Coastal upwelling variability in the California Current region, one of the four main eastern boundary current upwelling systems, is controlled by processes acting over a wide range of spatial and temporal scales. While the ensuing ecosystem response depends strongly on upwelled water properties, determining their exact physical and biogeochemical characteristics is notoriously difficult as it requires tracking water masses backward in space and time from the moment they upwell near the coast to their subsurface origin. Adjoint model simulations have been used successfully to track water masses in coastal upwelling systems and the work presented here extends these applications to determining the co-variability of physical and biogeochemical properties of source waters at spatial scales that resolve the known alongshore variability of coastal upwelling dynamics in the region. Notably, the results identify that the modulation of coastal upwelling efficiency by onshore/offshore geostrophic meanders is the dominant mechanism explaining alongshore variability in source depth and properties of upwelled waters. The simulations also reveal that source water properties vary seasonally in response to different balances between coastal upwelling intensity and biogeochemical processes. During spring, interannual variability of physical and biogeochemical properties is directly tied to the intensity of upwelling-favorable alongshore winds, whereas, during summer, biogeochemical properties respond more strongly to biological activity and subsequent organic matter remineralization at depth. Overall, the present work provides important insight into the mechanisms responsible for the alongshore mosaic and seasonal variation upwelled source water properties in the central California Current region.

This dataset is derived from adjoint passive tracer simulations for the central California Current upwelling system. The physical model is an implementation of the Regional Ocean Modelling System (ROMS; www.myroms.org) coupled to NEMUCSC, a customized version of the North Pacific Ecosystem Model for Understanding Regional Oceanography (NEMURO). The adjoint model is used to track backward-in-time passive tracer concentrations originating from nearshore locations where upwelled water masses are typically found in the central CCS. The simulations are implemented such that adjoint passive tracers at any time in the past represent the source locations of water masses initially present over a 14-day period in 0.5° alongshore bins extending 0–30 km offshore and 0–20 m deep. The adjoint simulations The dataset was produced with Ferret V7.1 (http://ferret.pmel.noaa.gov/Ferret) and is in NetCDF4 format (CF-1.4 compliant) with metadata describing each variable name and associated units.