Comparison of Backscattering Models at L-Band for Growing Corn

• Published in: IEEE geoscience and remote sensing letters 2011-01, Vol.8 (1), p.24-28
• Main Authors: Monsivais-Huertero, A, Judge, J
• Format: Article
• Language: English
• Subjects: Backscatter; Backscatter for dynamic vegetation; Coherence; coherent and incoherent scattering models; Coherent scattering; Corn; Drying; Frequency; L-band; Mathematical models; Microwave sensors; NASA; Radar scattering; Samarium; Sea surface; Seasons; Soil moisture; soil moisture (SM); soil moisture sensitivity; Stems; Terrain; Vegetation mapping
• ISSN: 1545-598X; 1558-0571
• DOI: 10.1109/LGRS.2010.2050459

The impact of incoherent and coherent formulations on estimates of terrain backscatter (σ terrain 0 ) at L-band for a growing season of corn is examined. The average root mean square difference (RMSD) between the two formulations over the growing season ranged between 3-4 dB, with higher RMSDs at HH polarization (pol), indicating the presence of coherent effects. In the incoherent model, the direct scattering from stems was the primary mechanism, while in the coherent formulation, the interactions between the stems and soil were the primary mechanisms due to the coherent effects. Both incoherent and coherent formulations estimated equally high sensitivities of σ terrain 0 to soil moisture (SM) during early stage under low vegetation conditions. During the early and mid stages, the σ terrain 0 estimated by both formulations exhibited higher sensitivities during dry conditions than wet conditions. In contrast, during the reproductive stage, the σ terrain 0 by the incoherent formulation was more sensitive to the SM at wet conditions than at dry conditions. Based upon the ALOS/SMAP accuracy for σ terrain 0 , the incoherent formulation exhibited the highest sensitivity during the early stage with detection of SM changes as low as 2 vol% for dry condition, whereas the coherent formulation exhibited the highest sensitivity during the mid stage with detection of SM changes as low as 2.5 vol%. The results of this study suggest that the coherent effects should be considered for defining accuracy of SM estimation algorithms for corn at L-band.