A Scheme to Estimate Diurnal Cycle of Evapotranspiration from Geostationary Meteorological Satellite Observations

Jing Lu,Li Jia,Chaolei Zheng,Ronglin Tang,Yazhen Jiang

doi:10.3390/w12092369

Abstract

The diurnal cycle of evapotranspiration (ET) is significant in studying the dynamics of land–atmosphere interactions. The diurnal ET cycle can be considered as an indicator of dry/wet surface conditions. However, the accuracy of current models in estimating the diurnal ET cycle is generally low. This study developed an improved scheme to estimate the diurnal cycle of ET by solving the surface energy balance equation combined with simplified parameterization, with daily ET as the constraint. Meteosat Second Generation (MSG) land surface temperature, and longwave and shortwave radiation products were the primary inputs. Daily ET was from the remote sensing-based ETMonitor model. The estimated instantaneous (30 min) ET from the improved scheme outperformed the official MSG instantaneous ET product when compared with in situ half-hourly measurements at 35 flux sites from the FLUXNET2015 dataset, and was also comparable with European Center for Medium-Range Weather Forecasts (ECMWF) ERA5 ET data, with an R2 of 0.617 and root mean square error (RMSE) of 65.8 W/m2 for the improved scheme. Results were largely improved compared with those without daily ET as the constraint. The improved method was stable for the estimation of ET’s diurnal cycle at the similar atmospheric conditions and the accuracy was comparative at different land cover surfaces. Errors in the input variables and the simplification of surface heat flux parameterization affected surface energy balance closure, which can lead to instability of the solution of constants in the simplified parameterization and further to the uncertainty of ET’s diurnal cycle estimation. Measurement errors, different source areas in measured variables, and inconsistent spatial representativeness between remote sensing and site measurements also impacted the evaluation.

Highlights

Turbulent fluxes of sensible and latent heat represent the principal means by which the surface cools and exchanges moisture with the atmosphere [1]
Latent heat flux due to ET is a significant component in the surface energy balance equation [2,3]
Were from Meteosat Second Generation (MSG) products, estimated instantaneous latent heat flux (LE) was strongly related to the MSG ET product, as expected, with an R2 of 0.841 and root mean square error (RMSE) of 33.2 W/m2 (Figure 10b)

Summary

Introduction

Turbulent fluxes of sensible and latent heat represent the principal means by which the surface cools and exchanges moisture with the atmosphere [1]. Evapotranspiration (ET) is the leading way of surface water loss from the land surface. Latent heat flux due to ET is a significant component in the surface energy balance equation [2,3]. ET and latent heat flux are different expressions of the same process and can be converted to each other. Accurate ET estimation is critical in many fields, such as atmospheric and hydrological processes, climate change, crop irrigation management, drought monitoring, and water resources management [4,5,6,7,8]. Current RS-based ET models are mainly classified as land surface

Objectives

Methods

Results

Conclusion