Modeling Climate Change Impacts on Alaskan Fisheries Kirstin Holsman [email protected] JISAO UW/AFSC NOAA Seattle, WA, USA K Aydin, G Gibbson, C Hansen, A Hermann, S Hjøllo, JA Jacobsen, G Oskarsson, AS Samuelsen, M Skern-Mauritzen, M Sigler, KR Utne, E Vendsen

Photo: Mark Holsman

IPCC Projections of Temperature in the 21st Century (delta T [1971-2000] ) NASA Center for Climate Simulation & Goddard Institute for Space Studies. 2013

5th IPCC Assessment Report (AR5)

Photo: Mark Holsman

5th IPCC Assessment Report (AR5)

4

Direct Climate Change Impacts?

(“easy” to model)

Indirect Climate Change Impacts? (not so “easy” to model)

Photo: Mark Holsman

Climate Change & Fisheries

Pinsky et al. (2013) Marine Taxa Track Local Climate Velocities. Science (13): 1239-1242 Cheung et al. (2013). Fisheries: Climate change at the dinner table. Nature 497, 365-368 6

Change in max catch potential: 2055-2005

Cheung et al. (2010). Global Change Biology (2010) 16: 24–35 Graphic: IPCC AR5 WII TSU 7

Barange et al. 2014. Impacts of climate change on marine ecosystem production in societies dependent on fisheries. Nature Climate Change (4): 212-216. 8

Present Change (1901-2012)

9

Arctic Sea Ice: September 2012

Image: NASA Earth Observatory image by Jesse Allen Data: National Snow and Ice Data Center

10

Arctic Sea Ice Volume Anomaly, v2

Image: Polar Science Center at the University of Washington

11

AR5 IPCC Projections: Sea Ice

12

Bering Sea & Climate Bering Sea “Cold Pool” 2001-2009

Winter

Summer

Graphic: J. Overland, P. Stabeno, M. Wang, C. Ladd, N. Bond, and S. Salo, PMEL/NOAA

13

EBS and Climate Predation is stronger in warm years (Coyle et al. 2011)

Recruitment & survival decline with increasing temperatures (Mueter et al. 2011, Coyle et al. 2011)

Temperature

Predators

Mueter et al. 2011

Coyle et al. 2011

14

Sources of Error 1. Observation error • • • •

Measurement error Spatial heterogeneity Temporal variability Reduce through replication

2. Process error • • •

“Noise” due to environmental variability Can be recreated using climate models Random simulation can get “avg” trend right

3. Model misspecification error • • • • •

Often result of spurious correlations Under or over estimate interactions More likely with indirect effects Experimental manipulation can reduce error Avg. from multiple models can help reduce error (“multi-model inference”)

15

high spatial & temporal resolution

IPCC

ROMS (A. Hermann & N. Bond)

NPZ (G. Gibson & A. Hermann)

low spatial & temporal resolution

MSMt (K. Holsman, K. Aydin, & J. Ianelli)

16

EBS & Climate Change “Future” Change in Ice Coverage: 2031-2040

CCMA

MIROC

ECHOG

“Future” Large Zoop. (mgC m-3): 2031-2040

CCMA

MIROC

ECHOG Slide Credit: Albert Hermann 17

0.7

0.6

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Cold pool % of survey area 0.8

ROMS

Data

0.5

0.4

0.3

0.2

0.1

0

18

high spatial & temporal resolution

IPCC

ROMS (A. Hermann & N. Bond)

NPZ (G. Gibson & A. Hermann)

low spatial & temporal resolution

MSMt (K. Holsman, K. Aydin, & J. Ianelli)

19

MSMt Recruitment

ROMS/ NPZ Indices

Mean R/S + Food Mean R/S

Mean R/S + Food Competition recruitment

productivity

carrying capacity environmental effects on carrying capacity 20

Recruitment: multi-species

Take home Correlation is consistent across models

Small Copepod Biomass

21

Recruitment: single-species

Take home Temp. covariate may be ok when recruitment success is direct function of environment (e.g., cross shelf transport) SST in fall

22

Multi-spp. + covar.

Single spp.+ covar.

Multi-spp.

Single spp.

Change in recruitment ROMS/ NPZ Indices

23

Multi-spp. + covar.

Single spp.+ covar.

Multi-spp.

Single spp.

Multi-spp. + covar.

Single spp.+ covar.

Multi-spp.

Single spp.

Change in recruitment Change in catch

24

Fisheries Management Implications Misspecification under-harvest preds & overharvest prey Include bottom-up and top-down dynamics Management > Recruitment >> Climate Manage for % B0 under future climate conditions

Take away 1. Indirect & direct effects of climate change (model types?)

2. Propagation of 3 types of error (how would you control error?)

3. Management implications (how to plan for uncertainty?)

Photo: Mark Holsman

Metric 1

Biomass limit

27

5th IPCC Assessment Report (AR5)

28