Abstract

Coastal Wetlands (CW) provide numerous imperative functions and provide an economic base for human societies. Therefore, it is imperative to track and quantify both short and long-term changes in these systems. In this dissertation, CW dynamics related to hydro-meteorological signals were investigated using a series of LANDSAT-derived normalized difference vegetation index (NDVI) data and hydro-meteorological time-series data in Apalachicola Bay, Florida, from 1984 to 2015. NDVI in forested wetlands exhibited more persistence compared to that for scrub and emergent wetlands. NDVI fluctuations generally lagged temperature by approximately three months, and water level by approximately two months. This analysis provided insight into long-term CW dynamics in the Northern Gulf of Mexico. Long-term studies like this are dependent on optical remote sensing data such as Landsat which is frequently partially obscured due to clouds and this can that makes the time-series sparse and unusable during meteorologically active seasons. Therefore, a multi-sensor, virtual constellation method is proposed and demonstrated to recover the information lost due to cloud cover. This method, named Tri-Sensor Fusion (TSF), produces a simulated constellation for NDVI by integrating data from three compatible satellite sensors. The visible and near-infrared (VNIR) bands of Landsat-8 (L8), Sentinel-2, and the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) were utilized to map NDVI and to compensate each satellite sensor's shortcomings in visible coverage area. The quantitative comparison results showed a Root Mean Squared Error (RMSE) and Coefficient of Determination (R2) of 0.0020 sr-1 and 0.88, respectively between true observed and fused L8 NDVI. Statistical test results and qualitative performance evaluation suggest that TSF was able to synthesize the missing pixels accurately in terms of the absolute magnitude of NDVI. The fusion improved the spatial coverage of CWs reasonably well and ultimately increases the continuity of NDVI data for long term studies.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

2019

Semester

Summer

Advisor

Singh, Arvind

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Civil, Environmental, and Construction Engineering

Degree Program

Civil Engineering

Format

application/pdf

Identifier

CFE0008105; DP0023244

URL

https://purls.library.ucf.edu/go/DP0023244

Language

English

Release Date

February 2020

Length of Campus-only Access

None

Access Status

Doctoral Dissertation (Open Access)

Download

Share

COinS