California's weather is characterized by extreme droughts and floods. This has resulted in overdraft of groundwater aquifers as growers turn to this source of water for irrigation during droughts. California produces 99% of all walnuts in the US, walnut growers are under extreme pressure to optimize crop water use. Water use estimates at the field scale are crucial for growers to refine irrigation scheduling decisions. In this study, two single source remote sensing-based energy balance models (pySEBAL [python-based Surface Energy Balance Algorithm for Land] and SEBS [Surface Energy Balance System algorithm]) for estimating evapotranspiration (ET a) were evaluated against in-situ ET a measurements from surface renewal in two young walnut orchards in California's Sacramento Valley. Strong correlations were obtained between the RS-based estimates and in-situ measurements for both pySEBAL and SEBS from 2017 to 2020, with R2 above 0.87, RMSE ranging from 0.79 to 1.05 mm, and NSE ranging from 0.79 to 0.88. SEBS out-performed pySEBAL on estimating time-series daily ET a for walnut at the field scale. During the mid-season characterized by high wind speed and high temperatures, pySEBAL and SEBS both underestimated ET a while the two models slightly overestimated ET a during the early growing season and post-harvest period. These results indicate the need for future research to focus on improving the performance of pySEBAL and SEBS when simulating time-series ET a under advection and sparse vegetation conditions, to provide more accurate ET a for both in-season and off-season irrigation water management. Comparisons of historical K c and RS-based K c were evaluated, and RS-based K c matched better with the actual water use requirements of developing 2nd to 4th leaf young walnuts orchards. We observed substantial spatiotemporal variability of RS-based K c in wo young walnut orchards at the CAPEX and Kauffman sites. For instance, K c values for the north and south halves of the CAPEX orchard were above 1.2 and 0.9–1.1 respectively. K c values for the north and south halves at the Kauffman site were in a range of 0.6–0.71 and 0.82–0.94 respectively. In other words, conventional irrigation management based on multiplying historical K c and reference evapotranspiration is inadequate for site-specific walnut irrigation management due to spatial and temporal variability in K c. ET a and K c mapping from two single source models were similar in the spatial patterns and provided an overall visual characterization of the spatial variability in crop water use. Overall, RS-based ET a models using freely accessible satellite images and open-source algorithms could be used as an alternative to expensive in-situ measurements for enhancing site-specific young walnut irrigation management. • RS-based ET a estimates for young walnut were compared to in-situ ET a measurements. • SEBS and pySEBAL both got reasonable ET a estimates, and SEBS performed better. • RS-based K c and ET a for developing young walnut captured its spatiotemporal variation. • RS-based ET a models with low cost could be a promising tool in the smart agriculture. • Upgraded irrigation system is necessary to provide enough control to irrigate based on RS-based ET a estimates. [ABSTRACT FROM AUTHOR]