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

Accueil du site → Projets de développement → Projets de recherche pour le Développement → 2013 → REMOTE SENSING OBSERVATIONS FOR MANAGING IMPACTS OF AGRICULTURAL DROUGHT ON CROP YIELDS IN HETEROGENEOUS LANDSCAPES

United States Department of Agriculture (USDA) 2013

REMOTE SENSING OBSERVATIONS FOR MANAGING IMPACTS OF AGRICULTURAL DROUGHT ON CROP YIELDS IN HETEROGENEOUS LANDSCAPES

Drought Crop Yield

United States Department of Agriculture (USDA) Research, Education & Economics Information System (REEIS)

Titre : REMOTE SENSING OBSERVATIONS FOR MANAGING IMPACTS OF AGRICULTURAL DROUGHT ON CROP YIELDS IN HETEROGENEOUS LANDSCAPES

Identification : FLA-ABE-005229

Pays : Etats Unis

Durée : Jan 15, 2013 à Sep 30, 2017

Domaine : soil moisture ; remote sensing ; microwave remote sensing agricultural drought

Partenaire : UNIVERSITY OF FLORIDA G022 MCCARTY HALL GAINESVILLE,FL 32611

Objectifs
The overall goal of this project is to develop a framework that utilizes new soil moisture product from the recent and upcoming satellite missions at 10- 50km with other satellite products to provide root zone estimates at meaningful spatial scales of 1km for quantifying impacts of agricultural drought on crop yields. The primary objectives of this project are (1) to implement an upscaling/downscaling methodology to obtain near-surface soil moisture at 1km (2) to conduct sensitivity studies and quantify errors in soil moisture for simulated data-poor scenarios (3) to obtain root zone soil moisture and crop yields at 1km through data assimilation.

Descriptif
The downscaling methodology will be evaluated using a synthetic experiment developed recently by the PI. The experiment uses coupled hydrology-crop growth models linked with microwave algorithms to generate land surface temperature (LST), leaf area index (LAI), crop yield, SM, and RZSM for two growing seasons of corn and cotton at different spatial scales. Such datasets are excellent to test data-driven methodologies, independent of model physics. The methodology will be implemented in three predominantly agricultural regions in the lower La Plata Basin in South America. The project will use remote sensing (RS) and in situ observations spanning two growing seasons from 1/10 to 4/12, during which ESA-SMOS observations are available to demonstrate the feasibility of the proposed approach. The downscaling methodology is based upon theoretic learning principles. Its implementation has two steps. First, a transformation function is created at fine resolution using in situ and fine-scale remotely sensed observations to probabilistically relate soil moisture to land surface and meteorological conditions. This provides and initial estimate of soil moisture at fine resolution. Second, the coarse scale satellite observations are merged with the initial estimate from step 1 using PRI to obtain an improved estimate of soil moisture. This remotely sensed soil moisture is assimilated using ensemble Kalman Filtering technique into a crop model to provide root zone soil moisture and crop yield

Présentation : USDA

Page publiée le 31 octobre 2015, mise à jour le 22 novembre 2017