As part of the recent Global Soil Wetness Project (GSWP), a dozen land-surface models were forced with the International Satellite Land-Surface Comparison Project (ISLSCP) Initiative 1 data. One goal was to produce soil wetness fields for use in soil moisture initialization in climate models, as well as other climate change studies. Another goal of the GSWP was to compare the water and energy balance variables from the different models. At the University of Arizona, we ran the Biosphere-Atmospheres Transfer Scheme (BATS), which at the time used the 3-layer soil model standard in BATS1e. Since then, a new 10-layer soil model has been added to BATS. This model was adapted from the National Center for Atmospheric Research (NCAR) Land Surface Model (LSM) and is likely to be very similar to the soil model that will be part of the new Common Land Model, currently under development. Therefore it is important to understand how the output from the two BATS soil models differs. In this study, we are repeating our GSWP simulations with the new 10-layer soil models and comparing those results to the earlier ones using the 3-layer soil model. As the two soil models have some differences in their underlying physics, changes in such variables as soil moisture, evaporation, runoff and skin temperature are expected. We are investigating these output differences at several temporal/spatial resolutions, from diurnal cycles at individual points to seasonal averages at global/continental scales