Presenter Information

Ina Goodman, University of Wyoming

Department

Department of Botany

First Advisor

Dr. Ramesh Sivanpillai

Description

Difference between map products derived from remotely sensed images and data collected from the field is common due to numerous aspects. Sun’s position with respect to the horizon, or sun angle, is related to the amount of light reflected by features on the Earth’s surface, which in turn influences how they appear in the images. Objective of this research was to assess how differences in sun incidence angle influence our ability to distinguish water bodies in remotely sensed images. Landsat images acquired from different months (or sun incidence angles) were downloaded from US Geological Survey for Bull Lake, Fontenelle, Key Hole, and Pilot Buette Reservoirs. Pixels in these images were classified using ISODATA algorithm, and resultant clusters assigned to water or non-water classes. Total surface area of each water body was calculated along with difficulties in interpreting and assigning the clusters to either class. Findings from this research will be useful in determining the average error of water body remote sensing data throughout the year in relation to water body geography and surrounding water body terrain. Specifically, findings will provide insights about whether the shadows cast upon the water body, due to the increasing and decreasing sun incidence angle throughout the year, led to greater difficulty classifying images; and consequently, have more uncertainty.

Comments

Oral Presentation, WyomingView

Share

COinS
 

Effect of Sun Incidence Angle on Classifying Water bodies in Landsat images

Difference between map products derived from remotely sensed images and data collected from the field is common due to numerous aspects. Sun’s position with respect to the horizon, or sun angle, is related to the amount of light reflected by features on the Earth’s surface, which in turn influences how they appear in the images. Objective of this research was to assess how differences in sun incidence angle influence our ability to distinguish water bodies in remotely sensed images. Landsat images acquired from different months (or sun incidence angles) were downloaded from US Geological Survey for Bull Lake, Fontenelle, Key Hole, and Pilot Buette Reservoirs. Pixels in these images were classified using ISODATA algorithm, and resultant clusters assigned to water or non-water classes. Total surface area of each water body was calculated along with difficulties in interpreting and assigning the clusters to either class. Findings from this research will be useful in determining the average error of water body remote sensing data throughout the year in relation to water body geography and surrounding water body terrain. Specifically, findings will provide insights about whether the shadows cast upon the water body, due to the increasing and decreasing sun incidence angle throughout the year, led to greater difficulty classifying images; and consequently, have more uncertainty.