Sensitivity of the Water Cycle

Andrew Seidel


A study was done by scientists K.M. Lau, C.H. Sui, and W.K. Tao from the laboratory for Atmospheres, Goddard Space Flight Center, Greenbelt Maryland. This study was done for two purposes. First was to test the credibility of the water vapor models that these men produced and the second was to then use these models to predict what would happen to the water cycle if the surface temperature rose to 30 degrees centigrade. Although preliminary, their models are consistent with other general circulation models with respect to changes in temperature (atmospheric) and water vapor.

The model they formulated is called the Goddard Cloud Ensemble Model(GCEM) and is based on mass conservation equations. They have one for water vapor and one for liquid water and ice. Both equations take into account the rates of evaporation, condensation, deposition, and sublimation. When these two equations are integrated, vertical soundings of water vapor, liquid water, and ice can be found. They set initial conditions and run the model. The conditions as they define are:

In order to study how the system will equilibrate, they extend the integration to 52 days. They run both a control case and what they call a warm case. The warm case is identical to the control except that the sea surface temperature is 30C instead of 28C. The reason why they run the model for 52 days is to insure equilibrium since most convective clusters last only about 1 day.

Before actually modeling, the validity of the models must be assured. To do this they run the control conditions and estimate the mean water budget vetically through the atmosphere. In other words they look at what happens to the rates of condensation, evaporation, deposition and sublimation. Through these observations they estimate that "66% of the total rainfall will come from the convective region and 34% from the stratiform region." This partition is consistant with observational data taken.

Now they run the model to see what the response is to an increased temperature. They measure only change in temperature (atmospheric) and change in moisture with time. The models show that the warm case is 2-3 degrees warmer than control and has 20% more moisture content than control.

Next they look at the difference between the control case and warm case in the vertical profile. Again they look only at water vapor and temperature(atmospheric). They observe from these models that there is a 10-15% change in both variables. One thing they found very interesting was that when they plotted relative water vapor on the same graph, they found the largest increase in the upper troposphere (8-15km above surface). They conclude that as a result of surface warming temperature rises thoughout the atmosphere but the largest increase in temperature (4C) occurs in the upper troposphere which again is consistent with other global climate models and satellite data.

They go with these models and predict what will happen to cloud formation and cloud moisture. What they see is a shift upwards in both in the atmosphere which makes sense. If the temperature is warmer then water will stay vaporized at slightly higher altitudes.

Stepping back for a moment, they do a mass balance between evaporation, moisture convergence, and the precipitation. They reason from this balance that:

They make a lot of assumptions in the GCEM. However these assumtions appear to be valid because they come close to both other models and observational data done by other researchers. The biggest source for error in this model will come they include a land mass on the surface instead of water? They admit that their results are applicable only to a convection region. This is a very good start however. If we can use this model to research the climate feedback process, then better models can be developed.


References: