Evaluation of high-resolution atmospheric and oceanic simulations of the California Current System

Abstract

12This paper is the first of two that present a 16-year reanalysis solution from a coupled physi-13 cal and biogeochemical model of the California Current System (CCS) along the U. S. West 14Coast and validate the solution with respect to mean and seasonal fields and, to a lesser de-15 gree, eddy variability. Its companion paper is Deutsch et al. (2019a). The intent is to construct 16 and demonstrate a modeling tool that will be used for mechanistic explanations, attributive 17 causal assessments, and forecasts of future evolution for circulation and biogeochemistry, with 18 particular attention to the increasing oceanic stratification, deoxygenation, and acidification. 19A well-resolved mesoscale (dx = 4 km) simulation of the CCS circulation is made with the 20 Regional Oceanic Modeling System over a reanalysis period of 16 years from 1995 to 2010. 21The oceanic solution is forced by a high-resolution (dx = 6 km) regional configuration of 22 the Weather and Research Forecast (WRF) atmospheric model. Both of these high-resolution 23 regional oceanic and atmospheric simulations are forced by lateral open boundary conditions 24 taken from larger-domain, coarser-resolution parent simulations that themselves have bound-25 ary conditions from the Mercator and Climate Forecast System reanalyses, respectively. We 26 first show good agreement between the simulated atmospheric forcing of the ocean and satellite 27 observations for the spatial patterns and seasonal variability of the cloud cover and for the sur-28 face fluxes of momentum, heat, and freshwater. The simulated oceanic physical fields are then 29 evaluated with satellite and in situ observations. The simulation reproduces the main struc-30 ture of the climatological upwelling front and cross-shore isopycnal slopes, the mean current 31 patterns (including the California Undercurrent), and the seasonal and interannual variabil-32 ity. It also shows agreement between the mesoscale eddy activity and the wind-work energy 33 exchange between the ocean and atmosphere modulated by influences of surface current on 34 surface stress. Finally, the impact of using a high frequency wind forcing is assessed for the 35 importance of synoptic wind variability to realistically represent oceanic mesoscale activity 36 and ageostrophic inertial currents. 1 38 Subtropical eastern boundary upwelling systems like the California Current System (CCS) are 39among the biologically most productive coastal environments (Carr and Kearns, 2003), supporting 40 some of the world's major fisheries (FAO, 2009). Seasonal upwelling (mainly during spring and 41 summer) of deep nutrient-rich water maintains high rates of productivity over broad scales (e.g., 42Chavez and Messie (2009)). Additionally, coastal currents and oceanic mesoscale variability con-43 tribute to cross-shore exchange of heat, salt, and biogeochemical materials between the open and 44 a dampening of the mesoscale activity; Renault et al. (2016d)), and high-frequency wind fluc-82 tuations. In this paper and its biogeochemical co...

Attributing Causes of Future Climate Change in the California Current System With Multimodel Downscaling

“…While few data sets exist at the appropriate scales of space and time to evaluate the impact of missing (sub)mesoscale processes across the CCS model domain (Capet et al, 2008), the closely related 4 km resolution ROMS‐BEC model has been extensively validated against available observational data for a broad suite of ocean physical and biogeochemical fields. It demonstrates a high fidelity in reproducing observations of the mean state, variability, and trends of key tracers including temperature, nutrient concentrations, NPP, and dissolved O 2 (Deutsch et al, 2020; Renault et al, 2020). Beyond the broad agreement between observations and both the 4 and 12 km resolution hindcasts, downscaled climate projections from both model resolutions agree well.…”

“…The model grid has 322 offshore cells and 450 alongshore cells for a horizontal resolution of 12 km and 42 vertical levels with terrain (i.e., bathymetry) following coordinates. The numerical models, configuration, and boundary conditions are as described in Renault et al (2020) and Deutsch et al (2020); the 4 km model presented in those works covers a subset of the 12 km model domain presented here.…”

Climate-driven aerobic habitat loss in the California Current System

“…Distributions of hydrographic variables such as temperature and O 2 over the CCS were reconstructed from Regional Ocean Modeling System (ROMS) simulations that were validated with historical hydrographic observations [ fig. S2 and Supplementary Materials; (15)(16)(17)(18)].…”

“…We used a regional model of the CCS circulation (ROMS) with a Biogeochemical Elemental Cycling model to hindcast physical and biogeochemical variables over the years 1995-2010 (15)(16)(17)(18). The hydrographic variability was extended further back in time by combining high-resolution model fields with long-term, lowerresolution hydrographic observations starting in 1956.…”

“…The future (100-year) changes of hydrographic variables were derived from Coupled Model Intercomparison Project Phase 5 Representative Concentration Pathway 8.5 projections, dynamically downscaled within ROMS for evaluation of future CCS conditions. The Supplementary Materials contains an overview of the model configuration and validation, which are covered in detail in other work (15)(16)(17)(18), and the specific earth system models and approach used in the dynamic downscalings. For the purposes of this work, the 100-year changes in temperature and O 2 were significant (P < 0.05) and consistent in the coastal CCS across the evaluated model forcings, justifying the use of ensemble-mean fields of hydrographic data to examine changes in aerobic habitat.…”