Comparison and Assessment of Three Advanced Land Surface Models in Simulating Terrestrial Water Storage Components over the United States

In preparation for the next generation NCEP North American Land Data Assimilation System (NLDAS), including upgrades or replacement of the land surface models (LSMs) executed therein, three advanced land surface models (CLM4.0, Noah-MP, and CLSM-F2.5) were executed from 1979 to 2014 within the NLDAS-based framework. To identify which model has a potential to be incorporated to the next generation NLDAS system, monthly total water storage anomaly (TWSA) and its individual water storage components were evaluated against satellite-based (e.g., GRACE) and in situ observations (e.g., USGS wells, NASMD soil moisture), as well as against reference reanalysis products (e.g., SNODAS), at basin-wide and statewide scales. In general, all three advanced LSMs are able to reasonably capture the monthly and interannual variability and magnitudes for TWSA. However, the contributions from individual water components to the TWSA anomalies are very dependent on the given model and basin. A major contributor to the TWSA is the anomaly of total column soil moisture content (SMCA) for CLM4.0 and Noah-MP, and the anomaly of groundwater storage anomaly (GWSA) for CLSM-F2.5. Other components such as the anomaly of snow water equivalent (SWEA) also play some role. For each individual water storage component, the models are able to capture broad features such as monthly and interannual variability. However, there are large inter-model differences and quantitative uncertainties in this study. This research seems to show that CLSM-F2.5 may have difficulty in use for the next generation NLDAS system.