Soil surface moisture estimation with ERS SAR.PRI images

The estimation of the superficial soil moisture is very important in environmental studies, especially in the field of the hydrology and agronomy. The soil humidity is one of the most important variables of soil physics. In agriculture, it is the main factor of growth and yield of the cultivations. Indeed, in unirrigated land regions the variability of the yield of one year to the other one depends on the difference between the humidity storage and the hydric requirements of the crops. As general rule, the more humid, the bigger yield [1]. Traditionally the water retained in the unsaturated layer of the soil was considered as resultant of a balance of other hydrological and meteorological parameters, of land cover and type of the soil and of relief. The results obtained with the hydrological models employees depend on the density of meteorological stations, the quality of the soil database and relief, and the updating of the land cover. This parameter is very dynamic in agricultural areas, and it influences in decisive mode in the surface hydric status. At the moment, the satellite remote sensors facilitate their estimation using diverse bands of the electromagnetic spectrum (visible, near and thermal infrared, and microwaves), due the particular interactions of the water molecule with the energy in those energy ranges. Given the orbital characteristics of each system, such estimation covers wide areas and it repeats with a certain frequency. Among the sensors that use the microwaves, there are the radars that emit an electromagnetic energy beam and they capture the part of the same one that is backscattered by the earth surface or target. The more used bands by the radars are the L (21 cm), C (5,3 cm) and X (2,8 cm.). In general, for longer wave longitude, bigger penetration. In particular, the microwaves (of 1 mm to 80 cm of wave longitude) present dependence of the properties and states of the soil (dielectric constant, density, salinity, roughness of the surface, vegetable covering, etc.). The microwaves are very sensitive to the dielectric constant of the system soil–water. When increasing the water content in the soil surface layers, the value of the dielectric constant also increases, producing an increment in the backscattering [2]. The important advantage on the other bands resides in that the microwaves have good penetration in the atmosphere (more than 90%), they cross clouds and they are “transparent” the fine and medium rains (up to 25 mm/h). Also, being active sensors, they become independent of the conditions of solar illumination, capturing so much day as night images. A limitation resides in that the detected humidity corresponds at the 5 to 10 cm superficial soil layer, depending on the band, polarization, angle of incidence and state of real humidity of the soil [3]. Numerous studies carried out with SAR images in band C demonstrated the effectiveness of these systems for the monitoring of the surface soil moisture [4] [5]. In Argentina, previous studies linking ERS-1 SAR backscattering intensity with water stored in the soil, proved the relation among these parameters [6].