Assessing the Potential of Diurnal Land Surface Temperature at High Spatial Resolution in Monitoring Heatwaves Over India

In India, the extreme heat events, characterized by prolonged periods of exceptionally high temperatures have increased in both frequency and intensity. These extreme heat events have impacted the public health leading to mortality and have induced crop stress resulting in reduced agricultural production and economic growth. Higher population density, increased agricultural activities and rapid urbanization make communities in India more exposed and vulnerable to heat waves. Traditionally, heatwaves are monitored and reported based on air temperature data observed at meteorological stations. However, this approach faces a major challenge due to the sparsity of meteorological stations, which often leads to the oversight heat events and its impacts at a local scale. The Land Surface Temperature (LST) observations from the remote sensing thermal infrared sensors offer spatially continuous measurements of surface temperature and are proven to provide insights into heatwave events. The LST temperature anomalies and Diurnal Temperature Range (DTR) are often studied to understand the spatiotemporal pattern of heatwaves. In practice, the Moderate Resolution Imaging Spectroradiometer (MODIS) data acquired during the afternoon and early morning overpasses are used to estimate the DTR. However, it is important to note that the LST observed during MODIS overpasses may not always reflect the true daily minimum and maximum temperature. Since, polar orbiting satellites acquire data once or twice daily, they cannot capture the diurnal variation of LST.

The geostationary satellites provide data every 15 or 30 mins, but they are coarser in resolution and can often miss the localised events. Thus, there is a critical need for high spatiotemporal resolution LST. As a first step, we have developed a hybrid approach that combines multivariate spatial disaggregation and diurnal temperature cycle modelling. This method allows us to derive high spatiotemporal LST data and its diurnal parameters from four LST observations acquired in a day from multiple TIR sensors such as MODIS and VIIRS. Further the diurnal parameters derived from the proposed methodology were compared during an heat wave year and a normal year for different land cover types (cropland, settlements and bareland) over two districts in Northern and central India. The results indicated a significant increase in minimum temperatures around dawn, maximum temperatures at noon, and temperature amplitude during heatwave events. The night time LST were also higher during heatwave for all selected land cover types. It was observed that time at which the maximum LST in a day occurred about 30 minutes earlier (~13.30 local time) during heatwave year than normal year (~14.00 local time). Further, the amplitude of the LST cycle over cropland significantly increased during heatwave year indicating heat stress on vegetation. Thus, diurnal parameters help in early detection of vegetation stress compared to satellite derived vegetation indices which usually detect stress after the occurrence of visible changes. In addition, the spatial disaggregation helped to identify the effect of heatwaves at a much finer spatial scale that were not visible at the 1km resolution of MODIS and VIIRS. In conclusion, our research demonstrates that diurnal LST at high spatial resolution can indicate the effects of heat stress at much finer spatial scales. Ongoing work focuses on extracting high-resolution diurnal parameters directly from geostationary satellite data which can give precise and timely insights for heatwave monitoring in India.