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

Titre : Estimation of evapotranspiration in forest areas of the Bermejo River basin based on models and passive microwave remote sensing data

Modelos y aproximaciones basadas en información satelital en microondas pasivas para la estimación de la evapotranspiración en áreas boscosas de la Cuenca del Río Bermejo

Auteur : Barraza Bernadas, Verónica D.

Université de soutenance : Universidad de Buenos Aires

Grade : Doctor 2015

Résumé
In this study, we evaluated and compared optical and passive microwave indices based on retrievals of surface conductance (Gs) and evapotranspiration (ET) following the Penman-Monteith (PM) approach. Given the wide range of satellite data from passive microwave systems available in the last decade (2002-2013) (AMSR-E, AMSR-2, Windsat, SMMI, SAC-D / Aquarius and MWR, SMOS), this thesis used data from passive microwave systems, optical and meterological data to obtain relevant infomation to estimate ET over forest areas. The methodology was based on the estimation of surface conductance from multisensor satellite data and then these estimates were combined with meteorological variables (obtained from meteorological station or re-analysis data) under the Penman-Monteith algorithm. Optical indices (such as, enhaced vegetation index (EVI) and normalized water index (NDWI)) and passive microwave indices (such as, frecuency index (FI) and polarization index (PI)) were used to estimate both surface conductance (Gs) and then ET. To evaluate the relationship between microwave and optical indices with biophysical variables relevant to estimate ET, such as the water content of the leaf canopy, two electromagnetic models were used (PROSAILH and Tor Vergata model). For the validation we used data from Eddy covariance flux tower of several forest types from FLUXNET (www.fluxnet.ornl.gov) and OZflux network (www.ozflux.org.au). Model performance was evaluated using the regression coefficient (r2), and the root mean square error (RMSE). The results obtained was compared to other ET products, such as MODIS MOD16A2, GLDAS/NOAH (Global Land Data Assimilation System) and MERRA (Modern-era retrospective analysis). These results are of particular interest to many ecosystems worldwide due to the spatially distributed meteorological data, becoming a robust alternative to estimate ET. Finally, ET was estimated for the northern Chaco forest using equations obtained in this thesis, combined with reanalysis data and surface models information.

Mots clés : INDICES OPTICOS ; INDICES DE MICROONDAS PASIVAS ; EVAPOTRANSPIRACION ; CONDUCTANCIA SUPERFICIAL ; PENMAN-MONTEITH ; OPTICAL INDICES ; PASSIVE MICROWAVE INDICES ; EVAPOTRANSPIRATION ; SURFACE CONDUCTANCE ; PENMAN-MONTEITH

Présentation (Biblioteca Digital)

Version intégrale