Complementarity of C-Band and Ku-Band Radar Data for Phytomass and Soil Moisture Estimation Over a Sahelian Area
FAYE, Gayane¹; FRISON, Pierre-Louis²; JARLAN, Lionel³; MOUGIN, Eric⁴; HIERNAUX, Pierre⁴; RUDANT, Jean-Paul^2
1Université Cheikh Anta Diop, SENEGAL; ²Université Paris-Est, FRANCE; ³CESBIO, FRANCE; ⁴Geosciences Environnement Toulouse, FRANCE

C band radar data acquired by scatterometers on board ERS or METOP are well suited to estimate surface parameters over Sahelian areas at regional scale [1] - [3].Sahelian pastoral areas are characterized by a long dry season, from October to June, with large extends of bare soils patches in the landscape, followed by a rainy season during which a layer of annual herbaceous grows. The vegetation development is mainly driven by rainfalls, which spatio-temporal pattern has become highly variable. Average annual rainfall increases from the northward Saharo-Sahelian to the southward Soudano-Sahelian regions, ranging from 100 mm to 600 mm, respectively. The high temporal frequency of scatterometers allows an accurate analysis of the temporal variations of the radar signal in relation with the seasonal variations of the surface. Past studies have shown that over Sahelian pastoral areas, the C band radar signal is sensitive to both soil moisture content and to vegetation mass. At the maximum of vegetation development, soil and vegetation contributions play equivalent roles. Consequently, an additional data set is used to estimate soil moisture content and herbaceous production. In the past, rainfalls estimated with infrared data acquired by the METEOSAT satellite were used as input for a Sahelian vegetation growth model, simulating the soil moisture content and the development of the herbaceous layer [4]. A radiative transfer model, taking into account the interaction between an electromagnetic wave and the surface simulates the radar signal [5].Coupling both models minimizes the simulations with radar observations and allowed to estimate the annual phytomass production at a regional scale [6]. In this study, the complementarity of C-band and Ku-band radar data are analyzed. Ku-band scatterometer data have been acquired by the SEAWIND sensor during the period extending from 1999 to 2009, while ASCAT sensor acquires data at C-band since 2007. A study area has been chosen in the Ferlo region, Senegal, where field data have been collected through regular ground survey. In particular, the vegetation production is assessed by measuring the herbaceous mass at maximum development in October. It provides an estimate of the annual production of the herbaceous layer. First results show that Ku band is more sensitive to vegetation layer than C band even if soil contribution remains significant. The ability of the two scatterometer data sets to recover the soil moisture and the annual herbaceous production is assessed for each separately and for the two combined. REFERENCE [1] Frison P.-L., Mougin E., Hiernaux P., 1998: Observations and interpretation of seasonnal ERS-1 Wind scatterometer response over the Sahel. Remote Sensing of Environment, vol.63, 233-242. [2] Frison P.-L. and Mougin E., 1996: Monitoring global vegetation dynamics with ERS-I wind scatterometer data. Int. J. Remote Sensing, vol. 17, n° 16, 3201-3218. [3] Magagi, R. D., and Kerr, Y. H. (1997), Characterization of surface parameters over arid and semi-arid areas by use of ERS-1 wind-scatterometer. Remote Sens. Rev. 15:133 - 155. [4] Mougin, E., Lo Seen, D., Rambal, S., Gaston, A., and Hiernaux, P. (1995), A regional Sahelian grassland model to be coupled with multispectral satellite data. I. Description and validation. Remote Sens. Environ. 52:181- 193. [5] Karam, M. A., Fung, A. K., Amar, F., et al. (1995), A microwave polarimetric scattering model for a forest canopy based on vector radiative transfer theory. Remote Sens. Environ. 53:16-30. [6] Jarlan L., Mazzega P., Mougin E., Lavenu F., Marty G., Frison P.-L., Hiernaux P., 2003: Mapping of Sahelian vegetation parameters from ERS scatterometer data with an evolutionnary algorithm. Remote Sensing of Environment, vol. 87, n°°, 72-84