The study of changes in rice evapotranspiration (ET) is important for agricultural water management. In this study, the Surface Energy Balance Algorithms for Land (SEBAL) and MODIS products were used to estimate long-term rice ET from 2003 to 2018 in the Sanjiang Plain, Northeast China, and the spatial and temporal variations in rice ET and related influencing factors were analyzed. The results showed that the SEBAL model accurately estimated the daily ET on the Sanjiang Plain. Rice ET in the Sanjiang Plain increased at a rate of 5.05 mm·year -1. The spatial distribution of rice ET varied significantly under different typical years. Temperature was the main factor affecting rice ET on the Sanjiang Plain, with a correlation coefficient of 0.67. The increase in rice ET per ha was positively correlated with rice area expansion, with a correlation coefficient of 0.79. The growth rate of rice ET slowed after 2011, which was related to the cooling effect of irrigation due to rice area expansion.