Experimental examination of the effect of observation geometry on land surface temperature estimates from remotely sensed ground based thermal imagery

Date
2011
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
In the world of civil and geotechnical engineering, remote sensing has the potential to replace in situ measurements of soils and sediments that are often unfeasible or lacking in spatial and temporal resolution. While visible band is, perhaps the most common form, it lacks the inherent properties of the thermal band which has proven itself in attaining material properties. Ground based applications have increased in popularity due to their relatively low price and ease of setup. However, these applications are performed at a relatively low elevation and objects of interest may be a large distance away. Therefore, the images are obtained at oblique incident angles. The oblique incident angles have an effect on the emissivity and through it the observed temperature. This project, unlike previous experiments, observed the effect of polar observation angle on emissivity, and through it the observed radiant temperatures, for both water and sand surfaces within a thermally controlled laboratory environment. The observations performed using a broadband (7.5-13μm) imager. The methodology was inverted and used to transform angular imagery to temperatures associated with observations normal to the material surface. This project was extended further with a field application. On a bank of the Wolf River in Mississippi, thermal imagery was captured. Thermal diffusivities were calculated for various regions of interest from the thermal imagery. The remotely determined thermal diffusivities were validated by values obtained from methods that exclusively used in situ temperature measurements. In summary, this project outlines the methods for using oblique, time- sequenced, remotely-sensed thermal imagery to obtain the thermal diffusivity of the observed material surface.
Description
Keywords
Citation