This study focused on the prediction of land-use changes in Nakhon Ratchasima city using a CA-Markov Model with GIS. Satellite images taken by Landsat-5 (1992), Landsat-7 (2002) and THEOS (2016) were used to predict land use in 2026. In 1992, the most proportion of land usage was built-up areas (47.76%) and followed by green areas (37.45%), bare lands (13.19%), and water bodies (1.60%), respectively. In 2002, the land use comprised built-up areas (56.04%), green areas (35.52%), bare lands (4.80%) and water bodies (3.63%). By 2016, urbanisation had changed the land use pattern, which comprised built-up areas (70.80%), green areas (20.78%), bare lands (6.37%), and water bodies (2.03%). The data were analysed using a change detection matrix and revealed an increase in built-up area at the expense of all other types, especially green areas. The results were in accordance with the prediction model created in two scenarios. Scenario 1 assumed city expansion following past trends, built-up areas (85.88%), green areas (11.67%), bare lands (2.15%), and water bodies (0.30%). Scenario 2 assumed city expansion in accordance with the national strategy, built-up areas (74.91%), green areas (15.77%), bare lands (8.48%), and water bodies (0.84%). The results indicated an expansion of built-up areas and a shrinking of green areas. In Scenario 2, urban expansion was less than in Scenario 1, and preserving the green area seemed more feasible due to governmental restrictions. The results indicated that planning the urbanisation according to the policies development plans, especially in specific areas, contributed to a more efficient urbanisation growth. The city should provide to promote the use of floor area ratio (FAR) and open space ratio (OSR) with urban planning measures as well as increasing the green areas.

The relationship between land surface temperature (LST) and land use in Nakhon Ratchasima was studied using data gathered from three satellite images from Landsat-5 (30th January 1992), Landsat-8 (9 May 2016) and THEOS (17th February 2016). There were four categories of land use: built-up area, green area, bare land and water sources. The split-window concept was used to estimate the LST. In 1992, Nakhon Ratchasima city in Thailand comprised 47.76% built-up area, 37.45% green area, 13.19% bare land and 1.60% water sources. By 2016, the built-up area had increased by 23.04%, the green area had decreased by 16.66%, bare land had decreased by 6.81%, but water sources had increased by 0.43%. Moreover, in 1992 the mean LST was 25.43 C for built-up areas, 24.44 C for green areas, 24.97 C for bare land and 24.75 C for water sources. However, by 2016 the LSTs had increased for each category: 28.74 C for built-up areas (+3.31 C), 27.20 C for green areas (+2.76 C), 28.11 C for bare land (+3.14 C) and 27.72 C for water sources (+2.97 C). The findings indicated that the LSTs increased with the pace of urbanization and changes in land use. Linear regression analysis revealed that built-up land had a positive correlation with LST, where a 1% increase in built-up area increased its LST by 0.146594 C.