An extensive study result is presented on the feasibility of SAR estimation of surface wind-speed distribution for coastal lagoon, specifically Lakes Shinji and Nakaumi, Japan. Four satellite-borne SARs, JERS-1, ERS-1, ERS-2 and Radarsat, are used to study the relation between SAR-derived normalized radar cross-section (NRCS) and in-situ surface winds, in order to study the dependence of SAR frequency and incidence angle on the sensitivity of NRCS change to wind speed. It is found that SAR NRCSs have good correlations with the in-situ wind data when C-band and relatively high incidence angle is used and when wind speed is moderate to high, indicating that such SARs have a potential to estimate high-spatial-resolution wind field. On the other hand, peculiar signatures are sometimes found especially when wind speed is low, indicating that the SAR is sensitive to very fine roughness conditions that may be caused by slicks, water flows and/or ship wakes. A simple spatial filtering is also tested to detect such signatures to extract the areas where SAR-estimated wind speed is reliable.
Rapid rates of population growth and urban expansion affect local and regional ecosystems, climate, and biogeochemistry. Ninety percent of future urbanization will be in low-income countries. In many low-income regions of the world, the only up to date maps of urban extents are those produced on a global scale. However, in some regions such as East Asia, urban expansion is proceeding so rapidly that only a few years can make a large difference on the ground. Therefore maps of settlement in low-income countries are often outdated, inaccurate or non-existent. This research improves our understanding of the methodological and validation requirements for global urban mapping from low-resolution remote sensing data. Maps derived thereby are then compared against urban boundaries derived from Landsat ETM+ imagery and the other five continental scale urban maps. Maps produced using gridded population density data, nighttime lights and MODIS data use less costly data and are simpler to produce yet proved to be more accurate.
This paper proposes a method for simultaneously estimating land surface temperature (LST) and spectral emissivity using hyper-spectral thermal infrared observations. The main approach is finding the value of LST and atmospheric condition which make estimated emissivity spectrum the most smooth. The proposed method could separate LST, spectral emissivity and atmosphetic effects simultaneously from only an observed radiance data. An evaluation function of the smoothness of emissivity spectrum was also developed. It is based on the characteristic of atmospheric spectrum shape and is robust against the measurement noise. The numerical simulation to verify the proposed method gave a root mean square error (RMSE) of LST of 0.7K and emissivity of 0.011. The method was also applied to Atmospheric Infrared Sounder (AIRS) observations.