Reservoirs are fundamental human‐built infrastructures that collect, store, and deliver fresh surface water in a timely manner for many purposes. Efficient reservoir operation requires policy makers and operators to understand how reservoir inflows are changing under different hydrological and climatic conditions to enable forecast‐informed operations. Over the last decade, the uses of Artificial Intelligence and Data Mining [AI & DM] techniques in assisting reservoir streamflow subseasonal to seasonal forecasts have been increasing. In this study, Random Forest [RF), Artificial Neural Network (ANN), and Support Vector Regression (SVR) are employed and compared with respect to their capabilities for predicting 1 month‐ahead reservoir inflows for two headwater reservoirs in USA and China. Both current and lagged hydrological information and 17 known climate phenomenon indices, i.e., PDO and ENSO, etc., are selected as predictors for simulating reservoir inflows. Results show (1) three methods are capable of providing monthly reservoir inflows with satisfactory statistics; (2) the results obtained by Random Forest have the best statistical performances compared with the other two methods; (3) another advantage of Random Forest algorithm is its capability of interpreting raw model inputs; (4) climate phenomenon indices are useful in assisting monthly or seasonal forecasts of reservoir inflow; and (5) different climate conditions are autocorrelated with up to several months, and the climatic information and their lags are cross correlated with local hydrological conditions in our case studies.

Vegetation plays an important role in regulating terrestrial water and energy fluxes. Recent research reported that China is the largest contributor to the global net increase in green foliage. Such large‐scale vegetation greenness change would inevitably affect the water budget across China. In this study, we evaluated the effects of vegetation greenness change on evapotranspiration (ET) and water yield in China from 1982 to 2014 with a process‐based conceptual hydrologic model and a detrending experiment of vegetation index. On the national scale, the change in vegetation greenness was dominated by an increasing trend (greening), which has led to an increase in ET and a decline in water yield from 1982 to 2014. The contribution of vegetation greenness change to the trend of annual ET and water yield is 0.25 mm/yr and −0.16 mm/yr, respectively. Spatially, the relative change in ET and runoff caused by vegetation greenness change is significantly greater in water‐limited areas than in energy‐limited areas. Vegetation greening exacerbated water scarcity in water‐limited areas. Therefore, the ecological restoration policy should fully consider the negative effect of vegetation greening on water yield in water‐limited regions and balance the conflicting water demands between the ecosystem and human beings.

Potential evapotranspiration (PET), which determines the upper limit of actual evapotranspiration (AET), is a necessary input in monthly hydrological models. In this study, the sensitivities of monthly hydrological models to different PET inputs are investigated in 37 catchments under different climatic conditions. Four types of PET estimation methods (i.e., Penman–Monteith, Hargreaves–Samani, Jensen–Haise, and Hamon) give significantly different PET values in the 37 catchments. However, similar runoff simulations are produced based on different PET inputs in both nonhumid and humid regions. It is found that parameter calibration of the hydrological model can eliminate the influences of different PET inputs on runoff simulations in both nonhumid and humid regions. However, the influences of parameter calibration on the simulated water balance components, including AET and water storage change (WSC), are different in nonhumid and humid regions. In nonhumid regions, simulated runoff, AET, and WSC are similar using different PET inputs. In humid regions, simulated AET and WSC using different PET inputs are significantly different, and simulated runoff and the sums of AET and WSC are similar to each other. It is suggested that the choice of PET inputs for monthly hydrological models be based on the selected region and the relevant hydrological practices. In runoff modeling, different PET inputs can produce similar runoff simulations in both nonhumid and humid regions. However, when estimating AET and WSC in humid regions, different PET inputs will result in significantly different AET and WSC simulations, which should be noted by model users.