DSpace Repository

MONITORING OF TREE SPECIES USING OBJECT BASED TIME SERIES ANALYSIS: -A CASE STUDY OF ACACIA XANTHOPHLOEA SPP IN LAKE NAKURU NATIONAL PARK, KENYA

Show simple item record

dc.contributor.author OSIO, ANNE ACHIENG
dc.date.accessioned 2024-02-26T08:55:53Z
dc.date.available 2024-02-26T08:55:53Z
dc.date.issued 2023-08
dc.identifier.uri http://hdl.handle.net/123456789/1947
dc.description PhD Thesis en_US
dc.description.abstract ABSTRACT The significance of wetlands as suppliers of ecosystem services to flora and wildlife in conservation zones is examined in this study. Acacia xanthophloea trees in Kenya's Lake Nakuru National Park are the subject of particular attention since they are in danger of going extinct as a result of repeated floods. This study attempts to detect and quantify the damage to the trees over time in order to comprehend the rate of their deterioration in order to address this issue. To identify and categorize Acacia xanthophloea trees in the Lake Nakuru riparian reserve, the study used an object-based image approach to determine the most efficient model for this task, a variety of spaceborne and aerial sensing technologies were used, including multi-temporal and multi-spatial. The overall goal of the study was to quantify Acacia xanthophloea and related plant species in terms of spectral, spatial, and contextual features in order to determine their degree of deterioration in both space and time. The study shows that the OBIA-Random Forest model and particular training techniques successfully and accurately identified Acacia xanthophloea trees. To identify riparian vegetation before and after floods, various algorithms were used, revealing changes in vegetation coverage brought on by flooding. The health and changes in vegetation degradation were monitored over time using NDVI (Normalized Difference Vegetation Index) maps. Classification of non-degraded forests, degraded forests, and submerged degraded forests was done using temporal profiles based on Landsat time series. The study also tested a variety of data processing methods, including the use of CNN frameworks on datasets based on UAVs and high-resolution Pleiades-1A images. The findings indicated alterations in Acacia xanthophloea coverage throughout various time periods and places, pointing to both gains and losses in tree populations. The importance of wetlands and Acacia xanthophloea trees in providing ecosystem services is highlighted by this study's findings. The study used a range of remote sensing tools and analysis techniques to identify, measure, and monitor the damage done to these trees over time. The results have consequences for conservation initiatives and offer guidance for managing and preserving biodiversity in ecosystems with similar wetlands. In order to stop the extinction of Acacia xanthophloea trees and improve general conservation practices, the technique chosen, incorporating the OBIA approach, provides a useful tool for swift response and management plans. en_US
dc.language.iso en en_US
dc.publisher Technical University of Kenya en_US
dc.subject OBIA, en_US
dc.subject Acacia xanthophloea, en_US
dc.subject Phenology, en_US
dc.subject Machine Learning, en_US
dc.subject Deep Learning, en_US
dc.subject Landsat, en_US
dc.subject ALOS PALSAR, en_US
dc.subject Pleiades, en_US
dc.subject Sentinel-1, en_US
dc.subject Sentinel-2, en_US
dc.subject Synthetic Aperture Radar, en_US
dc.subject Unmanned Aerial Vehicles, en_US
dc.subject Time series en_US
dc.title MONITORING OF TREE SPECIES USING OBJECT BASED TIME SERIES ANALYSIS: -A CASE STUDY OF ACACIA XANTHOPHLOEA SPP IN LAKE NAKURU NATIONAL PARK, KENYA en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search Repository


Browse

My Account