Abstract
This dissertation is framed as a retrospective research
concept that analysed and monitored land use change to assess the
impacts of urbanization on climate in cities. Multi-source datasets such
as remote sensing images, vector layers, topographical maps, historical
climatic variables and socioeconomic information were used for the
retrospective landscape studies. Land-use Land-cover (LULC) maps were
produced from historical Landsat series data using support vector
machine information extraction algorithm.
Subsequently, spatio-temporal settlement expansion analysis, change detection and urban growth modelling into the future was implemented to assess potential climate impacts due to urbanization. Thermodynamics of the urban landscape was investigated using relevant discrete historic climatic data and continuous thermal spatial datasets. Climate indices calculation and multiple spatial statistical approaches were used to analyse changes in air and land surface temperature and to detect urban warming and heat island impacts. Urban flood-risk assessment was investigated by integrating multisource geoinformation and morphometric analysis approach.
Using object-based image analysis, the generated urban density information and urban structural types were useful in demonstrating the relevance of integrated geospatial datasets analysis for land use and climate change studies. Two silent urbanization impacts identified in this study were, increased imperviousness which result in land surface temperature modification and urban flood-risk propagation. LULC was observed to have moderated urban micro climate in different urban landscapes of Abuja. The proposed disaggregation concept used in the morphometric analysis in this study also revealed the hydrological processes such as flood-risk can be perennial and imminent due to inadequate natural drainage densities and low bifurcation ratio status of the landscape. The output of synergizing multi-source geospatial datasets facilitated fine-scale human wellbeing and security vulnerability assessment for improved disaster risk reduction in the context of climate impacts in cities.
The main findings of this dissertation is a proof of concept of how integrated datasets and methodical research approach can be used for empirical climate and land use change science at local scale. Therefore, relevant institutions such the Federal Capital Development Authority and policy makers in other regions of Nigeria can adopt the concept demonstrated in this research for rapid assessment of urban landscapes to potential climate change impacts.