Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → États-Unis → 2019 → Multiscale Modeling of Crop Evapotranspiration And Carbon Dynamics in East-Central Texas Using Satellite Remote Sensing and Crop Growth Models

Texas A&M University (2019)

Multiscale Modeling of Crop Evapotranspiration And Carbon Dynamics in East-Central Texas Using Satellite Remote Sensing and Crop Growth Models

Menefee, Dorothy Scott

Titre : Multiscale Modeling of Crop Evapotranspiration And Carbon Dynamics in East-Central Texas Using Satellite Remote Sensing and Crop Growth Models

Auteur : Menefee, Dorothy Scott

Université de soutenance : Texas A&M University

Grade : Doctor of Philosophy (PhD) 2019

Résumé
Over the past few decades, there have been increasing concerns over the impacts of agricultural practices on global carbon and water dynamics. Understanding these dynamics in agroecosystems will help efforts to reduce carbon emissions and improve water use efficiency in agriculture. Carbon and water vapor exchange from a continuous conventionally-tilled cotton (Gossypium hirsutum) field and a continuous conventionally-tilled corn (Zea mays) field in College Station, Texas were evaluated using two eddy covariance (EC) systems that were installed in early 2017. Satellite imagery data from PlanetScope was incorporated to develop and validate gross primary productivity (GPP) models for both crops. Data from 2017 was used to develop the models and data from 2018 and 2019 was used to validate the models. The Decision Support System for Agrotechnology Transfer (DSSAT) software system was used to model crop growth and evapotranspiration (ET). In cotton, there were substantial differences in carbon fluxes between the years, which were driven by differences in meteorological conditions. A wet post-harvest season in 2018 spurred the growth of weeds, primarily volunteer cotton, morning glory (Ipomoea cordatoriloba), and Texas panicum (Uruchloa texana), resulting in substantial off-season carbon uptake in 2018 (374.2 g C m^-2 in 2018 compared to 100.1 g C m^-2 in 2017). The SAVI-based model was able to simulate GPP in corn production successfully ; with a standard error for the model’s validation of 1.74 g C m^-2 in 2018 and 1.50 g C m^-2 in 2019. The cotton GPP models performed adequately during the 2018 validation with an average standard error of 1.78 g C m^-2 ; however, there was significantly more error in the 2019 validation effort (2.36 g C m^-2 ). The DSSAT system was able to estimate ET in dryland corn ; however, the models had a tendency to underestimate ET, which was more pronounced in the Priestly-Taylor models. The average nRMSE was 0.36 mm for the Priestly-Taylor models and 0.35 mm for the FAO-56 models. There was mixed success in the ability of the DSSAT system to simulate ET in cotton. The model agreement with observed ET was low for 2019, with an average r^2 of 0.14.

Présentation

Page publiée le 28 mai 2021