Our Research: Regional climate

The goal of our regional climate research is to understand climate variability and change on spatial scales most relevant to humans and ecosystems. These scales may be much smaller than those that have been the historical focus of climate research, particularly in areas of significant surface heterogeneity such as intense topography, where profound climate variations may occur on scales of just few kilometers. Our tools for studying this problem include high-resolution (~ 5 kilometer) climate simulations focused on a particular region. Recently, we developed a regional coupled ocean-atmosphere model, consisting of WRF (atmosphere) and ROMS (ocean) components, to allow for simulation of a full suite of earth system processes. A critical element of all our studies is also validation of these simulations with available in situ and satellite data.

One of our laboratories for studying regional climate has been Southern California, whose large mountain complexes are responsible for interesting mesoscale climate dynamics. A starting point for work in this region has been a study of the modes of variabililty in the region in a high-resolution regional simulation, using some of the same techniques previously used to identify modes of variability at continental and hemispheric scales. We found that the region exhibits pronounced local modes of variability that are uncorrelated with larger-scale modes and that exhibit significant spatial structure within the region. In a separate study of diurnal variability in same simulation, we found that mesoscale dynamics in the form of tight links between temperature and atmospheric circulation account for large spatial variations in the climatological diurnal cycles of these variables. We've also demonstrated that mesoscale processes govern important aspects of air-sea interaction in the region, generating, for example, major coastal upwelling events.

Topography also introduces spatial structure into fields assumed to be smoothly-varying in global climate models. Chief among these are surface radiative fluxes, which can only be simulated precisely in areas of intense topography with 3D radiative transfer techniques. We used such a model to simulate spatial variability in clear sky fluxes in a mountainous region of NW Washington state and quantify errors in conventional radiative transfer schemes in overall fluxes and in flux components central to ecosystem processes, such as diffuse radiation.