Showing posts with label supervised classification. Show all posts
Showing posts with label supervised classification. Show all posts

Monday, 7 April 2025

Evaluation of Spatial Expansion of Rift Valley Lakes Using Satellite Data: A Brief Study | Chapter 5 | Geography, Earth Science and Environment: Research Highlights Vol. 9

Assessing and studying lake expansion can reveal important information about climate change and regional responses, normally a lake recedes during the dry season and floods during the wet season. In the recent years Lake Nakuru has had variations between dry and wet seasons water levels, suspected that it is caused by increasing watershed land conversion to intensive crop production and urbanization, both reduce the capacity of soils to absorb water; recharge ground water and thus increase seasonal flooding. The present work assessed the expansion and fluctuation of Lake Nakuru in Kenya by using satellite data and information. This study was carried out in Lake Nakuru area in Nakuru County using two methods. In the first method, data was obtained from the USGS website whereby Landsat 7 and 8 images were used, though images collected after May 31, 2003, had data gaps when the SLC failed, the images had about 78% of their pixels missing. In the second method shapefiles were used, to find out how the lake surface area had changed, it was done by using different methods.  Surface water magnitude was measured from optical sensors, such as Landsat. ENVI software was used to process and analyze data from the satellite images. The data was then used to create shapefile to get the area of the lake only. The shapefiles were classified using both Supervised and Unsupervised classification, and the area of the lake was obtained in hectares. The obtained area in hectares was recorded in a table and graphs were plotted to show the trend of the lake in the years 1972-2019. Furthermore, correlation was done by assuming the area of the shapefile before any classification is more accurate, therefore it was compared with the other results obtained by using different methods. Maximum like-lihood gave the best correlation values. For R2 it gave 0.8627 and R was 0.9312. The perfect correlation coefficient is 1, the graphs for correlation gave 0.91 for NDVI, 0.83 for K-MEANS, 0.93 for Maximum Likelihood and 0.92 for Spectral Angle Mapper.  In conclusion, the change was seen using line graph and that the Lake Nakuru area is expanding from the years 2010 to 2018.

 

Author (s) Details

 

Rose Yang Mulama
Department of Physics, University of Nairobi, Nairobi, Kenya.

 

Jephter Ongige Ondieki
School of Aerospace Engineering, Sapienza University of Rome, Roma, Italy.

 

Please see the book here:- https://doi.org/10.9734/bpi/geserh/v9/4942

Saturday, 15 March 2025

Land Use and Land Cover Transformations and their Implications for Conservation: A Study in Mukumbura (Ward 2), Mt Darwin, Zimbabwe, 2002-2022 | Chapter 1 | Geography, Earth Science and Environment: Research Highlights Vol. 7

Land use and land cover (LULC) change is a complex and dynamic process influenced by social, economic, and biophysical factors that can cause significant impacts on ecological processes and biodiversity conservation. Understanding trends of land use land cover (LULC) changes is important for biodiversity monitoring and conservation planning, identifying the areas affected by change and designing sustainable solutions to reduce the changes. The study aims to evaluate and quantify the historical changes in land use and land cover in Mukumbura (Ward 2), Mt Darwin, Zimbabwe, from 2002 to 2022. The objective of the study was to analyze the LULC changes in Ward 2 (Mukumbura), Mt Darwin, Northern Zimbabwe, for a period of 20 years using geospatial techniques. Landsat satellite images were processed using Google Earth Engine (GEE) and the supervised classification with maximum likelihood algorithm was employed to generate LULC maps between 2002 and 2022 with a five (5) year interval, investigating the following variables, forest cover, barren land, water cover and the fields. Findings revealed a substantial reduction in forest cover by 38.8%, water bodies (wetlands, ponds, and rivers) declined by 55.6%, whilst fields (crop/agricultural fields) increased by 93.3% and the barren land cover increased by 26.3% from 2002 to 2022. These findings point to substantial changes in LULC over the observed years. LULC changes have resulted in habitat fragmentation, reduced biodiversity, and the disruption of ecosystem functions. The study concludes that if these deforestation trends, cultivation, and settlement land expansion continue, the ward will have limited indigenous fruit trees. Therefore, the causes for LULC changes must be controlled, sustainable forest resources use practiced, hence the need to domesticate the Indigenous fruit trees in Arborloo toilets. Detailed research of the links between land-cover classes that increased in hectares and those that declined is recommended to establish the real drivers of LULC changes in the study area.

 

Author (s) Details

Musekiwa Innocent Maruza
Department of Wildlife Ecology and Conservation, Chinhoyi University of Technology, Chinhoyi, Zimbabwe.


Edson Gandiwa
Zimbabwe Parks and Wildlife Management Authority, Harare, Zimbabwe.

 

Never Muboko
Zimbabwe Parks and Wildlife Management Authority, Harare, Zimbabwe.

 

Ishmael Sango
Department of Wildlife Ecology and Conservation, Chinhoyi University of Technology, Chinhoyi, Zimbabwe.

 

Tawanda Tarakini
Department of Wildlife Ecology and Conservation, Chinhoyi University of Technology, Chinhoyi, Zimbabwe.

 

Nobert Tafadzwa Mukomberanwa
Department of Geoinformatics and Environmental Conservation, Chinhoyi University of Technology, Chinhoyi, Zimbabwe.

 

Please see the book here:- https://doi.org/10.9734/bpi/geserh/v7/4292