There is a growing interest in quantifying regional scale actual evapotranspiration (ET) for water resource management. Using moderate-resolution Imaging spectroradiometer (MODIS) data, in combination with meteorological records, an integrated approach is proposed to estimate daily ET over 1-km grid covering North China. The Surface Energy Balance System (SEBS) is adapted to partition the sensible heat from latent heat. A gap-filling method for ET estimation during cloudy intervals is accomplished by a combination of SEBS outputs and Penman-Monteith equation. The proposed scheme, named ETWatch, is calibrated and evaluated using measurements from four flux sites in the semi-humid regions with different soil types, land cover, and climate conditions in North China. The gap-filling method reproduces observed fluxes with satisfied correlation, yielding about a 22% MAPD in Yucheng agriculture site. The improved method could be used to study the impact of different agricultural water consuming patterns in North China.