Crop consumptive water use is recognized as a key element to understand regional water management performance. This study documents an attempt to apply a regional evapotranspiration model (SEBAL) and crop information for assessment of regional crop (summer maize and winter wheat) actual evapotranspiration (ETa) in Huang-Huai-Hai (3H) Plain, China. The average seasonal ETa of summer maize and winter wheat were 354.8 and 521.5 mm respectively in 3H Plain. A high-ETa belt of summer maize occurs in piedmont plain, while a low ETa area was found in the hill-irrigable land and dry land area. For winter wheat, a high-ETa area was located in the middle part of 3H Plain, including low plain-hydropenia irrigable land and dry land, hill-irrigable land and dry land, and basin-irrigable land and dry land. Spatial analysis demonstrated a linear relationship between crop ETa, normalized difference vegetation index (NDVI), and the land surface temperature (LST). A stronger relationship between ETa and NDVI was found in the metaphase and last phase than other crop growing phase, as indicated by higher correlation coefficient values. Additionally, higher correlation coefficients were detected between ETa and LST than that between ETa and NDVI, and this significant relationship ran through the entire crop growing season. ETa in the summer maize growing season showed a significant relationship with longitude, while ETa in the winter wheat growing season showed a significant relationship with latitude. The results of this study will serve as baseline information for water resources management of 3H Plain.

Climate change will have important implications in water shore regions, such as Huang-Huai-Hai (3H) plain, where expected warmer and drier conditions might augment crop water demand. Sensitivity analysis is important in understanding the relative importance of climatic variables to the variation in reference evapotranspiration (ET0). In this study, the 51-yr ET0 during winter wheat and summer maize growing season were calculated from a data set of daily climate variables in 40 meteorological stations. Sensitivity maps for key climate variables were estimated according to Kriging method and the spatial pattern of sensitivity coefficients for these key variables was plotted. In addition, the slopes of the linear regression lines for sensitivity coefficients were obtained. Results showed that ET0 during winter wheat growing season accounted for the largest proportion of annual ET0, due to its long phenological days, while ET0 was detected to decrease significantly with the magnitude of 0.5 mm yr-1 in summer maize growing season. Solar radiation is considered to be the most sensitive and primarily controlling variable for negative trend in ET0 for summer maize season, and higher sensitive coefficient value of ET0 to solar radiation and temperature were detected in east part and southwest part of 3H plain respectively. Relative humidity was demonstrated as the most sensitive factor for ET0 in winter wheat growing season and declining relativity humidity also primarily controlled a negative trend in ET0, furthermore the sensitivity coefficient to relative humidity increased from west to southeast. The eight sensitivity centrals were all found located in Shandong Province. These ET0 along with its sensitivity maps under winter wheat-summer maize rotation system can be applied to predict the agricultural water demand and will assist water resources planning and management for this region.