1. 1956-1970 1984-2000 Model Forced by: NCEP winds only Model Forced by: NCEP winds Local Surface Heat Flux Mean Advection of T’ CalCOFI Observations Coastal TS Diagrams for Observations and Model Model CASE A Model CASE C Observed Temperature Anomalies Observed Salinity Anomalies Geostrophic Along-shore Currents Warming Trend Low Frequency Salinity variations The Warming of the California Current: Dynamics and Ecosystem Implications Emanuele Di Lorenzo Georgia Institute of Technology, Atlanta, GA 30332 Arthur J. Miller SIO, La Jolla, CA 92093 Niklas Schneider IPRC, UH, Honolulu, HI 96822 James C. McWilliams UCLA, Los Angeles, CA 92093 ABSTRACT Long-term changes in the observed temperature and salinity along the Southern California coast are studied using a four-dimensional space-time analysis of the 52-year (1949-2000) California Cooperative Oceanic Fisheries Investigations (CalCOFI) hydrography combined with a sensitivity analysis of an eddy permitting primitive equation ocean model under various forcing scenarios. The goal of the study is to isolate and quantify the important physical and biological dynamics. North-East Pacific - Mean Surface Temperature California Current USA Model Domain CalCOFI In-Situ Observations 1950 – 2000 (high density sampling domain) C CONCLUSIONS for the PERIOD 1949 - 2000 Warming trend and surface heat fluxes - An overall warming trend of 1.3 C in the ocean surface, a deepening in the depth of the mean thermocline (18 m) and increased stratification between 1950 and 1999 are found to be primarily forced by large-scale decadal fluctuations in surface heat fluxes combined with horizontal advection by the mean currents. Trend in upwelling favorable winds - A cooling trend of –0.5C in SST is driven in the ocean model by the 50 year NCEP wind reanalysis, which contains a positive trend in upwelling favorable winds along the Southern California Coast. A net warming trend of +1C in SST occurs, however, when the effects of observed surface heat fluxes are included as forcing functions in the model. Increased stratification and reduced upwelling efficiency - Within 50 to 100 km of the coast, the ocean model simulations show that increased stratification/deepening of the thermocline associated with the warming reduces the efficiency of coastal upwelling in advecting subsurface waters to the ocean surface, counteracting any effects of the increased strength of the upwelling winds. Freshening of surface coastal salinity and Ecosystem Implications - The reduction in upwelling efficiency in the model leads to a freshening of surface coastal waters consistent with the observations. Because salinity and nutrients at the coast have similar distributions this must reflect a reduction of the nutrient supply at the coast, which is manifestly important in explaining the observed decline in zooplankton concentration. Increase in eddy variance - Model mesoscale eddy variance significantly increases in recent decades in response to both the stronger upwelling winds and the warmer upper ocean temperatures, suggesting that the stability properties of the SCCS have also changed. REFERENCES Di Lorenzo, E., A. J. Miller, N. Schneider, and J. C. McWilliams, 2004: The warming of the California Current: Dynamics and ecosystem implications. Journal of Physical Oceanography, in press. Schneider, N., E. Di Lorenzo, and P. Niiler, 2004: Salinity variations in the California Current, Journal of Physical Oceanography, submitted. McGowan, J., S. Bograd, R. J. Lynn, and A. J. Miller, 2003: The biological response to the 1977 regime shift in the California Current, Deep-Sea Res. Part II-Top. Stud. Oceanogr., 50, 2567-2582. Di Lorenzo, E., 2003: Seasonal dynamics of the surface circulation in the Southern California Current System. Deep-Sea Research II, 50 (14-16), 2371-2388. Section A 1) OBSERVATIONS of the WARMING TREND Vertical EOFs along Section A reveal a warming trend in the CCS that extends up to a depth of 150 m. This warming trend and increase in stratification, have been linked to the observed decline in Zooplankton (McGowan et al., 2003). Salinity anomalies are characterized by low frequency osciallations (companion study: Schneider et al., 2004). Figure (above) : Vertical EOFs 1 along SECTION A for the CalCOFI temperature and (c,d) salinity anomalies. The seasonal cycle is removed. (e,f) EOF 1 for geostrophic alongshore currents relative to 500 m. In (b) and (d) the black line represent the domain average. Figure (above) :EOF 1 for SST anomalies (left column) and temporal modulation (Principal Component, right column). The colored line in the right panel panel represent the model domain average SST anomaly. The figure shows the model SST response to the different forcing functions listed. CASE A Model Forced by: NCEP winds only The NCEP winds show a visible trend in upwelling favorable winds over the last 50 years, which have a net cooling effect on the system. CASE B Model Forced by: NCEP winds Local Surface Heat Flux Model open boundary cond. are set to climatology and do not allow for persistence of the heat gained through the Local (in the CalCOFI domain) surface fluxes. CASE C Model Forced by: NCEP winds Local Surface Heat Flux Mean Advection of T’ Heat fluxes (shown below) are forcing SSTa coherently and in phase over the entire North-East Pacific. This gives a persistence time scale to the locally forced SSTa anomaly that we model by introducing a mean advection of T’. 2) MODEL HINDCAST and the role of HEAT FLUXES The model response to the different forcing functions is evaluated to isolate the important terms in the heat budget that lead to the warming trend. We find that the surface heat fluxes are the dominant contributor to the warming (CASE C). These are strong enough to overcome the otherwise cooling trend forced by the trend in upwelling favorable coastal winds (CASE A). 3) SURFACE HEAT FLUXES and a 1D MODEL for TEMEPRATURE Over the last 50 year net heat flux variability is coherent and in phase over a broad portion of the North-East Pacific Ocean beyond the CalCOFI domain, as evident from the EOF 1 (data from Cayan private communication). Net Heat Fluxes (temporal) CalCOFI domain (spatial) Observed Coastal SSTa 1D Model Result A 1D Model of temperature forced with the observed net heat flux anomaly (shown above) is able to capture the warming trend and the sharp transitions toward warmer climate in the 1970, consistent with the observations. The broad spatial structure of the heat flux adds persistence to the locally forced SSTa in the CalCOFI domain. (ref. ocean model experiment: CASE C) 1D model: 4) INCREASED STRATIFICATION and REDUCED UPWELLING EFFICIENCY Observed TS diagrams show that the water masses are getting fresher and warmer after the 1970s. This is inconsistent with the observed trend of upwelling favorable winds that would lead to saltier and cooler water masses (Model CASE A). In order to correctly model the observed shift in water mass TS characteristics we need to include the effect of the warming trend (Model CASE C), which reduces the efficiency of the upwelling by increasing the stratification of the upper ocean and by depressing the isopycnals. 5) INCREASE in EDDY VARIANCE Figure (below): Model average along-shore (V) and cross-shore (U) velocity variance for the periods 1950-1970 (dark thick line) and 1980-2000 (gray thick line) as a function of distance from the coast. Each member of the average is plotted as a thin line to show the spread. Changes in variance are significant at the 95% level based on an F test with 80 degrees of freedom. V Along-shore velocity Model CASE C U Cross-shore velocity Model CASE C Temperature Salinity C