Search

Land surface temperature (LST) is an important indicator of the earth's environmental analysis which is significantly affected by changes in land use/land cover (LULC). The Landsat 5, 7 and 8 satellite data of January and September of the years 1991, 2001, 2011 and 2016 was used to study the impact of LULC changes on LST and its effect on crop yield (rice and wheat) in SBS Nagar, Ludhiana and Bathinda districts representing three different agro-climatic zones of Punjab (India). The Landsat satellite imageries were classified for the major LULC classes (water, vegetation [including agriculture and forest], built-up and bare soil) in each district using different spectral indices. The results showed that vegetation is the key land provider for built-up development in SBS Nagar and Ludhiana districts, whereas bare soil and water bodies were the main contributor of land for expansion of built-up and vegetation over the years in Bathinda district. Compared with 1991, the surface temperature of all the land cover features has increased over the years and the increase in temperature was higher for built-up features. Among different districts, the rate of increase was highest in Ludhiana district due to higher urban sprawl. The increase in temperature was higher in the direction where land use has been changed from agriculture, bare soil and forest to urban. The air temperature in agro-climatic regions of Punjab can be estimated from LST by multiplying it with a factor of 0.86. Normalized difference vegetation index (NDVI) was significantly positively correlated with rice and wheat yield, but significantly negatively correlated with LST. It is concluded that there is potential link between LULC and crop yield via impacts on horizontal surface temperature gradients which can be differentiated from Landsat data. [ABSTRACT FROM AUTHOR]

Accurate information on the extent of waterlogging is required for flood prediction, monitoring, relief and preventive measures. The rule-based classification algorithms were used for differentiating waterlogged areas from other ground features using Resourcesat-2 AWiFS satellite imagery (Indian Remote Sensing Satellite with spatial resolution of 56 m). Two spectral indices normalized difference water index (NDWI) and modified normalized difference water index (MNDWI) were used for extracting waterlogged areas in Sri Muktsar Sahib district of Punjab, India. These indices extracted the waterlogged areas (cropped areas inundated with water) but the water features were less enhanced in the NDWI-derived image (when compared with MNDWI-derived image) due to negative values of NDWI and, mixing of water with built up features. The water features were more enhanced with MNDWI and the values of MNDWI were positive for water features mixed with vegetation. The overall accuracy of waterlogged areas extracted from the MNDWI image was 96.9% with the Kappa coefficient of 0.89. The digital elevation model (DEM) was extracted from ASTER-GDEM. The relationships among depth to the water table recorded before the incessant rain in the region, DEM and classified MNDWI images explained the differences in the extent of waterlogging in various directions of the study area. These results suggest that MNDWI can be used to better delineate water features mixed with vegetation compared to NDWI. [ABSTRACT FROM PUBLISHER]