Abstracts of Some Important GENESIS Project Publications

Below are abstracts for a few important GENESIS project publications (detailed list of publications).


Bonan, G.B., Pollard, D. and Thompson, S.L., 1992: Effects of boreal forest vegetation on global climate. Nature, 359, 716-718.

Terrestrial ecosystems are thought to play an important role in determining regional and global climate; one example of this is in Amazonia, where destruction of the tropical rainforest leads to warmer and drier conditions. Boreal forest ecosystems may also affect climate. As temperatures rise, the amount of continental and oceanic snow and ice is reduced, so the land and ocean surfaces absorb greater amounts of solar radiation, reinforcing the warming in a "snow/ice/albedo" feedback which results in large climate sensitivity to radiative forcings. This sensitivity is moderated, however, by the presence of trees in northern latitudes, which mask the high reflectance of snow, leading to warmer winter temperatures than if trees were not present. Here we present results from a global climate model which show that the boreal forest warms both winter and summer air temperatures, relative to simulations in which the forest is replaced with bare ground or tundra vegetation. Our reuslts suggest that future redistributions of boreal forest and tundra vegetation (due, for example, to extensive logging, or the influence of global warming) could initiate important climate feedbacks, which could also extend to lower latitudes.


Wilson, K.M., Pollard, D., Hay, W.W., Thompson, S.L. and Wold, C.N., 1994: General circulation model simulations of Triassic climates: Preliminary results. Pangea: Paleoclimate, Tectonics and Sedimentation During Accretion, Zenith and Breakup of a Supercontinent. Special Paper 288, Geological Society of America.

The climates of two realistic geographic representations of the Triassic earth, corresponding in age to the Scythian (245 Ma) and the Carnian (225 Ma), are explored using a new atmospheric general circulation model (AGCM) call GENESIS. The GENESIS AGCM is coupled to a slab ocean 50 m thick, with prescribed heat transport; it also incorporates three types of cloud cover and new models for vegetation effects, soil hydrology, snow cover, and sea-ice formation and melting. Boundary conditions prescribed in the separate Scythian and Carnian experiments include realistic paleogeography and estimates of paleotopography, solar insolation, atmospheric CO2 concentration, vegetation and soil types, and oceanic heat flux. Seasonal simulations of Triassic climate were performed using a horizontal spectral resolution of R15 (4.5 degrees latitude by 7.5 degrees longitude) and 12 levels in the vertical for the atmosphere and 2° x 2° for the surface.

Results for both time intervals suggest that most of the seasonal precipitation fell on major highland areas of Pangea. Dry continental climates with very large seasonal temperature ranges (>45° C) were modeled in the dominantly lowland interior of Pangea. Carnian continental climates predicted by the AGCM were wetter than those of the Scythian; however, both time intervals were characterized by strongly monsoonal circulation. Comparison of these results with lithologic and fossil proxy climatic indicators suggests reasonably good correlations. However, the extreme temperature variations predicted for both Scythian and Carnian are somewhat difficult to reconcile with the fossil record, although accurate interpretation of fossil proxy climatic indicators is not a simple matter. Additional AGCM sensitivity studies may be necessary to resolve this problem.


Pollard, D. and Thompson, S.L., 1994: Sea-ice dynamics and CO2 sensitivity in a global climate model. Atmosphere-Ocean, 32, 449-467.

Present-day results and CO2 sensitivity are described for two versions of a global climate model (GENESIS) with and without sea-ice dynamics. Sea-ice dynamics is modelled using the cavitating-fluid method of Flato and Hibler (1990, 1992). The atmospheric general circulation model originated from the NCAR Community Climate Model version 1, but is heavily modified to include new treatments of clouds, penetrative convection, planetary boundary layer mixing, solar radiation, the diurnal cycle, and semi-Lagrangian transport of water vapor. The surface models include an explicit model of vegetation (similar to BATS and SiB), multilayer models of soil, snow and sea ice, and a slab ocean mixed layer. When sea-ice dynamics is turned off, the CO2-induced warming increases drastically around 60-80° S in winter and spring. This is due to much greater (and unrealistic) compactness of the Antarctic ice cover without dynamics, which is reduced considerably when CO2 is doubled and exposes more open ocean to the atmosphere. With dynamics, the winter ice is already quite dispersed for 1xCO2 so its compactness does not decrease as much when CO2 is doubled.


Thompson, S.L. and Pollard, D., 1995: A global climate model (GENESIS) with a Land-Surface-Transfer Scheme (LSX). Part 1: Present climate simulation. J. Climate, 8, 732-761.

The present-day climatology of a global climate model (GENESIS version 1.02) is described. The model includes a land-surface-transfer component (LSX) that accounts for the physical effects of vegetation. The atmospheric general circulation model is derived from the NCAR CCM1, modified to include semi-Lagrangian transport of water vapor, sub-grid plume convection, PBL mixing, a more complex cloud scheme and a diurnal cycle. The surface models consist of LSX, multilayer models of soil, snow, and sea ice, sea-ice dynamics and a slab mixed-layer ocean. Brief descriptions of the current model components are included in an appendix. GENESIS is on ongoing project to develop an earth system model prototype for global change research. The version 1.02 climate model has already proved useful in paleoclimate studies.

Results of present-day simulations are described using an atmospheric spectral resolution of R15 (4.5° latitude by 7.5° longitude) and a surface-model resolution of 2° by 2°. In general the quality of the simulations is comparable to that of previous coarse-grid models with predicted sea-surface temperatures. Most of the errors are attributed to coarse atmospheric resolution, inaccurate cloud parameterization, large ocean roughness length and lack of ocean dynamics.

The results are compared with those using a simplified bucket-soil model and crude parameterizations of surface albedo and roughness. Although quite similar results are obtained on global scales, significant regional differences including surface warming and drying occur in some regions of Amazonia and northern mid-latitude continental interiors.


Thompson, S.L. and Pollard, D., 1995: A global climate model (GENESIS) with a Land-Surface-Transfer Scheme (LSX). Part 2: CO2 sensitivity. J. Climate, 8, 1104-1121.

The sensitivity of the equilibrium climate to doubled atmospheric CO2 is investigated using the GENESIS global climate model version 1.02. The atmospheric general circulation model is a heavily modified version of the NCAR CCM1 and is coupled to a multi-canopy land-surface model (LSX), multi-layer models of soil, snow and sea ice, and a slab ocean mixed layer. Features that are relatively new in CO2 sensitivity studies include explicit sub-grid convective plumes, PBL mixing, a diurnal cycle, a complex land-surface model, sea-ice dynamics, and semi-Lagrangian transport of water vapor.

The global annual surface-air warming in the model is 2.1° C, with global precipitation increasing by 3.3%. Over most land areas, most of the changes in precipitation are insignificant at the 5% level compared to interannual variability. Decreases in soil moisture in summer are not as large as in most previous models, and only occur poleward of ~55° N in Siberia, northern Canada and Alaska. Sea-ice area in September recedes by 62% in the Arctic and by 43% in the Antarctic. The area of Northern Hemispheric permafrost decreases by 48%, while the the total area of Northern Hemispheric snowcover in January decreases by only 13%.

The effects of several modifications to the model physics are described. Replacing LSX and the multi-layer soil with a single-layer bucket model causes little change to CO2 sensitivities on global scales, and the regions of summer drying in northern high latitudes are reproduced although with somewhat greater amplitude. Compared to convective adjustment, penetrative plume convection increases the tropical Hadley Cell response, but decreases the global warming slightly by 0.1 to 0.3° , contrary to several previous GCM studies in which penetrative convection was associated with greater CO2 warming. Similarly, the use of a cruder parameterization for cloud amount changes the local patterns of cloud response but has slight effect on the global warming. We discuss implications of the greater global warming (3.2° C) found in an earlier version of the model, and suggest that it was due to more detailed interactions that no longer occur in the current version.


Pollard, D. and Thompson, S.L., 1995: Use of a land-surface-transfer scheme (LSX) in a global climate model: the response to doubling stomatal resistance. Global and Planetary Change, 10, 129-161.

One response of vegetation to future increases in atmospheric CO2 may be a widespread increase in stomatal resistance. Such a response would increase plant water usage efficiency while still allowing CO2 assimilation at current rates. The associated reduction in transpiration rates has the potential of causing significant modifications in climate on regional and global scales.

This paper describes the effects of a uniform doubling of the stomatal resistance parameterization in a global climate model (GENESIS). The model includes a land-surface transfer scheme (LSX) that accounts for the physical effects of vegetation, including stomatal resistance and transpiration, which is described in detail in an appendix. The atmospheric general circulationmodel is a heavily modified version of the NCAR Community Climate Model version 1 with new treatments of clouds, penetrative convection, planetary boundary layer mixing, solar radiation, the diurnal cycle, and semi-Lagrangian transport of water vapor. The other surface models include multi-layer models of soil, snow and sea ice, and a 50-m slab ocean mixed layer.

The effects of doubling the stomatal resistance parameterization are largest in heavily forested regions: tropical South America, and parts of the Northern Hemispheric boreal forests in Canada, Russia and Siberia in summer.The primary surface changes are a decrease in evapotranspiration, an increase in upward sensible heat flux, and a surface-air warming. Secondary effects include shifts in the ITCZ which cause large increases in precipitation, soil moisture and runoff in western tropical South America, and decreases in these quantities in northern subtropical Africa. Noticeable changes in relative humidity, cloudiness and meridional circulation occur throughout the troposphere. The global effects on atmospheric temperature and specific humidity are small fractions of those found in other doubled CO2 experiments. However,unlike doubled CO2 the signs of those changes combine to give relatively large reductions in relative humidity and cloudiness. It is suggested that the stomatal-resistance effect and other plant responses to large-scale environmental perturbations should be included in models of future climate.


Comments and Questions

12/8/97