A variety of remote sensing approaches have been developed to map evapotranspiration (ET) and vegetation stress using thermal infrared (TIR) and/or visible-to-shortwave-infrared (VSWIR) imagery from satellites. Especially at sub-field scales (100-m or finer), these data are valuable for informing deficit irrigation strategies used in many viticultural systems toward reducing water use and improving grape quality and yield. Widespread and routine generation of ET data at this scale has been enabled by cloud computing technologies, with the OpenET ensemble modeling system as an example of collaborative geospatial information development. The ensemble currently includes 6 ET models that have been ported to Google Earth Engine and represent a diversity of approaches using TIR and/or VSWIR inputs to map daily ET at 30-m resolution. This presentation evaluates the performance of OpenET over vineyards in the Central Valley of California. In particular, we examine the advantages and disadvantages of thermal- vs. surface reflectance-based models in terms of spatial and temporal response to changing soil moisture and vine conditions, as well as the value of ensemble averaging for producing robust vineyard water use estimates.