Land cover classification and quantitative analysis of multispectral data in mountainous regions is considerably
hampered by the influence of topography on the spectral response pattern. In the last years, different topographic
correction (TOC) algorithms have been proposed to correct illumination differences between sunny and shaded areas
observed by optical remote sensors. Although the available number of TOC methods is high, the evaluation of their
performance usually relies on the existence of precise land cover information, and a standardised and objective
evaluation procedure has not been proposed yet. Besides, previous TOC assessment studies only considered a limited set of illumination conditions, normally assuming favourable illumination conditions. This paper presents a multitemporal evaluation of TOC methods based on synthetically generated images in order to evaluate the influence of solar angles on the performance of TOC methods. These synthetic images represent the radiance an optical sensor would receive under specific geometric and temporal acquisition conditions and assuming a certain land-cover type. A method for creating synthetic images using state-of-the-art irradiance models has been tested for different periods of the year, which entails a variety of solar angles. Considering the real topography of a specific area a Synthetic Real image (SR) is obtained, and considering the relief of this area as being completely flat a Synthetic Horizontal image (SH) is obtained. The comparison between the corrected image obtained applying a TOC method to a SR image and the SH image of the same area, i.e. considered the ideal correction, allows assessing the performance of each TOC algorithm. This performance is quantitatively measured through the widely accepted Structural Similarity Index (SSIM) on four selected TOC methods, assessing their behaviour over the year. Among them, C- Correction has ranked first, giving satisfying results in the majority of the cases, while other algorithms show a good performance in summer but give worse results in winter.
Radar based surface soil moisture retrieval has been subject of intense research during the last decades. However, several difficulties hamper the operational estimation of soil moisture based on actually available space borne sensors. The main difficulty experienced so far consists of the parameterization of other surface characteristics, mainly roughness, which strongly influences the backscattering coefficient and harms the soil moisture inversion. This fact, along with the high spatial variability of the surface roughness parameters, makes it necessary to perform intensive roughness measurements in order to invert soil moisture values with an adequate accuracy, what reduces the applicability of the approach. This paper reviews an approach, proposed by Pauwels et al.<sup>8</sup>, in which a combined application of two well documented backscattering models, i.e. the Integral Equation Method model and the Oh model, is carried out following an iterative scheme. The approach can be applied to single configuration scenes acquired over homogeneous roughness conditions and yields estimates of both soil moisture and roughness parameters without performing ground measurements of soil moisture or roughness. The proposed algorithm was applied to a set of five RADARSAT-1 scenes acquired over Navarre (Spain) between February and April 2003. Inverted soil moisture and surface roughness parameters were compared to ground measured reference values over an experimental watershed. Results are encouraging and the possibility of simultaneously estimating both variables opens new application scenarios for radar remote sensing on the study of numerous processes at the soil surface.
The importance of soil moisture on many scientific fields like hydrology, meteorology, crop growth or soil erosion has been addressed frequently. Its characterisation has been a difficult task because of its high spatial and temporal variability. Several point based measurement techniques have been developed with different degree of success, but their conversion to spatially distributed values depends on complex geostatistical techniques. Furthermore, sensor installation and maintenance can be quite tedious. In this background, SAR remote sensing sensors provide valuable information on land surface parameters. The backscattering of the SAR signal depends amongst others on the dielectric constant of the observed surface, which is mainly related to the soil surface water content. It also gives spatially distributed information with a resolution adequate for different spatial scales: from medium or small watersheds to agricultural fields. Its periodicity can be appropriate for calibrating, on a monthly basis, the simulations of distributed hydrologic modelling tools. The present paper reports the first results of an ongoing research of which the main objective is the development of a simple methodology for the calibration of the soil moisture component of distributed hydrological models using SAR data. Five RADARSAT-1 images, acquired between 27/02/2003 and 02/04/2003 over the Navarre region (Northern Spain) have been processed. The calculated backscattering values have been compared to soil moisture and surface roughness ground measurements. Empirical linear regression models have been fitted at three different scales: point scale, field scale and catchment scale, showing acceptable correlation between calculated backscattering values and ground measured soil moisture specially at field and watershed scale. However, consistent trends have not been found probably due to differing local conditions such as surface roughness or vegetation cover. Seeking for a more consistent approach, the physically based Integral Equation Method (IEM) model has been applied. Yet, simulations run by the IEM have not been completely successful probably due to an inadequate characterisation of surface roughness.