Land surface temperature (LST) changes have been studied using different parameters and data sets to assess the heat effects in urban areas. Depending on landsat data, single window algorithm and split window algorithm are applied to generate LST output to assess the influence of urban green areas. During the present study, both these methods were used to retrieve LST from landsat TM, ETM+ and landsat 8 images. Correlation between LST and NDVI revealed that the latter is a good indicator of the former and that over 76 LST variability is due to randomly scattered distribution of the vegetation. The four base years, 1985, 1995, 2006 and 2015 showed a gradual increase in maximum LST as well as its expanse, but for a slight decrease in 1995. Out of the total area of 526.47 km2, very low ( 31oC) temperatures existed in 65.10 (342.76 km2) of the area in 1985. Subsequently, though very low, low and moderate temperature classes increased through 1995, 2006 and 2015, high and very high temperature classes became dominant in expanse. The R2 value between NDBI and LST was direct and positive. The higher the NDBI value, the higher was the LST. Temperature measured in the selected Public Parks compared to the surroundings disclosed up to 5o difference, indicating that Urban Public Parks play a vital role in mitigating urban heat island effects.
展开▼
