Evaporation capacity is the targeting index of water resources management and water disaster control under the global climate change. Attribution analysis of evaporation capacity concentrates on the annual scale in most researches. The verification of the corresponding attribution needs multiscale analysis. Daily meteorological data were selected at 819 stations in China from 1961 to 2015, in addition to potential evaporation (ETp). The trend and change points of annual ETp were excavated according to the time series analysis method at stations. Another special method was proposed based on diurnalscale total differential quadrature of Penman equation. This method was directly applied for the contributions of meteorological variables on the first order difference of daily ETp. It was further used for the accumulative contributions of meteorological elements on the changes in mean daily ETp. According to the contributions, the driving forces and corresponding reasons of the changes in evaporation capacity were analyzed. It was concluded that the mean value of the tendency rates was -33mm/（10a） at all stations. It spatially presented negative in Western and Eastern China and positive in Central China. Mean value of change points was 1987 at all stations. The driving force of decreasing ETp was decreasing net radiation in Eastern China. Corresponding accumulative contribution rate was 280. The driving force of decreasing ETp was increasing relative humidity in Western China. Corresponding accumulative contribution rate was 175. The driving forces of increasing ETp were increasing temperature and wind speed in Central China. Corresponding accumulative contribution rates were 355 and 121, respectively.