The arid climate in the middle reaches of the Heihe River Basin leads to intense surface evapotranspiration and considerable irrigation water use for farmland. Most current assessments of remote sensing evapotranspiration data focus on the comparison between the remote sensing data and the measured flux data, fewer efforts have been made from the perspective of monitoring irrigation water use. The STAEDM (with a temporal resolution of 10 days and a spatial resolution of 30 meters) data of the desert-oasis region in the middle reaches of the Heihe River Basin was compared with site observations, other evapotranspiration products (SSEBop and MOD16), and irrigation water use statistics. The results showed that compared with the observations, STAEDM had a better accuracy under the vegetated surface than under the arid surface. STAEDM was the most accurate (r2=0.88 at Daman Station), followed by SSEBop (r2=0.83), whereas MOD16 was the least accurate (r2=0.68). STAEDM had better simulation accuracy in terms of the seasonal variation pattern and magnitude of evapotranspiration. Furthermore, STAEDM had more detailed textures in the spatial distribution of evapotranspiration than SSEBop and MOD16 (with a spatial resolution of 1 km), which could clearly display the spatial characteristics of evapotranspiration on the complex desert oasis surface. A comparison of the irrigation water use data based on STAEDM and the Water Resources Bulletin showed a high correlation (r2=0.91), with a relative error of 15.4%. Although there was a slight overestimation in the numerical value, the inter-annual variation patterns showed good consistency. The assessment results indicate that STADEM data has high spatial and temporal continuity, which can be used to refine irrigation water use monitoring at field scale, and provide data support for the optimization of irrigation schemes.