CA-ANN based LULC prediction and influence assessment on LST-NDVI using multi-temporal satellite images

• Published in: Environmental earth sciences Vol. 83; no. 5; p. 144
• Main Authors: Raza, Danish, Khushi, Mudassar, SHU, Hong, Aslam, Hasnat, Saleem, Muhammad Sajid, Ahmad, Adeel, Mirza, Sahar, Saeed, Urooj, Khan, Sami Ullah
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01-03-2024; Springer Nature B.V
• Subjects: Agricultural land; Barren lands; Biogeosciences; Cellular automata; Earth and Environmental Science; Earth Sciences; Emissivity; Environmental Science and Engineering; Geochemistry; Geology; Green infrastructure; Heat; Heat waves; Hydrology/Water Resources; Land cover; Land surface temperature; Land use; Landsat; Landsat satellites; Original Article; Plant cover; Reflectance; Remote sensing; Satellite imagery; Satellite observation; Surface temperature; Temperature; Temperature differences; Temperature gradients; Temperature patterns; Terrestrial Pollution; Urban agriculture; Urban areas; Urban heat islands; Urban sprawl; Urbanization; Vegetation; Vegetation cover; Urban growth; Vegetation change; Maximum likelihood classification; Spatial correlation; Cellular Automata
• ISSN: 1866-6280; 1866-6299
• DOI: 10.1007/s12665-024-11467-8

The rapid urbanization observed in various regions worldwide has led to the transformation of agricultural, forest, and green spaces into grey areas. The expansion of these grey areas contributes to heat wave events and an overall increase of average temperatures in specific regions due to their high reflectance properties. This study focuses on analyzing the heat island phenomenon resulting from urban growth and vegetation change over the past 30 years (1990–2020) in three key districts, namely Sheikhupura, Chiniot, and Pakpattan, located in the Punjab province. Landsat data is utilized to assess the urban heat island (UHI) effect caused by land cover changes in these regions. The maximum likelihood classification method is employed to classify remote sensing data, enabling the identification of heat islands based on surface emissivity. The UHI has main influence on built-up land areas therefore, to forecast future scenarios of LULC and their impact on LST, the study employs cellular automata (CA) modeling and incorporates several spatial variables to predict LULC for the year 2050. Four primary land cover classes, namely vegetation including cropland, built-up, water, and barren land, are utilized to examine temperature differences across different land cover types, aiming to understand the impact of land cover changes on temperature patterns. The findings also indicate that barren land and urban areas exhibit the highest thermal reflectivity, contributing significantly to increased surface temperatures. Over the past three decades, the surface temperature has risen by 1.65 °C in Chiniot, 0.86 °C in Sheikhupura, and experienced a slight decrease of – 0.04 °C in Pakpattan. The temperature decrease in Pakpattan can be attributed to a substantial reduction in barren land, which has been converted into agricultural and urban areas. This indicates a negative correlation between land surface temperature (LST) and vegetation cover. The changes in land use and land cover (LULC) in the study area are found to be relatively insignificant.