docAtlantis Modelling of the Upper Gulf of California: EBM support for
download 
Report Transcription
Atlantis Modelling of the Upper Gulf of California: EBM support for
Cameron Ainsworth Northwest Fisheries Science Center (NOAA) Seattle, WA PICES Annual Meeting. Portland, OR. Oct. 2010 Samhouri, J.F., Ainsworth, C.H., Busch, D.S., Cheung, W.L., Okey, T.A. West Coast Vancouver Island Aquatic Management Board Trilogy of papers How will climate change impact marine ecosystems in the NE Pacific? NE Pacific study area • Prince William Sound (PWS) • Southeast Alaska (SEA) • Northern British Columbia (NBC) • West Coast Vancouver Island (WCVI) • Northern California Current (NCC) Review of 1st paper • 5 climate effects simulated using increased or decreased organism productivity (Ecosim forcing functions) • 3 levels of severity tested • Impacts tested alone and in combination (additive) Ocean acidification Species range shifts Cheung et al. 2009 (Fish and Fisheries) Plankton size structure Deoxygenation Primary productivity Ecopath with Ecosim Model Northern British Columbia Northern California Current Prince William Sound Southeast Alaska West Coast Vancouver Island # groups 53 65 48 40 15 Reference Ainsworth 2007 Field et al. 2006 Okey and Pauly 1999 Guenette et al. 2006 Martell 2002 Polovina, J.J., 1984. Coral Reefs, 3 (1): 1–11. ··· Christensen, V., Pauly, D., 1992.. Ecological Modeling, 61: 169–185. ··· Walters, C., Christensen, V., Pauly, D. 1997. Reviews in Fish Biology and Fisheries, 7: 139-172. Ecopath with Ecosim n n dBi = c g i ∑ f (B j , Bi ) − ∑ f (Bi , B j ) + I i − Bi (M i + Fi + ei ) dt j =1 j =1 Production Predation losses Fisheries and other mortality Ecopath master equations n Mass balance: Bi ⋅ ( P / B )i = Yi + ∑ B j ⋅ (Q / B) j ⋅ DC ji + Ei + BAi + Bi ( P / B) i ⋅ (1 − EEi ) j =1 Conservation of energy: B ⋅ (Q / B) = B ⋅ ( P / B) + (1 − GS ) ⋅ Q − (1 − TM ) ⋅ P + B ⋅ (Q / B) ⋅ GS Ecopath: Polovina, J.J., 1984. Coral reefs, 3(1): 1-11. Christensen, V. and D. Pauly, 1992. Ecological Modeling, 61: 169-185. Ecosim: Walters, C., V. Christensen and D. Pauly, 1997. Reviews in Fish Biology and Fisheries, 7: 139-172. Ecopath with Ecosim n n dBi = c g i ∑ f (B j , Bi ) − ∑ f (Bi , B j ) + I i − Bi (M i + Fi + ei ) dt j =1 j =1 Annual scalar multiplier of biological production Dissolved oxygen forcings (NBC) Range shift forcings (NBC) 1st paper: Individual climate effects Range shifts bring rockfish north and reduce competition for plankton Ocean acidification not good for calcified organisms 1st paper: Cumulative effects Less catch overall More low-value species jellyfish dogfish ratfish Doesn’t include human opportunism Now the current study Do community interactions change our predictions? Range shifts • Used bioclimatic envelope model (Cheung et al. 2009) to predict relative abundance of fished species in 0.5° x 0.5° grid • Predicts distribution based on species’ environmental preferences • Based on ocean temperature, salinity from GFDL coupled model 2.1 2010 2060 0 0 Low > Relative abundance > > > > > > > High > Pacific hake How we simulated range shifts Bioclimatic envelope model predictions Biomass (a focal species) Trophic interactions change the ‘single species’ trajectory Recreated using forcing functions All range shifts simultaneously Year Range shifts * Range shifts in Northern BC Did we get it right? Relative abundance 2.5 2.0 1.5 1.0 0.5 0.0 2000 2010 2020 2030 2040 2050 Year *Predicted by Cheung et al (2009) model in Ainsworth et al (2010) Coho salmon Chinook salmon Squid Ratfish Dogfish Forage fish Eulachon Adult herring Adult POP Rockfish Flatfish Adult halibut Adult Pacific cod Adult sablefish Adult lingcod Shallowwater benthic fish Skates Large crabs Commercial shrimp Some species increase, some decrease Range shifts Did we get it right? Not bad No change 22% Increase 17% Decrease 61% • Cheung’s single species methodology is recreated here by holding other groups static (no trophic interactions) Range shifts small pelagics rockfish No change 22% Increase 17% pelagic invertebrates large pelagics NBC flatfish elasmobranchs Decrease 61% demersals benthic invertebrates -40 -20 0 20 40 60 % difference vs. baseline Single species trajectory NCC SEA Range shifts small pelagics rockfish No change No change 22% 36% Increase 17% Increase pelagic invertebrates 35% large pelagics NBC flatfish elasmobranchs Decrease 29% 61% demersals benthic invertebrates -40 -20 0 20 40 60 % difference vs. baseline Now with all range shifts NCC SEA Findings • Community interactions might moderate range shifts across a wide range of exploited functional groups Ocean acidification Functional groups affected L S S Northern British Columbia Commercial shrimp Epifaunal invertebrates Euphausiids L S S L Northern California Current Benthic shrimp Euphausiids Infaunal invertebrates Pandalid shrimp S L Southeast Alaska Carnivorous epibenthic invertebrates Shrimps 2 Productivity Effect size 1870 1870 Linear forcing to -10% or -50% S 1 Aragonite Aragonite saturation saturation 2060 2060 ND ND Low Low L 0 Marginal Marginal Adequate Adequate Optimal Optimal 2010 2035 2060 Year 2008 Guinotte & Fabry Guinotte & Fabry 2008 Guinotte, J.M, and Fabry, V.J., 2008. Ocean acidification and its potential effects on marine ecosystems. In: Ostfeld, R.S., Schlesinger, W.H. (Eds.), The Year in Ecology and Conservation Biology 2008, Annals of the New York Academy of Sciences, pp. 320–342. Ocean acidification Manipulated groups only No change 39% Decrease 61% (manipulated groups and other groups) Single species trajectory Manipulated groups only Ocean acidification Manipulated groups only No change 39% Manipulated groups only Decrease 61% (manipulated groups and other groups) Now with all OA effects Multispecies effects reduce impacts on invertebrates relative to single-species expectations Ocean acidification Benthos under multispecies impacts 10.00% 4 Biomass (t/sqkm) Groups that depend on benthos p<0.01 8.00% 3 6.00% 2 4.00% 1 2.00% 0 0.00% p~0.2 p<0.01 ‐2.00% ‐1 ‐4.00% ‐2 ‐6.00% NBC NCC SEA NBC NCC SEA Multispecies OA Theseimpacts groups alleviate are responding, effects on butbenthos not how you would expect Makes them worse Findings • Community interactions may reduce OA impacts on affected invertebrates • Effects do not cascade up the food web as expected Primary production NCC NBC 15% 20% Anomaly 20% 10% 10% 5% 0% 0% ‐10% ‐5% ‐20% ‐10% 2010 2020 2030 2040 2050 2060 Year 2010 2020 2030 2040 2050 2060 Year y 30% 20% 15% 10% 5% 0% ‐5% ‐10% ‐15% SEA 2010 2020 2030 2040 2050 2060 Year • Geophysical Fluid Dynamics Laboratory (GFDL) Earth System Model (ESM2.1) • Includes Tracers of Ocean Phytoplankton with Allometric Zooplankton (TOPAZ) model of ocean ecosystems and biogeochemical cycles • Simulations based on the IPCC AR4 protocols (SRES A1B). Primary production NCC model 80 phytoplankton 70 Vs. GFDL predictions, variability of Phytoplankton variability decreased phytoplankton has increased 6% 15% 60 50 zooplankton 40 Zooplankton variability decreases 8% 30 20 10 0 2010 2020 2030 2040 2050 Compensatory feeding dynamics by high TLs reduce variation in plankton Zooplankton 4 month lag with phytoplankton (R = 0.74) NCC:show Zooplankton variability decreases 8% NBC: Zooplankton variability decreases 1% Foraging arena: Walters, C.J. and F. Juanes, 1993. CJFAS, 50: 2058-2070. Primary production Findings • Multispecies interactions increase variability of phytoplankton • Higher TL interactions reduce variability of prey Summary Do community interactions change our predictions? • Range shifts and ocean acidification impacts are moderated by multispecies interactions • Variability of primary production increases Thank you