrogen as a major N source. Consequently, Sphagnum mosses may outcompete vascular plants more easily and thereby enhance their ability to engineer the ecosystem (Van Breemen, 1995). Gerdol et al. (2006) stated that the cycling through mineralization of senescing tissues by heterotrophic bacteria is essential for the supply of N to the growing tissues. An interesting observation was the positive effect of high water tables on N cycling, a result that in the original paper was left unexplained (Gerdol et al., 2006). Significantly lower upward translocation of N to the growing capitula at low water tables was also observed by Aldous (2002b). A likely explanation for this phenomenon is that high water tables allow the occurrence of buoyancy flow, increasing recycling rates of nitrogen. The experiment performed in the Rancho Hambre bog demonstrates the transport of N by buoyancy-driven water flow in the Sphagnum fimbriatum sites and the subsequent uptake by the capitula in the upper Sphagnum layer. The increase in 15 N concentration in the treatment with unobstructed convective flow indicates the transport of 15 N by buoyancy flow. In two of the three replicates this increase was observed. Possibly in one case buoyancy flow did not occur, or did not result in the upward transport of 15 N. Multiple factors determine the variety in the occurrence of buoyancy flow. Although during the experiment the temperature difference between day and night were sufficient to induce buoyancy flow, this is not a guarantee for buoyancy flow cells to occur. Other factors co-determine the development of these cells. For example, a heterogeneity in the vertical hydraulic conductivity might locally hamper the development of buoyancy flow. Additionally, buoyancy flow occurs as adjacent cells with warm water going up and cold water going down (Adema et al., 2006). Sphagnum capitula residing in a downward flow do not receive nutrients from below despite the occurrence of buoyancy flow. Ecological implications CO 2 gradients. Lloyd et al. (1998) showed an increase with depth of the CO 2 concentration in a The transport of solutes 29

31 water-saturated Sphagnum core due to the uptake of CO 2 by photosynthetic activity of the capitula in the upper layer and the release of CO 2 by decomposition of organic material in the lower acrotelm. Since submerged Sphagnum species that inhabit peat hollows have been shown to be limited by CO 2 (Rice & Giles, 1996; Rice & Schuepp, 1995), buoyancy flow might be an important mechanism replenishing CO 2 around the capitula and enhancing photosynthesis by vertical transport of CO 2 from the deeper catotelm layer. Oxygen gradients. The photosynthetic activity results in high oxygen levels in the top Sphagnum layer and decreasing levels with depth (Adema et al., 2006; Lloyd et al., 1998). Lloyd et al. (1998) measured a steep oxygen gradient in the upper four centimeters of a water-saturated Sphagnum layer, decreasing from 300 to 0 µm. Mixing of water layers by buoyancy flow will result in a downward transport of oxygen. Adema et al. (2006) attributed a conspicuous change in oxygen concentration at 5 cm depth in a Sphagnum layer to the occurrence of buoyancy flow. Since the aerobic decomposition of organic material is significantly higher than the anaerobic decomposition (Bridgham et al., 1998; Waddington et al., 2001) the transport of oxygen to the lower parts of the acrotelm will increase decomposition rates thereby increasing the concentrations of CO 2 and nutrients like N. In turn the nutrients will become available to the growing Sphagnum when transported upwards by buoyancy flow. Moreover, the downward transport of O 2 will increase the CO 2 :O 2 ratio in the upper Sphagnum layer and will enhance photosynthetic performance even more since photosynthesis is inhibited by oxygen (see Chapter 6; Bowes & Salvucci, 1989; Raven, 2011; Raven et al., 2008; Skre & Oechel, 1981). Methane. Methane is anaerobically produced in large quantities in bogs (e.g. Gorham, 1991). Nevertheless, emissions of methane to the atmosphere are very low (Larmola et al., 2010) due to the oxidation of methane by methanotropic bacteria (Kip et al., 2010; Raghoebarsing et al., 2005). The mixing of methane and photosynthetically produced oxygen by buoyancy flow might results in lower methane emission rates, and affect global carbon cycling. In addition, the released CO 2 in this oxidation process has been shown to be a significant carbon source for Sphagnum (Raghoebarsing et al., 2005). Overall conclusion Net transport by buoyancy flow occurs when a vertical gradient exists. These gradients have been shown explicitly for CO 2, CH 4 and O 2 in a water-saturated Sphagnum layer (Lloyd et al., 1998) with large concentration differences in the upper 12 cm of the Sphagnum layer for CO 2 and CH 4, and O 2 decreasing to undetectable values at 2 cm depth. According to the models described by Rappoldt et al. (2003), the size of the buoyancy cells (dependent on the Rayleigh number of the Sphagnum layer) can be as large as 24 cm. Based on in situ temperature measurements in a pristine Sphagnum bog in Tierra del Fuego, Argentina, cell sizes of 3 to 26 cm were predicted. Therefore, due to buoyancy flow these solutes can be transported and gradients can even be reversed (figure 1). The maximal transport of nutrients per buoyancy event depends on the concentration in the lower layer. Rappoldt et al. (2003) showed in a model that the Sphagnum matrix is well mixed after 4 consecutive buoyancy events. The mesocosm experiment shows that the Sphagnum matrix was already mixed after one occurrence of buoyancy flow. 30 Solute transport in Sphagnum dominated bogs

32 The recycling of nutrients is of great importance in the functioning of Sphagnum bogs and has been suggested to explain the high carbon burial rates despite the low primary productivity, also known as the Paradox of Peatlands (Raghoebarsing et al., 2005). This chapter shows that buoyancy flow might be essential for the efficient recycling of nutrient by Sphagnum at least under waterlogged conditions. The transport of solutes 31

33 Chapter 3

34 Field characteristics of buoyancy-driven water flow and its global occurrence Wouter Patberg Erwin Adema Myra Boers Gert Jan Baaijens Christian Fritz Ab Grootjans Rodolfo Iturraspe Alfons Smolders Theo Elzenga

35 Abstract Nutrients enter Sphagnum bogs mainly by atmospheric deposition. However, to sustain their annual production, this must be supplemented with nutrients released by decomposition processes in the peat layer. The capitula are the actively growing parts of the Sphagnum plants and since they are spatially distinct from the decomposition processes, transport of the nutrients is essential. For Sphagnum dominated bogs, several nutrient transport mechanisms have been described; internal transport, external transport by diffusion, capillary action and buoyancy-driven water flow. In Chapter 2 it was shown that nutrients are transported more rapidly and in larger quantities by buoyancy flow than would be possible by diffusion or internal transport. Therefore, buoyancy-driven water flow is considered to be an important nutrient transport mechanism in water-saturated Sphagnum layers. However, this conclusion is based on laboratory experiments, which were conducted under controlled conditions, whereas in the field temporal and spatial variation in temperature and hydraulic conductivity might occur and affect the occurrence and size of the buoyancy cells. This chapter focuses on the occurrence of buoyancy flow in the field. Vertical temperature profiles and vertical hydraulic conductivities were measured in a pristine Sphagnum bog to validate the theoretical predictions in a natural situation. Subsequently, Ra numbers of the Sphagnum sites were calculated and the possible occurrence of buoyancy flow in the field was predicted using the model described by Rappoldt et al. (2003). The calculated Ra numbers indicate the Sphagnum fimbriatum layer to be suitable for the development of buoyancy flow, which is supported by the course of the vertical temperature profiles. Moreover, the predicted starting time of buoyancy flow development very well correlates to the observed starting time of buoyancy flow. Additionally, worldwide daily temperature data were analyzed to model the possible occurrence of buoyancy flow in peatlands throughout the world. The results from this GIS analysis indicate that many peatlands are subjected to temperature differences between day and night of 8 C or more, which is sufficient for the development of buoyancy flow. In the month July about 70% of the peatlands have at least 5 days with temperature differences between day and night suitable for the development of a buoyancy flow event. From this analysis it is apparent that buoyancy flow is a worldwide occurring phenomenon in peatlands. 34 Solute transport in Sphagnum dominated bogs

36 Introduction Sphagnum bogs receive nutrients mainly by precipitation (Van Breemen, 1995). However, to support their annual production, Sphagnum mosses are supported in their nutrient supply by nutrients released by decomposition processes in the peat (Aldous, 2002a; Bowden, 1987; Bridgham, 2002; Gerdol et al., 2006; Urban & Eisenreich, 1988). The capitula are actively growing parts of the Sphagnum plants (Johansson & Linder, 1980; Malmer, 1988; Malmer et al., 1994; Robroek et al., 2009; Rydin & Jeglum, 2006) and since they are spatially distinct from the decomposition processes, transport of the nutrients is essential. For Sphagnum dominated bogs several nutrient transport mechanisms have been described. Internal transport (Rydin & Clymo, 1989; Chapter 4), external transport by diffusion, by capillary action (Hayward & Clymo, 1982) or by buoyancy-driven water flow (Rappoldt et al., 2003; Adema et al., 2006; Chapter 2). Buoyancy-driven water flow is the vertical convective flow of water in a water-saturated Sphagnum layer, driven by the temperature difference between day and night (Baaijens, 1982; Rappoldt et al., 2003). During the night the upper water layer cools down more rapidly than the layers below, leading to relatively dense cold water on top of warmer water. Due to density differences between these two layers the colder and denser water will sink and the warmer water will rise. Consequently, buoyancy flow occurs as cells consisting of adjacent regions with upward and downward flow (Adema et al., 2006; Rappoldt et al., 2003). Evidence for the occurrence of buoyancy-driven water flow in a water-saturated Sphagnum layer was provided on theoretical and experimental grounds by Rappoldt et al. (2003). Whether an inversion in the vertical temperature profile leads to instability and the generation of buoyancy flow depends on the system s Rayleigh number, a dimensionless number that depends on several parameters, including vertical hydraulic conductivity and the temperature difference between day and night. Rappoldt et al. (2003) showed that buoyancy flow occurs if the system s Rayleigh (Ra) number exceeds a value of 25. For a typical peat moss layer, a temperature difference of 10 degrees between day and night will result in a Ra number of 80 which is suitable for the quick development of buoyancy flow (Rappoldt et al., 2003). Adema et al. (2006) provided evidence for the occurrence of buoyancy flow in a small Sphagnum dominated peat bog in the Netherlands. Based on the hydraulic conductivity (k) of the Sphagnum layer and a temperature difference between day and night of 8 C,, the calculated Ra number was sufficiently high to induce buoyancy flow. In Chapter 2, the convective transport of solutes by buoyancy-driven water flow is demonstrated in both a mesocosm experiment and a field situation. In the mesocosm experiment it was shown that buoyancy flow transports nutrients more rapidly and in larger quantities than would be possible by diffusion (Chapter 2) or by internal transport (Chapter 4). Therefore, buoyancy-driven water flow is considered to be an important nutrient transport mechanism in water-saturated Sphagnum layers. Whereas the laboratory experiments are all conducted under controlled conditions and with homogeneous samples, in the field temporal and spatial variation in temperature and hydraulic conductivity might occur and affect the occurrence and size of the buoyancy cells. Therefore, a series of vertical temperature profiles were measured in a pristine Sphagnum bog to validate the theoretical predictions in a natural situation. Based on the measured hydraulic conductivities and ambient day and night air temperatures, the Ra numbers of the Sphagnum sites were calculated. Field characteristics of buoyancy-driven water flow 35

37 Based on these Ra numbers, the possible occurrence of buoyancy flow in the field could be predicted using the model described by Rappoldt et al. (2003). Additionally, Geographical Information System software was used to analyze worldwide daily temperature data and to model the possible occurrence of buoyancy flow in peatlands throughout the world. Our findings will be discussed with respect to the supply and cycling of nutrients in Sphagnum dominated bogs. Materials and methods Rancho Hambre, Tierra del Fuego, Argentina Field measurements were performed in February 2005 in Rancho Hambre (54 45 S W), a Sphagnum magellanicum bog complex in the Argentinean province of Tierra del Fuego, extensively covered by pools of different size (Mataloni & Tell, 1996). A more detailed site description is given in Grootjans et al. (2010). In a Sphagnum fimbriatum pool site and a S. magellanicum lawn site, respectively three and ten intact peat monoliths were collected in 25 cm long PVC tubes with an internal diameter of 108 mm. The tubes were sealed by two PVC caps, stored at 4 C and analyzed the next day. For different lengths of the cores the vertical hydraulic conductivity (k) was measured (in situ) using the constant head method as described by Adema et al. (2006). Different lengths of the core were obtained by removing parts of the monolith from the bottom. This process was repeated until the hydraulic conductivity became too large to measure. For each length the hydraulic conductivity was measured in triplicate. From these measurements the hydraulic conductivity of each segment was calculated. In each core the depth of the transition from acrotelm to catotelm was determined. For a period of thirteen days temperature profiles were measured in a Sphagnum fimbriatum pool using eight copper-constantan-copper thermocouples connected to a Campbell Scientific AM25T solid state multiplexer with an internal reference RTD (Resistance Temperature Detector). The thermocouples were placed on 1, 6, 14, 25, 39, 56, 76, and 99 mm depth. The data were collected using a Campbell Scientific CR10x multi-channel data logger which measured every second of which the average was stored every minute. By using the temperature and the vertical hydraulic conductivity data, Ra numbers for the upper 12 cm of the cores were calculated for each day during this 13 day period (see Rappoldt et al., 2003). The following formula was used Ra = k α ΔT (t 0 /D eff ). In which Ra is the dimensionless Rayleigh number, k is the vertical hydraulic conductivity (m s -1 ); α is the thermal expansion coefficient (K -1 ); ΔT is the difference between maximum day and minimum temperature of the subsequent night ( C); t 0 is the duration of the low temperature phase in seconds (the value for t 0 was 43200s (12h; Rappoldt et al., 2003) and D eff is the thermal diffusivity (m 2 s -1 ). Both the thermal expansion coefficient (α) and the thermal diffusivity (D eff ) of water are temperature dependent. For each day the values of α and D eff were determined using the minimum, maximum and average temperature, resulting in three calculated Ra values per day. The time of 36 Solute transport in Sphagnum dominated bogs

38 onset of buoyancy flow development was derived from the telltale deviation from the monotonic decrease in temperature in the vertical temperature profile and was compared with the predicted time, based on the Ra numbers calculated according to Rappoldt et al. (2003). The global occurrence of buoyancy-driven water flow To quantify the potential global occurrence of buoyancy flow in peat bogs, air temperature data of peatlands around the world were analyzed by using a Geographical Information System (ArcGis software, ESRI, Redlands, CA, USA). The following digital maps were used: a map with the distribution of global peatland distribution (Yu et al., 2010). Daily day and night temperature data were obtained from NASA ( dataset MOD11C1 Version 005, format: HDF-EOS; Year: 2009; layers used: Daytime LST & Nighttime LST). Average monthly temperature data were obtained from the same website (dataset: MOD11C3 Version 005, format: HDF-EOS, year: 2009, layers used: Daytime LST). Nitrogen-deposition data were derived from the Oak Ridge National Laboratory Distributed Active Center (ORNL DAAC) site ( For all peatlands we calculated the daily temperature difference between the maximum day temperature and the minimum temperature of the subsequent night. The minimum temperature difference for buoyancy flow to occur is 8 C or more (Adema et al., 2006; Rappoldt et al., 2003). For each month we calculated the area of peatlands were 2, 5, 10 or 20 buoyancy flow events occurred. Only peatlands with possible Sphagnum growth were included in the analysis. We assumed that when the average monthly daytime temperature was lower than 8 o C, ice formation would prevent the development of buoyancy flow and/or plant growth would be restricted. Additionally, global nitrogen deposition data were combined with the distribution of buoyancy flow events. Based on the response of Sphagnum to increased nitrogen deposition as proposed by Lamers et al. (2000), three categories were distinguished: <12, and > 18 kg N ha -1 yr -1 (Lamers et al., 2000). Results Local characteristics of buoyancy flow, Rancho Hambre, Tierra del Fuego The hydraulic conductivity (k) values for cores of S. magellanicum and S. fimbriatum at different depths are shown in the figures 1a and b, respectively. The hydraulic conductivity in the upper layer of the S. fimbriatum cores is the highest (ranging from 0.3 to 0.5 m s -1 ) with k values decreasing with increasing core depth. In Sphagnum fimbriatum cores the depths at which the transition between acrotelm and acrotelm was found varied between 10 and 25 cm. In S. magellanicum the hydraulic conductivity is close to zero and does hardly vary with depth. In two of the S. magellanicum cores (4 and 9) the k values are higher in the upper layer (figure 1b). For all S. magellanicum cores the transition from acro- to catotelm was found at 25 cm depth. Temperature measurements Rancho Hambre The temperature measurements in a Sphagnum fimbriatum pool in the Rancho Hambre bog complex show a daily temperature cycle for several depths (figure 2a). The difference between day and night temperature decreases with depth and results in a reversal of the temperature gradient during the Field characteristics of buoyancy-driven water flow 37

39 night. The small temperature increases in the upper Sphagnum layer during nocturnal cooling (arrow in figure 2c) are clear indications that buoyancy-driven water flow has occurred (figure 2a). The small temperature increases in the upper water layer are the result of upward movement of warmer water and the sinking of cooler water (Adema et al., 2006; Rappoldt et al., 2003; Chapter 2). Figure 1. Vertical hydraulic conductivity (in m s-1) at different depths of S. fimbriatum (a) and S. magellanicum (b) cores. Symbols represent the average hydraulic conductivity (± SD) and are placed in the centre of the vertical line which represents the length and depth of the associating core segment. Figure 2b shows the calculated daily Ra numbers, ranging from 25 to 153, for each day for S. fimbriatum core 2. The Rayleigh numbers in figure 2b describe a state of the Sphagnum matrix calculated for a 24 hour period. However, it is very likely that the state of the Sphagnum matrix during a particular 24 h period is also influenced by the conditions during previous day or days. Therefore, the average Ra number for the complete period of 13 days was calculated as well. The Ra number for the Sphagnum matrix for the complete experimental period was calculated to be 81. Based on the calculated Rayleigh numbers, the earliest possible times of onset of buoyancy flow were determined according to the theoretical model described by Rappoldt et al. (2003). This model describes the relation between the Rayleigh number and the time of onset based on a 12 day/12 night temperature regime. Day/night temperature changes are either assumed to follow a block wave or a sine wave, resulting in slightly different times of onset between these approaches. 38 Solute transport in Sphagnum dominated bogs

40 Based on the sinus wave model, the times of onset were determined for the development of buoyancy flow in the S. fimbriatum matrix during the experimental period. These are indicated by the vertical lines in figure 2a. In general, at all days, the calculated time of onset is preceding the temperature peaks which are indicative for the occurrence of buoyancy flow. Figure 2c is a detailed view of the temperature profile at day 2 showing the typical temperature increase during nocturnal cooling due to buoyancy flow (arrow) and the calculated time of onset (vertical black line). Figure 2. a) Vertical temperature profile in a Sphagnum fimbriatum pool in the Rancho Hambre bog complex, Tierra del Fuego, Argentina. The measuring period lasted for 13 full days. Temperature was measured at 1 (red), 6 (light blue), 14 (orange), 25 (light green), 39 (dark blue), 56 (dark green), 76 (brown) and 99 mm (black) depth. The dashed vertical lines indicate the times of onset for the development of buoyancy flow determined by the theoretical model of Rappoldt et al. (2003). B) the calculated Ra numbers for the S. fimbriatum layer during the measuring period. For each day three Ra numbers were calculated using the thermal expansion coefficient (α) and the thermal diffusivity (D eff ) according to the minimum, maximum and average temperature of that day. C) Vertical temperature profile at day 2. The vertical line indicates the calculated time of onset which is preceding the temperature peak, pointed out by the arrow, which is indicative for the occurrence of buoyancy flow. Colors of the lines are as in figure a. Field characteristics of buoyancy-driven water flow 39

41 The global occurrence of buoyancy flow Figure 3 shows per month the relative amount of peatlands with a) the occurrence of 2 and 5 buoyancy flow events, b) the occurrence of < 2 events, c) without sufficient temperature data and d) without Sphagnum growth. In the months June to September in about 80% of the peatlands a buoyancy flow event occurs at least twice. In 40 to 70% of the peatlands it occurs at least 5 times and in 10 to 20% of the peatlands at least 10 times. Peatlands with occurrences of 20 times or more are scarce (in general 1%; 4% in April), however to establish that for a particular pixel 20 buoyancy flow events could have taken place, the number of days with missing data must be very low. Therefore it is possible that the number pixels with 20 buoyancy flow event is actually higher. Figure 3. The relative amount of peatlands per month where buoyancy flow events occurred <2, 2, 5 or where it was assumed that no Sphagnum growth could occur. Dark grey indicates the percentage of pixels with peatlands for which insufficient data were available to calculate the number of occurrences of a buoyancy flow event for that particular month. In figure 4 the distribution of peatlands throughout the world are shown. Indicated in green are the peatlands with the occurrence of at least 5 buoyancy flow events in July 2009, in red the peatlands with less then 5 buoyancy flow events, in orange peatlands without sufficient data for the analysis and in grey peatlands with an average growth temperature <8 C. In 70% of the peatlands a temperature difference of at least 8 C occurs at least 5 times (see also figure 3). In 28% of the peatlands insufficient temperature data are available. In only 0.5% of the peatlands buoyancy flow occurs less then 5 times. In southern Argentina and Chile the average monthly temperature is lower than 8 C and therefore not taken into account in the analysis. To obtain an indication whether Sphagnum growth indeed depends on the occurrence of buoyancy flow for transport and cannot sustain growth on atmospheric N-deposition, we also plotted the atmospheric nitrogen deposition (in kg ha -1 yr -1 ) in the year Based on the growth response of Sphagnum to increasing N input (Lamers et al., 2000) a distinction is made between areas with annual N loads of <12, and >18 kg ha -1 yr -1. The areas with the highest nitrogen load are North-Western Europe, East-Asia and small parts of North and South America. Remarkably, most peatlands do not coincide with these high N load areas and are thus likely to depend on buoyancy flow transported nitrogen. 40 Solute transport in Sphagnum dominated bogs

42 Figure 4. The occurrence of buoyancy flow in peatlands throughout the world for the month July Peatlands with the occurrence of at least 5 buoyancy flow events are green, peatlands with less than 5 buoyancy flow events are red, in grey peatlands the temperature was too low to allow growth and peatlands without sufficient temperature data for the analysis are in orange. The squares indicate an atmospheric nitrogen deposition 12 (green) or 18 kg ha-1 yr-1 (grey). Discussion Local characteristics of buoyancy-driven water flow Based on the vertical hydraulic conductivity of the Sphagnum fimbriatum cores and the temperature differences between day and night, the calculated daily Ra numbers varied from 25 to 153 and according to the model of Rappoldt et al. (2003) the S. fimbriatum layer is thus suitable for the development of buoyancy-driven water flow. Indeed, the typical small temperature increases during the nocturnal cooling in figure 2 very clearly indicates the regular occurrence of buoyancydriven water flow in the Sphagnum fimbriatum pool in Rancho Hambre. The calculated starting time of buoyancy flow development based on the Ra number, correlates, with exception of days 11 and 13 both with a Ra of 25, very well with time of the small temperature increases indicative of buoyancy flow. It should be noted that the calculated Rayleigh number describes a state of the Sphagnum matrix for a single day/night period, ignoring the temperature history during the previous days of the site. This could have been of influence at days 11 and 13, on which the buoyancy flow event occurs much faster than the modelled time of onset. These two days are characterized by a relatively low day temperature. As a consequence, during the cooling period at night the temperature of the surface layer drops fast below the still relatively high temperatures of the lower layers. Possibly, this explains the discrepancy with the model, which, based solely on the relatively small diurnal temperature difference and ignoring the previous days, predicts a very late time of onset of the buoyancy flow event. In agreement with the model is the observation that at day 8, which has a calculated Ra number of 153, no buoyancy flow event is apparent. At very high Ra numbers the model of Rappoldt et al. (2003) predicts that the size of the cells become very small and as a consequence are not detectable Field characteristics of buoyancy-driven water flow 41

43 anymore with the used sensor array. The vertical hydraulic conductivity values in the upper 12 cm of the S. magellanicum cores were in general (with the exception of two cores) too low to result in Ra numbers suitable for the development of buoyancy flow (data not shown). Furthermore, in the observed S. magellanicum lawns the water table is located about 20 cm below the top of the Sphagnum plants which form an insulating layer, preventing the development of a cool water surface layer and instability in the water column (Van der Molen & Wijmstra, 1994). If buoyancy flow would nevertheless occur, solutes transported from deeper layers to the upper water layer would still have to be transported to the capitula by capillary transport. In this case, buoyancy flow only acts as an auxiliary transport mechanism and its relative importance in the nutrient supply to the capitula is determined by the height of the capitula above the water level. Such a situation can be found in Sphagnum lawns and the transition zones between pools and hummocks. Global occurrence of buoyancy-driven water flow The results from the GIS analysis indicates that many peatlands are subjected to temperature differences between day and night of 8 C or more, several days each month during the growing season. At these sites, buoyancy flow events could occur, resulting in sufficient mixing of the water column and efficient recycling of nutrients. In the month July about 70% of the peatlands have at least 5 days and about 90% at least 2 days, with temperature differences between day and night suitable for the development of a buoyancy flow event. From this analysis it is apparent that buoyancy flow is occurring in peatlands on a global scale. For weather stations at latitudes above 50, the fraction of days with Ra numbers >100 during the months June and July was calculated by Rappoldt et al. (2003). From this analysis they concluded that buoyancy flow occurs on a large scale, but is more likely in continental areas than in coastal areas. This distinction between coastal and continental sites was not found in the present study. This is possibly due to the fact that in the current study a temperature difference of 8 o C was used as the sole criterion, while Rappoldt et al. (2003) calculated the Ra, taking into account the effect of temperature on thermal expansion coefficient and thermal diffusivity. This approach leads to different results at very low temperatures and does not affect our conclusions, since we have included only pixels with and average monthly temperature higher than 8 o C. Furthermore our study focused on Sphagnum dominated peatlands, while Rappoldt et al. (2003) included all weather station data available. The mesocosm experiment in Chapter 2 has shown buoyancy flow to be an efficient and rapid nutrient transport mechanism when compared to diffusion and internal transport. Uptake characteristics for ammonium of Sphagnum (Chapter 2) imply that Sphagnum can profit from stepwise increases in nitrogen availability. Therefore, even very infrequent occurrences of buoyancy flow can make a significant contribution to the nutrient availability in a water-saturated Sphagnum layer. The relative importance of transport of nutrients from deeper water layers to the top layer with the capitulum, is higher when atmospheric input is low. This is especially the case for nitrogen availability which is often considered growth limiting in Sphagnum bogs (Aerts et al., 1992; Bridgham et al., 1996; Gunnarsson & Rydin, 2000). Since areas with a high nitrogen deposition load only slightly overlap with the area covered with peatlands (figure 4), the importance 42 Solute transport in Sphagnum dominated bogs

44 of buoyancy flow in the transport and recycling of nitrogen in Sphagnum bogs, is hardly diminished by increased, anthropogenic N deposition. Therefore, we conclude that buoyancy flow is a worldwide occurring phenomenon which significantly contributes to the nutrient supply and nutrient recycling in Sphagnum bogs. The importance of buoyancy flow Buoyancy-driven water flow has been shown to act as an external nutrient transport mechanism in water-saturated Sphagnum habitats (Chapter 2), thereby contributing to the supply of nutrients to the Sphagnum capitula in the upper bog layer and thus to the efficient recycling of nutrients. Moreover, in comparison with diffusion and internal transport, buoyancy flow seems to be a quantitatively important nutrient transport mechanism (Chapter 2 and 4). The present study shows that buoyancy-driven water flow is a worldwide occurring phenomenon in peatlands. A Sphagnum bog can consist of hollows, lawns and hummocks. As buoyancy flow is restricted to the water layer in a Sphagnum bog, direct supply of nutrients from deeper layers to the capitulum by buoyancy flow only takes place in hollows. In lawns buoyancy flow can assist capillary driven nutrient transport and in hummocks buoyancy flow probably is relatively unimportant or does not occur. As the initial successional stage of a Sphagnum bog is the colonization of aquatic Sphagnum species of water bodies followed by the invasion of hummock forming species, buoyancy flow seems to be particularly important in the early stages of bog development. Efficient nutrient use might be of great importance in creating dominance over vascular plants by keeping the nutrient concentrations low in the lower acrotelm and thereby facilitating conditions beneficial to Sphagnum growth. With the regular occurrence of buoyancy flow, the nitrogen concentration in the acrotelmic water will be determined by the decomposition rate in catotelm, the depth of the buoyancy flow cells, the depletion in the top water layer by uptake and assimilation and dilution by rain water. One can easily imagine how such a complicated process, that prevents the build up of high nutrient levels in the deeper layers and thereby the establishment of vascular plants, can be disrupted by high nitrogen loads. Field characteristics of buoyancy-driven water flow 43

45 Chapter 4

46 Physiological evidence for internal acropetal transport of nitrogen in Sphagnum cuspidatum and S. fallax Wouter Patberg Bikila Warkineh Dullo Alfons Smolders Ab Grootjans Theo Elzenga

47 Abstract Several mechanisms for acropetal or upward transport of nutrients have been described for Sphagnum layers; diffusion, buoyancy-driven water flow, capillary transport and internal transport. This chapter focuses on the contribution of internal transport in the translocation of nitrogen from the catotelm, where nutrients are released by decomposition of organic material, to the acrotelm where growth of the Sphagnum plants takes place. The internal transport of nitrogen has often been assumed to be present, but has never been explicitly demonstrated before. The ability of Sphagnum to transport nitrogen internally was investigated by monitoring the transport of labeled nitrogen in two Sphagnum species, S. cuspidatum and S. fallax, in experiments in which diffusion and external transport were prevented. For both species a slow, but significant acropetal transport of nitrogen through an internal mechanism was observed. Within the time frame of the experiments, the internal transport of nitrogen was + only observed when nitrogen was supplied as NH 4 and not as NO 3-. The speed at which the internal transport took place was estimated at 5 mm day -1 and the equilibration between donor and acceptor parts of the plants was characterized by half time value of 17 days. Relative to other transport mechanisms, like buoyancy flow and capillary transport, internal transport represents only a minor fraction of the upward transport of externally supplied nitrogen to the capitula. The main importance of internal transport is therefore considered to be the reallocation of internally catabolized nitrogen compounds. 46 Solute transport in Sphagnum dominated bogs

48 Introduction Sphagnum mosses are the most dominant plant species in bogs. For their nutrient supply they depend mainly on atmospheric deposition. Bogs are, therefore ombrotrophic ecosystems and under natural conditions they are usually nitrogen-deficient (Bragazza et al., 2004; Bridgham et al., 2001b; Gunnarsson & Rydin, 2000; Li & Vitt, 1997). Sphagnum mosses are able to survive due to their very efficient nitrogen utilization (Bridgham, 2002; Li & Vitt, 1997). Sphagnum utilizes nitrogen from atmospheric deposition with an efficiency that ranges from 50 to 90% (Aldous, 2002a; Li & Vitt, 1997). Woodin & Lee (1987) even measured a retention of 100% of inorganic nitrogen at an unpolluted site, whereas chloride and sulphate were passing freely through the moss mat. The efficiency of nitrogen retention by Sphagnum clearly results in a competitive advantage of Sphagnum over vascular plants. Aldous (2002a), for instance, showed that vascular plants could take up less than 1% of N recently added by wet deposition. Nonetheless, under non-polluted conditions, the annual input of nitrogen by atmospheric deposition is not sufficient to support the observed primary production in Sphagnum dominated ecosystems (Aerts et al., 1999; Aldous, 2002a, b; Bowden, 1987; Bridgham, 2002; Damman, 1978, 1986; Morris, 1991; Pakarinen, 1978; Rosswall & Granhall, 1980; Urban & Eisenreich, 1988). Mass balance calculations have shown that direct retention of N from precipitation is less important than recycling of mineralized N to support Sphagnum growth (Aldous, 2002b; Bowden, 1987; Bridgham, 2002; Urban & Eisenreich, 1988). The highest mineralization rates have been found in the aerobic zone of the acrotelm at the border of the anaerobic catotelm (Bridgham et al., 1998; Malmer, 1993; Waddington et al., 2001). On the other hand, the highest metabolic activity and nutrient uptake in Sphagnum mosses was measured in the upper part of the plant, the capitulum (Aldous, 2002a; Johansson & Linder, 1980; Malmer, 1988; Malmer et al., 1994; Robroek et al., 2009; Rydin & Jeglum, 2006). Since the capitula are spatially separated from of the layer where mineralization occurs, transport of nitrogen to the capitula is necessary for the required recycling of nitrogen in a Sphagnum bog. Several nutrient transport mechanisms have been described for Sphagnum bogs; diffusion, buoyancy-driven water flow (Adema et al., 2006; Rappoldt et al., 2003; Chapter 3), capillary transport (Clymo & Hayward, 1982) and internal transport (Rydin & Clymo, 1989). Our study focuses on the contribution of internal transport in the upward translocation of nitrogen in a Sphagnum bog. Sphagnum plants are lacking both roots and a vascular transport system. Rydin & Clymo (1989), however, demonstrated the internal acropetal transport of carbon and phosphorus in Sphagnum fallax. They showed the presence of numerous plasmodesmata linking stem cells which create a possible symplastic (cytoplasm to cytoplasm flow, contrasting apoplastic transport which indicates the flow of solutes through the cell walls) transport pathway. Furthermore, Ligrone & Duckett (1998b) found cytological evidence for nutrient translocation in Sphagnum. In a light- and electron microscope study they revealed that the cells in the central region of Sphagnum stems have a highly specialized cytoplasmic organization which has only been described for assumed soluteconducting cells in mosses (Ligrone & Duckett, 1994). The presence of these cells, referred to as conducting parenchyma cells, strongly suggests a cellular specialization in symplastic transport (Ligrone & Duckett, 1998a). Internal transport of nitrogen has often been assumed (Aldous, 2002b; Bonnett et al., 2010; Bragazza et al., 2005; Gerdol et al., 2006; Limpens & Berendse, 2003; Limpens Physiological evidence for internal acropetal transport 47

49 & Heijmans, 2008; Malmer, 1988) but, to our knowledge, has never been demonstrated before. Here, the ability of Sphagnum to transport nitrogen internally is investigated. Sphagnum cuspidatum and S. fallax are used in experiments in which diffusion and capillary transport are excluded and the internal transport of labeled nitrogen is monitored. An attempt is made to elucidate the mechanism of transport, apo- or symplastic. By killing a small part of the stem cells symplastic transport is excluded while the transport of solutes through the apoplast is still possible. The contribution of internal transport in the nutrient supply to Sphagnum and to the nutrient cycling in a Sphagnum bog with respect to other nutrient transport mechanisms in a Sphagnum bog will be discussed. Materials and methods Plant material Sphagnum cuspidatum Ehrh. Ex Hoffm. plants were collected in a pool at the edge of a small bog in the nature reserve Dwingelderveld (N , E ) in the north of the Netherlands. Sphagnum fallax (klinggr.) Klinggr. plants were collected in a small bog adjacent to the Dwingelderveld (N , E ). The mean annual nitrogen deposition in the north of the Netherlands is 28 kg ha -1 yr -1 (RIVM, 2009). The plants were collected and stored overnight (dark, 4 C) in plastic containers. Before being used in the experiment the plants were cut to a length of 10 cm and rinsed three times with demineralized water. Only visually non-damaged and fully green plants were used in the experiments. Experimental design To measure the ability of Sphagnum to transport nitrogen internally, three experiments were performed in which the transport of solutes by diffusion and capillary transport was excluded. The experimental set up for these experiments consisted of two compartments divided by a barrier created by placing two square Petri-dishes (120*120mm; Greiner bio-one GmbH) against each other. Both compartments were filled with 85 ml of diluted artificial rainwater (100 times) containing 20 mm MES buffer (ph = 4.0). To one of the two compartments, the donor compartment, labeled nitrogen 25 µmol L NH 4 Cl (98 atom %; Sigma Aldrich Inc., product number ) or 10 µmol L -1 K 15 NO 3 (98 atom %; Sigma Aldrich Inc., ) was added. The other compartment was called the acceptor dish. Both solutions were not in contact with each other. Three plants with a length of 10 cm were placed over the rim dividing the two compartments. The 10 cm long plants had green stems and were therefore considered to be alive and healthy. Two Sphagnum species were used in this experiment, Sphagnum cuspidatum and S. fallax. The compartments were covered with a lid to reduce evapotranspiration. All experiments took place in a climate controlled room at 18±1 C and a 16L:8D photoperiod. Experiment 1: acropetal transport of nitrogen In the first experiment the internal transport of ammonium and nitrate from the stem to the capitula (acropetal transport) was measured. The upper 2 cm of the plants (including the capitulum) were placed in the acceptor dish. The 8 cm long stem was placed in the donor dish. Plants were harvested after 1, 2, 4 and 8 days. 48 Solute transport in Sphagnum dominated bogs

50 Experiment 2: Basipetal transport of nitrogen The potential transport of nitrogen from the capitula to the stem (basipetal transport) was determined by placing the upper 2 cm of the Sphagnum plants in the donor dish and the stem in the acceptor dish. Nitrogen was applied as ammonium ( 15 NH 4 Cl) or nitrate (K 15 NO 3 ). Plants were harvested at day eight. Experiment 3: Apoplastic transport of nitrogen To distinguish between apoplastic and symplastic transport of ammonium the plants were again placed with the stem in the donor compartment, but a small segment of the stem was killed by steam using a modification of the method described by Rydin & Clymo (1989). Two centimeter below the capitula a stem section of ca. 1 cm was treated with steam for 60 seconds. The adjoining sections of the stem were protected from the steam by temporarily covering them with cork. Vitality of the steam-treated and untreated stem sections was determined by FDA staining (Elzenga et al., 1991; Heslop-Harrison & Heslop-Harrison, 1970). Vital cells were distinguished from non-vital cells based on their fluorescein green appearances using a Zeiss fluorescence microscope. During the experiment, the steamed part of the stem became white whereas the untreated parts remained green. Plants were harvested at day four. In all three experiments leakage of 15 N from the donor to the acceptor compartment by capillary flow along the stem was excluded by covering the stem at the rim dividing the two compartments with white Vaseline (Lamers & Indemans, s Hertogenbosch). The efficiency of this waxy barrier was tested visually by using the dye Brilliant Blue (Coomassie Brilliant Blue G, No. B0770, Sigma Aldrich) as a marker. In the acceptor compartment no Brilliant Blue was observed when wax was used. A second test for leakage and contamination of the acceptor dishes with 15 N, three capitula (so called contamination control capitula ) were placed in the acceptor compartment during the experiment and were harvested together with the other capitula and stems. These capitula indicate a possible change in 15 N concentration in the capitula during the experiment when not exposed to a source of labeled nitrogen. All experiments were performed in triplicate resulting in 3*3 plants per sampling day. Analyses Upon harvesting the Sphagnum plants were separated into the upper first centimeter (the capitulum) and the undermost seven centimeters of the stem. A 2 cm segment bridging the rim was not included in the analysis to allow the grinding of the tissue without interference of possible wax remains. After harvesting the plants were three times thoroughly rinsed with demineralized water and dried for at least 48 hours at 80 C. Subsequently, the plants were grinded individually to a fine powder using a ball miller (Retsch MM2, Haan, Germany). The contamination capitula were pooled to reduce the number of samples. For each sample the %N and % 15 N (stable nitrogen isotope composition) was measured with a Carlo Erba NA 1500 elemental analyzer (Thermo Fisher Scientific Inc.) coupled online via a Finnigan Conflo III interface with a Thermo-Finnigan DeltaPlus massspectrometer. Additionally, the natural occurring 15 N concentration in the capitula and stems of the Sphagnum plants was determined (day=0 samples). The data are expressed as the amount of 15 N assimilated in the plants (in µmol g DW -1 and was Physiological evidence for internal acropetal transport 49

51 calculated by the formula (% 15 N * %N / 100)*10000/15, where %N is the percentage of total N in the sample, % 15 N is the percentage of 15 N of total N and 15 is the molecular weight of the stable N isotope. Statistical analyses In the acropetal experiment the 15 N concentrations of the capitula and contamination control capitula in the acceptor dishes at day 1, 2, 4 and 8 were compared to the values in the capitula and stems at day zero by using a one way ANOVA, with day as independent variable. A Dunnett s post-hoc test was performed in case of differences within factors. In both the basipetal and the apoplastic experiment, a t-test was used to determine differences in 15 N concentrations in capitula or stems between day 0 and the end of the incubation period. In general for each factor we had three replicates and three plants per replicate. Prior to analysis, all data were transformed (1/x) when necessary to meet the assumption of homogeneous variance. All statistical analysis were performed by using SPSS for Windows (version , 2007; SPSS Inc., Chicago, IL, USA). Hyperbolic curves were fitted to the ammonium uptake data using graphing software (Prism version 4.03, 2005; GraphPad Software, Inc., San Diego, CA, USA). Results For both Sphagnum cuspidatum and S. fallax a slow but significant acropetal transport of ammonium through an internal mechanism was observed (figure 1). A one way ANOVA showed a significant difference in 15 N concentration between days in the capitula of both S. cuspidatum (F(4,40)=12.389, p<.01) and S. fallax (F(4,39)=8.522). In the capitula of S. cuspidatum a significant increase of 15 N was observed after four (p=0.12) and eight days (p<.01); 4.8 ± 0.4 and 7.1 ± 0.9 µmol g DW -1, respectively, compared to 3.7 ± 0.3 µmol 15 N g DW -1 at the beginning of the experiment (day=0). For S. fallax a significant increase of 15 N in the capitula was observed at day eight (p=0.001); 4.0 ± 0.3 µmol g DW -1 compared to 2.8 ± 0.2 at day=0. The concentration of 15 N in the contamination control capitula remained very low throughout the experiment (average value; p>0.05), from which we concluded that no contamination occurred and that the concentration of 15 N in the Sphagnum mosses was not affected by the experimental procedure itself. Most obvious is the rapid uptake of 15 + NH 4 by the stems of both Sphagnum cuspidatum and S. fallax for which within one day respectively 64 and 84% of the maximum concentration is reached (figure 1). For nitrate the uptake by the stems of S. cuspidatum shows a similar pattern as ammonium; - 74% is taken up within the first day (figure 1). On the other hand, hardly any NO 3 was taken up by the stems of Sphagnum fallax. A one way ANOVA showed no significant difference in 15 N concentration between days in the capitula of S. cuspidatum (F(4,40)=1.940, p>.05) and S. fallax (F(4,40)=1.225, p>.05). Also no significant difference between 15 N concentration in the contamination control capitula between days was observed from which we assumed that no contamination occurred. 50 Solute transport in Sphagnum dominated bogs

52 Figure 1. Mean 15 N concentration (± SE) in the stems (grey circles), capitula (black circles) and contamination capitula (open circles; ± SD) in Sphagnum cuspidatum (a,c) and Sphagnum fallax (b,d) measured after 0, 1, 2, 4 and 8 days of incubation of the stems in 15 NH 4 Cl (a and b) or K 15 NO 3 (c and d). Because of large differences in 15 N concentration between the capitula and stems - these values are plotted distinctively on respectively the left and the right y-axes. Notice that the concentration of NO 3 in the donor compartment was lower than ammonium (10 vs. 25 µmol L -1 ). Significant increase of 15 N in the capitula compared to the concentration at the start of the experiment (day = 0) are indicated by an asterisk (p<0.05). When no error bar is visible the standard error is lower than the size of the circles. For clarity the untransformed data are shown. In the experiment designed to determine possible basipetal transport of nitrogen, the 15 - NO 3 and 15 + NH 4 was readily taken up from the experimental solution by the capitula in both Sphagnum cuspidatum and S. fallax within eight days (figure 2). However, no significant increase of 15 N in the stems was observed after 8 days in S. cuspidatum with either ammonium or nitrate (t(16)= and t(16)=0.163, resp., p>0.05) and also S. fallax showed no significant increase with NH 4 and NO 3 (t(16)= and t(16)=0.584, resp., p>0.05; figure 2). When a 1 cm segment of the stem was killed by steam no significant differences in 15 N concentration between the capitula at day 0 and day 4 were observed for both Sphagnum cuspidatum (t(15)= -0.76, p>0.05) and S. fallax (t(15)= , p>0.05; figure 3) and the acropetal transport as demonstrated in figure 1 is inhibited. Physiological evidence for internal acropetal transport 51

53 Figure 2. The average concentration of 15 N (± SE) per gram dry weight in the stems (black circles), capitula (white circles) and contamination capitula (± SD; grey circles) in Sphagnum cuspidatum and Sphagnum fallax measured after the incubation of the capitula in 15 NH 4 Cl or K 15 NO 3 for 0 and 8 days. No significant differences in 15 N concentration between days in the stems of both S. cuspidatum and S. fallax were observed. When no error bar is visible the standard error is lower than the size of the circles. For clarity the untransformed data are shown. Discussion Internal transport of nitrogen in Sphagnum For both Sphagnum cuspidatum and S. fallax a slow but significant acropetal transport of nitrogen through an internal mechanism was observed (figure 1). In this chapter we describe that live stems + - of S. cuspidatum and S. fallax do take up both, NH 4 and NO 3, although with different efficiencies, + and that nitrogen supplied as NH 4 is subsequently transported acropetally to the capitulum. The + - difference between NH 4 and NO 3 uptake efficiency will also have been determined by the concentrations in which both N species were applied, 25 and 10 µmol L -1 + for NH 4 and NO 3-, respectively. Yet, these values were chosen since they represent natural bog water concentrations under natural conditions. - + After uptake, both NO 3 and NH 4 are assimilated into the amino acid glutamine (Gln) and subsequently converted into other amino acids (Kahl et al., 1997; Rudolph et al., 1993). Therefore, it is assumed that transport of N takes place in the form of amino acids. When a segment of the stem is killed by steam, Sphagnum plants no longer show a significant 52 Solute transport in Sphagnum dominated bogs

54 Figure 3. The average concentration (± SE) of 15 N in the stems (black circles), capitula (open circles) and contamination capitula (± SD; grey circles) in Sphagnum cuspidatum and S. fallax measured after the incubation of the stems in 15 NH 4 Cl for 0 and 4 days. A one centimeter segment of the stem was killed by steam. No significant differences in 15 N concentration between days in the capitula of both S. cuspidatum and S. fallax were observed. When no error bar is visible the standard error is lower than the size of the circles. For clarity the untransformed data are shown. transport of 15 N to the capitula within four days. Killing a segment of the stem by steam, blocks symplastic transport of compounds, but should still allow apoplastic transport (Rydin & Clymo, 1989). Therefore, these results are indicative for the symplastic nature of acropetal transport of nitrogen, and are in full agreement with the findings of Ligrone and Duckett (1998b) and Rydin & Clymo (1989) who demonstrated cellular specializations of Sphagnum for symplastic transport. Nevertheless, the duration of four days might be too short to totally exclude a small contribution by + - apoplastic transport. An alternative hypothesis is the diffusive transport of NH 4 and NO 3 through the cell wall and the hyaline cells present in the epidermis of the stem of Sphagnum. However, diffusive extracellular transport would result in similar transport rates in both acropetal and basipetal direction. In contrast, 15 N is not being transported into the stem due to basipetal transport of 15 N in experiment 2. In our experiment we only observed net transport of nitrogen from stem to capitula. A possible explanation for this unidirect transport might be the higher sink strength for N of the capitulum, relative to the stem, which can be expected because growth is taking place exclusively in the capitulum. Since mineralized nitrogen is an important nitrogen source (Aerts et al., 1999; Aldous, 2002b; Bridgham, 2002; Gerdol et al., 2006; Morris, 1991; Urban & Eisenreich, 1988), that becomes available in the deeper layers, distinct from the growing capitulum in the top layer, the supply of nitrogen depends on upward-directed transport mechanisms (Aldous, 2002b; Bridgham, 2002). Uptake by the stem and subsequent internal transport to the growing parts is thus a possible pathway for mineralized nitrogen. Internal, stem-mediated transport can be involve in re-allocation of nitrogen from older, senescing tissue, a nutrient retention mechanism common in vascular plants (Aerts, 1990, 1995; Chapin, 1980; Vitousek, 1982) and in transport of nitrogen taken up from the external medium by the stems. This distinction must be taken into consideration when reviewing the importance of different processes in the transport of externally mineralized nutrients. Gerdol et al. (2006) stated that the mineralization of N is more important than the enzymatic reallocation of N. Physiological evidence for internal acropetal transport 53

55 Our experimental procedure only involves the translocation of the externally mineralized nitrogen. Higher C:N ratios in stems than in capitula are often observed in Sphagnum (e.g. Malmer, 1988) and are taken as an indication for the internal reallocation of N from the stem to capitula. This assumption, in combination with the internal transport of C and P (Rydin & Clymo, 1989), were reasons for the general acceptance of internal transport in Sphagnum (Bonnett et al., 2010; Bragazza et al., 2005; Gerdol et al., 2006; Limpens & Berendse, 2003; Limpens & Heijmans, 2008; Malmer, 1988). However, experimental evidence for the internal reallocation of N was lacking. Our findings provide physiological evidence for internal transport of N taken up from the external medium. There is, however, no reason to assume that the same mechanism does not transport N internally released from senescing tissue. + - NH 4 and NO 3 Within the time frame of the experiments the internal transport of nitrogen was only observed + when nitrogen was supplied as NH 4 and not as NO 3-. According to a review by Rudolph et al. (1993) - this difference might have a metabolic origin. After uptake NO 3 is reduced to NH 4+, which takes place in the chloroplast. The subsequent assimilation into amino acids also takes place in the + chloroplast. On the other hand, the assimilation of NH 4 can take place in the cytosol (Rudolph et al., 1993). Taking symplastic transport into account (see below), the N present in the cytosol is available for transport whereas the amino acids in the chloroplast are more or less fixed. - The uptake of NO 3 by the stems differed between the two species. The stems of S. fallax hardly - - took up NO 3 within the eight days of the experiment. Wiedermann et al. (2009) showed NO 3 to be taken up in small amounts by capitula of Sphagnum balticum and S. fuscum from a solution containing in total four N forms (NH 4+, alanine and glutamine and NO 3- ). According to Woodin & Lee (1987), most absorption of nitrate takes place in the capitulum and decreases down the stem. - With the stems hardly taking up NO 3 the transport to the capitula can not be expected. The rate of internal transport In earlier studies (Aldous, 2002b; Bridgham, 2002) the contribution of translocation to the nitrogen supply of the capitula was shown to be significant. However, these studies did not distinguish between different types of transport. Based on the uptake and internal transport of ammonium data by S. cuspidatum (figure 1) the rate by which nitrogen is transported internally can be estimated. Since already after one day the maximum concentration of 15 N in the stem is nearly reached, we can assume that the concentration in the stem is almost constant for the duration of the experiment. Since there is a lag period of almost 4 days before labeled N appears in the capitulum and we know the length of the stem segment between the donor compartment and the receiving capitulum (which is 2 cm), we can calculate the speed of the acropetal, symplastic transport process: 5 mm/d. From the increase in the capitulum in the four days following the lag, 2.3 µmol g DW -1 or 12% of the total N taken up by the stems, we can calculate a half time value of equilibration between donor and receptor sections of the plant: 17 days. Notice that this calculation is based on the assumption that the amount of 15 N in the capitula is a function of the amount of 15 N taken up by the stems in the preceding four days. It should be noted that the delay and apparent transport speed taken the assimilation + of NH 4 into amino acids is also included. This calculated half time value for N is higher than the estimated half time value of 11 days for the internal transport of C and P (Rydin & Clymo, 1989). 54 Solute transport in Sphagnum dominated bogs

56 Since the internal transport of nitrogen is a mechanism for efficient nitrogen use, the transport rate is expected to be reduced under high N content in the capitula (Bragazza et al., 2004). The plants used in these experiments were collected from an area with a high load of atmospheric nitrogen deposition. It has been demonstrated that Sphagnum mosses subjected to a high N supply accumulate elevated amounts of N (Nordin & Gunnarsson, 2000; Van der Heijden et al., 2000; Limpens & Berendse, 2003; Limpens et al. 2011). Moreover, The capitula of the Sphagnum cuspidatum plants used in our experiments also showed a relatively high N content: 15.3 ± 2.2 mg g -1. If indeed internal transport rates are negatively affected by nitrogen supply, our estimates are likely to be an underestimation for Sphagnum residing in non-polluted areas and internal transport might be more important. The importance of internal transport The uptake of ammonium by the living stems and the subsequent transport of N to the capitula shows that internal transport very likely functions as a mechanism in supplying the capitula with N. Moreover, it implies that Sphagnum mosses are able to compete with microbes and vascular plant roots for available soil nitrogen (see Bridgham, 2002) and thereby might function as a N retention step contributing to the efficient use of N by Sphagnum in ombrotrophic bogs. However, the importance of internal transport in the nutrient supply of Sphagnum and nutrient cycling in bogs depends on the transport rate relative to other transport mechanism. In a review of the internal transport in non-vascular plants (Raven, 2003) it is claimed that there is no evidence for symplastic transport in Sphagna faster than can be accounted for by diffusion. Next, the rate at which nitrogen is transported internally is very slow compared to the transport rates by buoyancy-driven water flow (Rappoldt et al., 2003). Chapter 2 shows buoyancy flow to be a fast and effective nutrient transport mechanism in bog water. The uptake kinetics of Sphagnum cuspidatum and S. fallax show the ability of Sphagnum to take up large amounts of ammonium relatively fast, indicating buoyancy flow to be a very effective nutrient transport mechanism in supplying the capitula with nitrogen. Therefore, with the regular occurrence of buoyancy flow, the supply of nitrogen by internal transport might be insignificant, compared to the supply of nitrogen by buoyancy flow. Indeed, buoyancy flow is restricted to water-saturated Sphagnum habitats and, because of its dependence on varying physical parameters like the difference in temperature between day and night, an irregularly occurring phenomenon. The importance of internal transport might therefore reside in its continuous character, continuously supplying Sphagnum with N which contrasts with the pulse wise supply of N by buoyancy flow (and of course precipitation). For Sphagnum species that form hummocks that extend above the water surface and do not benefit from buoyancy flow, internal transport is, next to capillary transport, a possible acropetal pathway for nutrients. Clymo (1973) estimated the average velocity by capillary flow to be 0.4 mm min -1. However, this rate, and the concomitant nutrient transport, is dependent of several factors, like evaporation, plant density and pore water nutrient concentrations (Clymo & Hayward, 1982). Moreover, during extracellular transport nutrients may be lost to microorganism or vascular plant roots. Compared to the external transport mechanisms (buoyancy flow and capillary transport) internal transport very likely represents a minor contribution in the upward transport of externally supplied N to the capitula. The main importance of internal transport is therefore considered to be the reallocation of internally catabolized nitrogen compounds. Physiological evidence for internal acropetal transport 55

57 Chapter 5

58 The importance of groundwater carbon dioxide in the restoration of Sphagnum bogs Wouter Patberg Gert Jan Baaijens Alfons Smolders Ab Grootjans Theo Elzenga

59 Abstract Essential for successful bog restoration is the re-establishment of Sphagnum mosses. High CO 2 availability has been shown to be of great importance for the growth of Sphagnum mosses. In well-developed Sphagnum bogs large amounts of CO 2 are produced by decomposition processes in the peat layer. In cut-over Sphagnum bogs this carbon source is often absent or strongly reduced. Therefore, for the successful restoration of cut-over Sphagnum bogs an alternative, additional carbon source might be essential for the re-establishment of Sphagnum mosses. This chapter focuses on the role of CO 2 in the development of Sphagnum bogs in a field situation. Study area is one of the largest wet heathland reserves in Western Europe and is characterized by many small damaged Sphagnum bogs. Rewetting measures resulted in large developmental differences between bogs; some bogs developed markedly well, whereas others did not. Of ten small bogs the developmental success was quantified using aerial photographs and surface water and groundwater samples were collected. In addition, the physiological characteristics of carbon dioxide uptake of two Sphagnum species were determined. Water chemistry analysis revealed that the total inorganic carbon concentration (TIC) in the nearby groundwater of the well-developed bogs, is significantly higher than that of poorly developed bogs. The CO 2 availability in the surface water of the investigated bogs was positively correlated to the inorganic carbon in the groundwater. It is concluded that the well-developing bogs are fed by a carbon-rich groundwater inflow from outside the bog. The CO 2 uptake kinetics of both Sphagnum species are characterized by a high compensation point and a low affinity, both indicating an adaptation to a high CO 2 availability. The present findings indicate that high carbon dioxide availability is a prerequisite for the successful re-establishment of Sphagnum mosses in peat bog restoration projects and that carbon-rich groundwater can apparently substitute for the decomposing peat layer as a source of CO 2. Therefore, the availability of CO 2 should be included in bog restoration feasibility studies. 58 Solute transport in Sphagnum dominated bogs

60 Introduction Due to the extensive exploitation for fuel, agriculture and forestry over many centuries, living (peat forming) mires have become endangered in most of North-Western Europe (Rochefort & Price, 2003). Even nowadays (extensive) peat extraction activities take place for commercial use in, for example, Canada, Scandinavia, Ireland and the Baltic states (Joosten, 2009). Due to the important role of peatlands in the global carbon cycle, and because of their unique ecological values, globally much effort is dedicated to the restoration of damaged mires. However, the restoration of large bog remnants in particular, has proven to be fairly complicated and not always successful (Grootjans et al., 2006; Money & Wheeler, 1999; Money et al., 2009). Essential for successful bog restoration is the re-establishment of Sphagnum mosses followed by the re-development of a functional acrotelm, leading to a self-sustaining system (Money & Wheeler, 1999; Money et al., 2009; Smolders et al., 2003). Since wet conditions are essential for Sphagnum growth (Robroek et al., 2009), the creation of suitable wet conditions is a prerequisite in restoring peatlands (e.g. Money et al. 2009). Often rewetting is realized by inundating large areas to ensure wet conditions throughout the year (Money & Wheeler, 1999; Smolders et al., 2003). The water layer can be colonized by aquatic Sphagnum species, especially Sphagnum cuspidatum, to form dense mats on which peat forming species like S. magellanicum and S. papillosum might establish (Money & Wheeler, 1999; Wheeler & Shaw, 1995). However, this method often results in large water bodies in which Sphagnum growth is severely hampered. The lack of success in re-colonization of aquatic Sphagnum species in rewetted bog remnants has been ascribed to the limited availability of light and/or CO 2 (Money & Wheeler, 1999; Smolders et al., 2001; Smolders et al., 2003; Wheeler & Shaw, 1995). Like most aquatic bryophytes (Raven et al., 1985) Sphagnum mosses are known to be obligate CO 2 users (Bain & Proctor, 1980) and are solely dependent on the diffusive supply of CO 2 to the site of carbon fixation (Rubisco - ribulose-1,5-bisphosphate carboxylase-oxygenase). In very wet conditions the Sphagnum mosses are surrounded by a thick water layer which lowers CO 2 conductivity resulting in a reduced photosynthetic rate (Silvola, 1990; Williams & Flanagan, 1996). Consequently, high rates of underwater photosynthesis can only be sustained when the leaves are exposed to high levels of CO 2 (Jauhiainen & Silvola, 1999; Paffen & Roelofs, 1991; Silvola, 1990; Smolders et al., 2003). In well developed bogs CO 2 is produced in large quantities by decomposition processes in the peat layer (Bridgham & Richardson, 1992; Glatzel et al., 2004; Smolders et al., 2001; Waddington et al., 2001). Carbon dioxide concentrations in the pore water can reach up to several millimolar (Smolders et al., 2001; Smolders et al., 2003), compensating low diffusion rates and ensuring sufficient substrate delivery for photosynthetic carbon fixation (Maberly & Madsen, 2002; Silvola, 1990). This so-called substrate-derived CO 2 has been shown to be an important carbon source for aquatic and emergent Sphagnum mosses (Baker & Boatman, 1990; Paffen & Roelofs, 1991; Riis & SandJensen, 1997; Roelofs, 1983; Smolders et al., 2001; Smolders et al., 2003). Under more reductive conditions the methane production in the catotelm may become higher than the production of CO 2. This methane can be oxidized by methanotrophic bacteria to CO 2, which then can be used as a carbon source by Sphagnum mosses (Kip et al., 2010). Cut-over peat bogs often lack this source of additional inorganic carbon. This type of damage The importance of groundwater 59

61 can often be found in North-Western Europe (Joosten, 2009). Peat extraction has removed the bulk of organic material. The highly decomposed, humified peat which is left behind, has only limited CO 2 (and methane) production rates (Bridgham & Richardson, 1992; Glatzel et al., 2004; Tomassen et al., 2004; Waddington et al., 2001). Therefore, for the successful restoration of cut-over Sphagnum bogs an additional carbon source might be essential for the re-establishment of Sphagnum mosses. This study focuses on the role of CO 2 in the development of Sphagnum bogs in a field situation. The study area is one of the largest wet heathland reserves in Western Europe and is characterized by numerous small Sphagnum bogs. They have been damaged by drainage and small scale peat excavations in the past. From 1988 onwards, rewetting measures have been carried out, but the developmental success has varied significantly between bogs; some bogs developed well, whereas others did not. It is hypothesized that in these hydrologically degraded bog remnants the restoration of Sphagnum growth is limited by the availability of CO 2. We expect that the well-developing bogs are being fed by lateral, carbon-rich, groundwater inflow. Groundwater with high inorganic carbon concentrations entering a bog, will release high amounts of CO 2 when it comes into contact with the more acidic water around the Sphagnum mass. The higher availability of CO 2 will stimulate the growth of aquatic Sphagnum mosses and subsequent bog development. We also determined the CO 2 requirement for two Sphagnum species (S. cuspidatum and S. fallax), which are abundant in well-developing bogs. Materials & Methods Study area Study area is the Dwingelderveld, one of Europe s largest wet heathland areas (about 3500 hectares), situated in the north of the Netherlands ( N E). The landscape consists of pine forest, wet and dry heathland and many small peat bogs scattered throughout the area (figure 1). The presence of boulder clay underneath the reserve is responsible for the generally wet character of the area. Wind erosion resulted in differences of up to 5 meters in height of the Pleistocene sand cover. During the second half of the last century most of the wet heathland and bogs became desiccated by drainage activities both in the reserve (for the benefit of pine plantations) and also in the surrounding brook valleys (for the benefit of agriculture). Many small bogs were also subjected to small scale peat cutting by farmers. These activities have ended around In 1988 large scale rewetting measures, i.e. closing of drainage ditches and cutting of trees, were initiated with variable results in developmental success of the different bogs. Some bogs developed well and are characterized by a luxurious growth of Sphagnum spp., whereas others did not and mainly consist of open water with marginal occurrence of Sphagnum mosses (Grootjans et al., 2003). An interesting observation by Grootjans et al. (2003) was that bogs with abundant Sphagnum growth were located in old erosion gullies, while bogs without Sphagnum-dominated succession were found outside or at the edge of these gullies. In these erosion gullies impermeable podsolic layers stretch out beyond the border of the bogs itself. The groundwater levels in the sandy hills are generally higher than in the gullies. Since the vertical conductance of these podsolic layers is very 60 Solute transport in Sphagnum dominated bogs

62 low, a horizontal sub-surface flow of groundwater towards the bogs is facilitated, prolonging the residence time of water in the soil and allowing the groundwater to become possibly enriched with inorganic carbon (Grootjans et al., 2003; figure 2). Figure 1. Aerial photograph of the research area in the Dwingelderveld. All bogs in this part of the nature reserve are outlined by a black line. The bogs used in this field study are indicated by a number corresponding to the numbers used in table 1. The numbers of the poorly developed bogs are underlined. The locations of the sampling sites are indicates by black circles. The white spot in the Diepveen (6) is open water as proven by field observations. Classifying peat bogs In 10 bogs 20 sampling sites were selected (figure 1; table 1). Aerial photographs of 1982 and 2006 were used to determine the developmental success of each of the sampling sites. The increase in surface area covered by Sphagnum mosses was used as a measure for the success of bog development. With the use of image analysis software (ImageJ, version 1.41o, National Institute of Health, USA) the area of open water per sampling site in both 1982 and 2006 was calculated based on grey scale differences between vegetation and open water. The developmental success was determined by calculating the relative decrease of open water surface between 1982 and The following formula was used: (%OW %OW 2006 ) / %OW 1982, where %OW is the surface percentage of the bog occupied by open water in 1982 or In this aerial photograph analysis the vegetation was assumed to be bog vegetation dominated by Sphagnum mosses. This was validated by field observations. In some cases, Cootjes Veen, Groote Veen East and Veerles Veen, the aerial photograph analysis had to be adjusted. Here the vegetation was dominated by vascular plants (e.g. Molinia caerulea) instead of Sphagnum mosses. Consequently these bogs were classified as poorly developed bogs. The importance of groundwater 61

63 Table 1. Names, coordinates, the percentage open water in 1982 and 2006, the decrease of open water and the developmental success of the investigated bogs in the Dwingelderveld. The developmental success of the bogs was determined by calculating the relative decrease of open water surface in 2006 compared to the surface open water in A + indicates a well developed bog and a - indicates a poorly developed bog. For three bogs the developmental success was not determined by aerial photographs as indicated by an *. See the materials and methods section for a more detailed explanation. The numbers in the first column correspond with the numbers in figure 1 Surface open water % Decrease open water Develop mental success # Name Coordinates (%) 1 Barkmans Veen N E Groote Veen N E Reigersplas N E Adderveen N E Cootjes Veen N E * - 6 Diepveen N E Groote Veen East N E * - 7 Kliploo N E Schurenberg N E Veerles Veen N E * - 10 Zandveen N E Water sampling and analysis Groundwater and surface water samples were taken at the sampling sites in February and April 2007, August and October 2008 and September Groundwater samples were collected using piezometers (Ø32 mm PVC tubes with nylon filters) and a peristaltic pump. The piezometers were placed just outside the bogs on the slopes of the gully and always with the filters above the impervious layer. The exact position of the piezometers was chosen such that they intercepted the groundwater flowing into the bogs, based on the results of a previous hydrological study of the area (Verschoor et al., 2003). One day before sampling the water present in the piezometers was discarded to allow refilling with fresh groundwater. Surface water samples were taken in the bog close to the piezometers. 30 ml airtight bottles were filled by gently submersing them in the surface water. Water samples were transported in a cool box to the laboratory where ph and TIC measurements were performed immediately. The remaining samples were stored frozen until further analyses. Precipitation data of the Dwingelderveld were obtained from the Royal Netherlands Meteorological Institute ( station number 327 Dwingeloo ). The concentration of Total Inorganic Carbon (TIC) in the water samples was determined by measuring the CO 2, released after acidifying the samples to a ph <3, using an Infra-Red Gas Analyzer (IRGA, ABB Advance Optima). The ph of the water samples was determined using a combined ph electrode with an Ag AgCl internal reference (Cole Parmer Instrument Company, Illinois, USA) and a PHM 64 ph meter (Radiometer, Copenhagen). The concentrations of CO 2 and bicarbonate in the water samples were calculated based on the ph and the TIC concentration (Prins & Elzenga, 1989). 62 Solute transport in Sphagnum dominated bogs

64 Concentrations of nitrate (NO 3 ), ammonium (NH 4 ) and chloride (Cl) were measured colourimetrically according to Geurts et al. (2008) and potassium (K) by flame photometry by using an Auto Analyzer 3 system (Bran+Luebbe, Norderstedt, Germany). Aluminium (Al), calcium (Ca), iron (Fe), magnesium (Mg), manganese (Mn), sodium (Na), total phosphorus (P), sulphur (S), silicon (Si) and zinc (Zn) were measured using an ICP Spectrometer (IRIS Intrepid II, Thermo Electron Corporation, Franklin, MA). Figure 2. A cross section of a part of the study area in the Dwingelderveld. The small bogs are situated in the gullies which are surrounded by the higher Pleistocene sand cover. A layer of boulder clay is present underneath the whole area, resulting in wet conditions throughout the area. The bogs are characterized by the presence of podsolic layers, responsible for the wet conditions and bog development. Note the lack of peat development in the bog in the forefront which lies outside the gully. Groundwater samples were taken adjacent to the border of the bogs, as indicated by the piezometer. CO 2 uptake characteristics Carbon dioxide uptake characteristics of Sphagnum cuspidatum Ehrh. Ex Hoffm. and S. fallax (klinggr.) Klinggr., two pioneer Sphagnum species important in the initial stage of bog formation (Money, 1995; Smolders et al., 2003), were determined by measuring the photosynthetic activity (A) at different CO 2 concentrations at saturating light conditions (1500 µmol m -2 s -1 ; Hansatech Quantitherm Light meter). Plants were collected in April 2008 in two small Sphagnum bogs in the Dwingelderveld area. A capitulum was placed in a closed thermostatic cuvette containing 1 ml of measuring solution (see below) which was stirred continuously. The photosynthetic evolution of oxygen was measured by a Clark electrode located at the bottom of the cuvette in combination with a millivolt recorder. The temperature of the measuring solution was 21 C. Any inorganic carbon present in the hyaline cells or adhering water was removed by illuminating the capitula with 1000 µmol m -2 s -1 for at least 60 minutes while keeping them in a CO 2 free medium. When the capitula showed a steady, low rate of oxygen uptake it was assumed that no significant CO 2 stores were left in the hyaline cells. The solution was then removed from the cuvette by using a syringe and replaced by a solution with the desired CO 2 concentration. Different CO 2 concentrations ranging from 0 to 800 µmol L -1 in 10 times diluted artificial rainwater (Smolders et al., 2001) containing 20 mm MES ph = 5.5 were created of which 1 ml was used in the cuvette and The importance of groundwater 63

65 1 ml was used to exactly determine the Total Inorganic Carbon concentration, by using an infrared gas analyzer (CO 2 analyzer model no. S151, QUBIT Systems Inc., Kingston, ON, Canada). The ph of the measuring solution was determined using a ph microelectrode (type MI-406, Microelectrodes Inc., Bedford, NH, USA) in combination with an Ag AgCl micro-reference electrode (type MI-401, Microelectrodes Inc., Bedford, NH, USA) and a millivolt meter. The concentrations of CO 2 and bicarbonate in the water samples were calculated based on the ph and the TIC concentration (Prins & Elzenga, 1989). Per measurement one capitulum was used and each capitulum was used for a maximum of three measurements. Large branches were trimmed to fit in the cuvette. After usage capitula were frozen at -80 C, ground to a powder and the chlorophyll concentration was determined according to Lichtenthaler (1987). Statistical analysis Data were tested for normality using a Kolmogorov-Smirnov test and equality of variance using Levene s test. The assumption of homogeneity of variance was not always met, not even after transformation of the data. According to Heath (1995), the analysis of variance appears not to be greatly affected by heterogeneity in variance if sample sizes are more or less equal. Therefore, we decided to continue our analysis using non-transformed data. Differences in groundwater TIC concentration and surface water CO 2 concentration between well and poorly developed bogs, were tested using a mixed model, with bog development (well and poor) as fixed factor and sampling date as repeated measure. Differences between sampling dates were determined by using Bonferroni s post-hoc test. The Pearson correlation coefficient was determined for the relation between groundwater TIC and surface water CO 2 concentrations (both log transformed). The mixed model and the Pearson correlation test were performed using SPSS for Windows (version , 2007; SPSS Inc., Chicago, IL, USA). Additionally, the data were analyzed by applying principal component analyses (PCA) by using Aabel (version 3.0.3; Gigawiz Ltd. Co., Tulsa, OK, USA). All data were normalized. A hyperbolic curve was fitted to the CO 2 uptake data using graphing software (Prism version 4.03, 2005; Graph- Pad Software, Inc., San Diego, CA, USA). Results Classification of peat bogs Based on the aerial photographs and field observations the developmental success of the selected bogs was classified into two categories; well developed bogs, with an open water decrease of at least 78% and poorly developed bogs with a decrease of open water less than 20% (table 1). The Adderveen, however, is an exception with an open water decrease of 42%. For most sites the field observations confirmed the classification based on the aerial photograph analysis: well developed bogs showed luxurious and dominant growth of Sphagnum spp. without or with little open water, whereas in poorly developed bogs marginal Sphagnum spp. growth was accompanied by the presence of vascular plants and open water. Open water surface decreased in all sites between 1982 and 2006, except for Diepveen, where the open water surface slightly increased with 8%. 64 Solute transport in Sphagnum dominated bogs

66 Table 2. Water analysis of the groundwater and surface water of well (+) and poorly (-) developed bogs; average values for all sampling dates are given with standard deviation (SD). Ion concentrations are given in µmol L -1 Groundwater Surface water mean SD mean SD mean SD mean SD TIC HCO CO ph NH NO K P Al Ca Cl Fe Mg Mn Na S Si Zn Water chemistry Bog water chemistry data are shown in table 2. The difference in groundwater composition between the well developed and poorly developed bogs is illustrated by a principal component analysis (figure 3). The groundwater samples from well and poorly developed bogs appear as clusters in the PCA diagram along the first principal component axis that is dominated by high TIC, iron (Fe) and silicium (Si), indicating that good Sphagnum development is associated with groundwater rich in inorganic carbon. The groundwater is relatively acid and CO 2 is the main inorganic carbon species (table 2). The chemical composition of groundwater and surface water is clearly different, with concentrations of most multivalent minerals and inorganic carbon being higher in the groundwater (table 2). Groundwater TIC concentrations are shown per location in figure 4. The average groundwater TIC concentration near the well developed bogs was 4983 ± 802 µmol L -1 and ranged from 2620 to 6215 µmol L -1. The poorly developed bogs, with an average TIC concentration of 2765 ± 1426 µmol L -1, showed a much wider range in TIC concentration both between measurements at The importance of groundwater 65

67 individual locations and between locations. A significant main effect of category on groundwater TIC concentrations was found (F(1,12) = , p=0.001). In the well developed bogs the average CO 2 concentration in the surface water was 1215 ± 730 µmol L -1 and for the poorly developed bogs 743 ± 625 µmol L -1 (table 2). For surface water CO 2 concentration also a significant main effect of category was found, F(1,13) = 6.063, p= Figure 3. Principal Component Analysis (PCA) biplot of all groundwater samples and selected environmental variables. Each symbol represents a sampling location at one of the sampling dates. Black circles are the well developed bogs and the open circles represent the poorly developed bogs. The first axis explains 27% of the variation and second principal component accounts for 20 % of the variation. Figure 4. Box plot showing the total inorganic carbon (TIC) concentration in the groundwater in µmol L -1 per location of both well (gray boxes) and poorly (white boxes) developed bogs. Box plots are composed of minimum, maximum, 25%, 75% quartiles and the median. Where used, north, south, east and west, indicate sampling sites at one bog. 66 Solute transport in Sphagnum dominated bogs

68 Figure 5. Groundwater total inorganic carbon concentrations (TIC, solid lines) and surface water CO 2 concentrations (dashed lines) for the well developed bogs (filled symbols) and the poorly developed bogs (open symbols) shown per sampling date. Symbols represent mean values in µmol L -1. Bars represent standard deviations and are one sided for readability of the graph. Different letters mean significant differences between sampling dates, tested for groundwater TIC and surface water CO 2 concentrations separately. Figure 5 shows the changes over time in the total inorganic carbon and CO 2 concentration of both groundwater and surface water of well and poorly developed bogs. A significant main effect of date on both groundwater TIC and surface water CO 2 concentrations was found (F(4,48) = and F(4,52) = 6.126, respectively, p<0.01); the groundwater TIC concentration was significantly higher on August than in February (p<0.05). For the surface water CO 2 concentrations significant higher values were found in April and September (figure 5). However, no significant interaction effect of category * date was found for both groundwater TIC (F(4,48) = 0.566, p>0.05) and surface water CO 2 concentrations (F(4,52) = 1.317, p>0.05). In other words, groundwater TIC and surface water CO 2 concentrations in well and poorly developed bogs, respectively, were not affected differently by date. The logarithm of the groundwater TIC concentration was positively and significantly correlated to the logarithm of the CO 2 concentrations in the surface water. However, this correlation was the strongest in the well developed bogs; Pearson s r=0.521 (p<0.01) compared to a value of r= (p<0.05) in the poorly developed bogs (figure 6). The importance of groundwater 67

69 Figure 6. The relation between the total inorganic carbon (TIC) concentration in the groundwater and the CO 2 concentration in the surface water for well (filled circles) and poorly (open circles) developed bogs. Water samples collected in April and August are in black, others in gray. Lines are regression lines for well (solid line, r 2 =0.272) and poorly (dashed line, r 2 =0.082) developed bogs. For clarity of presentation, the non-transformed data are shown. Figure 7. Dose-response curves of photosynthetic activity as a function of the carbon dioxide concentration for Sphagnum cuspidatum (filled symbols; r 2 =0.935) and S. fallax (open symbols; r 2 = 0.974). CO 2 uptake characteristics The photosynthetic rate of Sphagnum cuspidatum and S. fallax as a function of CO 2 concentration are shown in figure 7. A hyperbolic curve according to the formula A=V max *[CO 2 ]/(K m +[CO 2 ]) + c was fitted to the data. From the curve the compensation point (Γ) for CO 2 was calculated (table 3). Table 3. Carbon dioxide uptake characteristic (±SD) for Sphagnum cuspidatum and S. fallax. See text for explanation of the parameters S. cuspidatum S. fallax V max (nmol O 2 s -1 mg chl -1 ) 19.8 ± ± 2.6 K m (µmol CO 2 L -1 ) ± ± 52.5 Γ (µmol CO 2 L -1 ) c -1.4 ± ± Solute transport in Sphagnum dominated bogs

70 Discussion Restoration success The situation in the Dwingelderveld is characterized by a number of small Sphagnum bogs that were subjected to identical restoration measures, but differing in groundwater influence and developmental success. This offered us the opportunity to study the importance of CO 2 in the development of Sphagnum bogs in a field situation. The aerial photograph analysis revealed distinct differences in restoration success. The results were based on a quantitative approach; the occurrence of a high cover of Sphagnum species in general. The success of the restoration was also evaluated by Everts et al. (2002) using total species composition. Their results were in agreement with the results from our approach using aerial photographs. Evidence for influence of local groundwater flows The PCA analysis using all main groundwater chemical data showed that well developed and poorly developed bogs separate well, and that total inorganic carbon (TIC) differences are largely responsible for that; the groundwater nearby the well developed bogs contains a significantly higher TIC concentration than groundwater nearby the poorly developed bogs (figure 4). We also found that CO 2 concentrations in the surface water of the well developed bogs were significantly higher than in the poorly developed bogs. Moreover, the CO 2 availability in the surface water of the investigated bogs is positively correlated to the inorganic carbon concentration in the groundwater (figure 6). Carbon dioxide is very likely released from the carbon-rich groundwater upon entering the acidic bog environment resulting in an increased CO 2 availability stimulating Sphagnum growth. Higher groundwater levels in the surrounding sandy areas will result in a flow of the local groundwater towards the bogs. The concave shaped impermeable layer, essential for bog formation, will result in the uni-directional flow of the local groundwater towards the bogs. An alternative hypothesis is that the high inorganic carbon concentrations found in the groundwater nearby the well developed bogs has its origin in the decomposition of organic material in the bog. However, the chemical signature of the groundwater indicates an origin from outside the bog. This was particularly clear in the silicium values, which are generally much higher in water that has been in contact with mineral sediments for a longer period (Engelen & Jones, 1986). Moreover, the groundwater near the Diepveen and the Zandveen which are poorly developed bogs, contained high TIC concentrations despite the absence of accumulated organic material (figure 4). Additionally, the placement of the piezometers on the concave shaped impermeable layers ensures the sampling of inflowing water. Therefore, the results presented are in agreement with the hypothesis that the well-developing bogs are fed by a lateral, carbon-rich, groundwater inflow. To investigate differences between well and poorly developed bogs, a principal component analysis was performed based on the chemical composition of the groundwater. Since the developing Sphagnum capitula are in close contact with surface water and only indirectly influenced by groundwater this seems counter-intuitive. However, surface water composition is more influenced by short term changes compared to the more stable composition of groundwater. Next, fluxes of CO 2 are more important than the resulting concentrations. The growth of Sphagnum mosses resulting from the release of carbon rich groundwater will generate some positive feedbacks which will The importance of groundwater 69

71 result in an enhancement of the submerged Sphagnum growth (figure 8). The resulting accumulation of organic matter will enhance the internal generation of CO 2 from decomposition processes. Furthermore, it will result in a decrease of the water depth which will increase light availability. Figure 8. Schematic view of the positive feedbacks concerning CO 2 availability in a Sphagnum dominated bog resulting from the input of carbon rich groundwater. In a well developed bog the decomposition of accumulated organic matter results in a high availability of CO 2 stimulating Sphagnum growth, which in turn increases the accumulation of organic matter. During the initial stages of bog development organic matter is absent and the inflow of carbon rich groundwater can substitute for the organic matter as a source of CO 2, stimulating Sphagnum growth and thereby inducing the internal positive feedback mechanism concerning CO 2 availability. Additionally, the accumulation of organic matter decreases water depths, increasing light and CO 2 availability stimulating Sphagnum growth as well. Seasonality of groundwater input Some seasonality in groundwater input appears to be visible. In April and August we observed increased CO 2 concentrations in the well developed bogs compared to the poorly developed bogs (figure 5). In autumn and winter we did not observe differences in CO 2 in the surface water of the bogs. In spring and (early) summer lowered surface water levels (due to evaporation) will result in an increase of the hydrological gradient resulting in an increased inflow of groundwater into the bogs. In autumn and winter, this gradient will be decreased by higher surface water levels due to an increase in precipitation. Additionally, strong rainfalls will lead to an increased run off towards the bogs of very superficial groundwater which is relatively poor in CO 2. Interestingly, this period from April to August represent the growing season of the Sphagnum mosses (Clymo, 1970). High CO 2 consumption in the well developed and growing Sphagnum bogs, could even have caused a lower than expected CO 2 concentrations difference between well and poorly developed bogs. Why do some Sphagnum species require high CO 2 concentrations? Smolders et al. (2003) showed very poor growth of five Sphagnum species under inundated conditions on strongly humified, black peat. Carbon dioxide concentrations in the water layer remained very low in this situation (<20 µmol L -1 ). The same species, however, developed very 70 Solute transport in Sphagnum dominated bogs

72 well on weakly humified, white peat. In short, low carbon availability in combination with low diffusion rates of CO 2 in water severely reduces CO 2 availability and limitation of Sphagnum growth is very likely to occur. The requirement for high CO 2 availability by Sphagnum can partly be explained by the mechanisms of CO 2 uptake that are determined by the physiological characteristics of both Sphagnum cuspidatum and S. fallax (table 3; figure 7). Both Sphagnum species are characterized by high CO 2 compensation values, the CO 2 concentration at which CO 2 fixation by photosynthesis balances CO 2 loss by respiration. Air-equilibrated water contains a CO 2 concentration of µmol L -1 between 25 and 10 C. The high compensation values of S. cuspidatum and S. fallax imply that under air-saturated conditions no, or extremely limited, net carbon accumulation can occur. In the acidic bog environment, where no reservoir of bicarbonate is present to replace the CO 2 that is taken up, this is especially relevant and Sphagnum growth will not occur when CO 2 is provided exclusively through equilibration with air. The high K m values of and µm CO 2 for S. cuspidatum and S. fallax, respectively, further indicate that even when CO 2 is present at a concentration that is higher than air-saturated, carbon utilization is not optimal. In the investigated Sphagnum species CO 2 concentrations up to µm are still not saturating (figure 7). Like most aquatic plants lacking a carbon concentrating mechanism the kinetic properties of CO 2 uptake of the investigated Sphagnum species indicate an adaptation to a high CO 2 availability (Raven et al., 1998). Under natural conditions the stagnant bog water will result in thick boundary layers and long diffusion path lengths when compared to the well stirred conditions during the measurements, lowering the apparent affinity for CO 2 concentration even more. Additionally, photosynthetically produced oxygen will accumulate in the boundary layer and because of the competition between CO 2 and O 2 at the site of Rubisco, photosynthetic rate is negatively affected by this increase in O 2 (Bowes & Salvucci, 1989). The importance of diffusion rates on CO 2 availability for Sphagnum has been shown by Baker and Boatman (1985). They showed the ability of Sphagnum cuspidatum to form smaller and thinner leafs under low CO 2 availability. This will reduce the boundary layer resistance and facilitate CO 2 uptake. Paffen & Roelofs (1991) concluded that a dissolved CO 2 concentration of at least 750 µmol L -1 is necessary for the optimal growth of Sphagnum cuspidatum and the subsequent formation of floating vegetation. This was shown by the remarkable low average CO 2 concentrations in the surface water of the poorly developing sites Schurenberg North, Cootjes Veen South and Groote Veen East; 233, 287 and 122 µmol L -1, respectively. At those sites Sphagnum growth was severely hampered. For the successful re-establishment of aquatic Sphagnum species an additional carbon source seems to be crucial. This field study shows that the inflow of carbon-rich groundwater can substitute for the peat layer as a source of CO 2 during the initial stages of bog development. Other explanations for restricted Sphagnum growth in peat ponds Of course, many factors affect the re-establishment of Sphagnum and subsequent bog development (Money et al., 2009). In the majority of the studied bogs, Sphagnum growth appears to be limited by the availability of CO 2. However, there are several interesting exceptions. This is illustrated by the high CO 2 availability in some of the poorly developed bogs (figure 4). Nutrient and light limitation might be responsible for failing Sphagnum re-colonization (e.g. Money & Wheeler, 1999; Smolders et al., 2003). A shortage in nutrients would probably prevent an increase in produc- The importance of groundwater 71

73 tion at enhanced CO 2 concentrations in natural ecosystems (Kramer, 1981). The Diepveen and the Zandveen are two bogs lacking the re-colonization of Sphagnum under conditions of apparently sufficient inorganic carbon (figure 4). However, nutrient limitation seems not to be responsible for this since no clear difference in groundwater or surface water chemistry compared to the other bogs was found (data not shown). Smolders et al. (2003) concluded that the availability of both light and CO 2 have to be sufficient to enable submerged Sphagnum to reach high photosynthetic and growth rates. These factors might indeed affect the Sphagnum development in both Diepveen and Zandveen South in which the water depth is generally several meters in the centre of the bog and on average >50 cm at the edges, very probably hampering Sphagnum growth due to the reduced light availability. Additionally, physical constraints like wind and wave action possibly severely hamper Sphagnum growth in open water bodies (Money et al., 2009). The lack of re-colonization of Sphagnum mosses and hampered growth of already established Sphagnum mosses has often been ascribed to high levels of atmospheric nitrogen deposition (Lamers et al., 2000; Money & Wheeler, 1999; Twenhoven, 1992). The ammonium availability in the surface water at all sampling sites (table 2) reflects the high nitrogen loads as present in the north of the Netherlands, on average 28 kg ha -1 yr -1 (Limpens et al., 2003; RIVM, 2009). However, since no significant differences in nitrogen availability are found between well and poorly developed bogs (data not shown) the current nitrogen deposition is not the determining factor for bog developmental success. Moreover, Tomassen (2004), suggested that bog vitality is much less affected by high nitrogen deposition if other environmental factors, such as water table and the availability of other nutrients (such as CO 2 ), are optimal. Thresholds in the restoration of bogs The present study demonstrates that Sphagnum bogs in the Dwingelderveld are part of the total landscape hydrology instead of being hydrologically distinct entities. This might be the case for many damaged Sphagnum bogs and it implies a landscape approach for successful bog restoration. The current findings clearly show that high CO 2 availability is a pre-requisite for the successful re-establishment of Sphagnum mosses and subsequent bog development. Therefore, CO 2 availability should be included in bog restoration feasibility studies. 72 Solute transport in Sphagnum dominated bogs

74

75 Chapter 6

76 Photosynthesis of three Sphagnum species after acclimatization to high and low carbon dioxide availability Wouter Patberg Jan Erik van der Heide Theo Elzenga

77 Abstract A high CO 2 availability stimulates the growth of both aquatic and emergent Sphagnum species. As shown in the previous chapter, the physiological parameters of CO 2 uptake by Sphagnum also show an adaptation to a high CO 2 availability; a high CO 2 compensation point and a low affinity for CO 2. However, from literature it is known that Sphagnum is able to acclimatize to different CO 2 levels. For example, culturing plants under high CO 2 availability, results in lower photosynthetic rates compared to plants that are grown under CO 2 -limiting conditions. In this chapter, the physiology of carbon uptake by Sphagnum (substrate specificity, affinity and plasticity of carbon assimilation) was determined for three Sphagnum species grown for long periods at high and low CO 2 availability. The CO 2 compensation point and the K m values of the high and low CO 2 grown S. cuspidatum plants indicate that primary production is limited under air-equilibrated conditions. Remarkably, in S. cuspidatum the low CO 2 treated plants were capable of higher photosynthetic rates compared to the high CO 2 treated plants at similar, high CO 2 concentrations. This difference was not found for S. fallax and S. magellanicum. Possibly, this reflects the difference in habitat: S. cuspidatum is a submerged aquatic species, while S. fallax and S. magellanicum are both emergent species. Considering the high CO 2 compensation point, the low affinity for CO 2, the absence of a carbon concentrating mechanism and the limited morphological and physiological plasticity of the plants when exposed to a low external CO 2 concentration, primary production by Sphagnum is expected to be extremely low when solely supplied with atmospheric CO 2. This agrees with our findings in Chapter 5 that when an organic layer is lacking, i.e. during the initial stages of bog development, an alternative external CO 2 source seems to be essential for the successful (re-)establishment of Sphagnum. 76 Solute transport in Sphagnum dominated bogs

78 Introduction Bogs ecosystems are very wet and acidic and are often dominated by mosses of the genus Sphagnum (Clymo & Hayward, 1982). The decomposition of organic material in bogs is slower than the photosynthetic fixation of CO 2, resulting in the accumulation of peat (Clymo et al., 1998). Since Sphagnum bogs function as carbon sink they play an important role in global carbon cycling (Bridgham et al., 2001a; Gorham, 1991). In contrast to vascular plants Sphagnum mosses lack a cuticle and stomates to regulate photosynthesis (Proctor, 2008). Sphagnum mosses are surrounded by an external water film through which gas exchange for photosynthesis is taking place. Since the diffusion of CO 2 is about 10 4 times lower in water than in air, the diffusional barrier formed by the external water films reduces the supply of CO 2 to the carbon assimilating cells and, consequently, the photosynthetic rate (Bowes & Salvucci, 1989; Rice & Giles, 1996; Silvola, 1990; Williams & Flanagan, 1996). The photosynthetic rate of Sphagnum mosses has been shown to be a compromise between external water content and the availability of CO 2 (Schipperges & Rydin, 1998; Silvola, 1990; Titus et al., 1983). At low water contents, dehydration inhibits photosynthesis whereas at very high water contents Sphagnum species may suffer from carbon limitation due to very thick boundary layers (Jauhiainen & Silvola, 1999; Rice & Giles, 1996; Silvola, 1990; Titus et al., 1983; Williams & Flanagan, 1996). For Sphagna, a morphological difference between aquatic and non-aquatic species was demonstrated by Rice and Schuepp (1995); aquatic Sphagnum species have, compared to non-aquatic taxa, longer and thinner leaves and consequently a thinner boundary layer. The growth of submersed aquatic macrophytes is often limited by CO 2 (Raven et al., 1985; Rice & Schuepp, 1995). To overcome the diffusion barrier many aquatic plant species make use of a carbon concentrating mechanism (CCM), which enhances the accumulation of carbon (Maberly & Madsen, 2002). The mechanism most often found is the utilization of bicarbonate (HCO 3- ) as a carbon source in photosynthesis (Prins & Elzenga, 1989). Most aquatic bryophytes, however, lack such a CCM and are known to be pure CO 2 users (Bain & Proctor, 1980; Raven et al., 1998; Raven et al., 1985). By performing ph drift experiments, Bain and Proctor (1980) demonstrated that Sphagnum cuspidatum is a pure CO 2 user and exclusively depends on diffusion of CO 2 to the site of carbon fixation. Due to the diffusional barrier presented by the water layer surrounding the plants high rates of photosynthesis can only be sustained when the leaves are exposed to high levels of CO 2 (Jauhiainen & Silvola, 1999; Raven et al., 1985; Silvola, 1990). A high CO 2 availability has been shown to stimulate the growth of both aquatic and emergent Sphagnum species (Baker & Boatman, 1990; Jauhiainen & Silvola, 1999; Paffen & Roelofs, 1991; Riis & SandJensen, 1997; Roelofs, 1983; Smolders et al., 2001; Smolders et al., 2003). Chapter 5 describes a field study, which demonstrates the importance of a high CO 2 availability for the successful reestablishment of Sphagnum and subsequent bog development. Despite the obvious importance of a high CO 2 availability for Sphagnum, the physiological background of this apparent high CO 2 requirement of Sphagnum has never been established. The physiological characteristics of aquatic plants lacking a CCM indicate an adaptation to high CO 2 availability: a high CO 2 compensation point concentration and a low affinity for CO 2. With these characteristics it is likely that Sphagnum will be limited by the diffusion of CO 2 under air equilibrated conditions (Raven et al., 1985). Chapter 5 shows the need for high CO 2 availability Photosynthesis of three Sphagnum species 77

79 based on the physiological background of carbon uptake by Sphagnum cuspidatum and S. recurvum. However, the Sphagnum plants used in that experiment were grown under ambient CO 2 conditions. For Sphagnum fuscum, a hummock forming species, acclimation to high CO 2 levels has been shown. Culturing plants under high CO 2 availability results in low photosynthetic rates compared to plants that are grown under CO 2 -limiting conditions (Jauhiainen & Silvola, 1999). In the present study, the physiological background of carbon uptake by Sphagnum (substrate specificity, affinity and plasticity of carbon assimilation) was determined for plants grown for a long period at high and low CO 2 availability. Three Sphagnum species (Sphagnum cuspidatum, S. fallax and S. magellanicum) were grown for four months under high or low CO 2 availability. The Sphagnum species used in this study occupy different ecological niches. Sphagnum fallax and S. magellanicum both are emergent and grow above the water surface, while S. cuspidatum is growing completely submerged. From an evolutionary perspective, emergent species might be adapted to water holding capacity and less to low CO 2 levels (Rice & Schuepp, 1995). The high and low CO 2 S. cuspidatum plants were used to measure the photosynthetic response at different CO 2 concentrations. At CO 2 concentrations close to the saturation level, low CO 2 grown Sphagnum cuspidatum plants exhibited a higher photosynthetic rates compared to the high CO 2 grown plants. At this CO 2 level, the photosynthetic rate of S. fallax and S. magellanicum was measured as well. However, differences in photosynthetic rate between treatments were not observed. In addition, supplemental to the ph drift experiments performed on S. cuspidatum by Bain and Proctor (1980) similar ph drift experiments were carried out to test carbon utilization by S. fallax and S. magellanicum; both species were shown to be pure CO 2 users as well. Materials and methods ph drift experiment For the ph drift experiment Sphagnum fallax (klinggr.) Klinggr. and S. magellanicum Brid. were collected in a small bog in the Dwingelderveld, a nature reserve in the north of the Netherlands (N , E ). The upper two cm of ten plants were incubated in a closed 250 ml Erlenmeyer flask completely filled with ten times diluted artificial rainwater (Smolders et al., 2001) supplemented with 1 mm NaHCO 3. The flasks were kept at 20 C by placing them in a water bath. The solution was continuously and slowly stirred. The flasks were illuminated by a halogen lamp (FL 103, Walz, Effeltrich Germany) with a light intensity of approximately 350 µmol m -2 s -1. The ph of the solution was measured continuously for at least 6 hours using a combined ph electrode with an Ag AgCl internal reference electrode (Cole Parmer Instrument Company, Illinois, USA) in combination with a home made amplifier (input impedance Ohm) and a Campbell Scientific CR10X data logger. The experiment was terminated when no further ph increase could be observed. At the end of the experiment, three 1 ml samples were taken from each Erlenmeyer flask for total inorganic carbon (TIC) measurements. TIC concentration was measured using an infrared gas analyzer (CO 2 analyzer model no. S151, - Qubit Systems Inc., Kingston, ON, Canada). The ratio between HCO 3 and CO 2 in the TIC samples was calculated based on the ph (Prins & Elzenga, 1989). For each species the ph drift experiment was performed in triplicate. Carbon uptake, CO 2 compensation point and half-saturation constant (K m ) for CO 2 uptake by S. fallax and S. magellanicum were calculated according to Maberly and Spence (1983). 78 Solute transport in Sphagnum dominated bogs

80 CO 2 uptake experiment Sphagnum cores were collected in April 2010 from two small bogs in the Dwingelderveld. Three species of Sphagnum were collected, Sphagnum cuspidatum Ehrh. Ex Hoffm., S. fallax (klinggr.) Klinggr. and S. magellanicum Brid.. Per species four homogeneous 10 cm thick sections of 18 cm by 25 cm were cut out of the Sphagnum carpet and gently placed in a glass container (18*25 cm, height 10 cm) and transported to a greenhouse. During transport the water table in the containers was kept at approximately 2 cm below the capitula. In the greenhouse the cores were cut to a depth of 5 cm and placed in plastic nets which, in turn, were mounted in the same glass container. The sides of the containers were covered with black plastic sheets to keep out the light. The containers were filled with artificial rainwater (according to Smolders et al. (2001) except for NH 4 NO 3 of which 100 µmol L -1 was used). The S. cuspidatum cores were placed in the containers with the capitula at the water level, S. fallax with their capitula ~1 cm and S. magellanicum ~3 cm above the water table. Growth was compensated for by lowering the nets with the Sphagnum cores in order to keep the water level equal with respect to the top of the capitula. During the experiment the containers were continuously fed with artificial rainwater (~20 ml hr -1 ) from 40 liter containers by using black norprene tubes (l = 400 mm, 4.8 mm outer and 1.6 mm inner diameter; Saint-Gobain Performance Plastics, Verneret, France) in combination with a peristaltic pump (Masterflex L/S model , Cole Parmer Instrument company). The solution left the container through an overflow located 3 cm below the rim. For each species, two containers were fed by artificial rainwater bubbled with carbon dioxide (Carbon Dioxide 4.0, Linde AG, Munich, Germany) and two containers with airequilibrated artificial rainwater, resulting in a high and low CO 2 treatment, respectively. Vascular plants were removed from the Sphagnum cores on a regular basis. In the greenhouse, natural light was supplemented with high pressure sodium lamps to induce a 14 hour photoperiod. The plants acclimated for two weeks before the treatments started. After four months the photosynthetic rate of the Sphagnum species at different CO 2 concentrations was measured (see below). During the culture period the CO 2 concentration and ph of the pore water in the containers were measured every two months. Water samples were collected using 10 cm long Teflon Rhizons (Eijkelkamp, Agrisearch, Giesbeek, the Netherlands) which were placed diagonally in the middle of the Sphagnum cores. The total inorganic carbon (TIC) concentrations were measured using an Infra Red Gas Analyzer (IRGA; Li-7000 CO 2 / H 2 O analyzer, Li-Cor, Inc., USA); The ph was measured using a Metrohm 780 ph meter (Metrohm, Herisau, Switzerland) together with a combined Metrohm glass electrode (Metrohm ). The CO 2 concentration in the water samples was subsequently calculated based on the ph and the TIC concentration (Prins & Elzenga, 1989). During the growth period the pigment content of the mosses was determined three times with regular intervals. Per container five capitula were randomly collected, pooled and pigments were determined according to Lichtenthaler (1987). CO 2 uptake characteristics were determined by measuring photosynthetic activity (A) at different CO 2 concentrations. Photosynthetic activity was determined by the photosynthetic evolution of oxygen by a capitulum placed in a closed thermostatic cuvette containing 1 ml of measuring buffer (see below) at 18 ± 0.2 C and saturating light conditions (1500 µmol m -2 s -1 ; Hansatech Quantitherm Light meter). The solution in the cuvette was stirred continuously. Per measurement one capitulum was used and each capitulum was used for only one measurements. Large branches were trimmed to fit in the cuvette. Oxygen was measured by a Clark electrode located at the Photosynthesis of three Sphagnum species 79

81 bottom of the cuvette in combination with a millivolt recorder (Kipp and Zonen BD40; Delft, the Netherlands) connected to a Graphtec GL200 midi logger (Graphtec Corp., Yokohama, Japan) on which data were logged every second. The measuring buffer consisted of 10 times diluted artificial rainwater (Smolders et al., 2001) and 20 mm MES set at ph = 4.0 using NaOH. During the initial phase of the experiment, the oxygen concentration in the measuring buffer was in equilibrium with air (21%) which seemed to reduce photosynthetic rate considerably. Therefore, the oxygen concentration was reduced by flushing the measuring buffer with N 2 which resulted in an average (±SD) oxygen concentration in the buffer of 9% (±3). Different CO 2 concentrations were obtained by diluting 20 to 600 µl of 0.1 M NaHCO 3 solution with 35 ml of buffer of which 1.5 ml was injected (Hamilton 2.5 ml syringe) into the cuvette whereof 0.5 ml was instantly retaken for the immediate determination of the CO 2 concentration. Total inorganic carbon (TIC) concentrations were measured using an Infra Red Gas Analyzer (IRGA; Li-7000 Co2/H2O analyzer, Li-Cor, Inc., USA). After usage the fresh weight of the capitula was determined and the plant material was subsequently frozen at -80 C for until determination of the chlorophyll concentration according to Lichtenthaler (1987). Photosynthetic activity is expressed as the amount of oxygen produced (nmol O 2 s -1 ) on a fresh weight (g FW) and chlorophyll (mg) basis. Prior to each measurement residual inorganic carbon from the culture medium, present in the hyaline cells or adhering water was reduced as much as possible by illuminating the capitula with 1000 µmol m -2 s -1 for at least 60 minutes while keeping them in a CO 2 free medium. Statistical analysis Hyperbolic curves were fitted to the carbon dioxide uptake data, using graphing software (Prism version 4.03, 2005; GraphPad Software, Inc., San Diego, CA, USA). Per species t-tests were performed to indicate differences between high and low CO 2 availability on photosynthetic performance by using SPSS for Windows (version , 2007; SPSS Inc., Chicago, IL, USA). Results ph Drift experiment If only free CO 2 is taken up, the final ph value of a ph drift experiment will be between 8 and 10, - whereas in case of HCO 3 uptake the ph can reach a value between 11 and 12 (Bain & Proctor, 1980; - Prins & Elzenga, 1989). So, a distinction between a CO 2 and a HCO 3 user can be made, based on this difference in maximum ph. In the ph drift experiment maximum ph values of 8.1 ± 0.03 and 8.4 ± 0.18 were measured for S. fallax and S. magellanicum, respectively, indicating that these Sphagnum species are not able to utilize - HCO 3 as a carbon source, and are therefore defined as strictly CO 2 users. Figure 1 shows the photosynthetic activity (A) (expressed as relative carbon uptake compared to the maximum carbon uptake at the beginning of the experiment) of Sphagnum fallax and S. magellanicum as a function of CO 2 availability during the ph drift experiment. Hyperbolic curves according to the formula A=V max *[CO 2 ]/ (K m +[CO 2 ]) + c were fitted to the data (see figure 1). From the curves the CO 2 compensation point concentrations (Γ), the affinity for CO 2 (K m(co2) ) and photorespiration (c) were determined, see table Solute transport in Sphagnum dominated bogs

82 Figure 1. Photosynthetic activity as a function of CO 2 concentration of Sphagnum fallax (a) and S. magellanicum (b) during the ph drift experiment. Photosynthetic activity is expressed as the CO 2 uptake relative to the maximum CO 2 uptake at the beginning of the ph drift experiment (in %). Table 1. The affinity for CO 2 (K m ), CO 2 compensation point (Γ) and photorespiration (c) (±SD)for Sphagnum fallax and S. magellanicum determined from the relation between CO 2 availability and photosynthetic activity as shown in figure 1 S. fallax S. magellanicum K m (µmol CO 2 L -1 ) 14.4 ± ± 2.9 Γ (µmol CO 2 L -1 ) 7.2 ± ± 0.9 c 60% ± % ± 3.9 CO 2 uptake experiment CO 2 concentration, ph and chlorophyll content during growth phase Per treatment there were two containers per species. However, per species and treatment plants from only one container were used in the photosynthesis measurements. For these containers the average, minimum and maximum CO 2 concentrations and average ph values measured in the pore water during the growth phase of the experiment are shown in table 2. Photosynthesis of three Sphagnum species 81

83 Table 2. Average, minimal and maximum carbon dioxide concentrations (n=3; in µmol L -1 ) and average ph (n between brackets) in the pore water of the containers during the growth phase of the experiment. + = high and - = low CO 2 treatment [CO 2 ] (µmol L -1 ) S. cuspidatum S. fallax S. magellanicum mean min. max. ph (3) (2) (3) (2) (2) (2) The pigment content of the three Sphagnum species after growth for four months at the two different CO 2 treatments are given in figure 2. Per species and treatment the pigment concentration of five pooled capitula was determined. For all three species the plants grown at the low CO 2 regime contained higher pigment concentrations (chl a, b and carotene) compared to plants grown at high CO 2. This higher pigment content in the low CO 2 treated plants might be indicative of an increased investment in the photosynthetic apparatus under more severe CO 2 limitation. Figure 2. Pigment content of Sphagnum cuspidatum, S. fallax and S. magellanicum grown at high and low CO 2 availability for a period of four months. The white bars represent chlorophyll a, light grey bars chlorophyll b and the dark grey bars carotene, all in mg g FW -1. Per species and pigment type, the left bar represents the high (+) and the right bar the low (-) carbon dioxide treatment. Each bar represents one sample of five pooled capitula. Photosynthetic response to CO2 The photosynthetic rate of Sphagnum cuspidatum grown at high and low CO 2 availability as a function of CO 2 concentration is shown in figure 3. Hyperbolic curves according to the formula A=V max *[CO 2 ]/(K m +[CO 2 ]) + c were fitted to the data. From the curves the CO 2 compensation point concentrations (Γ) were calculated. The maximum photosynthetic rate (V max ), the affinity for CO 2 (K m ) and CO 2 compensation point are presented in Table 3. Based on visual inspection and to reduce the degrees of freedom in the fitting procedure, one K m value was fitted for both treatments. 82 Solute transport in Sphagnum dominated bogs

84 Table 3. CO 2 uptake characteristics (±SE) for Sphagnum cuspidatum grown for three months at high carbon dioxide availability (+) and at low carbon dioxide availability (-). See text for explanation of the parameters + - V max (nmol O 2 s -1 FW -1 ) 2.71 ± ± 0.27 K m (µmol CO 2 L -1 ) ± ± 3.52 Γ (µmol CO 2 L -1 ) c ± ± 0.16 For pure CO 2 users the CO 2 compensation point is the lower limit for net C fixation. The CO 2 compensation point concentrations for the high and low CO 2 grown S. cuspidatum plants are and µmol CO 2 L -1, respectively (table 3). Considering that air-equilibrated water contains a CO 2 concentration of µmol L -1 between 25 and 10 C, the CO 2 compensation point found for S. cuspidatum (both treatments) indicates that primary production is limited by CO 2 diffusion under air-equilibrated conditions. Under air-equilibrated conditions no net growth can be expected, since the minimal carbon gain during the light period is expected to be more than offset by the carbon loss due to respiration during the dark period. The K m value for S. cuspidatum grown at high and low CO 2 availability is µmol L -1 (table 3), indicative for the fixation of CO 2 by Rubisco (lacking a CCM). Consequently, Rubisco is far from being saturated in air-equilibrated conditions and high CO 2 concentrations are needed for optimal photosynthesis. More remarkably is the difference in photosynthetic rate between high and low grown S. cuspidatum at higher CO 2 concentrations (figure 3 and 4). Based on fresh weight (figure 4a) the photosynthetic rate of S. cuspidatum is on average significantly higher in the low CO 2 treatment (1.24 ± 0.23 nmol O 2 s -1 g FW -1 ), than in the high CO 2 treatment (0.75 ± 0,18; t(6)=-3.411, p<0.05). The exact CO 2 concentrations at which photosynthetic rate were measured are mentioned in the accompanying text. The average photosynthetic rates of S. fallax in the high and low CO 2 treatments were 1.47 ± 0.53 and 3.08 ± 1.45, respectively. For S. magellanicum the average photosynthetic rate was 1.66 ± 1.04 for the high and 2.33 ± 1.29 for the low CO 2 treatment. For both S. fallax and S. magellanicum the average rates of photosynthesis were highest in the low CO 2 treatment but no significant differences between treatments were observed; t(6)=-2.343, p=0.058 and t(9)=-0.934, p=0.375, respectively. Photosynthesis of three Sphagnum species 83

85 Figure 3. Net photosynthetic rate of Sphagnum cuspidatum grown at high and low CO 2 availability, measured at different CO 2 concentrations at 18±0.2 C. Photosynthetic rate is expressed as the amount of oxygen released based on fresh weight. The data are fitted to hyperbolic curves of the form A=V max * [CO 2 ] / (K m + [CO 2 ]) + c, where V max is the maximum photosynthetic rate, K m the CO 2 concentration at which half of the maximum rate of photosynthesis is reached and c is a constant. Figure 4. Box plots showing the photosynthetic rate (in nmol O 2 s -1 g FW -1 ) of S. cuspidatum, S. fallax and S. magellanicum based on fresh weight (a) and based on chlorophyll content (b). Plants were grown for four months at a high (+) or low (-) carbon dioxide availability. The average (± SD) carbon dioxide concentrations at which photosynthetic rates were measured were 177 ± 9 µmol L -1 (+) and 150 ± 10 (-) for S. cuspidatum; 151 ± 4 (+) and 153 ± 9 (-) for S. fallax and 158 ± 6 (+) and 159 ± 11 (-) for S. magellanicum. The box plots are composed of minimum, maximum, 25%, 75% quartiles and the median. Significant differences between treatments are indicated by an asterisk (p<0.05). The amount of used capitula is noted between brackets in figure a. 84 Solute transport in Sphagnum dominated bogs

86 Based on chlorophyll content (figure 4b) the photosynthetic rates also show a significant difference between treatments for S. cuspidatum, t(6)=-2.858, p<0.05) with the low CO 2 grown plants giving a higher rate (26.57 ± 4.16 nmol O 2 s -1 mg Chl -1 ) than the high CO 2 plants (16.77 ± 5.46). For S. fallax and S. magellanicum the differences in photosynthetic rate between treatments differed not significantly, t(6)=-0.516, p=0.624 and t(9)=1.467, p=0.177, respectively. For S. fallax the photosynthetic rates based on chlorophyll content were on average lower in the high CO 2 treatment than the average rate measured in the low CO 2 treated plants; ± and ± 33.25, respectively. On the other hand, the average photosynthetic rates of S. magellanicum based on chlorophyll content were highest in the high CO 2 treatment, ± compared to ± in the low CO 2 treatment. Discussion Kinetic properties of CO 2 uptake by Sphagnum Like most aquatic bryophytes, Sphagnum has been shown to be a pure CO 2 user. For S. cuspidatum this was shown by Bain and Proctor (1980) and the results presented here show it for S. fallax and S. magellanicum. CO 2 uptake by S. cuspidatum, as shown here, is characterized by a high CO 2 compensation point and a K m value consistent with Rubisco being the primary CO 2 fixing enzyme and with the absence of a carbon concentrating mechanism (table 1 and 2) (Raven et al., 1985). For submerged aquatic macrophytes a low affinity for CO 2 is common and K m values are usually in the range of µm CO 2 (Bowes & Salvucci, 1989; Raven et al., 1985). The K m values for S. cuspidatum, S. fallax and S. magellanicum found in this study, are lower (11.6, 14.4 and 24.5 µmol L -1, resp.) and very likely a consequence of the low oxygen levels used and the vigorous stirring during the photosynthesis measurements. Due to the competition of CO 2 and O 2 at the site of Rubisco, K m values and CO 2 compensation points are higher with increasing O 2 concentrations. Therefore, the presented K m values are very likely lower compared to K m values at ambient oxygen levels. This is supported by the higher calculated K m values (133.2 and µmol L -1 for S. cuspidatum and S. fallax, resp.) obtained from the CO 2 response curve in Chapter 5, which were determined at ambient O 2 levels. The difference in K m values with differing O 2 concentration stresses the influence of O 2 availability on CO 2 uptake. This is especially of interest when regarding the high O 2 concentrations in the upper Sphagnum layer due to photosynthetic activity (Lloyd et al., 1998). In the stagnant bog water, the boundary layers will be substantially thicker, compared to the well-stirred conditions in the presented experiments, reducing the affinity for CO 2 even more. The K m values determined from the ph drift experiments are likely to be an underestimation, compared to the values in the natural situation, due to the reservoir of HCO 3-, present at high ph, that can replenish the CO 2 that is taken up and which is not present in the acidic bog situation. Despite the variability in K m and compensation point due to O 2 availability (and other environmental conditions as light and temperature (e.g. Maberly & Spence, 1983)), these kinetic properties are suggestive of optimization for operation at high CO 2 concentrations at the active site of Rubisco, a condition that is not being met under air-equilibrated conditions. The CO 2 acquisition characteristics are typical for an aquatic plant lacking a CCM (Raven et al., 1985). Photosynthesis of three Sphagnum species 85

87 Plasticity of CO 2 assimilation characteristics Main focus of this chapter, however, is the difference in CO 2 uptake characteristics between Sphagnum plants subjected to different CO 2 concentrations. In principle, for a submerged plant that can only utilize CO 2 there are three possible strategies to increase carbon fixation rates at external concentrations of CO 2 that are slightly higher than the compensation concentration: 1. Increase in the affinity of the primary CO 2 fixing enzyme (lower K m(co2) ); 2. Increase in the photosynthetic capacity (higher V max ) and 3. Decrease in dark respiration (less negative c). In figure 5 the effects of these three strategies on net photosynthesis at low external CO 2 concentration are illustrated. Figure 5. Visualization of the photosynthetic response as a function of CO 2 concentration for the three possible strategies that can be applied by a submerged Sphagnum plant, in order to increase carbon fixation rates at low external CO 2 concentrations. The strategies are: (i) increasing the affinity of the primary CO 2 fixing enzyme (lower K m ); (ii) increasing the photosynthetic capacity (higher V max ) and (iii) decreasing dark respiration (less photorespiration). The strategies curves are plotted relative to a control curve, using arbitrary units on both axes. At air-equilibrated conditions, water contains a CO 2 concentration of µmol L -1 between 25 and 10 C which is indicated by the grey box. Between treatments, no (significant) difference in CO 2 affinity (K m ) and compensation point (Γ) by S. cuspidatum were observed. Increased affinity and lowering the CO 2 compensation point are strategies shown by micro-organisms and submersed angiosperms to gain more carbon in response to a decreased DIC availability (Bowes, 1996). Despite the fact that Sphagnum lacks a biophysical and biochemical mechanism to increase the concentration of C at the site of fixation by Rubisco, S. cuspidatum shows some adaptation to CO 2 availability. However, our results indicate that S. cuspidatum does increase its photosynthetic capacity under CO 2 -limiting conditions, possibly to increase carbon fixation at low external CO 2 availability. This is shown by the difference in V max between S. cuspidatum grown in the high and the low CO 2 treatments; the low CO 2 treated plants were capable of higher photosynthetic rates compared to the high CO 2 plants at similar CO 2 concentrations. This might be caused by an increase in the chlorophyll concentration under CO 2 limiting conditions 86 Solute transport in Sphagnum dominated bogs

88 compared to high CO 2 availability. This was shown for all three Sphagnum species (figure 2). The higher concentrations of chlorophyll probably allows the mosses to gain more CO 2 (Rice, 1995) and thus lead to an increase in photosynthetic performance. For S. magellanicum this inverse correlation between chlorophyll content and CO 2 availability was already shown by Smolders et al. (2001). Jauhiainen and Silvola (1999) showed a reduced photosynthetic efficiency for Sphagnum fuscum when grown under high CO 2 availability, compared to plants grown under low CO 2 conditions. Our findings are in line with the general pattern to down-regulate photosynthetic performance at high carbon availability (e.g. Maberly & Madsen, 2002). On the other hand, increased dry weight by the production of non-structural carbohydrates under high CO 2 levels (Van Der Heijden et al., 2000) might cause the lower photosynthetic rate based on weight (figure 4). However, these increases are small on a dry mass base and therefore not taken into account in this study. Possible morphological adaptation to increased CO 2 uptake Neither S. fallax, nor S. magellanicum were able to enhance photosynthetic performance in comparison with the high CO 2 grown plants, despite CO 2 limitation (indicated by an increased chlorophyll content in the low CO 2 treatment). An explanation could be found in the morphological differences between the Sphagnum species. A mechanism to increase photosynthetic performance under low CO 2 conditions is to enhance the supply of CO 2 by reducing the boundary layer thickness. Aquatic plant species commonly change morphology in order to reduce boundary layer resistance under CO 2 -limited conditions (Bowes & Salvucci, 1989; Rice & Schuepp, 1995). Sphagnum cuspidatum was shown to be able to form thinner leaves when grown submerged at low CO 2 availability (Baker & Boatman, 1985; Rydin & McDonald, 1985). In the present study, leaf morphology was not determined. However, since S. cuspidatum was grown submerged, a morphological adaptation of S. cuspidatum to low CO 2 levels very likely contributes to the significant higher photosynthetic rate of plants grown under low CO 2 levels compared to the high CO 2 plants. When growing emergent, like S. fallax and S. magellanicum in this experiment, the formation of thinner leaves is not to be expected, since this will not results in a reduced boundary layer thickness when growing under low CO 2 availability. Moreover, for emergent growing Sphagnum plants, plant morphology must compromise the possible conflicting requirements of water holding capacity (and conduction) and free gas exchange for photosynthesis (Proctor, 2008). Since the formation of thinner leaves will result in a decreased water holding capacity, this morphological adaptation is not to be expected in S. fallax and S. magellanicum. However, when based on chlorophyll content, the photosynthetic rate of S. fallax and S. magellanicum grown at low CO 2 availability were on average lower than the rate of the high CO 2 plants, which is opposite to the differences between treatments when based on fresh weight (figure 3), indicative for an increased photosynthetic performance at low CO 2 availability probably due to the increased investment in the photosynthetic apparatus. There is a substantial difference between species, and to a lesser extent between treatments, in the variation of the photosynthetic rate (figure 4); S. fallax and S. magellanicum show a variation of at least 300% whereas S. cuspidatum exhibits a more narrow range. Sphagnum fallax and S. magellanicum grown under low CO 2 levels show a greater variation in photosynthetic rate than the plants grown under high CO 2 (figure 4). Due to their emergent growth S. fallax and S. magellanicum were not directly in contact with the CO 2 concentrations in the water, possibly leading to varying CO 2 availability throughout the containers and consequently less uniform adaptations by the plants. Photosynthesis of three Sphagnum species 87

89 Ecological consequences of low affinity CO 2 uptake Because of the competition between CO 2 and O 2 at the site of Rubisco, photosynthetic rate, and consequently K m and the compensation point, is affected by the ratio between CO 2 and O 2 (Bowes & Salvucci, 1989; Maberly & Spence, 1983). For Sphagnum this was shown by Skre & Oechel (1981) and here it became evident during the set up of this experiment; in an air-equilibrated measuring buffer containing 21% O 2 the photosynthetic rate of Sphagnum cuspidatum was severely inhibited at a CO 2 concentration of 20 µmol L -1. Reduced O 2 levels resulted in an increased photosynthetic performance. Due to the production of O 2 and the uptake of CO 2 as a result of photosynthetic activity in combination with the low gas diffusion rates in water, the CO 2 /O 2 ratios are very likely to decrease rapidly having negative consequences for photosynthetic performance. Methanotrophic bacteria, globally occurring in symbiosis with Sphagnum mosses (Kip et al., 2010; Larmola et al., 2010), have been shown to be of great importance by providing Sphagnum with substantial amounts of CO 2 by oxidizing CH 4 using oxygen derived from photosynthesis (Raghoebarsing et al., 2005). We believe that the consumption of oxygen by these methanotrophs, which reduces the O 2 concentration in the close vicinity of the photosynthesizing cells, is at least as important for the successful growth of aquatic Sphagnum species. Considering the low affinity for CO 2, the absence of a carbon concentrating mechanism and the limited morphological and physiological reactions of the plants to low external CO 2 concentration, primary production by Sphagnum is expected to be extremely low when solely supplied with atmospheric CO 2. Growth of Sphagnum therefore requires an additional CO 2 source. Due to aerobic decomposition processes in the peat layer, additional CO 2 will be produced in Sphagnum bogs (Bridgham & Richardson, 1992; Glatzel et al., 2004; Lamers et al., 1999; Smolders et al., 2001; Waddington et al., 2001) and as a consequence pore water CO 2 concentrations can reach up to several millimoles (Smolders et al., 2001; Chapter 5), thereby ameliorating the effects of low diffusion rates (Maberly & Madsen, 2002; Silvola, 1990). This agrees with our findings in Chapter 5 that when an organic layer is lacking, i.e. during the initial stages of bog development, an alternative external CO 2 source seems to be essential for the successful (re-)establishment of Sphagnum. 88 Solute transport in Sphagnum dominated bogs

90

91 Chapter 7

92 Summary and synthesis

93

94 The importance of buoyancy-driven water flow in Sphagnum dominated bogs The requirement of nutrient transport The motivation for this thesis is based on the findings by Baaijens (1982) and Rappoldt et al. (2003). They report on a phenomenon called buoyancy-driven water flow, which is the occurrence of convective flow in water-saturated Sphagnum layers when the temperature difference between day and night is sufficiently large. During the night, the surface of the peat moss layer cools and results in a relatively denser and colder water layer on top of warm water. When the density difference become large enough the cold water in the top layer sinks and warm water rises. It was hypothesized that in this flow of water, solutes will be transported as well. Therefore, buoyancy-driven water flow was proposed as a newly discovered mechanism for the translocation of nutrients in a Sphagnum dominated peat bog. In bogs, the mineralization of organic matter has been shown to be the most important nutrient source for Sphagnum (Aerts et al., 1999; Aldous, 2002a, b; Bowden, 1987; Bridgham, 2002; Damman, 1978, 1986; Morris, 1991; Pakarinen, 1978; Rosswall & Granhall, 1980; Urban & Eisenreich, 1988). In contrast, the highest metabolic activity and nutrient uptake takes place in the capitula (Aldous, 2002a; Johansson & Linder, 1980; Malmer, 1988; Malmer et al., 1994; Robroek et al., 2009; Rydin & Jeglum, 2006). The spatial distinction between mineralization and capitula requires an efficient nutrient transport system. Diffusion, internal transport and capillary transport were the known nutrient transport mechanisms in Sphagnum bogs. Complementary to these mechanisms, buoyancy-driven water flow was hypothesized to be a possible external nutrient transport mechanism, redistributing nutrients from lower Sphagnum layers to the capitula, and vice versa. Evidence for the development of buoyancy-driven water flow in a water-saturated Sphagnum layer was provided, based on theoretical and experimental grounds, by Rappoldt et al. (2003). However, direct evidence for the transport of solutes was lacking and the importance of buoyancy flow in nutrient transport in Sphagnum bogs remained unclear. This thesis provides direct evidence for the transport of solutes by buoyancy flow. Moreover, it is demonstrated that buoyancy flow transports nutrients in such quantities that it, relative to other transport mechanisms, plays an important role in the redistribution of nutrients in a water-saturated Sphagnum layer. The transport of solutes by buoyancy flow The mesocosm experiment in Chapter 2 unequivocally demonstrates the transport of solutes by buoyancy-driven water flow in a water-saturated Sphagnum matrix. Moreover, the experiment shows that due to buoyancy flow a reversal of the gradient can take place in a relatively short period of time. The findings of this mesocosm experiment indicate that buoyancy-driven water flow acts as an efficient external nutrient transport mechanism in water-saturated Sphagnum habitats and thereby can contribute to the supply of nutrients to the Sphagnum capitula in the upper bog layer and the recycling of nutrients. The uptake capacity of ammonium by the capitula In a water-saturated Sphagnum layer, a stepwise increase of solutes near the capitula can be Summary and synthesis 93

95 induced due to the reversal of the gradient by buoyancy flow (Chapter 2, figure 1). The importance of buoyancy flow as a nutrient transport mechanism in supplying the capitula is also determined by its ability to absorb and take up pulses of high nutrient concentrations. Sphagnum has been shown to be opportunistic in its N uptake (Twenhoven, 1992; Woodin et al., 1985). The strong cation exchange capacity of the cell wall of Sphagnum is often regarded as an efficient mechanism to retain cations when supplied by rain water (Bates, 1992; Buscher et al., 1990). Ammonium, the dominant form of nitrogen in bog water, has been shown to be retained very efficiently by Sphagnum (Li & Vitt, 1997; Wiedermann et al., 2009; Williams et al., 1999; Jauhiainen et al., 1998; Twenhoven, 1992). Additionally, the very short lag phase of the substrate inducible enzyme nitrate reductase enables Sphagnum assimilating even short pulses of nitrogen (Woodin et al., 1985). Moreover, Sphagnum is very well able to deal with pulse-wise supply of N by the accumulation of a surplus of N in N-rich amino acids like arginine and asparagine (Baxter et al., 1992; Karsisto, 1996; Limpens & Berendse, 2003; Nordin & Gunnarsson, 2000). The observed uptake kinetics for ammonium by Sphagnum cuspidatum and S. fallax (Chapter 2) very well fit the opportunistic nitrogen uptake characteristics. Ammonium uptake is not saturated by concentrations up to 100 µmol L -1 (Chapter 2, figure 2). The time over which the uptake rates can be maintained also determines the ability of Sphagnum to benefit from the high nutrient availability caused by buoyancy flow. In a separate experiment the time dependence of uptake was + determined for NH 4 (figure 1). It was shown that, when exposed to NH 4+, for 190 hours, the capitula of S. cuspidatum and S. fallax take up 17 and 13% of the final value within one hour and 76 and 64% within 24 hours, respectively. Together, the uptake characteristics enable Sphagnum to benefit from a stepwise increase in ammonium (or cation) availability which is the case after precipitation and buoyancy flow events. Figure 1. The increase of 15 N in capitula of S. cuspidatum (open symbols) and S. fallax (filled symbols; in µmol g DW -1 ) over time when incubated in 100 µmol 15 NH 4 Cl L -1. Each symbol represents the average of three capitula, which were incubated in a square Petri-dish filled with 50 ml of experimental solution (see materials and methods section in Chapter 4) containing the labeled nitrogen, 20 mm MES (ph=4.0) and 100 times diluted artificial rainwater. Error bars represent standard deviations. The experiment took place in a climate controlled room at 18±1 C and a 16L:8D photoperiod and a light intensity of 185 µmol m -2 s -1 ). 94 Solute transport in Sphagnum dominated bogs

96 In Chapter 2 the role of the cation exchange sites in ammonium uptake was demonstrated as well. The adsorption of ammonium by the cell wall is most important at lower concentrations. With increasing concentrations the relative importance of adsorption to total uptake decreases; the cell wall will saturate and increased uptake will take place by intracellular uptake. Compared to the cation exchange, the active intracellular uptake of ammonium is a slow process. These observations support the general assumption of the cell wall functioning as a temporal extension of nutrient availability for intracellular uptake (Buscher et al., 1990; Clymo, 1963; Hajek & Adamec, 2009; Jauhiainen et al., 1998). The internal transport of nitrogen The relative importance of buoyancy flow should be weighed against the contribution of the other transport mechanisms occurring in a water-saturated Sphagnum layer: diffusion and internal transport. The mesocosm experiment demonstrated that the transport of solutes by buoyancy flow can be much faster than is possible by diffusion alone. Moreover, buoyancy flow can transport more solutes upwards than is possible by diffusion. In Chapter 4 the internal transport rate of nitrogen in Sphagnum was determined. The experiments were performed using two Sphagnum species, S. cuspidatum and S. fallax, occupying respectively hollows and pools, both wet habitats where buoyancy flow is likely to occur. Until this present study it was generally assumed that nitrogen was translocated by an internal transport mechanism, but direct evidence for such a mechanism was lacking. In Chapter 4 physiological evidence for the internal acropetal transport of nitrogen in Sphagnum is provided. The findings in Chapter 4 are indicative for symplastic transport of nitrogen, which is in line with the findings of Ligrone and Duckett (1998b) and Rydin & Clymo (1989) who demonstrated cellular specializations in Sphagnum for symplasmic transport. No basipetal transport of nitrogen was observed. Therefore, it seems to be a mechanism that supplies the capitulum with nitrogen and thereby contributes to the efficient use of N. The amount of N transported internally to the capitula is low compared to the amounts potentially transported upwards by buoyancy flow and, subsequently, taken up by the capitula. The uptake kinetics of the capitula show a much faster uptake rate for ammonium than can be supplied by internal transport (Chapter 2, figure 2). When exposed to 25 µmol 15 + NH 4 L -1 the uptake by the capitula of S. cuspidatum is 5.6 ± 2.1 µmol g DW -1 hr -1, whereas the transport of this amount by internal transport takes at least four days. Therefore, with the regular occurrence of buoyancy flow, the supply of nitrogen by internal transport will be insignificant, compared to the supply of nitrogen by buoyancy flow. Thus, in comparison with diffusion and internal transport, buoyancy flow seems to be a quantitatively important nutrient transport mechanism in a water-saturated Sphagnum habitats. Buoyancy flow is restricted to water-saturated Sphagnum habitats and, because of its dependence on varying physical parameters (i.e. the difference in temperature between day and night) an irregularly occurring phenomenon. The importance of internal transport might therefore reside in its continuous character, supplying the capitulum slowly, but steadily, with N and contrasting with the pulsed supply of N by buoyancy flow and atmospheric deposition. Moreover, during extracellular transport nutrients may be lost to microorganism or vascular plant roots. For Sphagnum species that form hummocks that extend above the water surface and do not Summary and synthesis 95

97 benefit from buoyancy flow, internal transport is, next to capillary transport, a possible acropetal pathway for nutrients. Clymo (1973) estimated the average velocity by capillary flow to be 0.4 mm min -1. However, this rate, and the concomitant nutrient transport, is dependent on several external factors, like evaporation, plant density and pore water nutrient concentrations (Clymo & Hayward, 1982). Based on the uptake by the stems and internal transport to the capitula of ammonium by S. cuspidatum, the rate by which nitrogen is transported internally was estimated at 5 mm day -1, which is in accordance with a half time value of equilibration between the stem and capitula of 17 days. Compared to external transport mechanism (buoyancy flow and capillary transport), internal transport is very likely of limited importance for the upward transport of externally supplied N to the capitula. The main function of internal transport is therefore assumed to be the reallocation of internally broken down N. Since the internal transport of nitrogen is a mechanism for efficient nitrogen use, the transport rate is expected to be reduced under high N availability (Bragazza et al., 2004). The transport rate of solutes by buoyancy-driven water flow is independent from the internal concentration, thus also supplying the capitula with nitrogen (and other nutrients) when there is a low demand. The assimilation of N in amino acids by Sphagnum enables the Sphagnum plants to take up the N and store it for later use. With the regular occurrence of buoyancy flow, and thereby the supply of N to the capitula, the sink strength for N of the capitula will be reduced and consequently the rate of internal transport as well. Thus, the relatively low contribution of internal transport in the transport of N in a water-saturated Sphagnum layer will very likely be reduced even more in the presence of buoyancy flow. Ecological importance of buoyancy flow Chapter 3 clearly shows the regular occurrence of buoyancy-driven water flow in a Sphagnum pool in a field situation. Moreover, the theoretical models of Rappoldt et al. (2003) on the development of buoyancy flow can be applied to the field situation. Based on the Ra numbers, calculated from the vertical hydraulic conductivity of the Sphagnum cores and the difference in temperature between day and night, the occurrence and starting times of buoyancy flow development was very well predictable. The results from the GIS study indicate that many peatlands throughout the world are subjected, several days each month during the growth season, to temperature difference between day and night, which are suitable for the development of buoyancy flow. Sphagnum bogs can consist of a patchwork of hollows, lawns and hummocks. As buoyancy flow is restricted to the water layer in a Sphagnum bog, direct supply of nutrients from deeper layers to the capitulum by buoyancy flow only takes place in hollows. The transport of nitrogen by buoyancy-driven water flow and the subsequent uptake by the capitula in the upper Sphagnum layer in a field situation was demonstrated by the field experiment performed in the Rancho Hambre bog complex, Argentina (Chapter 2; figure 2). The increase in 15 N concentration in the S. fimbriatum capitula in the treatment with unobstructed convective flow, indicates the upward transport of 15 N by buoyancy flow. As expected, in the S. magellanicum sites, however, no increase in 15 N in the capitula was observed. Moreover, in the observed S. magellanicum lawns (Chapter 3) the water table is located about 20 cm below the top of the Sphagnum plants which form an insulating layer (Van der Molen & Wijm- 96 Solute transport in Sphagnum dominated bogs

98 stra, 1994), preventing the development of a cool water surface layer and instability in the water column. If buoyancy flow would nevertheless occur, solutes transported from deeper layers to the upper water layer would still have to be transported to the capitula by capillary transport (or by diffusion in case of, for example, CO 2 ). In this case buoyancy flow only acts as an auxiliary transport mechanism and its relative importance in the nutrient supply to the capitula is determined by the height of the capitula above the water level. Such a situation could be found in Sphagnum lawns and the transition zones between pools and hummocks. As the initial successional stage of a Sphagnum bog is the colonization of aquatic Sphagnum species of water bodies followed by the invasion of hummock forming species, buoyancy flow seems to be particularly important in the early stages of bog development. An overview of the (expected) relative importance of buoyancy flow, diffusion, internal transport and capillary transport in three different Sphagnum habitats (hollow, lawn and hummock) are presented in figure 2. Figure 2. A schematic cross section of three different Sphagnum habitats (hollows, lawns and hummocks) indicating the relative importance (in %) of buoyancy flow (light grey), diffusion (dotted grey), internal transport (dark grey) and capillary transport (medium grey) in the redistribution and cycling of nutrients. The three habitats differ in the height of the capitula (indicated by the asterisks) relative to the water table level (indicated by the thick dashed horizontal lines). In hollows, buoyancy flow is the major mechanism by which nutrients are transported. Since the capitula are at the water level in hollows, buoyancy flow can directly contribute to the supply of nutrients to the capitula. Diffusion and internal transport take place as well in hollows but has been shown to be less rapid and effective than buoyancy flow. Even though buoyancy flow might occur irregularly, the opportunistic nutrient uptake capacity of Sphagnum will result in a significant contribution of buoyancy flow even at a few occurrences of buoyancy flow. Since buoyancy flow is restricted to the water layer its contribution will be less in lawns and very likely totally absent in hummocks. In contrast, capillary transport only takes place above the water table and therefore only contributes to nutrient transport in lawns and hummocks. Since the plant density is higher in hummocks than in lawns, capillary transport will be faster and consequently more important in hummocks than in lawn. Moreover, in lawns buoyancy flow might occur, reducing the relative contribution of capillary transport in lawns compared to hummocks. With increasing heights of the capitula above the water table the relative importance of diffusion will be less and completely reduced to zero in hummocks. Internal transport has been shown to be a slow mechanism. As a consequence, internal transport very likely plays a minor role in the distribution of nutrients in especially hollows and hummocks because of the presence of the faster mechanisms buoyancy flow and capillary transport, respectively. In lawns the relative importance of internal transport is expected to be greatest since buoyancy flow is less effective and capillary transport will be less fast due to the low plant density in these habitats. Summary and synthesis 97

99 Nitrogen The main N source for Sphagnum has been shown to be re-mineralized N (Aerts et al., 1999; Aldous, 2002b; Bridgham, 2002; Gerdol et al., 2006; Morris, 1991; Urban & Eisenreich, 1988). The importance of re-mineralization of nitrogen for Sphagnum growth has been demonstrated in situ by Urban & Eisenreich (1988). They calculated the assimilation of nitrogen by plants (primarily Sphagnum) to be 66 kg ha -1 yr -1, whereas only 14.6 kg N ha -1 yr -1 was supplied by total inputs. The remainder was supplied by mineralization of the peat. Maybe one would expect that vascular plants may be better competitors for this source of nitrogen by scavenging the peat for mineralized N with their fine roots (Backeus, 1990; Jackson et al., 1990) and as a consequence vascular plants will outcompete Sphagnum mosses. This is, however, not the case. Instead, Sphagnum is capable of very efficient (re-)use of mineralized nitrogen, creating a low nutrient environment which consequently contributes to their dominance over vascular plants. As already mentioned by Gerdol et al. (2006), these findings are contradictory to the general idea that Sphagnum and vascular plants utilize spatially distinct nutrient pools, with Sphagnum relying on N from precipitation and vascular plants on mineralization of senescing organic matter in the deeper acrotelm (Malmer et al., 1994; Pastor et al., 2002). The importance of nutrient transport for efficient nutrient recycling is generally accepted. For example, Aldous (2002b) mentions the translocation of nitrogen as a key process in bog nutrient cycling. Blodeau et al. (2006) states that Sphagnum mosses are the dominant species in northern peatlands, in part because they have the capability to conserve nitrogen by transferring it from lower, inactive parts of their stem to apices where new biomass is formed (Aldous, 2002a, b; Malmer, 1988). However, the contribution of the different nutrient transport mechanisms in nutrient recycling was never determined. In this thesis we demonstrate that buoyancy-driven water flow is an important mechanism contributing to the recycling of mineralized nutrients in Sphagnum bogs. Consequently, Sphagnum mosses may outcompete vascular plants more easily and thereby enhance their ability to engineer the ecosystem (Van Breemen, 1995). Carbon dioxide In waterlogged Sphagnum habitats, the stagnant bog water will result in thick boundary layers and long diffusion path, lengths which reduces the supply of CO 2 to the carbon assimilating cells and, consequently, the photosynthetic rate (Bowes & Salvucci, 1989; Rice & Giles, 1996; Silvola, 1990; Williams & Flanagan, 1996). Consequently, high rates of underwater photosynthesis can only be sustained when the leaves are exposed to high levels of CO 2 (Jauhiainen & Silvola, 1999; Paffen & Roelofs, 1991; Silvola, 1990; Smolders et al., 2003). Submerged Sphagnum species that inhabit peat hollows have been shown to be limited by CO 2 (Chapter 5; Rice & Giles, 1996; Rice & Schuepp, 1995). On the other hand, high carbon dioxide concentrations haven been shown to stimulate Sphagnum growth (e.g. Smolders et al., 2001; Chapter 5). However, doubling atmospheric CO 2 concentrations appear to have rather limited effects on the growth of Sphagnum (Heijmans et al., 2001; Hoosbeek et al., 2001; Jauhiainen et al., 1994; Toet et al., 2006). The high CO 2 requirements for Sphagnum is determined by physiological characteristics of the CO 2 uptake mechanism of Sphagnum, as shown in Chapters 5 and 6. Sphagnum mosses have been shown to be pure CO 2 users (Bain & Proctor, 1980; Chapter 6) and therefore exclusively depend on 98 Solute transport in Sphagnum dominated bogs

100 diffusion of CO 2 to the site of carbon fixation. The kinetic properties of CO 2 uptake indicate an adaptation to a high CO 2 availability. The investigated Sphagnum species are characterized by a high CO 2 compensation value, the CO 2 concentration at which CO 2 fixation by photosynthesis balances CO 2 loss by respiration. Air-equilibrated water contains a CO 2 concentration of µmol L -1 between 25 and 10 C. The high compensation value of Sphagnum implies that under air-saturated conditions no, or extremely limited, net carbon accumulation can occur. In the acidic bog environment, where no reservoir of bicarbonate is present to replace the CO 2 that is taken up, this is especially relevant and Sphagnum growth will not occur when CO 2 is provided exclusively through equilibration with air. Furthermore, The high K m values of and µm CO 2 for S. cuspidatum and S. fallax, respectively, further indicate that even when CO 2 is present at a concentration that is higher than air-saturated, carbon utilization is not optimal. Remarkably, Sphagnum cuspidatum was shown to be able to form thinner leaves when grown submerged at low CO 2 availability (Baker & Boatman, 1985; Rydin & McDonald, 1985), which will reduce the boundary layer resistance and facilitate CO 2 uptake. Moreover, in Chapter 6 it is demonstrated that S. cuspidatum is able to increase its photosynthetic capacity under CO 2 -limiting conditions, possibly to increase carbon fixation at low external CO 2 availability. Very likely due to an increased Rubisco concentration under CO 2 limiting conditions compared to high CO 2 availability. Buoyancy flow will result in net transport when a vertical gradient exists, as for example is the case for nutrients. Uptake and assimilation of CO 2 will result in depletion in the zone of the capitula where most photosynthetic activity takes place. In contrast, in the lower acrotelm CO 2 is released as a consequence of the decomposition of organic material. An increasing CO 2 concentration with depth has been shown in a water-saturated Sphagnum layer (Lloyd et al., 1998). Therefore, buoyancy flow is very likely an important mechanism also in replenishing CO 2 in the upper Sphagnum layer and enhancing photosynthesis. Oxygen Photosynthetic activity in the top layer of the Sphagnum matrix will also result in the production of oxygen. Due to the thick boundary layers and the low diffusion rates of oxygen in water the oxygen will accumulate in the upper Sphagnum layer during the day, resulting in a decreasing gradient with depth (Adema et al., 2006; Lloyd et al., 1998). Lloyd et al. (1998) measured a steep oxygen gradient in the upper four centimeters of a water-saturated Sphagnum layer, decreasing from 300 to 0 µm. The mixing of the water layers by buoyancy flow will result in a net downward transport of oxygen. Adema et al. (2006) attributed a conspicuous change in oxygen concentration at 5 cm depth in a Sphagnum layer to the occurrence of buoyancy flow. Because of the competition between CO 2 and O 2 at the site of Rubisco, photosynthetic rate is negatively affected by the accumulation of O 2 in the surroundings of the capitula (Chapter 6; Bowes & Salvucci, 1989; Raven, 2011; Raven et al., 2008; Skre & Oechel, 1981). Together with the upwards transport of carbon dioxide by buoyancy flow, the CO 2 :O 2 ratio in the upper Sphagnum layer will increase and thereby enabling the mosses to photosynthesize. However, during the day the CO 2 :O 2 ratio decreases again due to photosynthetic activity. Therefore, photosynthetic rates of aquatic Sphagnum species might be highest in the beginning of the day. The downward transport of oxygen very likely will increase decomposition rates since Summary and synthesis 99

101 the aerobic decomposition of organic material is significantly higher than the anaerobic decomposition (Bridgham et al., 1998; Waddington et al., 2001). Consequently, this might result in a positive feedback mechanism in which the concentrations of CO 2 and nutrients like N and P will be increased by the downward supply of oxygen. These nutrients will become available to the growing Sphagnum when transported upwards by buoyancy flow which in turn will stimulate photosynthesis. Consequently, under N-limiting conditions, the transport of oxygen by buoyancy flow might even determine primary production. Moreover, the negative effect of oxygen on photosynthesis might give rise to another important feature of the symbiosis between Sphagnum and methanotrophic bacteria. The consumption of oxygen by the methanotrophs, which reduces the O 2 concentration in the close vicinity of the photosynthesizing cells and thereby enhance photosynthesis. This might at least be as important for the successful growth of aquatic Sphagnum species as the concomitant CO 2 release. Methane Inseparable from the availability and redistribution of O 2 and CO 2 in Sphagnum bogs is the presence of methane. Under more reductive conditions the methane production in the catotelm may become higher than the production of CO 2. This methane can be oxidized by methanotrophic bacteria to CO 2, which then can be used as a carbon source by Sphagnum mosses (Kip et al., 2010; Raghoebarsing et al., 2005). Methane is anaerobically produced in large quantities in bogs (e.g. Gorham, 1991). Nevertheless, emissions of methane to the atmosphere are relatively low (Larmola et al., 2010) due to the activity of methanotropic bacteria. The mixing of methane and photosynthetically produced oxygen by buoyancy flow might have an important role in 1) the CO 2 supply to the capitula and 2) the low methane emission rates, thereby playing a significant role in the global carbon cycling. Other determinants for the importance of buoyancy flow Basically, the nutrient concentration in acrotelmic water will be determined by the decomposition rate in the catotelm and the depletion in the top water layer by uptake and assimilation and dilution by rain water. As discussed above, buoyancy flow has its effect on the distribution of multiple nutrients which mutually interact. Not mentioned are phosphorus and potassium, which have been shown to limit Sphagnum growth (Aerts et al., 1992; Bridgham et al., 1996; Hoosbeek et al., 2002) and the organic peat layer being an important source for these nutrients (Bates, 1992; Damman, 1978, 1986). The effect of buoyancy flow on the nutrient concentration and distribution will be determined by two other factors. First, the depth of the buoyancy cells: the deeper the cells the more mixing takes place which very likely results in a higher nutrient availability. Second, the frequency of buoyancy flow events. Because of its dependence on the difference in temperature between day and night, buoyancy flow is an irregularly occurring mechanism. The more buoyancy flow events occur the more the water layer will be mixed, which results in higher decomposition rates, the amount of transported nutrients and photosynthetic activity (see above). At low frequencies, gradients are allowed to be build up and relatively large amounts of solutes might be transported at once. Consequently, the transport and nutrient availability might not be in synchrony with the requirements of Sphagnum for primary production. For nitrogen this might not pose a problem, since the pulse-wise availability of N is buffered by the opportunistic uptake and 100 Solute transport in Sphagnum dominated bogs

102 assimilation characteristics of Sphagnum (Chapter 2). In contrast, photosynthesis will very likely benefit more from a frequent occurrence of buoyancy flow. Photosynthetic activity will result in lowered CO 2 and increased O 2 levels in the surroundings of the capitula, which in turn will inhibit photosynthesis. Since this inhibition can occur within the period of one day, overall photosynthetic rates will be enhanced optimally with a diurnal occurrence of buoyancy flow. The relative importance of transport of nutrients from deeper water layers to the top layer with the capitula, is higher when atmospheric input is low. This is especially relevant for nitrogen since atmospheric N deposition has increased significantly during recent decades (Vitousek, 1982). An increased N deposition results in higher ammonium concentrations in the bog water (Lamers et al., 2000). Under such conditions the decreasing nitrogen gradient with depth will be reduced or even be completely diminished, thereby reducing the net upward transport of nitrogen by buoyancy flow to zero. Under high N availability, Sphagnum growth has been shown to shift from a nitrogen to a phosphorus limitation (Aerts et al., 1992). Consequently, the importance of buoyancy-driven water flow as a nutrient transport mechanism will also shift from nitrogen to phosphorus. Moreover, the availability of CO 2 might become important under high N loads as well. However, since areas with a high nitrogen deposition load only slightly overlap with the area covered with peatlands (Chapter 3, figure 4), the importance of buoyancy flow in the transport and recycling of nitrogen in Sphagnum bogs, is hardly diminished by increased, anthropogenic N deposition. Restoration and conservation of Sphagnum bogs Throughout the world, Sphagnum bogs have become greatly endangered and consequently much effort is dedicated to the restoration of damaged bogs and the conservation of bog remnants. Therefore, studying the functioning of Sphagnum bogs is inseparable to the restoration and conservation of these ecosystems. Chapter 5 focuses on the restoration of Sphagnum bogs. As mentioned in the previous paragraph, several studies have reported on the stimulation of Sphagnum growth by high CO 2 concentrations (Baker & Boatman, 1990; Jauhiainen & Silvola, 1999; Paffen & Roelofs, 1991; Riis & SandJensen, 1997; Roelofs, 1983; Smolders et al., 2001; Smolders et al., 2003). Chapter 5 and 6 show the physiological background of the high CO 2 needs of Sphagnum. In addition to these studies, Chapter 5 reports on a field study in which the significance of CO 2 availability for the successful re-establishment of Sphagnum and subsequent bog development is demonstrated. Study area was the Dwingelderveld, a nature reserve in the north of the Netherlands characterized by several small damaged Sphagnum bogs throughout the area. After rewetting measures were taken, the developmental success between these bogs varied significantly; some bogs developed well, whereas others did not. It was shown that the poorly developed bogs were limited in CO 2, whereas the successful re-establishment of Sphagnum in the well developed bogs was correlated with high CO 2 availability. Groundwater essential for bog development; a paradox The findings in Chapter 5 clearly show that high CO 2 availability is a pre-requisite for the successful re-establishment of Sphagnum mosses and subsequent bog development. In well developed bogs CO 2 is sufficiently available due to the decomposition of organic material in the peat layer Summary and synthesis 101

103 (Bridgham & Richardson, 1992; Glatzel et al., 2004; Smolders et al., 2001; Waddington et al., 2001). However, damaged bogs subjected to peat cuttings or drainage often lack such a carbon source and an additional CO 2 source is needed for Sphagnum growth. Therefore, one of the conclusions of Chapter 5 is that CO 2 availability should be included in bog restoration feasibility studies. Remarkably, water chemistry analysis revealed that the well developed bogs in the Dwingelderveld (Chapter 5) received carbon rich groundwater from outside the bogs, increasing CO 2 concentrations in the bog stimulating Sphagnum growth and thereby bog development. Interestingly, this dependence of Sphagnum growth on groundwater input hides a paradox; bogs exists due to being isolated from the groundwater. In such an ombrotrophic, low nutrient environment Sphagnum mosses have a competitive advantage over vascular plants. Moreover, groundwater is often characterized by a high ph and a high Ca 2+ concentration and has been shown to be toxic for most Sphagnum species (Skene, 1915; Clymo & Hayward, 1982). The need for carbon under high N loads A large number of studies have focussed on the negative effects of an increased atmospheric nitrogen deposition on the growth of Sphagnum and the functioning of Sphagnum bogs (Limpens et al., 2011). The lack of re-colonization of Sphagnum mosses and hampered growth of already established Sphagnum mosses has often been ascribed to high levels of atmospheric nitrogen deposition (Lamers et al., 2000; Li & Vitt, 1994; Money & Wheeler, 1999; Twenhoven, 1992). Chapter 5 demonstrates that the successful restoration of Sphagnum bogs is possible under high nitrogen loads. It is hypothesized that CO 2 can compensate the negative effects of a high nitrogen deposition on an ecological as a physiological level. Sphagnum mosses lack a regulatory mechanism for nitrogen uptake. As a consequence, internal nitrogen concentration increases with increasing nitrogen deposition rates (Bragazza et al., 2005; Lamers et al., 2000; Limpens et al., 2011). However, under high levels of N deposition levels, Sphagnum is not able to filter out all the N from precipitation and nitrogen leaches to the rhizosphere were it becomes available for vascular plants, thereby making the bog vulnerable to invasions by competitive vascular plants that require a high N supply. The reduced growth of Sphagnum is often attributed to the shading by vascular plants (Berendse et al., 2001; Heijmans et al., 2001; Lamers et al., 2000; Limpens et al., 2011). + After uptake nitrate is reduced to NH 4 prior to assimilation, while ammonium is directly assimilated into glutamine (Rudolph et al., 1993). Subsequently, glutamine is converted into other amino acids (Kahl et al., 1997; Rudolph et al., 1993). With increasing N deposition, plants are no longer N-limited, but will still take up N (Lamers et al., 2000). Continued assimilation of N leads to accumulation of free amino acids (Baxter et al., 1992; Karsisto, 1996; Nordin & Gunnarsson, 2000). Under these conditions a decrease in Sphagnum growth was observed (Baxter et al., 1992; Nordin & Gunnarsson, 2000), possibly the result of the accumulation of amino acids requiring carbon and energy (Baxter et al., 1992; Nordin & Gunnarsson, 2000). Hence, under high N availability an additional need of carbon is very likely. Paffen & Roelofs (1991) showed that high ammonium concentrations in the water layer had no major effects on the growth of submerged growing S. cuspidatum when simultaneously a high concentration of 1000 µmol CO 2 L -1 was applied. High CO 2 levels might enable Sphagnum to increase their competitive strength over vascular plants by enabling continued growth and N assimilation, keeping the nitrogen concentrations low 102 Solute transport in Sphagnum dominated bogs

104 in the pore water and thereby gain competitive strength over vascular plants. The importance of carbon under high N availability is supported by the following findings. + In Sphagnum, under normal conditions, NH 4 is stored in amino acids having relatively high C:N ratios like glutamine (5:2) (Kahl et al., 1997; Nordin & Gunnarsson, 2000; Rudolph et al., 1993). However, under increased nitrogen loads, N is accumulated in amino acids with lower C:N ratios, mostly arginine (Nordin & Gunnarsson, 2000), which has a C:N ratio of 3:2, the lowest all amino acids. This shift in amino acid accumulation suggests an economical use of carbon under high N loads. Thus, the high carbon dioxide needs of Sphagnum are very likely enhanced under high N availability. Moreover, the interaction between CO 2 availability and high levels of atmospheric N deposition might have an effect on nutrient uptake as well. The accumulation of N in N-rich rich amino acids has been shown to be at the expense of C rich amino acid like phenylalanine (Smolders et al., 2001), which is a precursor of cell wall compounds like polymeric uronic acids and phenolic compounds. A study performed by Richter and Dainty (1989) on the cell wall ion exchange capacity of Sphagnum russowii suggests that polymeric uronic acids account for over half the cation exchange capacity (CEC) and phenolic compounds for about 25%. With its high CEC, the cell wall also plays an important role in nutrient uptake (see Chapter 2). As mentioned above, high N loads reduces the amounts of amino acid important in cell wall assimilation and composition. This might result in changes in CEC and consequently in nutrient uptake processes. Additionally, high concentrations + of NH 4 in bog water (as a consequence of high N loads) directly affect cation composition at the cell wall and possibly thereby the nutrient balance in Sphagnum. Concluding remarks Sphagnum mosses are known for their ability to engineer their environment (Van Breemen, 1995). One of these engineering abilities is to maintain low nutrient concentrations in the pore water, preventing increased vascular plant cover and keeping the competitive advantage. Depending on Sphagnum habitat, different transport mechanism play a role in this efficient scavenging for nutrients. In this thesis we demonstrated that buoyancy-driven water flow plays an important role in the distribution of nutrients in Sphagnum bogs. Buoyancy flow might also have an important influence on primary production and decomposition rates by preventing the building up of gradients that have a negative feedback on these processes (N, CO 2 and O 2 ). In this context, the disruption of the nutrient balance in these ecosystems by high N atmospheric deposition loads are easily understood. Buoyancy flow, a worldwide occurring, but poorly studied phenomenon in Sphagnum bogs will have to be taken account when we want to understand bog functioning. Summary and synthesis 103

105

106 References

107

108 Adema, E.B., Baaijens, G.J., van Belle, J., Rappoldt, C., Grootjans, A.P., & Smolders, A.J.P. (2006) Field evidence for buoyancy-driven water flow in a Sphagnum dominated peat bog. Journal Of Hydrology, 327, Aerts, R. (1990) Nutrient use efficiency in evergreen and deciduous species from heathlands. Oecologia, 84, Aerts, R., Wallen, B., & Malmer, N. (1992) Growth-limiting nutrients in Sphagnum-dominated bogs subject to low and high atmospheric nitrogen supply. Journal Of Ecology, 80, Aerts, R., Verhoeven, J.T.A., & Whigham, D.F. (1999) Plant-mediated controls on nutrient cycling in temperate fens and bogs. Ecology, 80, Aldous, A.R. (2002a) Nitrogen translocation in Sphagnum mosses: effects of atmospheric nitrogen deposition. New Phytologist, 156, Aldous, A.R. (2002b) Nitrogen retention by Sphagnum mosses: responses to atmospheric nitrogen deposition and drought. Canadian Journal Of Botany-Revue Canadienne De Botanique, 80, Andrus, R.E., Wagner, D.J., & Titus, J.E. (1983) Vertical zonation of Sphagnum mosses along hummock-hollow gradients. Canadian Journal Of Botany-Revue Canadienne De Botanique, 61, Baaijens, G.J. (1982) Water en levensgemeenschappen. In: Water in Drenthe. Symposiumverslag, Assen: Backeus, I. (1990) Production and depth distribution of fine roots in a boreal open bog. Annales Botanici Fennici, 27, Bain, J.T. & Proctor, M.C.F. (1980) The requirement of aquatic bryophytes for free CO 2 as an inorganic carbon source - some experimental evidence. New Phytologist, 86, Baker, R.G. & Boatman, D.J. (1985) The effect of carbon dioxide on the growth and vegetative reproduction of Sphagnum cuspidatum in aqueous solutions. Journal Of Bryology, 13, 399-&. Baker, R.G.E. & Boatman, D.J. (1990) Some effects of nitrogen, phosphorus, potassium and carbon dioxide concentration on the morphology and vegetative reproduction of Sphagnum cuspidatum Ehrh. New Phytologist, 116, Bates, J.W. (1992) Mineral nutrient acquisition and retention by bryophytes. Journal Of Bryology, 17, Baxter, R., Emes, M.J., & Lee, J.A. (1992) Effects of an experimentally applied increase in ammonium on growth and amino acid metabolism of Sphagnum cuspidatum Ehrh Ex Hoffm from differently polluted areas. New Phytologist, 120, Berendse, F., Van Breemen, N., Rydin, H., Buttler, A., Heijmans, M., Hoosbeek, M.R., Lee, J.A., Mitchell, E., Saarinen, T., Vasander, H., & Wallen, B. (2001) Raised atmospheric CO 2 levels and increased N deposition cause shifts in plant species composition and production in Sphagnum bogs. Global Change Biology, 7, Blodau, C., Basiliko, N., Mayer, B., & Moore, T.R. (2006) The fate of experimentally deposited nitrogen in mesocosms from two Canadian peatlands. Science Of The Total Environment, 364, Bonnett, S.A.F., Ostle, N., & Freeman, C. (2010) Short-term effect of deep shade and enhanced nitrogen supply on Sphagnum capillifolium morphophysiology. Plant Ecology, 207, References 107

109 Boudreau, B.P. (1997) Diagenetic models and their implementation: modeling transport and reactions in aquatic sediments. Springer, Berlin. Bowden, W.B. (1987) The biogeochemistry of nitrogen in fresh water wetlands. Biogeochemistry, 4, Bowes, G. & Salvucci, M.E. (1989) Plasticity in the photosynthetic carbon metabolism of submersed aquatic macrophytes. Aquatic Botany, 34, Bowes, G. (1996). Photosynthetic responses to changing atmospheric carbon dioxide concentration. In: Photosynthesis and the environment (eds N.R. Baker), pp Kluwer Academic Publishers. Bragazza, L., Tahvanainen, T., Kutnar, L., Rydin, H., Limpens, J., Hajek, M., Grosvernier, P., Hajek, T., Hajkova, P., Hansen, I., Iacumin, P., & Gerdol, R. (2004) Nutritional constraints in ombrotrophic Sphagnum plants under increasing atmospheric nitrogen deposition in Europe. New Phytologist, 163, Bragazza, L., Limpens, J., Gerdol, R., Grosvernier, P., Hajek, M., Hajek, T., Hajkova, P., Hansen, I., Iacumin, P., Kutnar, L., Rydin, H., & Tahvanainen, T. (2005) Nitrogen concentration and delta N-15 signature of ombrotrophic Sphagnum mosses at different N deposition levels in Europe. Global Change Biology, 11, Bridgham, S.D. & Richardson, C.J. (1992) Mechanisms controlling soil respiration (CO 2 and CH 4 ) in southern peatlands. Soil Biology & Biochemistry, 24, Bridgham, S.D., Pastor, J., Janssens, J.A., Chapin, C., & Malterer, T.J. (1996) Multiple limiting gradients in peatlands: A call for a new paradigm. Wetlands, 16, Bridgham, S.D., Updegraff, K., & Pastor, J. (1998) Carbon, nitrogen, and phosphorus mineralization in northern wetlands. Ecology, 79, Bridgham, S.D., Ping, C.-L., Richardson, J.L., & Updegraff, K. (2001a). Soils of northern peatlands: Histosoils an gelisoils. In: Wetland soils: genesis, hydrology, landscapes and classification. (eds J.L. Richardson & M.J. Vepraskas), pp Boca Raton, FL, USA: CRC Press. Bridgham, S.D., Updegraff, K., & Pastor, J. (2001b) A comparison of nutrient availability indices along an ombrotrophic-minerotrophic gradient in Minnesota wetlands. Soil Science Society Of America Journal, 65, Bridgham, S.D. (2002) Nitrogen, translocation and Sphagnum mosses. New Phytologist, 156, Brown, D.H. (1982). Mineral nutrition. In: Bryophyte Ecology (eds A.J.E. Smith), pp Chapman & Hall, London. Brown, D.H. & Bates, J.W. (1990) Bryophytes and nutrient cycling. Botanical Journal Of The Linnean Society, 104, Buscher, P., Koedam, N., & Vanspeybroeck, D. (1990) Cation-exchange properties and adaptation to soil acidity in bryophytes. New Phytologist, 115, Chapin, F.S. (1980) The mineral-nutrition of wild plants. Annual Review Of Ecology And Systematics, 11, Clymo, R.S. (1963) Ion exchange in Sphagnum and its relation to bog ecology. Annals Of Botany, 27, 309-&. Clymo, R.S. (1970) Growth of Sphagnum - Methods of measurement. Journal Of Ecology, 58, 13-&. Clymo, R.S. (1973) Growth of Sphagnum - Some effects of environment. Journal Of Ecology, 61, Solute transport in Sphagnum dominated bogs

110 Clymo, R.S. & Hayward, P.M. (1982). The ecology of Sphagnum. In: Bryophyte ecology (eds A.J.E. Smith), pp London: Chapman & Hall. Clymo, R.S. (1984) The limits to peat bog growth. Philosophical Transactions Of The Royal Society Of London Series B-Biological Sciences, 303, Clymo, R.S., Turunen, J., & Tolonen, K. (1998) Carbon accumulation in peatland. Oikos, 81, Damman, A.W.H. (1978) Distribution and movement of elements in ombrotrophic peat bogs. Oikos, 30, Damman, A.W.H. (1986) Hydrology, development, and biogeochemistry of ombrogenous peat bogs with special reference to nutrient relocation in a western Newfoundland bog. Canadian Journal Of Botany-Revue Canadienne De Botanique, 64, Daniels, R.E. & Eddy, A. (1985) Handbook of European Sphagna. Institute of Terrestrial Ecology, Abbots Ripton. Elzenga, J.T.M., Keller, C.P., & Van Volkenburgh, E. (1991) Patch clamping protoplasts from vascular plants - method for the quick isolation of protoplasts having a high success rate of gigaseal formation. Plant Physiology, 97, Engelen, G.B. & Jones, G.P. (1986) Developments in the analysis of groundwater flow systems. IAHS Publications 163. Wallingford. Everts, F.H., Baaijens, G.J., Grootjans, A.P., Verschoor, A.J., & De Vries, N.P.J. (2002). Hoogveenontwikkeling in veentjes en kleinschalige hoogveencomplexen op het Dwingelderveld. Report University of Groningen/ EGG consult Everts & de Vries. University of Groningen. Gerdol, R., Bragazza, L., & Brancaleoni, L. (2006) Microbial nitrogen cycling interacts with exogenous nitrogen supply in affecting growth of Sphagnum papillosum. Environmental And Experimental Botany, 57, 1-8. Geurts, J.J.M., Smolders, A.J.P., Verhoeven, J.T.A., Roelofs, J.G.M., & Lamers, L.P.M. (2008) Sediment Fe:PO 4 ratio as a diagnostic and prognostic tool for the restoration of macrophyte biodiversity in fen waters. Freshwater Biology, 53, Glatzel, S., Basiliko, N., & Moore, T. (2004) Carbon dioxide and methane production potentials of peats from natural, harvested, and restored sites, eastern Quebec, Canada. Wetlands, 24, Gorham, E. (1991) Northern Peatlands - Role in the carbon cycle and probable responses to climatic warming. Ecological Applications, 1, Grootjans, A.P., Adema, E., Baaijens, G.J., Rappoldt, C., & Verschoor, A.J. (2003) Mechanisms behind restoration of small bog ecosystems in a cover sand landscape. Archiv fur Naturschutz und Landschaftsforschung, Grootjans, A.P., Van Diggelen, R., & Bakker, J.P. (2006). Restoration of mires and wet grasslands. In: Restoration Ecology (eds J. Van Andel & J. Aronson), pp Blackwell Science Ltd. Grootjans, A.P., Iturraspe, R., Lanting, A., Fritz, C., & Joosten, H. (2010) Ecohydrological features of some contrasting mires in Tierra del Fuego, Argentina. Mires and Peat, 6, Gunnarsson, U. & Rydin, H. (2000) Nitrogen fertilization reduces Sphagnum production in bog communities. New Phytologist, 147, Hajek, T. & Adamec, L. (2009) Mineral nutrient economy in competing species of Sphagnum mosses. Ecological Research, 24, References 109

111 Hayward, P.M. & Clymo, R.S. (1982) Profiles of water content and pore size in Sphagnum and peat, and their relation to peat bog ecology. Proceedings Of The Royal Society Of London Series B-Biological Sciences, 215, Heath, D. (1995) An introduction to experimental design and statistics for biology UCL Press Limited, London. Heijmans, M., Berendse, F., Arp, W.J., Masselink, A.K., Klees, H., de Visser, W., & van Breemen, N. (2001) Effects of elevated carbon dioxide and increased nitrogen deposition on bog vegetation in the Netherlands. Journal Of Ecology, 89, Heslop-Harrison, J. & Heslop-Harrison, Y. (1970) Evaluation of pollen viability by enzymatically induced fluorescence: intracellular hydrolysis of fluorescein-diacetate. Stain Technology, 45, Hoosbeek, M.R., van Breemen, N., Berendse, F., Grosvernier, P., Vasander, H., & Wallen, B. (2001) Limited effect of increased atmospheric CO 2 concentration on ombrotrophic bog vegetation. New Phytologist, 150, Hoosbeek, M.R., Van Breemen, N., Vasander, H., Buttler, A., & Berendse, F. (2002) Potassium limits potential growth of bog vegetation under elevated atmospheric CO 2 and N deposition. Global Change Biology, 8, Ingram, H.A.P. (1978) Soil layers in mires - Function and terminology. Journal Of Soil Science, 29, Jackson, R.B., Manwaring, J.H., & Caldwell, M.M. (1990) Rapid physiological adjustment of roots to localized soil enrichment. Nature, 344, Jauhiainen, J., Vasander, H., & Silvola, J. (1994) Response Of Sphagnum fuscum to N deposition and increased CO 2. Journal Of Bryology, 18, Jauhiainen, J., Wallen, B., & Malmer, N. (1998) Potential NH 4 and NO 3 uptake in seven Sphagnum species. New Phytologist, 138, Jauhiainen, J. & Silvola, J. (1999) Photosynthesis of Sphagnum fuscum at long-term raised CO 2 concentrations. Annales Botanici Fennici, 36, Johansson, L.G. & Linder, S. (1980). Photosynthesis of Sphagnum in different microhabitats on a subarctic mire. In: Ecology of subarctic mire. Ecological bulletins (eds M. Sonesson), Vol. 30, pp Joosten, H. (2009). Human impacts: farming, fire, forestry and fuel. In: The Wetlands Handbook (eds E. Maltby & T. Barker), pp Blackwell Publishing Ltd. Kahl, S., Gerendas, J., Heeschen, V., Ratcliffe, R.G., & Rudolph, H. (1997) Ammonium assimilation in bryophytes. L-glutamine synthetase from Sphagnum fallax. Physiologia Plantarum, 101, Karsisto, M., Kitunen, V., Jauhiainen, J., Vasander, H. (1996). Effect of N deposition in free amino acid concentrations in two Sphagnum species. In: Northern peatlands in global climatic change (eds R. Laiho, Laine J., Vasander, H.), pp Academy of Finland, Helsinki. Kilham, P. (1982) The biogeochemistry of bog ecosystems and the chemical ecology of Sphagnum. Mich. Bot., 21, Kip, N., van Winden, J.F., Pan, Y., Bodrossy, L., Reichart, G. J., Smolders, A.J.P., Jetten, M.S.M., Damste, J.S.S., & Op den Camp, H.J.M. (2010) Global prevalence of methane oxidation by symbiotic bacteria in peat moss ecosystems. Nature Geoscience, 3, Solute transport in Sphagnum dominated bogs

112 Kramer, P.J. (1981) Carbon dioxide concentration, photosynthesis, and dry matter production. Bioscience, 31, Lamers, L.P.M., Farhoush, C., Van Groenendael, J.M., & Roelofs, J.G.M. (1999) Calcareous groundwater raises bogs; the concept of ombrotrophy revisited. Journal Of Ecology, 87, Lamers, L.P.M., Bobbink, R., & Roelofs, J.G.M. (2000) Natural nitrogen filter fails in polluted raised bogs. Global Change Biology, 6, Larmola, T., Tuittila, E.S., Tiirola, M., Nykanen, H., Martikainen, P.J., Yrjala, K., Tuomivirta, T., & Fritze, H. (2010) The role of Sphagnum mosses in the methane cycling of a boreal mire. Ecology, 91, Li, Y.H. & Vitt, D.H. (1994) The dynamics of moss establishment - temporal responses to nutrient gradients. Bryologist, 97, Li, Y.H. & Vitt, D.H. (1997) Patterns of retention and utilization of aerially deposited nitrogen in boreal peatlands. Ecoscience, 4, Lichtenthaler, H.K. (1987) Chlorophylls and carotenoids - pigments of photosynthetic biomembranes. Methods In Enzymology, 148, Ligrone, R. & Duckett, J.G. (1994) Cytoplasmic polarity and endoplasmic microtubules associated with the nucleus and organelles are ubiquitous features of food-conducting cells in bryoid mosses (Bryophyta). New Phytologist, 127, Ligrone, R. & Duckett, J.G. (1998a) Development of the leafy shoot in Sphagnum (Bryophyta) involves the activity of both apical and subapical meristems. New Phytologist, 140, Ligrone, R. & Duckett, J.G. (1998b) The leafy stems of Sphagnum (Bryophyta) contain highly differentiated polarized cells with axial arrays of endoplasmic microtubules. New Phytologist, 140, Limpens, J. & Berendse, F. (2003) Growth reduction of Sphagnum magellanicum subjected to high nitrogen deposition: the role of amino acid nitrogen concentration. Oecologia, 135, Limpens, J., Berendse, F., & Klees, H. (2003) N deposition affects N availability in interstitial water, growth of Sphagnum and invasion of vascular plants in bog vegetation. New Phytologist, 157, Limpens, J. & Heijmans, M. (2008) Swift recovery of Sphagnum nutrient concentrations after excess supply. Oecologia, 157, Limpens, J., Granath, G., Gunnarsson, U., Aerts, R., Bayley, S., Bragazza, L., Bubier, J., Buttler, A., van den Berg, L.J.L., Francez, A.J., Gerdol, R., Grosvernier, P., Heijmans, M.M.P.D., Hoosbeek, M.R., Hotes, S., Ilomets, M., Leith, I., Mitchell, E.A.D., Moore, T., Nilsson, M.B., Nordbakken, J.F., Rochefort, L., Rydin, H., Sheppard, L.J., Thormann, M., Wiedermann, M.M., Williams, B.L., & Xu, B. (2011) Climatic modifiers of the response to nitrogen deposition in peat-forming Sphagnum mosses: a meta-analysis. New Phytologist, 191, Lloyd, D., Thomas, K.L., Hayes, A., Hill, B., Hales, B.A., Edwards, C., Saunders, J.R., Ritchie, D.A., & Upton, M. (1998) Micro-ecology of peat: minimally invasive analysis using confocal laser scanning microscopy, membrane inlet mass spectrometry and PCR amplification of methanogen-specific gene sequences. Fems Microbiology Ecology, 25, Maberly, S.C. & Spence, D.H.N. (1983) Photosynthetic inorganic carbon use by fresh-water plants. Journal Of Ecology, 71, References 111

113 Maberly, S.C. & Madsen, T.V. (2002) Freshwater angiosperm carbon concentrating mechanisms: processes and patterns. Functional Plant Biology, 29, Malmer, N. (1988) Patterns in the growth and the accumulation of inorganic constituents in the Sphagnum cover on ombrotrophic bogs in Scandinavia. Oikos, 53, Malmer, N. (1993) Mineral nutrients in vegetation and surface layers of Sphagnum-dominated peat-forming systems. Advances in Bryology, 5, Malmer, N., Svensson, B.M., & Wallen, B. (1994) Interactions between Sphagnum mosses and field layer vascular plants in the development of peat-forming systems. Folia Geobotanica & Phytotaxonomica, 29, Mataloni, G. & Tell, G. (1996) Comparative analysis of the phytoplankton communities of a peat bog from Tierra del Fuego (Argentina). Hydrobiologia, 325, Money, R.P. (1995). Re-establishment of a Sphagnum-dominated flora on cut-over lowland raised bogs. In: Restoration of temperate wetlands (eds B.D. Wheeler, S.C. Shaw, W.J. Fojt & R.A. Robertson), pp John Wiley & Sons, Chichester. Money, R.P. & Wheeler, B.D. (1999) Some critical questions concerning the restorability of damaged raised bogs. Applied Vegetation Science, 2, Money, R.P., Wheeler, B.D., Baird, A.J., & Heathwaite, A.L. (2009). Replumbing wetlands - managing water for the restoration of bogs and fens. In: The Wetlands Handbook (eds E. Maltby & T. Barker), pp Blackwell Publishing Ltd. Morris, J.T. (1991) Effects of nitrogen loading on wetland ecosystems with particular reference to atmospheric deposition. Annual Review Of Ecology And Systematics, 22, Nordin, A. & Gunnarsson, U. (2000) Amino acid accumulation and growth of Sphagnum under different levels of N deposition. Ecoscience, 7, Paffen, B.G.P. & Roelofs, J.G.M. (1991) Impact of carbon dioxide and ammonium on the growth of submerged Sphagnum cuspidatum. Aquatic Botany, 40, Pakarinen, P. (1978) Production and nutrient ecology of 3 Sphagnum species in southern Finnish raised bogs. Annales Botanici Fennici, 15, Pastor, J., Peckham, B., Bridgham, S., Weltzin, J., & Chen, J.Q. (2002) Plant community dynamics, nutrient cycling, and alternative stable equilibria in peatlands. American Naturalist, 160, Prins, H.B.A. & Elzenga, J.T.M. (1989) Bicarbonate utilization - function and mechanism. Aquatic Botany, 34, Proctor, M.C.F. (2008). Physiological ecology. In: Bryophyte Biology (eds B. Goffinet & A.J. Shaw). Cambridge University Press, Cambridge. Raghoebarsing, A.A., Smolders, A.J.P., Schmid, M.C., Rijpstra, W.I.C., Wolters-Arts, M., Derksen, J., Jetten, M.S.M., Schouten, S., Damste, J.S.S., Lamers, L.P.M., Roelofs, J.G.M., den Camp, H., & Strous, M. (2005) Methanotrophic symbionts provide carbon for photosynthesis in peat bogs. Nature, 436, Rappoldt, C., Pieters, G., Adema, E.B., Baaijens, G.J., Grootjans, A.P., & van Duijn, C.J. (2003) Buoyancy-driven flow in a peat moss layer as a mechanism for solute transport. Proceedings Of The National Academy Of Sciences Of The United States Of America, 100, Raven, J.A., Osborne, B.A., & Johnston, A.M. (1985) Uptake of CO 2 by aquatic vegetation. Plant Cell And Environment, 8, Solute transport in Sphagnum dominated bogs

114 Raven, J.A., Griffiths, H., Smith, E.C., & Vaughn, K.C. (1998). New perspectives in the biophysics and physiology of bryophytes. In: Bryology for the twenty-first century (eds J.W. Bates, N.W. Ashton & J.G. Duckett), pp British Bryological Society. Raven, J.A. (2003) Long-distance transport in non-vascular plants. Plant Cell And Environment, 26, Raven, J.A., Cockell, C.S., & La Rocha, C.L. (2008) The evolution of inorganic carbon concentrating mechanisms in photosynthesis. Philosophical Transactions Of The Royal Society B-Biological Sciences, 363, Raven, J.A. (2011) The cost of photoinhibition. Physiologia Plantarum, 142, Redinger, K. (1934) Studien zur Ökologie der Moorschienken. Beihefte zum Botanischen Centralblatt, 52, Rice, S.K. & Schuepp, P.H. (1995) On the ecological and evolutionary significance of branch and leaf morphology in aquatic Sphagnum (Sphagnaceae). American Journal Of Botany, 82, Rice, S.K. (1995) Patterns of allocation and growth in aquatic Sphagnum species. Canadian Journal Of Botany-Revue Canadienne De Botanique, 73, Rice, S.K. & Giles, L. (1996) The influence of water content and leaf anatomy on carbon isotope discrimination and photosynthesis in Sphagnum. Plant Cell And Environment, 19, Richter, C. & Dainty, J. (1989) Ion behavior in plant-cell walls. 1. Characterization of the Sphagnum russowii cell wall ion exchanger. Canadian Journal Of Botany-Revue Canadienne De Botanique, 67, Riis, T. & SandJensen, K. (1997) Growth reconstruction and photosynthesis of aquatic mosses: Influence of light, temperature and carbon dioxide at depth. Journal Of Ecology, 85, RIVM (2009) Jaaroverzicht Luchtkwaliteit 2008, RIVM report /2009, RIVM Bilthoven. Robroek, B.J.M., Schouten, M.G.C., Limpens, J., Berendse, F., & Poorter, H. (2009) Interactive effects of water table and precipitation on net CO 2 assimilation of three co-occurring Sphagnum mosses differing in distribution above the water table. Global Change Biology, 15, Rochefort, L. & Price, J. (2003) Restoration of Sphagnum dominated peatlands. Wetlands Ecology and Management, 11, 1-2. Roelofs, J.G.M. (1983) Impact of acidification and eutrophication on macrophyte communities in soft waters in the Netherlands. 1. Field observations. Aquatic Botany, 17, Rosswall, T. & Granhall, U. (1980) Ecology of a subarctic mire. Nitrogen cycling in a subarctic mire. Ecol. Bull., 30, Rudolph, H., Hohlfeld, J., Jacubowski, S., Von der Lage, P., Matlok, H., & Schmidt, H. (1993) Nitrogen metabolism of Sphagnum. Advances in Bryology, 5, Rydin, H. & McDonald, A.J.S. (1985) Photosynthesis in Sphagnum at different water contents. Journal Of Bryology, 13, Rydin, H. & Clymo, R.S. (1989) Transport of carbon and phosphorus-compounds about Sphagnum. Proceedings Of The Royal Society Of London Series B-Biological Sciences, 237, Rydin, H. & Jeglum, J.K. (2006) The biology of peatlands. Oxford University Press, Oxford. Schipperges, B. & Rydin, H. (1998) Response of photosynthesis of Sphagnum species from contrasting microhabitats to tissue water content and repeated desiccation. New Phytologist, 140, References 113

115 Silvola, J. (1990) Combined effects of varying water content and CO 2 concentration on photosynthesis in Sphagnum fuscum. Holarctic Ecology, 13, Skene, M. (1915) The acidity of Sphagnum and its relation to chalk an mineral salts. Annals of Botany, 29, Skre, O. & Oechel, W.C. (1981) Moss functioning in different taiga ecosystems in interior Alaska. 1. Seasonal, phenotypic, and drought effects on photosynthesis and response patterns. Oecologia, 48, Smolders, A.J.P., Tomassen, H.B.M., Pijnappel, H.W., Lamers, L.P.M., & Roelofs, J.G.M. (2001) Substrate-derived CO 2 is important in the development of Sphagnum spp. New Phytologist, 152, Smolders, A.J.P., Tomassen, H.B.M., van Mullekom, M., Lamers, L.P.M., & Roelofs, J.G.M. (2003) Mechanisms involved in the re-establishment of Sphagnum-dominated vegetation in rewetted bog remnants. Wetlands Ecology and Management, 11, 403. Titus, J.E., Wagner, D.J., & Stephens, M.D. (1983) Contrasting water relations of photosynthesis for 2 Sphagnum mosses. Ecology, 64, Toet, S., Cornelissen, J.H.C., Aerts, R., van Logtestijn, R.S.P., de Beus, M., & Stoevelaar, R. (2006) Moss responses to elevated CO 2 and variation in hydrology in a temperate lowland peatland. Plant Ecology, 182, Tomassen, H.B.M., Smolders, A.J.P., Lamers, L.P.M., & Roelofs, J.G.M. (2004) Development of floating rafts after the rewetting of cut-over bogs: the importance of peat quality. Biogeochemistry, 71, Turetsky, M.R. & Wieder, R.K. (1999) Boreal bog Sphagnum refixes soil-produced and respired (CO 2 )-C-14. Ecoscience, 6, Twenhoven, F.L. (1992) Competition between 2 Sphagnum species under different deposition levels. Journal Of Bryology, 17, Urban, N.R. & Eisenreich, S.J. (1988) Nitrogen cycling in a forested Minnesota bog. Canadian Journal Of Botany-Revue Canadienne De Botanique, 66, Van Breemen, N. (1995) How Sphagnum bogs down other plants. Trends In Ecology & Evolution, 10, Van der Heijden, E., Verbeek, S.K., & Kuiper, P.J.C. (2000) Elevated atmospheric CO 2 and increased nitrogen deposition: effects on C and N metabolism and growth of the peat moss Sphagnum recurvum P. Beauv. var. mucronatum (Russ.) Warnst. Global Change Biology, 6, Van der Molen, P.C. & Wijmstra, T.A. (1994) The thermal regime of hummock-hollow complexes on Clara Bog, Co Offaly. Biology And Environment-Proceedings Of The Royal Irish Academy, 94B, Verschoor, A.J., Baaijens, G.J., Everts, F.H., Grootjans, A.P., Rooke, W., Schaaf, v.d.s., & Vries, d.n.p.j. (2003). Hoogveenontwikkeling in veentjes en kleinschalige hoogveencomplexen op het Dwingelerveld; een landschapsbenadering. Deel 2: Landschapsontwikkeling en hydrologie (in Dutch). Expertisecentrum LNV. Vitousek, P. (1982) Nutrient cycling and nutrient use efficiency. American Naturalist, 119, Waddington, J.M., Rotenberg, P.A., & Warren, F.J. (2001) Peat CO 2 production in a natural and cutover peatland: Implications for restoration. Biogeochemistry, 54, Solute transport in Sphagnum dominated bogs

116 Wheeler, B.D. & Shaw, S.C. (1995) Restoration of damaged peatlands. Department of the Environment, University of Sheffield, UK&HMSO, London, UK. Wiedermann, M.M., Gunnarsson, U., Ericson, L., & Nordin, A. (2009) Ecophysiological adjustment of two Sphagnum species in response to anthropogenic nitrogen deposition. New Phytologist, 181, Williams, B., Silcock, D., & Young, M. (1999) Seasonal dynamics of N in two Sphagnum moss species and the underlying peat treated with (NH 4 NO 3 )-N-15-N-15. Biogeochemistry, 45, Williams, T.G. & Flanagan, L.B. (1996) Effect of changes in water content on photosynthesis, transpiration and discrimination against (CO 2 )-C-13 and (COO)-O-18-O-16 in Pleurozium and Sphagnum. Oecologia, 108, Woodin, S., Press, M.C., & Lee, J.A. (1985) Nitrate reductase-activity in Sphagnum fuscum in relation to wet deposition of nitrate from the atmosphere. New Phytologist, 99, Woodin, S.J. & Lee, J.A. (1987) The fate of some components of acidic deposition in ombrotrophic mires. Environmental Pollution, 45, Yavitt, J.B., Williams, C.J., & Wieder, R.K. (1997) Production of methane and carbon dioxide in peatland ecosystems across north America: Effects of temperature, aeration, and organic chemistry of peat. Geomicrobiology Journal, 14, Yu, Z.C., Loisel, J., Brosseau, D.P., Beilman, D.W., & Hunt, S.J. Global peatland dynamics since the Last Glacial Maximum. Geophysical Research Letters, 37. References 115

117 116 Solute transport in Sphagnum dominated bogs

118 Solute transport in Sphagnum dominated bogs 117