The increased societal attention to climate change, drought and flood early warning systems, ecosystem monitoring and biodiversity conservation has led to a large demand for estimating, modelling, mapping, and forecasting evapotranspiration (ET) as a key water flux at the soil-vegetation-atmosphere interface. Cutting-edge techniques, which increase the efficiency of processing large datasets, such as artificial intelligence (AI), data fusion, sharpening algorithms, and the integration of physical- and process-based models with empirical and statistical approaches, including machine learning, are essential for bridging different spatial and temporal scales while addressing and communicating method-specific uncertainties.

This session will focus on various ET estimation methods, including sap flow or soil heat pulse sensors, lysimeters, eddy covariance stations, scintillometers, and remote sensing-based methods. We will also explore emerging techniques such as AI, data fusion, sharpening algorithms, machine learning, and cloud computing. Additionally, we will cover detailed evaluations of scale dependencies, approaches for handling uncertainties and systematic biases, and assessment of the representativeness of the estimates.

We welcome contributions that (1) assess and compare various in-situ and remote sensing methods, (2) analyse trends and spatio-temporal patterns in ET data, including sources of error and uncertainty, (3) bridge scales between different in-situ measurements, modelled and remotely sensed ET, addressing validation, calibration, and upscaling challenges, and (4) evaluate the challenges and opportunities associated with fusing data products, improving processing pipelines, applying AI methods, cloud computing and new technologies.