THE GRAS-MODEL (GRASSLAND-SUCCESSION-MODEL)

Transcription

1 THE GRAS-MODEL (GRASSLAND-SUCCESSION-MODEL) A simulation model for the succession of grassland biotopes under various management regimes (developed for the Eifel National Park) Silvana Siehoff, Thomas G. Preuss, Hans Toni Ratte, Martina Ross-Nickoll, Gottfried Lennartz Overview Context Objectives Study site Model description National Park grasses & herbs bushes trees gras n bush n tree n megaherbivore Results cell soil wild game cutting Conclusion Initial state Non-interference Grazing by wisents Non-interference (without red deer) of AE of CA of AE of FR or H pastures & meadows increase of RU increase of RU spread of forest Year 1

2 Context Establishment of the Eifel National Park 24 Management of the open grasslands in the former military training site Vogelsang Model as support for decision making concerning goals and perspectives for the grasslands Objectives Simulation of grassland succession under various land use forms (hay-making, grazing, non-interference) Tool to extract and analyse processes relevant for succession Decision support system for all actors that deal with management of grasslands on the landscape level 2

3 Study site Eifel National Park: established 24 1,8 ha 7% forest Simulated area: former military training site Vogelsang 1,5 ha of semi-natural grassland Elevation: 4 m (NE) above sea level 58 m (SW) above sea level Climate gradient: NE: 8 C, 7-8 mm precipitation SW: 6 C, -12 mm precipitation Eifel National Park Study site Soils: acid brown soils (Cambisol), ph CaCl2 4-5 shallow depth Predominant plant communities: mountain hay meadows mountain pastures fallow grasslands shrubberies (common broom (Cytisus scoparius), blackthorn (Prunus spinosa) or black- and raspberries (Rubus spp.)) forests (spruce (Picea abies), birch (Betula pendula), beech (Fagus sylvatica) or oak (Quercus robur)) 3

4 Dataset: Vegetation mapping Analysis of species composition most relevant succession stages Coexistence of different succession stages within the Eifel National Park (space for time substitution) most important processes of succession Information about the speed of processes: data about historical land use in the area (historical maps and aerial photographs) Calibration of processes on well-known vegetation developments Important for grassland succession: Neighbourhood relationships between different habitats Cellular automaton Modelled area: divided into cells Each cell: defined by position, width and length, soil type, land use form and set of representative species Competition of species for space: within each cell and between adjacent cells raster polygon 1 cell Emergence of vegetation succession from competition 4

5 Spatial explicit data as model input Eifel National Park digital map: Polygon map with about 25 polygons of biotope types simulated area digital map: Raster map with about 5, cells (1m x 1m) spatially explicit input data (biotope type, land use, soil properties, ) Cellular automaton competition for space grasses & herbs bushes trees gras n Pflanze 2 Pflanze 1 bush n Pflanze 2 Pflanze 1 tree n Pflanze 2 Pflanze 1 national park wild game cell megaherbivores soil hay-making 5

6 Factors influencing species growth i.e. competition strength: cutting grazing / stamping Interspecific competition Management and wildlife Soil properties wild pig soil Calculating species growth according to land use dcg g cg dt Time [years] UG f (grazing) 1-9 logistic growth - cg g g max - cg g gmax f (cutting) f (grazing) f (stamping) < f(x) < 1 - I G U 9 G 8 6 Growth at a given grazing intensity 4 U G : 6 U G : 7 2 U G : 8 U G : Time [years] U G : 1 U G : 2 U G : 3 U G : 4 U G : 5 U x : Utilization number (Briemle et al. 22) similar to Ellenberg indicator numbers but for utilization 6

7 Resulting species competition strength according to land use Time [years] I G :% 8 I G :% UG: 5 UG: 5 Species 1 Species Time [years] Time [years] U G: 3 U G: 5 I G :% U G: 1 U G: Results Grassland succession at non-interference (including 22 red deer /ha) GT-p becomes FR-p GT-r becomes H-n of AE LC-r CA Year PA Year 2 Year 6 increase of RU spread of BP and PA only in former LC Grasslands N poor N rich GT-p GT-r mountain meadows, nutrient poor or rich: Geranio-Trisetetum FC-p LC-r mountain pasture, nutrient poor: (Festuco-Cynosuretum), nutrient rich: (Lolio-Cynosuretum) Fallow grasslands with stands: nutrient poor: red fescue (Festuca rubra), FR-p H-n AE-n CA nutrient rich: Hemikryptophytes like Dactylis glomerata, Holcus lanatus, developing into stands of tall oat grass (Arrenatherum elatius) or thistle (Cirsium arvense) Shrubberies CS RS common broom (Cytisus scoparius) or black- and raspberries (Rubus spp.) Forests BP FS QR PA Birch (Betula pendula), beech (Fagus sylvatica), oak (Quercus robur), spruce (Picea abies) 7

8 Results Initial state GT-p Non-interference (22 red deer / ha.2 LU /ha) Grazing by wisents (incl. red deer,.2 LU /ha) increase of CS Non-interference (without red deer,. LU /ha) GT-r of FR or H of AE pastures & meadows stand of CA of AE FR-p / H-n with CS Grasslands N poor GT-p GT-r FC-p LC-r N rich FR-p H-n AE-n CA Shrubberies Geranio-Trisetetum Festuco-Cynosuretum, Lolio- Cynosuretum Dominance of Festuca rubra, Hemikryptophytes, Arrenatherum elatius, Cirsium arvense CS RS Cytisus scoparius, Rubus spp. Forests Year BP FS QR PA PA LC-r Betula pendula, Fagus sylvatica, Quercus robur, Picea abies increase of RU No spread of forests Red deer in combination with a dense grass-sward inhibit growth of seedlings. Grasslands: stands of single grass species Spread of Rubus spp. increase of RU and CS Few shrubberies and trees More trees although red deer and wisents feed on seedlings, because wisents disturb sward Meadows & pastures, few stands Spread of Rubus spp., C. scoparius and thistle few PA & BP spread of forest Forest development Without any herbivores feeding on the seedlings, the grasses sward cannot inhibit growth of trees (mainly birch), fast spread esp. at pastures. Grasslands: stands of single species Results Forest development under different management scenarios year year Fastest spreading tree: birch Competition between full grown trees within forest and wisents impact on full grown trees: not yet included Initial state Noninterference with red deer Wisents + red deer Noninterference without red deer Succession towards forest, not succession within forest is modelled Other conifers Spruce Other deciduous Birch Oak Beech 8

9 Immobilization of litter Conclusion Objectives: Simulation of grassland succession Plausible predictions of vegetation dynamics (according to experts of the area). The processes included are appropriate to simulate interspecific competition (model verification). Model validation: will be performed for the first years. Tool to analyse processes relevant for succession Analysis of single processes, e.g. impact of red deer on forests formation. Decision support system Many sometimes compensating processes included (e.g. grazing: feeding on seedlings & disturbance of the sward). high spatial resolution of simulation result. With the outcome from a chosen management procedure, one can easily accomplish an efficiency control of ecological restoration. Still to check: transfer to other sites. Should be easily accomplished due to the usage of the utilization numbers. Further developments Red deer Heck cattle Wisent Konik horse N-Deposition NH 4 NO 3 Megaherbivore and wildlife dispersal and behaviour N-Reabsorption Biomass: Roots + Shoots N-Fixation Denitrification of NO 3 Nitrogen cycle Immobilization of NH 4 Litter: Roots + Shoots Uptake of NO 3 Mineralization N total Nitrification NH 4 NO 3 Leaching of NO 3 9

10 Thank you for your attention! Literatur Briemle G., Nitsche S., Nitsche L. (22). Nutzungswertzahlen für Gefäßpflanzen des Grünlandes. Schriftenreihe für Vegetationskunde 38, Lennartz G., Preuss T., Fürste A., Theißen B., Toschki A., Strauss T., Tischler B., Schäffer A., Roß-Nickoll M. (26). Abschlussbericht Modellierung von Landschaftsentwicklungszenarien für die Managementzone im Nationalpark Eifel unter Berücksichtigung verschiedener Managementvarianten einschließlich spezieller Untersuchungen der Heuschrecken, Tagfalter und Vögel sowie der Ableitung von Monitoringparametern für die Offenlandentwicklung, Lehrstuhl für Umweltbiologie und chemodynamik der RWTH Aachen und Forschungsinstitut für Ökosystemanalyse und bewertung e.v. (gaiac) im Auftrag der Landesanstalt für Ökologie, Bodenordnung und Forsten Nordrhein-Westfalen. 1

11 List of representative species Grasses & herbs Bushes Trees Festuca rubra agg. Arrhenatherum elatius Cynosurus cristatus Lolium perenne Trifolium repens Cirsium arvense Groups: Hemicryptophytes ( Horstpflanzen ) Hemicryptophytes ( Rosettenpflanzen und Therophyten ) Hemicryptophytes ( Schaftfplanzen ) Hemicryptophytes ( Kletterpflanzen ) Rubus spec Prunus spec Cytisus scoparius Betula pendula Fagus sylvatica Quercus robur Picea abies 11