Growth and development of Pseudocalanus spp. in the northern Gulf of Alaska

Transcription

1 Growth and development of Pseudocalanus spp. in the northern Gulf of Alaska HUI LIU AND RUSSELL R. HOPCROFT* INSTITUTE OF MARINE SCIENCE, UNIVERSITY OF ALASKA FAIRBANKS, AK , USA PRESENT ADDRESS: COOPERATIVE INSTITUTE FOR MARINE RESOURCES STUDIES, OREGON STATE UNIVERSITY, 2030 SE MARINE SCIENCE DRIVE, NEWPORT, OR 97365, USA *CORRESPONDING AUTHOR: Received January 19, 2008; accepted in principle April 4, 2008; accepted for publication April 10, 2008; published online April 25, 2008 Corresponding editor: Roger Harris Pseudocalanus are the numerically dominant calanoid species in coastal subarctic Pacific waters. We examined their juvenile growth rates, and explored their relationships to temperature, chlorophyll a and body size for Pseudocalanus spp. from 2002 to 2004 in the northern Gulf of Alaska. Generally, the monthly mean growth rates increased from (SE) day 21 in March to day 21 in August, declining slightly to day 21 in October. Typically, growth rates at most stations were around 0.05 day 21, with no consistent or significant pattern between stations. After standardization to 5 and 108C, the mean growth rates were day 21 and day 21 respectively, with growth rate decreasing with increasing development stage. Unlike other local calanoid copepod species, Pseudocalanus species tend to be more temperature-dependent than food-dependent, with composite statistical models describing at most 30% of the observed variability in growth rate. Interestingly, development time was comparable to other co-occurring calanoid copepods; however, growth rates of Pseudocalanus spp. were considerably lower. We demonstrate this with a new multi-species model that describes the growth rates of other egg-scattering copepods in this ecosystem, but to which Pseudocalanus does not fit. Thus, the egg-carrying Pseudocalanus species appear to employ a life history strategy optimized for slow growth at low chlorophyll that keeps individuals relatively small, and may therefore reduce visual predation upon them. INTRODUCTION Calanoid copepods are generally considered the most important components in the marine pelagic ecosystem being grazers on phytoplankton and other protists, and prey for animals at higher trophic levels (Mauchline, 1998). Our knowledge of their life parameters (e.g. development time, growth rate and egg production) provides fundamental information on the energy and matter transformation in pelagic food webs. Growth is a key component in the determination of secondary production, and thus in appreciating the roles of copepods in the flow of matter and energy in the sea (Kiørboe, 1997). At present, we know relatively little of the rate of growth by juvenile copepods in nature. In the subarctic North Pacific, this deficiency has been addressed for the larger species in this region (Liu and Hopcroft, 2006a, b, 2007), but the lack of field-derived vital rates for small-bodied copepods, such as Pseudocalanus, remains. Copepods of the genus Pseudocalanus contain small particle-feeding species, which frequently dominate the zooplankton collections in temperate-boreal neritic waters of the Northern hemisphere (Corkett and McLaren, 1978). On the Gulf of Alaska shelf, and the adjacent Bering Sea, Pseudocalanus numerically rank the second most abundant copepods after Oithona among the crustacean zooplankton assemblages (Cooney, 1986; Incze et al., 1997; Cooney et al., 2001; Coyle and Pinchuk, 2002, 2003, 2005), are the top secondary producers (Coyle and Pinchuk, 2003), and important prey doi: /plankt/fbn046, available online at # The Author Published by Oxford University Press. All rights reserved. For permissions, please journals.permissions@oxfordjournals.org

2 items for both larval and juvenile fish in the northern Gulf of Alaska and the adjacent Bering Sea (Kendall and Nakatani, 1991; Hilgruber et al., 1995, Schabetsberger et al., 2003; Napp et al., 2005). Until recently, Pseudocalanus were often considered as a complex of copepods, and now we appreciate that at least seven species coexist in this genus, with two common species (P. minutus and P. newmani) occurring in the temperate-boreal waters of the northwestern Pacific (Frost, 1989). The most recent study reports that P. mimus, P. newmani and, to a limited extent, P. minutus are the three common species in the coastal Gulf of Alaska (Napp et al., 2005). Unlike congeneric species in the northern Atlantic waters, Pseudocalanus species in the subarctic Pacific have only received a few field studies on egg production (Dagg et al., 1984; Paul et al., 1990; Siefert, 1994; Ban et al., 2000; Napp et al., 2005), with in situ somatic growth studied once in the southeastern Bering Sea (Vidal and Smith, 1986), while laboratory rates have been examined for P. newmani (Lee et al., 2003) and for Pseudocalanus sp. (Vidal, 1980). Even in more temperate-pacific waters, few studies have considered field estimates of somatic growth, although several studies have considered egg production (Gómez-Gutiérrez and Peterson, 1999; Peterson et al., 2002; Halsband-Lenk et al., 2005). Thus, despite their importance, our knowledge of field-derived growth rates of Pseudocalanus spp. throughout the North Pacific remains limited. This study aims to determine the juvenile growth rates of this ecologically important small-bodied eggcarrying copepod, and explore the fundamental relationships to temperature, food concentrations and body size through a multiple year field study in the northern Gulf of Alaska. As the finale to studying copepod growth in this region, we also test the hypothesis that small-bodied copepods grow faster than large ones (e.g. Hopcroft et al., 1998; Hirst and Bunker, 2003), and that egg-carrying species may grow slower than broadcast spawning species (Hirst and Lampitt, 1998; Hirst and Bunker, 2003). Currently, the very methodology appropriate for growth rate determination has become debated (Hirst et al., 2005, Kimmerer et al., 2007), with the likelihood that ever greater effort will be needed to determine these essential rates in the future. METHODS Experiments were executed as part of the US Northeast Pacific GLOBEC program in the Coastal Gulf of Alaska (Weingartner et al., 2002). For this species, we employed six cruises conducted in March, April, May, June/July, August, and October of 2002 and 2003, plus cruises from March, May and June/July of Experimental work was set up at four stations along the Seward line from inshore to just past the shelf break (i.e. GAK1, 4, 9, 13), plus one station along the western inner passage of Prince William Sound (PWS-either KIP2 or PWS2) where the depth is m (see Fig. 1 in Liu and Hopcroft, 2006b). Ambient temperature and size-fractioned chlorophyll a concentrations were measured concurrently, and have been presented previously (see Fig. 3 in Liu and Hopcroft, 2006b). Experimental methodology is identical to that employed successfully for other species in this area (Liu and Hopcroft, 2006a, b, 2007), where more detailed methodology is presented. In brief, copepod collections were sequentially sorted into artificial cohorts (Kimmerer and McKinnon, 1987; Peterson et al., 1991; Hopcroft and Roff, 1998; Hopcroft et al., 1998) by passage through submerged screens of the following mesh sizes: 1300, 1000, 800, 600, 500, 400, 300, 200, 150 and 100 mm. The sample was constantly diluted with pre-screened water cooled at ambient seawater temperature, and as each cohort size-class was created, it was placed into an incubator at ambient seawater temperature. Prior to incubation, each size-fraction was gently homogenized and evenly divided. One half was concentrated and preserved in 5% buffered seawater formalin as the time zero sample (T-0), and the other half equally divided among several 20 L carboys previously filled with prescreened seawater. After 5 days (in March, April and May) and 4 days (in June/July, August and October), the carboys were filtered through 45 mm sieves, copepods were pooled by the original size fractions and preserved immediately in 5% buffered seawater formalin as the final sample (T-5 or T-4). All preserved material was stained with Rose Bengal. It is clear that three species occur in these collections (Napp et al., 2005); however, laboratory sorting of early copepodites of Pseudocalanus is problematic. Thus, preserved copepodites were only identified to developmental stages. A minimum of 30, and up to 100 animals in both the initial and final sample pair were used for the data analysis depending on their availability with the experiments. Copepodite prosome lengths were digitally measured (Roff and Hopcroft, 1986), and the progression of the cohorts was determined by changes in the stage and body size. Development time was calculated as the reciprocal of the molting rate (MR) observed in each cohort. Copepodite dry weights (DWs) were predicted from a length weight relationship developed for Pseudocalanus spp. in this study area: log 10 DW ¼ 27.62þ2.85*log 10 PL, where PL (prosome length) is in mm, and DW is in mg (Fig. 1). To convert 924

3 H. LIU AND R. R. HOPCROFT j GROWTH OF PSEUDOCALANUS IN THE GULF OF ALASKA DW to carbon weights, 40% carbon content in DW for copepods is assumed (Båmstedt, 1986). The growth rate (day 21 ) of a given cohort over the incubation time t (days) was computed from the equation g¼(ln W t 2ln W 0 )t 21 (Runge and Roff, 2000). To some extent, methodological biases inevitably exist in this study, which we have discussed previously (Liu and Hopcroft, 2007). First, the MR method is valid only for population with uniform age structure within stage (Hirst et al., 2005), and it is possible that this was not always true. The modified MR method (Hirst et al., 2005) would be more appropriate but cannot be employed because we generally have a mixture of stages, and frequently lack of knowledge of developmental times of adjoining stages within each experiment. Second, our reliance on length weight regression rather than direct weight measurement could cause bias for calculation of growth rates (Kimmerer et al., 2007); however, it is painful to trade off directly measurement of weights against destruction of samples. Third, our failure to account lipid storage for later copepodite stage (C5) could introduce bias in total growth, although not necessarily structural growth, particularly in spring when oil sacs were most obvious. The coexistence of both P. mimusand P. newmani in our samples may also introduce some biases because these species have differences in prosome length and different spatial distributions (Napp et al., 2005). Nonetheless, at individual stations both species occur consistently, with their prosome lengths co-varying seasonally (Napp et al., 2005; Hopcroft et al., unpublished), as does their weightspecific egg production (Hopcroft et al., unpublished). Thus, we expect any biases introduced by different proportion of these species in different experiments to introduce relatively little, if any, systematic bias in our analysis. A multiple regression model to elucidate the relative effects of developmental stage, body weight, temperature, total chlorophyll a concentration and their interactive influences on growth rates of Pseudocalanus spp. was conducted (SAS system V9). We continually tested the utility of the composite non-linear model (Liu and Hopcroft, 2006b), for Pseudocalanus spp. by combined features of SAS (V8) and R (2.6.0) with equivalent r 2 calculated from appropriate model sum of squares (Anderson- Sprecher, 1994). For ease of comparison to the literature, growth rates were standardized to both 5 and 108C using Q 10 of 2.7 for food-satiated juvenile broadcast spawners (Hirst and Bunker, 2003). For other analyses and all figure-making, we used Sigmaplot (V10). RESULTS Growth rate and development time of Pseudocalanus spp. Spatiotemporally, all developmental stages of Pseudocalanus species occurred year-around (Fig. 2), suggesting multiple overlapping generations annually. Fig. 1. Relationship between prosome length PL (mm) and dry weight (DW) (mg) for Pseudocalanus spp. C1 C6 in the northern Gulf of Alaska. Fig. 2. Stage durations (upper panels) and growth rates (lower panels) of Pseudocalanus spp. in the northern Gulf of Alaska Stage is the average of the population at start of incubation. 925

4 Seasonally, the monthly mean growth rates of Pseudocalanus spp. increased from March to August, then declined slightly in October, while the monthly mean development time followed a reciprocal trend. Monthly variability in growth rate and development time within or between years was insignificant (Fig. 3). Average growth rate and development time across developmental stages can be affected by C5s because of their slow somatic growth, and it is clear that the inclusion of copepodite stage C5 can lessen the overall patterns (Fig. 3). Generally, copepodite growth and development (C1 C5) in the spring (March to May) were faster in 2003 than in 2002 (Fig. 3). The slowest growth rate, with corresponding stage duration (in parenthesis), was day 21 (22 days) and occurred in March of The fastest spring rates of day 21 (8.5 days) and day 21 (15.3 days) appeared in March and May of 2004, respectively, likely due to the Fig. 3. Seasonal variability in growth rate and development time of Pseudocalanus spp. in the northern Gulf of Alaska favorable condition caused by the later than typical cruise dates. After peaking in summer (July or August), the trend of growth rates started to slow down to October. During the summer and fall, the fastest growth and development for C1 C4 was day 21 (3.6 days) in July 2002, while the slowest (including C1 C5) was day 21 (29 days) in same month of These patterns become clear when compressed across years (Table I). Effects of temperature Overall, the effect of temperature on growth rates of Pseudocalanus spp. was positively significant (r 2 ¼ 0.13, P, ), while similar analysis for each individual stage was positively significant for C1 (r 2 ¼ 0.81, P, ), C2 (r 2 ¼ 0.26, P, 0.007) and C3 (r 2 ¼ 0.23, P, 0.02), and insignificant for C4 (r 2 ¼ 0.017, P. 0.45) and C5 (r 2 ¼ 0.16, P. 0.20). Typically, body weights tend to be negatively related to temperature in this study (Fig. 4). For each individual stage, the relationship between body weight and temperature was insignificant at C1, C3 and C5 (r 2 ¼ ), except for C2 (r 2 ¼ 0.27, P, 0.006) and C4 (r 2 ¼ 0.25, P, 0.003). However, the overall pattern was significantly negative (r 2 ¼ 0.083, P, 0.002). There was no consistent spatial variability in growth rate and development time after standardization to 58C (Q 10 of 2.7) during each study year except for obvious noise caused by C5s (Fig. 5). Typically, at most stations, growth rates ranged around 0.05 day 21 with development time from 9 to 15 days, with the annual variability in growth rate at each station insignificant except for station PWS2 in 2004 (Fig. 5). Although, the mean growth rates at all sampling stations were variable (Table I), differences were not significant (one-way ANOVA, P. 0.05). After standardization to 5 and 108C, the mean across-stage growth rates and development time were day 21 and 15.8 days at 58C and day 21 and 9.6 days at 108C, respectively, and exhibited a clear trend of declining growth and development with the development stage (Table II). Functional relationships to growth rate A significant multiple regression model for growth rates of Pseudocalanus spp. (r 2 ¼ , P, ) included developmental stage (partial r 2 ¼ ), temperature (partial r 2 ¼ ) and body weight (partial r 2 ¼ ) as explanatory variables, but no relationship to either total chlorophyll a or sized-fractionated chlorophyll a.5 mm (Table III). After standardization to 926

5 H. LIU AND R. R. HOPCROFT j GROWTH OF PSEUDOCALANUS IN THE GULF OF ALASKA Table I: Temporal and spatial comparison of growth rates (SE, in parenthesis) (day 21 ) for Pseudocalanus spp in the northern Gulf of Alaska Growth rate (SE) Species Temp (8C) March April May July August October Average Pseudocalanus spp (0.007) (0.005) (0.005) (0.011) (0.016) (0.009) (0.004) PWS2 GAK1 GAK4 GAK9 GAK13 Pseudocalanus spp (0.006) (0.007) (0.007) (0.011) (0.011) (0.004) 58C, growth rate was negatively correlated to body size for C1 C5 (r 2 ¼ 0.05, P ¼ 0.016). Within each stage, growth rates appeared positively related to the body size except for the negative relationship observed for C1 (r 2 ¼ 0.38, P, 0.02); however, the positive relationship between growth rate and body size was statistically significant for C2 (r 2 ¼ 0.27, P, 0.006) and C3 (r 2 ¼ 0.16, P, 0.05), but not for C4 and C5 (Fig. 6). The growth rates standardized to 5 and 108C were insignificantly correlated with the total ambient chlorophyll a concentration in the form of a Michaelis Menten relationship (r 2 ¼ 0.009, P ¼ 0.332; Fig. 6; Table III). The significance of this relationship improved markedly when chlorophyll was restricted to particles.5 mm, for which g max and K chl were day 21 and mg m 23 at 58C, and day 21 and mg m 23 at 108C, respectively. The large impact on K chl occurs because small cells contribute most at low chlorophyll concentration, while saturated rates are (by definition) independent of chlorophyll concentration. A composite non-linear model incorporating body size into the classical Michaelis Menten model Fig. 4. Effect of temperature on growth rate and body weight of Pseudocalanus spp. in the northern Gulf of Alaska Fig. 5. Spatial patterns in growth and development time of Pseudocalanus spp. in the northern Gulf of Alaska Data were corrected to 58C (Q 10 ¼ 2.7). Error bars are standard errors. 927

6 Table II: Growth rates (day 21 ) and development times (days, in parenthesis) for Pseudocalanus species in the subarctic Pacific and the Oregon coast determined by field and laboratory studies Species Temp (8C) Growth rate and developmental time C1 C2 C3 C4 C5 EPR Average Source Pseudocalanus spp. Pseudocalanus spp. Pseudocalanus spp. Pseudocalanus mimus Pseudocalanus spp. (lab) Pseudocalanus newmani (lab) indicated that body size was negatively related to growth rate, with g max and K chl (.5 mm) of day 21 and mg m 23 at 58C, and day 21 and mg m 23 at 108C, respectively. In comparison to the Michaelis Menten model, saturated growth rates of the new model were higher under high food conditions, and more variability (10 16%) was explained by the composite non-linear models (Table III). Incorporation of developmental stage further improved both chlorophyll models, explaining 32 33% of the variations, and notably contained negative relationships to stage, positive relationship to body size and no significant relationship to chlorophyll. DISCUSSION 5 a (13.9) (10.2) (10.9) (16.1) (40.5) (15.8) This study 10 a (8.5) (6.2) (6.6) (9.8) (24.7) (9.6) This study (4 5) 0.13 (4 5) 0.13 (4 5) 0.13 (4 5) 0.13 (4 5) Vidal and Smith (1986) Peterson et al. (2002) Vidal (1980) 10 (3.4) (3.0) (3.7) (2.8) (3.9) 0.14 Lee et al. (2003) a Corrected to these temperature by Q 10 of 2.7. Growth rates of Pseudocalanus spp. The overall patterns of growth rate are generally consistent with field-derived patterns demonstrated elsewhere (e.g. Hirst and Bunker, 2003), and locally for other species (Liu and Hopcroft, 2006a, b, 2007). However, estimated growth rates in this study ( day 21 with mean of day 21 ) were low compared to other estimated rates for Pseudocalanus species in the subarctic Pacific (Table II). For example, a growth rate of 0.13 day 21 at temperatures ranging from 0.5 to 68C was measured in the southeastern Bering Sea (Vidal and Smith, 1986), and 0.22 day 21 at 11.88C was measured off the Oregon coast (Peterson et al., 2002). Interestingly, our estimated juvenile growth rate corrected to 108C was day 21, which was consistent with the egg production rate of 0.06 day 21 at 11.88C estimated off the Oregon coast (Peterson et al., 2002). The temperature corrected rate of day 21 at 58C was within the range of day 21 for P. acuspes measured at 48C in the central Baltic Sea (Renz et al., 2007). For the two previous studies in subarctic waters, the low temperatures during April and May in the southeastern Bering Sea (Vidal and Smith, 1986), as well as the high chlorophyll a concentrations mostly in June off the Oregon coastal upwelling (Peterson et al., 2002), are favorable for copepod growth, in contrast to the wide seasonally variable conditions in the Gulf of Alaska (see Fig. 3 in Liu and Hopcroft, 2006b). During periods of high chlorophyll a, contribution from microzooplankton is less important, and chlorophyll a provides an adequate estimate of food availability. To some extent, both previously estimated growth rates likely represent the upper limit of growth rates in the field, and this is confirmed by the similarity of these Bering Sea and Oregon rates to those determined for these species under food-satiated laboratory conditions (Table II). Estimates in this study may reasonably represent more typical levels under food-limitation. Interestingly, in comparison with other common large calanoid copepod species in this area, the stage durations of Pseudocalanus spp. were not significantly different from four other species (Fig. 7), while growth rates of Pseudocalanus spp. were significantly lower (Table IV). Similar to this study, low growth rates with comparable stage durations were observed for P. acuspes in the central Baltic Sea (Renz et al., 2007). 928

7 929 Table III: Relationships between growth rate (Gr, g day 21 ), initial stage (Stg), temperature (T, 8C) and total chlorophyll a concentration (Chl, mg m 23 ) for Pseudocalanus spp. in the northern Gulf of Alaska Coefficients (p, partial r 2 ) Model Equation n T (8C) a 1 a 2 a 3 a 4 a 5 r 2 (p) Multiple regression Gr¼a 1 þa 2 Stgþa 3 Tþa 4 log BW þa 5 [Total Chl] (C1 C5) ; ; (0.0109); Total Chloro a b g max K chl Michaelis Menten Gr¼Chl[g max ]/(ChlþK ch ) (C1 C5) (,0.001) (.0.1) (0.332) Gr¼Chl[g max ]/(ChlþK ch ) (C1-C5) (,0.001) (.0.1) (0.332) Composite non-linear Gr¼b*log BWþChl[g max ]/(ChlþK ch ) (C1 C5) (,0.01) (,0.001) (.0.1) Gr¼b*log BWþChl[g max ]/(ChlþK ch ) (C1 C5) (,0.01) (,0.001) (.0.1) Chloro.5 mm Michaelis Menten Composite non-linear Gr¼a*stageþb*log BWþChl[g max ]/ (ChlþK ch ) (C1 C5) (,0.001) (,0.001) (,0.001) (.0.1) Gr¼Chl[g max ]/(ChlþK ch ) (C1 C5) (,0.001) (,0.1) (0.023) Gr¼Chl[g max ]/(ChlþK ch ) (C1 C5) (,0.001) (,0.1) (0.023) Gr¼b*log BWþChl[g max ]/(ChlþK ch ) (,0.001) (,0.001) (,0.05) (C1 C5) Gr¼b*log BWþChl[g max ]/(ChlþK ch ) (,0.001) (,0.001) (,0.05) (C1 C5) Gr¼a*stageþb*log BWþChl[g max ]/ (ChlþK ch ) (C1 C5) (,0.001) (,0.01) (,0.001) (.0.1) Downloaded H. from LIU AND R. R. HOPCROFT j GROWTH OF PSEUDOCALANUS IN THE GULF by OF guest ALASKA on 20 November 2018

8 Fig. 6. Relationship between temperature-corrected growth rates, body size and chlorophyll a (size.5 mm) for Pseudocalanus spp. in the northern Gulf of Alaska The implication is that despite annually high abundance and biomass in this area (Coyle and Pinchuk, 2003, 2005), Pseudocalanus grows slowly. This mean that either Pseudocalanus is more efficient at utilizing resources, has higher fecundity, or experience lower mortality than co-occurring copepod species to maintain their prominence in this ecosystem. If it is true, then Pseudocalanus spp. exhibits a unique life strategy compared to other larger local species (i.e. Neocalanus flemingeri/plumchrus, Metridia pacifica, Calanus marshallae, C. pacificus). All of them grow fast (Liu and Hopcroft, 2006a, b, 2007), because of their relatively short optimal growth season during spring in this area. Cyclopoid females may have a higher mortality when bearing sacs than without them due to a higher susceptibility to their visual predators (Kiørboe and Sabatini, 1994); however, they are generally less visible to visual predators than large-bodied species. At the same time, mortality of eggs and nauplii may be low compared to broadcast spawners. As egg-carrying calanoid species, Pseudocalanus likely face the same situation in the field, and concurrent egg production rates of this species (Napp et al., 2005; Hopcroft et al., unpublished) suggest specific egg production of this species is also lower than local broadcast spawning species, but higher than observed somatic rates. We speculate that Pseudocalanus spp. possibly takes advantage of slow growth to stay small, thereby minimizing visual predation, but has per capita recruitment to copepodites (and ultimately adults). Further observations on these life history strategy trade-offs of eggcarrying and small-bodied copepods are clearly needed to explore such possibilities. Effects of temperature on growth rates Temperatures strongly regulate rates of copepod growth (McLaren, 1978; Huntley and Lopez, 1992). Globally, juvenile growth rates are more temperature-dependent for sac-spawners than for broadcasters (Hirst and Lampitt, 1998; Hirst and Bunker, 2003), but temperature-dependent growth rates are invariably confounded with food concentrations and body sizes (Vidal, 1980; Hirst and Bunker, 2003; Liu and Hopcroft, 2006b). Secondary production of Pseudocalanus spp. in particular has been shown to be controlled more by temperature than by food (McLaren, 1978; Davis, 1984; Frost, 1985). In this study, effects of temperature on growth rate and body size of Pseudocalanus were obvious. Growth rates showed a significantly positive relationship to temperature for stages C1 C5 combined, as well as individual stage of C1, C2 and C3. Temperature was also statistically significant in the selected multiple linear regression model (Table III). Moreover, in this study, body sizes of Pseudocalanus spp. tend to be negatively temperature-dependent across stages. Previous studies reported that Pseudocalanus species exhibit large seasonal variability in body length, even within the same species (Yamaguchi et al., 1998; Napp et al., 2005; Renz and Hirche, 2006). Both higher temperature and lower phytoplankton concentrations are major variables causing smaller size in adult copepods (McLaren, 1974; Corkett and McLaren, 1978; Vidal, 1980). Thus, the trend of temperature-dependent body size was largely caused by variable conditions in the Gulf of Alaska, in particular the higher temperatures and low chlorophyll a concentrations during summer. Fig. 7. Comparison of the development time for the dominant copepods in the northern Gulf of Alaska. Data were corrected to 58C (Q 10 ¼ 2.7). Error bars are 95% CI. Chlorophyll a, body size and growth rates Standardizing growth rates to a fixed temperature helps reveal underlying relationships between growth rates, food concentrations and body size. 930

9 H. LIU AND R. R. HOPCROFT j GROWTH OF PSEUDOCALANUS IN THE GULF OF ALASKA Table IV: Comparison of the Q 10 standardized growth rate (day 21 ) of the dominant calanoid copepod species in the northern Gulf of Alaska Species Temp (8C) Growth rate Specific Egg production rate Location Source Pseudocalanus spp a Gulf of Alaska This study Calanus marshallae a Gulf of Alaska Liu and Hopcroft (2007) Calanus pacificus a Gulf of Alaska Liu and Hopcroft (2007) Metridia pacifica Gulf of Alaska Liu and Hopcroft (2006b), Hopcroft et al. (2005) Metridia okhotensis Gulf of Alaska Hopcroft et al. (2005) Neocalanus flemingeri/plumchrus Gulf of Alaska Liu and Hopcroft (2006a) Centropages abdominalis b 0.07 b Gulf of Alaska Slater and Hopcroft (2005) a Hopcroft et al. (unpublished). b Growth rate corrected from original 7.0 to 5.08C (Q 10 ¼ 2.7). Unsurprisingly, standardized rates were negatively related to body size, due in part to the severity of food-limited growth increasing with body sizes (Vidal, 1980; McKinnon, 1996; Hopcroft et al., 1998; Hirst and Bunker, 2003). Within each individual stage, however, growth rates and body size tend to be positively correlated (Fig. 6). This is because under favorable food conditions, within each stage animals growing fast tend to be larger than animals growing more slowly at that stage (Liu and Hopcroft, 2006a). The relationships were significant for C2 and C3, but not for C1, C4 and C5, which might suggest that the growth of younger stages was more optimal than for later stages. Interestingly, similar patterns of copepod growth rates and their body size are observed for other large calanoid copepods in this area (Liu and Hopcroft, 2006a, b, 2007), suggesting this kind of pattern may exist for all marine copepods. Standardized growth rates showed significant Michaelis Menten relationships with chlorophyll a concentration (.5 mm); however, the variability in growth rates explained by chlorophyll a was low (Table III). Furthermore, chlorophyll a concentration failed to be chosen in the best fitted model in the multiple regression analysis. A similar result is found for sac-spawners copepod species in general (Hirst and Bunker, 2003). Originally, Pseudocalanus species were characterized as primarily herbivorous (Corkett and McLaren, 1978), whereas in more recent studies diatoms, flagellates, dinoflagellates, ciliates, heterogeneous particulate matter and even sinking particles can be food source for P. acuspes in the central Baltic Sea (Peters et al., 2006; Renz and Hirche, 2006). Clearly, not all ingested particles contain chlorophyll a, hence although it is the most commonly measured index of food for aquatic grazers, it is an inadequate estimate of total food availability (e.g. Hirst Fig. 8. Comparison of growth rates predicted by global models with measured rates for Pseudocalanus spp. in the northern Gulf of Alaska. Model specifications: Temp, Huntley and Lopez (1992); T&BS, Hirst and Lampitt (1998); all other models in Hirst and Bunker (2003), J-sac, for juvenile sac spawner; A-sac, for adult sac spawners; all, for all data; MM-J-sac, Michaelis Menten relationship for juvenile sac-spawners at 158C; MM-A-sac, Michaelis Menten relationship for adult sac-spawners at 158C; MM-Pseudo, Michaelis Menten relationship for adult Pseudocalanus spp. at 158C. 931

10 Fig. 9. Multi-species composite growth models at 58C for dominant copepods in the northern Gulf of Alaska. and Bunker, 2003). To further complicate matters, laboratory observations suggest that deleterious chemicals in some diatom species are harmful for early survivals as well as egg production of copepods (e.g. Ianora et al., 2003). Although some paradoxical results are observed in the field (Frost, 2005), Pseudocalanus reproduction does have the potential to be impacted by diatom diets (Halsband-Lenk et al., 2005), but it remains unclear if somatic growth can also be impacted. Models In the absence of direct measurements, models become necessary in the estimation of copepod growth and production (i.e. Huntley and Lopez, 1992; Hirst and Lampitt, 1998; Hirst and Bunker, 2003; Bunker and Hirst, 2004). Clearly, any global growth model is limited in its precision for any given species in a specific ecosystem. That is why minor discrepancies, and even large errors, occur when global models are tested under various environments (Richardson et al., 2001; Peterson et al., 2002; Rey-Rassat et al., 2004). In our recent studies, global models on copepod growth rates have been exhaustively tested for other larger dominant species in this area (Liu and Hopcroft, 2006a, b, 2007), and can now be considered for the smaller-bodied and egg-carrying Pseudocalanus species. Generally, global models reasonably match with directly measured rates for Pseudocalanus spp., except for large errors occurring for animals at stage C5 (Fig. 8). The obvious overestimation by the simple temperature-dependent model (Huntley and Lopez, 1992) is due to its food-satiated 932

11 H. LIU AND R. R. HOPCROFT j GROWTH OF PSEUDOCALANUS IN THE GULF OF ALASKA Fig. 10. Validation of composite non-linear models for dominant copepods in the northern Gulf of Alaska. underpinning. Noticeably, two general models of Hirst and Bunker ( juvenile sac-spawners and for all data; Fig. 8) matched well with the directly measured rates, while as expected the model for adult sac-spawners (Hirst and Bunker, 2003) underestimated juvenile growths (Fig. 8). In comparison to the three classical Michaelis Menten models in Hirst and Bunker (Hirst and Bunker, 2003), including a model for adult Pseudocalanus specifically, their models reasonably match our measured rates for early stages, but errors tend to be large for later stages (Fig. 8). The lack of body size as a parameter in their (Hirst and Bunker, 2003) Michaelis Menten models is likely a major cause for these discrepancies, and further refinement of their models is suggested. Global models become a shortcut to the complexities and uncertainties in experimental study through the use of a few easily measured variables (i.e. temperature, chlorophyll a and body size) into a mathematical function, but may not adequately capture the confounding effects between temperature, food source and body size on growth of marine copepods (Vidal, 1980; Hopcroft et al., 1998; Hirst and Bunker, 2003; Dzierzbicka-Glowacka, 2004; Liu and Hopcroft, 2006b, 2007). To disentangle these synergistic effects, we developed the composite non-linear model (Liu and Hopcroft, 2006b). This model exhibited higher explanatory power in growth rates of Pseudocalanus spp. from both chlorophyll a and body size than by chlorophyll a alone (Table III), consistent with results for other copepod species in this area (Liu and Hopcroft, 2006b, 2007). In this study, further addition of stage to the composite model had a profound impact on the utility of the model (Table III). In this four-parameter model, developmental stage now takes the negative slope as we observed in the univariate analysis, and the sign of the body-weight parameter changes from the negative across-stage pattern to encompass the positive within-stage relationship discussed previously for Pseudocalanus as well as for other local species (Liu and Hopcroft, 2006a, b, 2007). Nonetheless, the fourparameter model continues to illustrate the poor utility of chlorophyll in describing growth rate in contrast to the other broadcast spawning local species (ibid; Slater and Hopcroft, 2005). Thus far, the three-parameter composite model shows a great promise for describing the growth rates of each of the dominant copepods in this ecosystem (Figs 9 and 10). Through its use, it is clear that although the broadcast spawning species are comparable in their rates of growth, the egg-carrying Pseudocalanus clearly has lower somatic growth. This later conclusion is consistent with the recent generalization from global growth rates syntheses that broadcaster-spawners and egg-carriers differ notably in their rates of somatic growth (Hirst and Lampitt, 1998; Hirst and Bunker, 2003). We suggest that further refinement of global models, by incorporation of non-linear relationship such as allowed in the composite model, is still required to improve their parameterization and utility. ACKNOWLEDGEMENTS We thank the captain and crew of the R/V Alpha Helix and R/V Wecoma as well as Amanda Byrd, Mike Foy and Alexei Pinchuk for assistance in experimental setup and execution at sea. Alexei Pinchuk also provided invaluable assistance by terminating experiments still running post-cruise. Cheryl Clarke provided significant laboratory support. Dean Stockwell and Terry Whitledge kindly provided ambient chlorophyll a concentrations from the GLOBEC LTOP program. FUNDING This is contribution number 587 of the US GLOBEC program, jointly funded by the National Science Foundation and the National Oceanic and Atmospheric Administration under NSF Grant OCE REFERENCES Anderson-Sprecher, R. (1994) Model comparisons and R 2. Am. Statist., 48,

12 Båmstedt, U. (1986) Chemical composition and energy content. In Corner, E. D. S. and O Hara, S. C. M. (eds), The Biological Chemistry of Marine Copepods. Clarendon, pp Ban, S., Lee, H.-W., Shinada, A. et al. (2000) In situ egg production and hatching success of the marine copepod Pseudocalanus newmani in Funka Bay and adjacent waters off southwestern Hokkaido, Japan; associated to diatom bloom. J. Plankton Res., 22, Bunker, A. J. and Hirst, A. G. (2004) Fecundity of marine planktonic copepods: global rates and patters in relation to chlorophyll a, temperature and body weight. Mar. Ecol. Prog. Ser., 279, Cooney, R. T. (1986) Zooplankton. In Hood, D. W. and Zimmerman, S. T. (eds), The Gulf of Alaska: Physical Environment and Biological Resources. U.S. Government Printing Office, Washington, pp Cooney, R. T., Coyle, K. O., Stockmar, E. et al. (2001) Seasonality in surface-layer net zooplankton communities in Prince William Sound, Alaska. Fish. Oceanogr., 10(Suppl. 1), Corkett, C. J. and McLaren, I. A. (1978) The Biology of Pseudocalanus. Adv. Mar. Biol., 15, Coyle, K. O. and Pinchuk, A. I. (2002) Climate-related differences in zooplankton density and growth on the inner shelf of the southeastern Bering Sea. Prog. Oceanogr., 55, Coyle, K. O. and Pinchuk, A. I. (2003) Annual cycle of zooplankton abundance, biomass and production on the northern Gulf of Alaska shelf, October 1997 through October Fish. Oceanogr., 12, Coyle, K. O. and Pinchuk, A. I. (2005) Cross-shelf distribution of zooplankton relative to water masses on the northern Gulf of Alaska shelf. Deep-Sea Res. II, 52, Dagg, M. J., Clarke, M. E., Nishiyama, T. et al. (1984) Production and standing stock of copepod nauplii, food items for larvae of the walleye Pollock, Theragra chalcogramma, in the southeastern Bering Sea. Mar. Ecol. Prog. Ser., 19, Davis, C. S. (1984) Food concentrations on Georges Bank: nonlimiting effect on development and survival of laboratory reared Pseudocalanus sp. & Paracalanus parvus (Copepoda: Calanoida). Mar. Biol., 82, Dzierzbicka-Glowacka, L. (2004) Growth and development of copepodite stages of Pseudocalanus spp. J. Plankton Res., 26, Frost, B. W. (1985) Food limitation of the planktonic marine copepods Calanus pacificus and Pseudocalanus sp. in a temperate fjord. Arch. Hydrobiol. Beih. Ergeb. Limnol., 21, Frost, B. W. (1989) A taxonomy of the marine calanoid copepod genus Pseudocalanus. Can. J. Zool., 67, Frost, B. W. (2005) Diatom blooms and copepod responses: paradigm or paradox? Prog. Oceanogr., 67, Gómez-Gutiérrez, J. and Peterson, W. T. (1999) Egg production rates of eight calanoid copepod species during the summer 1997 off Newport Oregon, USA. J. Plankton Res., 21, Halsband-Lenk, C., Pierson, J. J. and Leising, A. W. (2005) Reproduction of Pseudocalanus newmani (Copepoda: Calanoida) is deleteriously affected by diatom blooms a field study. Prog. Oceanogr., 67, Hilgruber, N., Haldorson, L. J. and Paul, A. J. (1995) Feeding and selectivity of larval walleye pollock Theragra chalcogramma in the oceanic domain of the Bering Sea. Mar. Ecol. Prog. Ser., 120, Hirst, A. G. and Lampitt, R. S. (1998) Toward a global model of in situ weight-specific growth in marine planktonic copepods. Mar. Biol., 132, Hirst, A. G. and Bunker, A. J. (2003) Growth of marine planktonic copepods: global rates and patterns in relation to chlorophyll a, temperature, and body weight. Limnol. Oceanogr., 48, Hirst, A. G., Peterson, W. T. and Rothery, P. (2005) Errors in juvenile copepod growth rate estimates are widespread: problems with the Moult Rate method. Mar. Ecol. Prog. Ser., 296, Hopcroft, R. R. and Roff, J. C. (1998) Zooplankton growth rates: the influence of female size and resources on egg production of tropical marine copepods. Mar. Biol., 132, Hopcroft, R. R., Roff, J. C., Webber, M. K. et al. (1998) Zooplankton growth rates: the influence of size and resources tropical marine copepodites. Mar. Biol., 132, Hopcroft, R. R., Clarke, C., Byrd, A. G. et al. (2005) The paradox of Metridia spp. Egg production rates: a new technique and measurements from the coastal Gulf of Alaska. Mar. Ecol. Prog. Ser., 286, Hopcroft, R. R., Napp, J. M., Clarke, C. et al. Egg production rates of Pseudocalanus mimus, Pseudocalanus minutus and Pseudocalanus newmani in the Gulf of Alaska. Huntley, M. E. and Lopez, M. D. G. (1992) Temperature-dependent production of marine copepods: a global synthesis. Am. Nat., 140, Ianora, A., Poulet, S. and Miralto, A. (2003) The effects of diatoms on copepod reproduction: a review. Phycologia, 42, Incze, L. S., Siefert, D. W. and Napp, J. M. (1997) Mesozooplankton of Shelikof Strait, Alaska: abundance and community composition. Cont. Shelf Res., 17, Kendall, A. W., Jr and Nakatani, T. (1991) Comparisons of early life history characteristics of walleye Pollock Theragra chalcogramma in Shelikof Strait, Gulf of Alaska, and Funka Bay, Hokkaido, Japan. Fish. Bull., 90, Kiørboe, T. and Sabatini, M. (1994) Reproductive and life cycle strategies in egg-carrying cyclopoid and free-spawning calanoid copepods. J. Plankton Res., 16, Kiørboe, T. (1997) Population regulation and role of mesozooplankton in shaping marine pelagic food webs. Hydrobiologia, 363, Kimmerer, W. J. and McKinnon, A. D. (1987) Growth, mortality, and secondary production of the copepod Acartia tranteri in Westernport bay, Australia. Limnol. Oceanogr., 32, Kimmerer, W. J., Hirst, A. G., Hopcroft, R. R. et al. (2007)Measurement of Juvenile copepod growth rates: corrections, inter-comparisons and recommendations. Mar. Ecol. Prog. Ser., 336, Lee, H. -W., Ban, S., Ikeda, T. et al. (2003) Effect of temperature on development, growth and reproduction in the marine copepod Pseudocalanus newmani at satiating food condition. J. Plankton Res., 25, Liu, H. and Hopcroft, R. R. (2006a) Growth and development of Neocalanus flemingeri/plumchrus in the northern Gulf of Alaska: validation of the artificial cohort method in cold waters. J. Plankton Res., 28, Liu, H. and Hopcroft, R. R. (2006b) Growth and development of Metridia pacifica (Copepoda: Calanoid) in the northern Gulf of Alaska. J. Plankton Res., 28, Liu, H. and Hopcroft, R. R. (2007) A comparison of seasonal growth and development of Calanus marshallae and C. pacificus in the northern Gulf of Alaska. J. Plankton Res., 29, Mauchline, J. (1998) The Biology of Calanoid Copepods. Advances in Marine Biology, Vol. 33. Academic Press, London. 934

13 H. LIU AND R. R. HOPCROFT j GROWTH OF PSEUDOCALANUS IN THE GULF OF ALASKA McLaren, I. A. (1974) Demographic strategy of vertical migration by a marine copepod. Am. Nat., 108, McLaren, I. A. (1978) Generation lengths of some temperate marine copepods: estimation, prediction, and implications. J. Fish. Res. Bd. Can., 35, McKinnon, A. D. (1996) Growth and development in the subtropical copepod Acrocalanus gibber. Limnol. Oceanogr., 41, Napp, J. M., Hopcroft, R. R., Baier, C. T. et al. (2005) Distribution and species-specific egg production of Pseudocalanus in the Gulf of Alaska. J. Plankton Res., 27, Paul, A. J., Coyle, K. O. and Ziemann, D. A. (1990) Variations in egg production rates by Pseudocalanus spp. in a subarctic Alaskan bay during the onset of feeding by larval fish. J. Crustac. Biol., 10, Peters, J., Renz, J., van Beusekom, J. et al. (2006) Trophodynamics and seasonal cycle of the copepod Pseudocalanus acuspes in the Central Baltic Sea (Bornholm Basin): evidence from lipid composition. Mar. Biol., 149, Peterson, W. T., Tiselius, P. and Kiørboe, T. (1991) Copepod egg production, moulting and growth rates, and secondary production, in the Skagerrak in August J. Plankton Res., 13, Peterson, W. T., Gómez-Gutiérrez, J. and Morgan, C. A. (2002) Cross-shelf variation in calanoid copepod production during summer 1996 off the Oregon coast, USA. Mar. Biol., 141, Renz, J. and Hirche, J. (2006) Life cycle of Pseudocalanus acuspes Giesbrecht (Copepoda, Calanoida) in the Central Baltic Sea: I. Seasonal and spatial distribution. Mar. Biol., 148, Renz, J., Peters, J. and Hirche, J. (2007) Life cycle of Pseudocalanus acuspes Giesbrecht (Copepoda, Calanoida) in the Central Baltic Sea: II. Reproduction, growth and secondary production. Mar. Biol., 151, Rey-Rassat, C., Bonnet, D., Irigoien, X. et al. (2004) Secondary production of Calanus helgolandicus in the Western English Channel. J. Exp. Mar. Biol. Ecol., 313, Richardson, A. J., Verheye, H. M., Herbert, V. et al. (2001) Egg production, somatic growth and productivity of copepods in the Benguela Current system and Angola-Benguela Front. S. Afr. J. Sci., 97, Roff, J. C. and Hopcroft, R. R. (1986) High precision microcomputer based measuring system for ecological research. Can. J. Fish. Aquat. Sci., 43, Runge, J. A. and Roff, J. C. (2000) The measurement of growth and reproductive rates. In Harris, R. P., Weibe, P. H., Lenz, J., Skjoldal, H. R., Huntley, M. et al.(eds), ICES Zooplankton Methodology Manual. Academic Press, London, pp Schabetsberger, R., Sztatecny, M., Drozdowski, G. et al. (2003) Size dependent, spatial, and temporal variability of juvenile walleye Pollock (Theragra chalcogramma) feeding at a structural front in the southeast Bering Sea. P.S.Z.N. Mar. Ecol., 24, Siefert, D. L. W. (1994) The importance off sampler mesh size when estimating total daily egg production by Pseudocalanus spp. in Shelikof Strait, Alaska. J. Plankton Res., 16, Slater, L. M. and Hopcroft, R. R. (2005) Development, growth and egg production of Centropages abdominalis in the eastern subarctic Pacific. J. Plankton Res., 27, Vidal, J. (1980) Physioecology of Zooplankton. I. Effects of phytoplankton concentration, temperature, and body size on the growth rate of Calanus pacificus and Pseudocalanus sp. Mar. Biol., 56, Vidal, J. and Smith, S. L. (1986) Biomass, growth and development of populations of herbivorous zooplankton in the southeastern Bering Sea during spring. Deep-Sea Res., 33, Weingartner, T. J., Coyle, K. O., Finney, B. et al. (2002) The Northeast Pacific GLOBEC Program: Coastal Gulf of Alaska. Oceanography, 15, Yamaguchi, A., Ikeda, T. and Shiga, N. (1998) Population structure and life cycle of Pseudocalanus minutus and Pseudocalanus newmani (Copepoda: Calanoida) in Toyama Bay, southern Japan Sea. Plankton Biol. Ecol., 45,