Heliyon (Jul 2023)
Detailed urban roughness parametrization for anthropogenic heat flux estimation using earth observation data
Abstract
Anthropogenic Heat (AH) emissions modify the energy balance in urban areas and is crucial for urban microclimate modelling and improved weather forecast modelling. Therefore, the present study conducted on Delhi and its surroundings firstly aims to estimate AH using Earth Observation (EO) data of Landsat 8 then, evaluate the impact of detailed urban roughness parameterization on the estimation of AH and further validate the obtained flux values with ground based observations of Large Aperture Scintillometer (LAS) setup. The study has been conducted over three time periods for October 2017, March 2018 and June 2018 by processing six Landsat tiles. Three methods have been employed on EO data for AH computation i.e. single urban roughness value for entire study area (Method 1), LULC based roughness values adopted from the literature (Method 2) and lastly, detailed pixel-by-pixel varying roughness values calculated from fine scale urban parameterization (Method 3) for each time period, The average AH values are higher for the month of June 2018: 359.91, 368.57 and 359.16 W/m2 as compared to month of March 2018 (322.44, 330.84 and 298.35 W/m2) and October 2017 (318.00, 331.04 and 306.71 W/m2) for method 1, method 2 and method 3 respectively. Net radiation and Sensible Heat Flux shows a good correspondence with in-situ measurements for most of the tiles and method 3 shows better spatial distribution of fluxes as compared to other two methods. However, due to difference in approach for estimation of fluxes (LAS setup computes latent heat flux as residual while EO based approach computes AH as residual), conclusive results could not be drawn with respect to aptness of a single method. Accurate estimation of AH in urban areas can assist further in formulating policies, regulations and action plans related to mitigation and control of heat stress, climate change and improved weather forecasting.
