Graduation Date

Fall 2021

Document Type

Thesis

Program

Master of Science degree with a major in Natural Resources: option Environmental Science and Management

Committee Chair Name

Dr. James Graham

Committee Chair Affiliation

HSU Faculty or Staff

Second Committee Member Name

Dr. Kerry Byrne

Second Committee Member Affiliation

HSU Faculty or Staff

Third Committee Member Name

Dr. Buddhika Madurapperuma

Third Committee Member Affiliation

HSU Faculty or Staff

Subject Categories

Natural Resources

Abstract

The functionality of tropical forest ecosystems and their productivity is highly related to the timing of phenological events. Understanding forest responses to major climate events is crucial for predicting the potential impacts of climate change. This research utilized Landsat satellite data and ground-based Forest Inventory and Analysis (FIA) plot data to investigate the dynamics of Puerto Rico and the U.S. Virgin Islands’ (PRVI) tropical forests after two major hurricanes in 2017. Analyzing these two datasets allowed for validation of the remote sensing methodology with field data and for the investigation of whether this is an appropriate approach for estimating forest health in areas lacking in-situ data. I performed extensive cloud masking processes on the satellite imagery to produce masked, repaired, near cloud-free imagery, which were used to extract phenology metrics and produce annual phenology curves. FIA data was used to estimate forest percent mortality and change in aboveground live biomass (AGLB). Simple and multiple linear regression were used to explore the relationship between the FIA data and the remote sensing derived phenology metrics to analyze and compare trends. Phenology metrics showed a consistent trend of an initial decrease in index values the first year after the hurricanes, followed by a spike in values the second year after. Consistent trends were seen after the hurricanes of a decrease in AGLB, an increase in mortality, and a decrease in phenology values the first year, followed by increase in values the second year after. Significant changes were found in AGLB and in the phenology metrics before and after the hurricanes, however there were no significant linear relationships found between the FIA data and the remote sensing data. Meaningful phenology curves were successfully generated when analyzing a small region with only one forest type and no data gaps. The results, therefore, help in constructing a base understanding of PRVI’s tropical forests dynamic relative to climate change and give a clearer indication of the capabilities of the remotely sensed data. Furthermore, this research demonstrated approaches and techniques that can be further applied to larger, global sustainability goals to sustain living systems in times of climate variability and change.

Citation Style

APA

Share

 
COinS