Mapping of coupling hot spots of satellite derived latent heat flux in indian agro-climatic regions

Abstract. This study focuses on the understanding and mapping of coupling hotspots of LE versus terrestrial and meteorological parameters. Single source surface energy balance model was used to derive surface energy balance parameters. Agro climatic region wise monthly information of terrestrial, energy balance and meteorological parameters were derived during June- September from decadal analysis of MODIS data (2003–2012) over India (68–100°E, 5–40°N) at 5 km spatial resolution. Information on rainfall was obtained from gridded rainfall data (1° × 1° spatial resolution) from Indian Meteorological Department (IMD). The spatiotemporal variability of the parameters such as rainfall, evapotranspiration (ET), evaporative fraction (EF), soil water index (SWI), land surface temperature (LST) and air temperature (T_a) showed strong influence on seasonal LE fluctuation. LE showed positive linear coupling with ET (0.52 2 ≤ 0.91), EF (0.79 ≤ R² ≤ 0.96), SWI (0.80 ≤ R² ≤ 0.93) and negative exponential coupling with LST (0.63 ≤ R² ≤ 0.87), T_a (0.55 < R² ≤ 0.83). The pixel based knowledge of the parameters was incorporated into hierarchical decision rule algorithm and pixel-by-pixel segmentation of monthly coupling of LE versus parameters (ET, EF, SWI, LST, T_a) was generated. The rainfall zonations in a spatiotemporal domain were done based on the LE couplings that clearly demarcated the highest (West Coast Plains and Hills Region, Himalayan region), moderate (Gangetic Plains and Hills Regions, and the Plateau and Hills Regions) and lowest rainfall (Western dry region) areas. The transition of zone-wise availability of rainfall (both surplus and deficient) can be very well understood from the seasonal dynamics of the LE couplings.