Date of Award

1-1-2018

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

Department of Information Science

Content Description

1 online resource (ii, viii, 126 pages) : illustrations (some color), maps (some color)

Dissertation/Thesis Chair

Alexander Buyantuev

Committee Members

John B Davis, James Mower

Keywords

Image fusion, Phenology, Remote Sensing, Urban Heat Island, Plant phenology, Urban agriculture, Crops, Landsat satellites, Earth resources technology satellites, Remote sensing

Subject Categories

Environmental Sciences | Remote Sensing

Abstract

This dissertation centrally focuses on developing and validating the use of fused satellite imagery to monitor the effects of the urban area on plant phenology, specifically the timing of the start and end of the growing season (SOS and EOS, respectively). In the first paper, Chapter 2, data from the Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat were fused together using the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) to produce a time series of a high spatial and high temporal resolution vegetation index. From this time series, the SOS and EOS were extracted and compared between the urban and exurban developed areas. This paper used two versions of the fusion software and the results from both versions generally supported the hypothesis that the urban areas have an earlier SOS, later EOS, and therefore a longer length of season (LOS). In addition, the 30m Landsat-like resolution of the fused imagery allowed differentiation between urban and exurban areas at a much finer spatial resolution, allowing for a better understanding of heterogeneous effects than would be possible at a coarser resolution of MODIS (500m). However, two issues were identified. First, the two software versions showed minor differences in results. Without in-depth validation of the software, it was not known which version produced the most reliable results and therefore which version should be adopted in future research. Second, the paper did not attempt to compare the results with ground conditions. Therefore there was no evidence to support that the observed parameters were actually related to true SOS or EOS, or if they were related to some other phenomena. The second and third papers investigated these two issues. The second paper focused on a technical comparison of the ability of the original and the most recent version (v1.2.2) of the STARFM software to predict high temporal and high spatial resolution fused images. The results show that v1.2.2 can be adopted. The third paper used v1.2.2 to produce fused images for three US cities that coincided with in situ citizen science phenology data of the red maple (Acer rubrum). Comparison of the phenophase dates derived from the fused imagery with these in situ data showed that the amount of relationship varied between the two depending on the phenophase and city. In addition, although usually less than half of the variation in the phenology produced from the fused data was explained by the observed phenology of the red maple, there were statistically significant relationships between the two.

Download

Share

COinS