http://197.159.135.214/jspui/handle/123456789/915 2026-04-23T15:11:16Z 2026-04-23T15:11:16Z Bayah, Ernest Kwame http://197.159.135.214/jspui/handle/123456789/982 2024-09-09T08:49:14Z 2023-01-01T00:00:00Z

Comparative Analysis of Machine Learning Models for High-Resolution Mapping of Soil Organic Carbon Stocks Using Remote Sensing Variables in Northern Ghana Bayah, Ernest Kwame Accurate and comprehensive knowledge of spatial soil characteristics is crucial for environmental modelling, risk assessment, and decision-making. The utilization of Remote Sensing data for Digital Soil Mapping has proven to be a cost-effective and time-efficient alternative to traditional soil mapping methods. However, the potential of Remote Sensing data in enhancing understanding of local-scale soil information in West Africa remains largely untapped. This research aimed to explore the use of satellite data, and laboratory-analysed soil samples to map the distribution of organic carbon (SOC) in Northeastern Ghana. Three statistical prediction models, namely Random Forest, Xtreme Gradient Boosting, and Naïve Bayes were employed and compared. To ensure robustness, internal validation was performed using cross-validation techniques. Analysis of model performance statistics indicated that the RF and XG techniques exhibited slightly superior performance compared to the Naïve Bayes Algorithm, with RF yielding the highest accuracy in most cases. One limitation of Naïve Bayes was its inability to effectively capture non-linear relationships between dependent and independent variables, leading to less accurate predictions of soil properties in unsampled locations. Among the spectral predictors, precipitation data was found to be the most significant in Random Forest and Xtreme Gradient Boosting models, while Soil Organic Matter, Soil Bulk Density, Biomes, and NDVI emerged as prominent terrain/climatic variables in predicting soil properties. Furthermore, the results highlighted Precipitation, Soil Bulk Density, Soil Organic Matter, and Land Surface Temperature as significant predictors in the Naïve Bayes Algorithm. With the growing availability of freely accessible Remote Sensing data, the enhancement of soil information at local and regional scales in data-scarce regions like West Africa can be achieved with relatively minimal financial and human resources. A Thesis submitted to the West African Science Service Center on Climate Change and Adapted Land Use and Université Joseph KI-ZERBO, Burkina Faso in partial fulfillment of the requirements for the Master of Science Degree in Informatics for Climate Change

2023-01-01T00:00:00Z Garba, Nabassebeguelogo Juste http://197.159.135.214/jspui/handle/123456789/934 2024-09-04T09:27:27Z 2022-01-01T00:00:00Z

Improving the quality of Gridded Precipitation Datasets over Burkina Faso Using merging Methods Garba, Nabassebeguelogo Juste The management of climate risks such as droughts, floods and heat waves requires high quality historical climate data that offers good spatial and temporal distribution. To achieve this, rain gauges are installed to provide the most reliable measurement of rainfall. However, the rain gauge network is generally very poor in developing countries, leading to many uncertainties, especially in areas where no rain gauge is installed. To fill this gaps, rainfall estimation based on satellites seems to be a good and cost-effective alternative because they supply information for these areas at a relative low cost. However, these datasets are subject to systematic and random errors inherent to the observation method; therefore, there is a need to adjust them before their use for operational applications and decision making. This study proposes a rigorous method in three-step to improve satellite estimation data for Burkina Faso. The first step is devoted to assessing the accuracy of seven satellite precipitation datasets. Then the best dataset is bias corrected using Empirical Quantile Mapping (EQM) and Time and space-variant (TSV) biasadjustment approaches. The final step is to generate blended datasets between the best corrected datasets and in-situ gauges data to produce more robust estimates of precipitation datasets. This blending was performed with Regression kriging (RK) and Mean Field Bias (MFB) with two interpolation techniques namely Shepard and Spheremap. The main results of the study are the following: The evaluation revealed that TAMSAT and CHIRPS were the best for daily and monthly time scales respectively. EQM method outperformed TSV at daily scale, while at the monthly scale the TSV was more suitable for bias correction. Morever, RK-Spheremap was the best of the four methods for merging satellite and in situ data at both time scales. Thus, the approach proposed in this study has improved the correlation coefficients improved the correlation of the daily data by 85.2% (from 0.147 to 0.999), the Bias by 12.4% (from 0.875 to 0.999) and the RMSE by 95.6% (from 26.494 to 1.175). Concerning the monthly dataset, the correlation coefficients are enhanced by 8.4% (from 0.916 to 1), the Bias by 2.3% (from 0.977 to 1) and the RMSE by 99.9% (from 35.654 to 0.042). This study may help in improving floods and droughts monotoring, as well as climate model validation over Burkina Faso. A Thesis submitted to the West African Science Service Center on Climate Change and Adapted Land Use and Université Joseph KI-ZERBO, Burkina Faso in partial fulfillment of the requirements for the Master of Science Degree in Informatics for Climate Change

2022-01-01T00:00:00Z Degbey, Codjo David http://197.159.135.214/jspui/handle/123456789/933 2024-09-04T09:16:34Z 2022-07-12T00:00:00Z

Artificial Intelligence-based Spatiotemporal Assessment of Historical and Future Shoreline Dynamics: Case Study of Benin Republic Degbey, Codjo David Global Sea Level Rise (SLR), especially caused by global warming, highly threatens the coastal countries by causing coastal erosions, biodiversity loss, coastal flooding, etc. The study of shoreline dynamics is a key aspect of coastal area management, and is useful for reducing risk and vulnerability to climate change within coastal ecosystems. Artificial Intelligence (AI) is an efficient approach to this type of problem because of its advanced multidimensional data extraction and computational tools. This study uses AI methodologies combined with Remote Sensing (RS) and Geographic Information Systems (GIS) tools to perform a spatiotemporal analysis of the shoreline dynamics of the Republic of Benin from 2001 to 2021. Preprocessed multispectral and multitemporal Landsat 5, 7 and 8 surface reflectance images as well as the NASA Shuttle Radar Topography Mission (SRTM) digital elevation data were extracted from the Data Catalog of Google Earth Engine (GEE). Then, the cloud computing platform of GEE was used to perform binary (Sea Water / Non-Sea Water) supervised image classification. The Machine Learning algorithms used were Support Vector Machines (SVM) and Random Forest (RF). QGIS was used to extract historical shorelines from the respective classified images in 5-year intervals. Net Shoreline Movement (NSM), End Point Rate (EPR), Linear Regression Rate (LRR) and Weighted Linear regression Rate (WLR) were the shoreline change statistics derived from Digital Shoreline Analysis System (DSAS), a plugin of ArcGIS. Then, the Kalman Filter model, implemented in the DSAS plugin, was used to provide a “beta” forecast of the shoreline shape respectively for the years 2031 and 2041, with a confidence interval of 90%. A geostatistical analysis of the results mainly showed, for the study period, a clear coastal erosion trend at average rates of -2.17 ± 0.28 m (EPR), -1.51 ± 1.68 m (LRR). A sectorial and deeper analysis revealed that, in terms of NSM, the West side of the Port infrastructure underwent more accretion (41.67% against 25.71% at the East) while the East side experienced more erosion (74.29% against 58.33% at the West). Assuming a Business-as-Usual (BaU) scenario, the projected highest erosion values were found to be -226.42 m (2031) and -146.14 m (2041) in the locality of Avloh plage (Grand-Popo), -68.11 m (2031) and -182.55 m (2041) in the locality of Hillacondji (Grand-Popo), -146.14 m (2031) and -307.30 m (2041) in the locality of Agblangandan (Cotonou). Finally, a web application was designed and implemented to facilitate an easy data access and visualization of results by non-experts and decision makers. A Thesis submitted to the West African Science Service Center on Climate Change and Adapted Land Use and Université Joseph KI-ZERBO, Burkina Faso in partial fulfillment of the requirements for the Master of Science Degree in Informatics for Climate Change

2022-07-12T00:00:00Z Doumbia, Abdramane http://197.159.135.214/jspui/handle/123456789/932 2024-09-04T09:11:04Z 2021-01-01T00:00:00Z

Effect of migration on farmers’ income and food security in the Kayes region, Mali Doumbia, Abdramane In several studies, literature authors have investigated migration. However, discourses on migration are more than just about migration; rather than employed as a factual reference to a particular more or less well-defined social phenomenon (Mueller et al., 2020). However, West Africa’s countries which are more vulnerable to CC has long been characterized by high levels of mobility, a trend that far predates the current configuration of borders established during the colonial era. Mali a country in West Africa is particularly known to be vulnerable due to high climate variability, high reliance on rain-fed agriculture, and limited economic and institutional capacity to respond to CC (Sultan & Gaetani, 2016). This leads Mali be both a country of origin and transit for migrants in the West African region(Sultan & Gaetani, 2016). Some regions of Mali are more affected by migration depending on the different factors. The Kayes region is particularly known for migration patterns both within Africa and to France, some of which date back to the beginning of France’s colonial rule in the late 1800s(REACH, 2020). A region which dominated by the Agriculture, so there is a need is to assess the effect of migration on farmers’ income and food security in that region. A survey has been administrated to 97 households in the Cercle of Kayes: Kayes, Nioro, Diema, Yelimane. It has been established that migration flows, both interregional and international, are explained by the constraints existing in the areas of origin: living conditions, income, and the development potential of these regions. Also, the research found a positive correlation between migration, temperature maximal, and precipitation and the SPEI in Kayes. It is also well seen that, a decrease in the minimal value of SPEI from 2000 to 2020 refers to a severe drought in that period in which the migration rate increased considerably. The impacts of migration in the Kayes region could be positive and negative on both countries’ destination and origin. For a sustainable, the negative impacts dominate the positive ones. In order to decrease the trend of migration in Kayes, the research recommends to (i) conduct information and communication activities that raise awareness of behavior change through local radio communication system(ii) strengthen farmers' land ownership status in Kayes, (iii) economic growth is essential to improve household income, (iv) Promote literacy for women and men. A Thesis submitted to the West African Science Service Center on Climate Change and Adapted Land Use and Université Joseph KI-ZERBO, Burkina Faso in partial fulfillment of the requirements for the Master of Science Degree in Informatics for Climate Change

2021-01-01T00:00:00Z