[1] The augmented Noah land surface model described in the first part of the two-part series was evaluated here over global river basins. Across various climate zones, global-scale tests can reveal a model's weaknesses and strengths that a local-scale testing cannot. In addition, global-scale tests are more challenging than local-and catchment-scale tests. Given constant model parameters (e. g., runoff parameters) across global river basins, global-scale tests are more stringent. We assessed model performance against various satellite and ground-based observations over global river basins through six experiments that mimic a transition from the original Noah LSM to the fully augmented version. The model shows transitional improvements in modeling runoff, soil moisture, snow, and skin temperature, despite considerable increase in computational time by the fully augmented Noah-MP version compared to the original Noah LSM. The dynamic vegetation model favorably captures seasonal and spatial variability of leaf area index and green vegetation fraction. We also conducted 36 ensemble experiments with 36 combinations of optional schemes for runoff, leaf dynamics, stomatal resistance, and the b factor. Runoff schemes play a dominant and different role in controlling soil moisture and its relationship with evapotranspiration compared to ecological processes such as the b factor, vegetation dynamics, and stomatal resistance. The 36-member ensemble mean of runoff performs better than any single member over the world's 50 largest river basins, suggesting a great potential of land-based ensemble simulations for climate prediction.

, toward the goal of better estimation of snowpack (in particular, snow water equivalent and snow depth) via incorporating both Gravity Recovery and Climate Experiment (GRACE) terrestrial water storage (TWS) and Moderate Resolution Imaging Spectroradiometer (MODIS) snow cover fraction (SCF) information into the Community Land Model. The different properties associated with the SCF and TWS observations are accommodated through a unified approach using the ensemble Kalman filter and smoother. Results show that this multisensor approach can provide significant improvements over a MODIS-only approach, for example, in the Saint Lawrence, Fraser, Mackenzie, Churchill & Nelson, and Yukon river basins and the southwestern rim of Hudson Bay. At middle latitudes, for example, the North Central and Missouri river basins, the inclusion of GRACE information preserves the advantages (compared with the open loop) shown in the MODIS-only run. However, in some highlatitude areas and given months the open loop run shows a comparable or even better performance, implying considerable room for refinements of the multisensor algorithm. In addition, ensemble-based metrics are calculated and interpreted domainwide. They indicate the potential importance of accurate representation of snow water equivalent autocovariance in assimilating TWS observations and the regional and/or seasonal dependence of GRACE's capability to reduce ensemble variance. These analyses contribute to clarifying the effects of GRACE's special features (e.g., a vertical integral of different land water changes, coarse spatial and temporal resolution) in the snow data assimilation system.

[1] Carbon (C) in tropical peatlands over Southeast Asia and Amazonia, if released to the atmosphere, can substantially increase the growth rate of atmospheric carbon dioxide. Over Southeast Asia, where the most extensive tropical peatlands in the world occur, 11 climate models for the IPCC Fourth Assessment show an overall decrease of rainfall in future dry seasons. Over Amazonia, future rainfall changes in dry seasons are highly uncertain; five models predict increased rainfall, and the remaining models predict the opposite. We have further examined the UKMO-HadCM3, GISS-ER, and GFDL-CM2.1 models. Over Southeast Asia, all three models predict similar decreases of rainfall and evaporative fraction, implying an increase of water table depth and surface dryness during the dry season south of the equator. Such changes would potentially switch peat ecosystems from acting as C sinks to C sources. Over Amazonia, the two models with the best simulations of current rainfall produce conflicting results for the future of peat stability.

No abstract