Estimating urban sensible heat flux using satellite-based data

Abstract

A model for calculating sensible heat flux (Q H ) – a primary component of the urban surface energy budget is presented here. Remote sensing data from the NOAA GOES-16 satellite and a highresolution land cover dataset are used as inputs to calculate the spatio-temporal variability in urban sensible heat flux. The primary motivation for this model is to present a cost-effective approach to calculate Q H independent of traditional flux observations and computational methods. The GOES-16 satellite data, which has a moderate spatial and high temporal resolution (2 km square at 5 minute intervals) enables the estimation of Q H over highly heterogeneous urban areas. The model is constructed using an iterative algorithm that uses surface layer turbulence parameterization to solve for Q H as a function of the enterprise GOES-16 Land Surface Temperature product, an urban air temperature model, publicly-accessible ground observations, and the National Land Cover Database (NLCD). Preliminary model validation was performed over a five-month period in 2019. Three (3) ground flux stations in the New York City metro area with varying degrees of urbanization were used for model validation. Statistics from validation found an RMSE of 42.9 W-m -2 , a mean bias of 12.9 W-m -2 , and an R 2 of 0.80. Validation results demonstrate that the algorithm shows good correlation with observed values, suggesting that satellite data can be used as an accessible and cost-effective option to estimate Q H in urban areas.