what do global climate models say about increasing variance in the california current upwelling ecosystem

what do global climate models say about increasing variance in the california current upwelling ecosystem marisol garcía-reyes farallon institute for advanced ecosystem research william sydeman, ryan rykaczewski, allison wiener, isaac schroeder & steven bograd PICES \ 2013 Annual Meeting Oct 11-20 \ Nanaimo, BC, Canada

king et al. 2011 (from agostini 2005) california current upwelling ecosystem

Global Change Biology (2013) 19, 1662 1675, doi: 10.1111/gcb.12165 Increasing variance in North Pacific climate relates to unprecedented ecosystem variability off California WILLIAM J. SYDEMAN*, JARROD A. SANTORA*, SARAH ANN THOMPSON*, BALDO MARINOVIC and EMANUELE DI LORENZO

ecosystem sensitivity to winter upwelling variability Biological Principal Component Environmental Principal Component sst & winds egg lay-date, reproductive success, growth garcía-reyes et al., 2013 cordell bank website

choosing the indicators

climate indicators driving ecosystem variability upwelling: winds and temperature winter time

upwelling winds driver + winter + H Meridional wind speed (m/s) + Latitude H + L summer Longitude Bograd et al., 2002

pcui (m3 s 1 100 Area (x106 km2) a 42 39 36 0.4 33 48.3 46.2 44.6 43.3 0.4 0.4 37.8 36.6 35.1 32.7 48.5 46.5 44.5 42.5 4 38.5 36.5 34.5 0.6 0.7 0.6 0.6 0.7 PC1bio (normalized) Area (x106 km2) 1000 2 PC1bio (normalized) 45 48.3 46.2 44.6 43.3 41.7 3!! January February means0 of x, y, A, and p speciﬁcally, a 0.7 max with pairwise 0.6 1970 1975 1980 1985 1990 1995 20000 2005 2010 0.8 0.65, 0.7 of!4, 0.78, and 0.80.78, respectively. Sr correlations model (r2 = 0 0.7 Year 0.7 0.8 Of the four NPH variables, the January February mean of A wintertime N 1 high ﬂedglin hasfigure the best ﬁt with pcui. Thus, area is the only variable 3. Time series and bivariate plots between the preconditioning retained when comparing to atmospheric patterns and bi-, growth, all po the leading January February meannph of the NPH s area and (c, d) PC1 Schroeder et al., 2013 bio 2 1 = 7, p < 0.01) best ological production. A linear model (r of productivit 0.6 0.4 0.6 sensitive to wintertime upwelling. The black line in the bi indices that are describes the relationship 0.6 between pcui and 0.4the January February mean of A (Figure 3b). Similar relationships between n 48 1 2000 Correlation 33 pcui (m3 s 1 100 m 1) pcui (m3 s 1 100 m 1) pcui (m3 s 1 100 m 1) winter upwelling and the NPH 3 1 3000 6 amplitu terms ofð! latitudevariability ð!yþ shows more variability averages NPH position [9] Monthly ositioning, latitude ypositioning, Þ shows more than ] Monthly averages ofofnph positionand and am [9than longitude, especially during winter. Thus, the NPH center signiﬁcantly correlate with monthly averages of UI, sea lev especially during winter.as far Thus, center signiﬁcantly correlate of stronge UI, s # 2000 N or as far north as 36# N and SST can be located souththe as 23NPH over the with coursemonthly of the year.averages However,4 the # # orbeaslocated far north as 36 N (126 cated as far south as 23 and# W) SSTcorrelations over the among coursethese of the year.occur However, s in winter; or itncan in a coastal position variables between the Janua # 1000 2 The seasonal the four and March or fartherin offshore (157 W). (Figure 2). When the wintertime is locat W) of correlations or it can be located a coastal position (126#cycle among these variables occur NPH between variables!x,!y, A, and pmax explain 38.0%, 51.3%, to the northwest and has larger amplitude coastal upwellin # W). seasonal cycle of variance, the fourrespectively. offshore (157NPH and0march (Figureand 2). coastal Whenseathe wintertime is 0 NPH Tim 34.5%, The and 25.8% of the overall is enhanced, level and SST decline. SCHROEDER ET AL.: NPH 51.3%, ables!x,!y, A, andthepseasonal to with the northwest and amplitude coastal up in January are signiﬁcantly correlat When signals 38.0%, are removed, all correlate series of!y, A, andhas pmax larger max explain in February and Marc with theand time coastal series of!ysea, A, and pmax and (n = 528, p < 0.01); longitude ð!xþ negatively nd 25.8% of one theanother overall variance, respectively. is4000 enhanced, level SST decline 1 8 signiﬁca 4000 c athus, NPH amplitude and latitudinal position show relates to!y, A, and pmax (Sr = $0.22, $0.32, and $0.44,! in January are signiﬁcantly co seasonal signals are removed, all correlate with series of y, A, and p max respectively) such that the NPH center tends to be farther correlations from January into early spring. So, when the NP 3000 0 6 upwellin 3000 February and with the istime series of!y,ita, and topmax her (n = 528, pnorth, < 0.01); longitude ð!xþpressure, negatively have higher maximum and a larger area strong in January, is likely lead in to enhanced when located farther fromand land$0.44, (to the west). TheNPH in early spring. and latitudinal position show sig!y, A, and pmax (Srit =is $0.22, $0.32, Thus, amplitude 2000 2000 N 4ranges fro highest correlation among variables is between A and pmax [10] Our pre-conditioning index (pcui) at 39# 1 ly) such that (Sthe= NPH centerthantends to be farther correlations from 3 $1 January $1into early spring.3 So, $1 when t! 0.81), greater those between y and the amplitude s to 100 m in 1983 to 4153 m s to 100 m 223 m r 1000 2 1000 2 ve higher maximum and a larger area is stronginin2007 January, is likely to lead to enhanced up (Figureit 3a). The pcui correlates (p < 0.01) variables (Spressure, r = 0.31 for A and 0.43 for pmax). is located farther from land (to the west). The in early spring. 0 0 30 1980 1985 1990 index 1995 2000 2005 at 2010 Our1975 pre-conditioning (pcui) 39# N rang orrelation among variables is between A and pmax [10] 1970 3 $1 $1Year 3 $1 48!y and the amplitude 1), greater than those between s to 100 m in 1983 to 4153 m s to 1 223 m 0.6 45 4000 1 0.7 2007c (Figure 3a). The pcui correlates (p < 0 p ). in (Sr = 0.31 for A and 0.43 for max 42 Figure 3. Time series and0.6 bivariate plots between the preconditioning cu 0.4 Longitude 39 Latitude Area Maximum 0.7 3000 0 0.7, the leading prin January February mean of0.7the NPH s0.7area and (c, d) PC1 bio 0.8 36 0.4 0.4 pressure indices that are sensitive to wintertime upwelling. The black 1 line in the biva

climate indicators driving ecosystem variability upwelling: winds and temperature winter time

climate indicators driving ecosystem variability upwelling: winds and temperature sea level pressure winter time spatial and temporal scales adequate for the use of global climate models

Alongshore wind 43 Cape Blanco change in variance 42 N27 41 N22 California Upwelling Variability i Cape Mendocino 40 Latitude ( o N) 39 38 37 N14 Point Arena N13 Bodega Bay Point Reyes San Francisco Bay Gulf of Farallones N26 Point Ano Nuevo N12 N42 Monterrey Bay 36 Point Sur N28 35 N11 past Point Conception!125!124!123!122!121!120 Longitude ( o W) future García-Reyes & Sydeman. 2012, PICES Anomalies in CUI Macias et al. 2012 Figure 4. Analysis of the annual signal at 396N. a) Climatologic (gray) and SSA (black) annual cycles of the CUI difference between both annual cycles. doi:10.1371/journal.pone.0030436.g004 coherent along the coast, indicating that the processes governing the dynamics of this signal could reverse phase or be altered from of this region to determine the areas tha each climatic index (as in [52]). We also lo

mechanism for increased variance unknown natural pacific climate oscillations anthropogenic climate change regional changes

how to test a change in variance long time series, with enough temporal resolution past future regional models global models local processes large scale ones are prescribed global change & large scale processes no local processes

mechanism for increased variance unknown natural pacific climate oscillations anthropogenic climate change regional changes GCM models IPCC AR5 (CMIP5)

IPCC AR5 - CMIP5 38 models output, 21 different models RCP8.5 period: 2006-2095 Modeling Center (or Group) Institute ID Model Name Commonwealth Scientific and Industrial Research Organization (CSIRO) and Bureau of Meteorology (BOM), Australia Beijing Climate Center, China Meteorological Administration Instituto Nacional de Pesquisas Espaciais (National Institute for Space Research) College of Global Change and Earth System Science, Beijing Normal University Canadian Centre for Climate Modelling and Analysis CSIRO-BOM BCC ACCESS1.0 ACCESS1.3 BCC-CSM1.1 BCC-CSM1.1(m) INPE BESM OA 2.3 * GCESS CCCMA BNU-ESM CanESM2 CanCM4 CanAM4 University of Miami - RSMAS RSMAS CCSM4(RSMAS)* National Center for Atmospheric Research NCAR CCSM4 Community Earth System Model Contributors Center for Ocean-Land-Atmosphere Studies and National Centers for Environmental Prediction Centro Euro-Mediterraneo per I Cambiamenti Climatici Centre National de Recherches Météorologiques / Centre Européen de Recherche et Formation Avancée en Calcul Scientifique Commonwealth Scientific and Industrial Research Organization in collaboration with Queensland Climate Change Centre of Excellence NSF-DOE- NCAR COLA and NCEP CMCC CNRM- CERFACS CSIRO-QCCCE CESM1(BGC) CESM1(CAM5) CESM1(CAM5.1,FV2) CESM1(FASTCHEM) CESM1(WACCM) CFSv2-2011 CMCC-CESM CMCC-CM CMCC-CMS CNRM-CM5 CNRM-CM5-2 CSIRO-Mk3.6.0 LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences LASG-IAP FGOALS-gl FGOALS-s2 The First Institute of Oceanography, SOA, China FIO FIO-ESM NASA Global Modeling and Assimilation Office NASA GMAO GEOS-5 NOAA Geophysical Fluid Dynamics Laboratory NASA Goddard Institute for Space Studies National Institute of Meteorological Research/Korea Meteorological Administration Met Office Hadley Centre (additional HadGEM2-ES realizations contributed by Instituto Nacional de Pesquisas Espaciais) NOAA GFDL NASA GISS NIMR/KMA MOHC (additional realizations by INPE) GFDL-CM2.1 GFDL-CM3 GFDL-ESM2G GFDL-ESM2M GFDL-HIRAM-C180 GFDL-HIRAM-C360 GISS-E2-H GISS-E2-H-CC GISS-E2-R GISS-E2-R-CC HadGEM2-AO HadCM3 HadGEM2-CC HadGEM2-ES HadGEM2-A Institute for Numerical Mathematics INM INM-CM4 Institut Pierre-Simon Laplace Japan Agency for Marine-Earth Science and Technology, Atmosphere and Ocean Research Institute (The University of Tokyo), and National Institute for Environmental Studies Atmosphere and Ocean Research Institute (The University of Tokyo), National Institute for Environmental Studies, and Japan Agency for Marine-Earth Science and Technology Max-Planck-Institut für Meteorologie (Max Planck Institute for Meteorology) Meteorological Research Institute IPSL MIROC MIROC MPI-M MRI IPSL-CM5A-LR IPSL-CM5A-MR IPSL-CM5B-LR MIROC-ESM MIROC-ESM-CHEM MIROC4h MIROC5 MPI-ESM-MR MPI-ESM-LR MPI-ESM-P MRI-AGCM3.2H MRI-AGCM3.2S MRI-CGCM3 MRI-ESM1 EC-EARTH consortium EC-EARTH EC-EARTH LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences and CESS,Tsinghua University LASG-CESS FGOALS-g2 Nonhydrostatic Icosahedral Atmospheric Model Group Norwegian Climate Centre NICAM NCC NICAM.09 NorESM1-M NorESM1-ME

CMIP5 models resolution

Dec-Feb 37 o N 27 o N 37 o N 27 o N 37 o N 27 o N 37 o N 27 o N Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 150 o W 130 o W 150 o W 130 o W 150 o W 130 o W x 10 6 7 6 5 4 3 2 1 Area (km 2 ) e position and amplitude of the North Pacific High (1967 2010) with respect to month. Each dot indicates the NPH for Schroeder a given year; color et denoted al. 2013 the area of the 1020 hpa contour. The black contour is the climatological ar. 542 Apr-Aug

not all models are made equal

not all models are made equal best representation of the NPH 2nd-round models

terms ofð! latitudevariability ð!yþ shows more variability[9than averages NPH position amplitu [9] Monthly ositioning, latitude ypositioning, Þ shows more than ] Monthly averages ofofnph positionandand am longitude, especially during winter. Thus, the NPH center signiﬁcantly correlate with monthly averages of UI, sea lev especially during winter.as far Thus, center signiﬁcantly correlate of stronge UI, s # N or as far north as 36# N and SST can be located souththe as 23NPH over the with coursemonthly of the year.averages However, the # # orbeaslocated far north as 36 N (126 cated as far south as 23 and# W) SSTcorrelations over the among coursethese of the year.occur However, s in winter; or itncan in a coastal position variables between the Janua The seasonal the four and March or fartherin offshore (157# W). (Figure 2). When the wintertime is locat W) of correlations or it can be located a coastal position (126#cycle among these variables occur NPH between variables!x,!y, A, and pmax explain 38.0%, 51.3%, to the northwest and has larger amplitude coastal upwellin # W). seasonal cycle of variance, the fourrespectively. offshore (157NPH and March (Figureand 2). coastal Whenseathe wintertime NPH Tim is 34.5%, The and 25.8% of the overall is enhanced, level and SST decline. 51.3%, ables!x,!y, A, andthepseasonal to with the northwest and amplitude coastal up in January are signiﬁcantly correlat When signals 38.0%, are removed, all correlate series of!y, A, andhas pmax larger max explain in February Marc with theand time coastal series of!ysea, A, and pmax and (n = 528, p < 0.01); longitude ð!xþ negatively nd 25.8% of one theanother overall variance, respectively. is enhanced, level SST and decline relates to!y, A, and pmax (Sr = $0.22, $0.32, and $0.44, Thus, NPH amplitude and latitudinal position show signiﬁca!y, A, andfrom in January signiﬁcantly co seasonal signals are removed, allnph correlate withto beseries pmax respectively) such that the center tends farther of correlations January into earlyare spring. So, when the NP February and with the istime series of!y,ita, and topmax her (n = 528, pnorth, < 0.01); longitude ð!xþpressure, negatively have higher maximum and a larger area strong in January, is likely lead in to enhanced upwellin when located farther fromand land$0.44, (to the west). The in early spring. and!y, A, and pmax (Srit =is $0.22, $0.32, Thus, NPH amplitude latitudinal position# show sig maximum SLP correlation among variables is between A and pmax [10] Our pre-conditioning index (pcui) at 39 N ranges fro ly) such that highest the NPH center tends to be farther correlations from January into early spring. So, when t (Sr = 0.81), greater than those between!y and the amplitude 223 m3 s$1 to 100 m$1 in 1983 to 4153 m3 s$1 to 100 m ve higher maximum and a larger area is stronginin2007 January, is likely to lead to enhanced up (Figureit 3a). The pcui correlates (p < 0.01) variables (Spressure, r = 0.31 for A and 0.43 for pmax). is located farther from land (to the west). The in early spring. latitude of maximum SLP orrelation among variables is between A and pmax [10] Our pre-conditioning index (pcui) at 39# N rang 3 $1 $1 3 $1 48!y and the amplitude 1), greater than those between s to 100 m in 1983 to 4153 m s to 1 223 m 0.6 45 for pmax). in0.7 2007 (Figure 3a). The pcui correlates (p < 0 (Sr = 0.31 for A and 0.43 longitude 42 latitude parameters + 39 36 0.4 33 42 39 36 0.4 33 48.3 46.2 44.6 43.3 41.7 0.4 36.6 35.1 32.7 0.7 0.7 0.8 0.4 1 48.3 46.2 44.6 43.3 41.7 0.4 37.8 0.7 0.7 0.6 0.7 0.7 0.7 0.6 0.6 0.8 0.8 0.7 0.7 0 0.7 0.8 1 48.5 46.5 44.5 42.5 4 38.5 36.5 34.5 32.5 0.6 0.6 Schroeder et al. 2013 0.4 0.6 0.6 0.4 1 ion 45 0.6 Correlation 48 0.4

Maximum SLP over NPH region latitude SLP (hpa) + years longitude

Maximum SLP over NPH region Summer SLP (hpa) SLP Climatology Winter SLP (hpa) SLP (hpa) years Feb Apr Jun Aug Oct Dec

Difference Maximum and Minimum SLP Models ensemble SLP (hpa) years

Difference Maximum and Minimum SLP Models ensemble SLP (hpa) -0.1hPa/decade years

Difference Maximum and Minimum SLP Models ensemble # models SLP (hpa) trend/decade years -0.1hPa/decade

Latitude of SLP Climatology Maximum SLP Latitude ( o N) Summer Feb Apr Jun Aug Oct Dec Latitude ( o N) latitude + longitude Winter years

37 o N 27 o N 37 o N 27 o N 37 o N 27 o N 37 o N 27 o N Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 150 o W 130 o W 150 o W 130 o W 150 o W 130 o W x 10 6 7 6 5 4 3 2 1 Area (km 2 ) Schroeder et al. 2013 e position and amplitude of the North Pacific High (1967 2010) with respect to month. Each dot indicates the NPH for a given year; color denoted the area of the 1020 hpa contour. The black contour is the climatological ar. Latitude ( o N) 542 Winter

Latitude ( o N) std 2006-2036 2064-2094 Winter

no change in SLP variability

how well models represent variability & its change? no change in SLP variability

how well models represent variability & its change? no change in SLP variability temperature? SLP <=> winds? <=>?

how well models represent variability & its change? no change in SLP variability temperature? SLP <=> winds? - is lack of change or models skill? - match with observed increasing variability in winds? - ensemble method appropriate? <=>?