Improvements and Sensitivities of Key Inputs to the ALEXI/DisALEXI Modeling Scheme for California Specialty Crops

Abstract:

An accurate spatial representation of evapotranspiration (ET) by way of satellite remote sensing is crucial for optimizing irrigation scheduling and efficiency in agricultural regions prone to limited water supplies such as the Central Valley of California. Agriculture is a key industry in California’s economy, generating billions of dollars in revenue each year through its impressive quantities and diversity of products, including fruits, vegetables, grapes, nuts and grains. This spatial diversity makes it difficult to model ET as surface conditions and plant biomass patterns change drastically over relatively small regions. In the current study, we look to improve and test the sensitivity of physically-derived inputs to the ALEXI/DisALEXI + STARFM modeling scheme in order to better represent the unique canopy architecture, row structure and interrow spacing of these specialty crops grown in California. Specifically, we test improvements to field-scale remotely-sensed Leaf Area Index (LAI) and an operationally focused Land Surface Temperature (LST) product capable of near-real-time ET estimates. We also test the value of adding crop specific information to the modeling structure via a more detailed crop-type/landuse map. These additions allow fields such as grape and almond orchards, which are commonly found adjacent to one another, to be more accurately represented in their differing biophysical characteristics, ultimately improving energy balance modeling and deriving more robust ET estimates.