Satellite passive microwave remote sensing for estimating diurnal variation of leaf water content, as a proxy of evapotranspiration, in the Dry Chaco Forest, Argentina.

VERONICA BARRAZA; FRANCISCO MATÍAS GRINGS; PAOLO FERRAZZOLI; ANIBAL CARBAJO; ROBERTO FERNANDEZ; HAYDEE KARSZENBAUM

San Francisco

Congreso; American Geophysical Union Fall meeting 2012; 2012

American Geophysical Union

CONTROL ID: 1488291 TITLE: Satellite passive microwave remote sensing for estimating diurnal variation of leaf water content, as a proxy of evapotranspiration, in the Dry Chaco Forest, Argentina. AUTHORS (FIRST NAME, LAST NAME): Veronica Barraza Bernadas1, Francisco Grings1, Paolo Ferrazzoli2, Anibal Carbajo3, Roberto Fernandez4, Haydee Karszenbaum1 INSTITUTIONS (ALL): 1. Teledeteccion cuantitativa, Instituto de astronomia y fisica del espacio, Ciudad Autonoma de Buenos Aires, Argentina. 2. DISP, Tor Vergata University, Roma, Italy. 3. Ecología de Enfermedades Transmitidas por Vectores, 3IA, San Martin, Argentina. 4. IFEVA-Ecología, Facultad de Agronomía, Ciudad autonoma de Buenos Aires, Argentina. ABSTRACT BODY: Evapotranspiration (ET) is a key component of water cycle, which is strongly linked with environmental condition and vegetation functioning. Since it is very difficult to robustly estimate it from remote sensing data at regional scale it is usually inferred from other proxies using water balance. This work describes a procedure to estimate ET in a dry forest by monitoring diurnal variation of leaf water content (LWC), using multitemporal passive microwave remote sensing observations. Hourly observations provide the opportunity to monitor repetitive diurnal variations of passive microwave observations, which can only be accounted by changes in LWC (which is itself related to water vapor that enters to the atmosphere from land surface). To this end, we calculated the vegetation frequency index (FI) as FI= 2*(TBKa-TBX)/ ((TBKa +TBX)), where TBKa and TBX indicate brightness temperatures at 37 and 10.6 GHz respectively. There is both theoretical and experimental evidence that link this index to microwave to LWC. The index was computed for vertical polarization, because it presents higher correlation with vegetation state. At diurnal temporal scale, changes in LWC are commonly very small. Nevertheless, it was previously shown that passive remote sensing data (FI computed using Ku and Ka bands) acquired at different hours can be used to estimate the seasonal changes in ET. In this work, we present a procedure based on the hourly changes of FI, which are interpreted as changes in LWC. In order to present a quantitative estimation, the discrete forest model described in (Ferrazzoli and Guerriero, 1996) has been used to simulate the variations of FI with LWC. To illustrate the procedure, AMSR-E and WINDSAT data from 2007-2009 at X and Ka bands were used, and up to four observations per day at four different local times (2.30 am, 7.00 am, 2.30 pm and 7.00 pm) were analyzed. The region addressed is the area of the Dry Chaco forest located in Bermejo River Basin in Argentina (22-27°S, and 58-66°W).This area is characterized for being an open dry forest (20% of tree crown cover), with mean annual temperatures between 20 and 22 °C, mean summer temperatures between 24 and 27 °C and minimal annual rainfall (500 mm). The annual behavior of diurnal LWC shows a range increase in summer and a decrease in winter, being correlated with vegetation annual growing season (foliation/defoliation). For summertime, our results show a decrease of FI values from 7.00 am to 2.30 pm and an increase between 2.30 pm to 7.00 pm. According to the interaction model, these observed changes of FI corresponded to an increase in LWC from ~0.4 g/g to 0.5 g/g in six hours (7.00 am - 2.30 pm) and then a similar decrease for the 2.30 pm to 7.00 pm period. We hypothesize that this daily variations (an increase of LWC during the sunny hours) could be related to more water being available at the leaves, to cope with evaporation needs in order to achieve positive diurnal water balance during the hot diurnal hours. This could be a specific adaptation to high temperature and drought environmental conditions, like the ones present in Chaco. Ferrazzoli P., Guerriero L.,(1996)"Passive microwave remote sensing of forests: a model investigation", IEEE Transactions on Geoscience and Remote Sensing, vol.34, n. 2, pp.433-443. KEYWORDS: [0480] BIOGEOSCIENCES / Remote sensing.