Physical processes -- by themselves and through interaction with fluid-dynamical ones -- are important factors in the maintenace and variability of the observed state of the climate system. The TCD group engages in modeling various aspects of these processes, through the study of: energy balance models, ice-sheet models, and coupling of slow physical processes on the paleoclimate time scale, hydrological cycle, interaction between thermal and dynamical processes, planetary boundary layer, radiative-convective equilibria .

Radiative-convective model

Radiative-convective (RCM: vertical variation, horizontally averaged) and energy-balance (EBM: horizontal variation; vertically integrated) models are the two simplest types of models which capture the basic physical processes in the atmosphere. These models can reveal the underlying processes and mechanisms, as well as help improve more comprehensive models. Using the two-stream approximation for radiative transfer in a simple column model with temperature-dependent surface albedo, an RCM provides a tool to study: 1) radiative multiple equilibria and their stability, and 2) their sensitivity to changes in the atmosphere's optical properties. The figure shows multiple equilibria of the model's temperature profile: a) present (surface albedo=0.3; stable), b) totally ice-coverd (albedo=0.9; stable), and c) intermediate (albedo=0.525; unstable) climate along with the general circulation model's (GCM's) horizontally averaged temperatude profile (from Li et al. 1997).

Variability in United States surface-air temperatures

Changes in surface-air temperature are one of most "obvious" indications of climate change. Continental- or regional-scale temperature variations may be related to climatic variations with similar periodicities, either global or centered in other regions. The upper figure shows the time series of average temperatures over traditional climate divisions from Georgia (GA), Arizona (AZ), Idaho (ID) plus 10C, and New Hampshire (NH) plus 10C. The lower figure shows the contour plots of temperature-index reconstructions in 1915 and 1975, dashed where negative; contour interval is 0.05 standard deviations (of overall variability) times 10 (from Dettinger et al. 1995). The point is that there is substantial spatial variability in climate trends: even over a relatively small fraction of the globe, like the contiguous United States, interdecadal temperature changes can proceed in different directions at different times and in different subregions.

Quaternary glaciation cycles

As we attempt to penetrate further into the remote past, evidence for climate change becomes sketchier, in both temporal and spatial detail. To help understand this limited information on dramatic alternations between a relatively ice-covered and ice-free planet, simple models can capture the major feedbacks between the atmosphere, ice sheets, oceans and biota. The figure shows both a perspective from space of such a model's surface features (left) and its vertical cross-section (right; from Ghil, 1994). The model solution's power spectrum is compared with that of an isotopic proxy record for air temperature from the Vostok ice core (from Yiou et al., 1994).