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

Accueil du site → Doctorat → États-Unis → 2010 → Modeling evapotranspiration in sub-Saharan Africa : A tool for food security analysis

University of California, Santa Barbara (2010)

Modeling evapotranspiration in sub-Saharan Africa : A tool for food security analysis

Marshall Michael Thomas

Titre : Modeling evapotranspiration in sub-Saharan Africa : A tool for food security analysis

Auteur : Michael Thomas Marshall

Université de soutenance : University of California, Santa Barbara

Grade : Doctor of Philosophy (PhD) 2010

Résumé
In the Sahel, a climatically sensitive region where rain-fed agriculture is the primary livelihood, expected climate-induced decreases in water supply will increase food insecurity. Studies on changes in the water cycle in sub-Saharan Africa are few, due in part to poor calibration of modeled actual evapotranspiration (Ea), a key input in continental-scale hydrologic models. In this work, readily and freely available remote sensing and surface reanalysis data are combined to create Ea model estimates from 1979-2009 for sub-Saharan Africa. Chapter II summarizes the modern (2000- 2009) phase of the analysis, which yielded Ea estimates with a lower bias and higher correlations (particularly in the humid tropics) when compared against observed measurements. This model is most appropriate for near real-time monitoring applications as it uses the latest remote sensing and surface reanalysis data. In chapter III, the transpiration component of the model, which includes a dynamic vegetation parameterization appropriate for use in semi-arid regions, is used to develop a crop stress index that is applied to tracking district-level crop yield estimates. Chapter IV summarizes the findings of the historical (1979-2009) analysis, which summarizes Ea and climate-(surface temperature and precipitation) related trends and implications for mitigating food insecurity in the region. In chapter IV, several land surface models are compared using runoff data from 1979-2000. Trend analysis and harmonic regression were used to determine the timing and magnitude of Ea and climate-related trends from the model with the highest performance. The most significant trends for food security were found in the Sahel. Large decreases in Ea corresponded to decreases in precipitation (soil moisture supply) and increases in surface temperature (atmospheric demand). Several sub-regions were compared to understand the contribution of precipitation reductions and surface temperature increases to decreases in Ea. Although precipitation plays an important role in these sub-regions, as has been shown in previous studies, surface temperature appears to be significant as well, particularly at higher latitudes where temperatures are more extreme. This reveals, under expected climate warming that a shift in the energy balance in the Sahel will further inhibit agricultural development and increase food insecurity in the region.

Mots Clés : Climate reanalysis, Evapotranspiration, Food security, Remote sensing, Sub-Saharan Africa

Présentation (PROQUEST)

Page publiée le 31 décembre 2011, mise à jour le 1er septembre 2017