Are Remote Sensing Evapotranspiration Models Reliable Across South American Ecoregions?

Thiago Rangel Rodrigues ,Gabriela Posse ,Mauricio Galleguillos ,Eduardo Soares De Souza ,Bergson Guedes Bezerra ,Anne Verhoef ,Natalia Edith Tonti ,Joshua B. Fisher ,A. Moreno,Thiago Valentim Marques ,Suany Campos ,Diego G. Miralles ,José Romualdo De Sousa Lima ,Rodolfo Nóbrega ,Dominik Rains ,Jamil Alexandre Ayach Anache ,Rodolfo Souza ,Antônio Celso Dantas Antonino ,M. S. B. De Moura ,Valéria Peixoto Borges ,David Holl ,Antônio Alves Meira Neto ,Djair Alves De Melo ,Rafael Rosolem ,C. M. S. Silva,Edson Wendland ,Claudio F. Pérez ,Lars Kutzbach ,Jorge F. Perez‐Quezada ,Paulo Tarso Sanches De Oliveira ,Brecht Martens ,José S. Nogueira ,M. I. Gassman,O. M. R. Cabral

doi:10.1029/2020wr028752

Abstract

AbstractMany remote sensing‐based evapotranspiration (RSBET) algorithms have been proposed in the past decades and evaluated using flux tower data, mainly over North America and Europe. Model evaluation across South America has been done locally or using only a single algorithm at a time. Here, we provide the first evaluation of multiple RSBET models, at a daily scale, across a wide variety of biomes, climate zones, and land uses in South America. We used meteorological data from 25 flux towers to force four RSBET models: Priestley–Taylor Jet Propulsion Laboratory (PT‐JPL), Global Land Evaporation Amsterdam Model (GLEAM), Penman–Monteith Mu model (PM‐MOD), and Penman–Monteith Nagler model (PM‐VI). was predicted satisfactorily by all four models, with correlations consistently higher () for GLEAM and PT‐JPL, and PM‐MOD and PM‐VI presenting overall better responses in terms of percent bias (%). As for PM‐VI, this outcome is expected, given that the model requires calibration with local data. Model skill seems to be unrelated to land‐use but instead presented some dependency on biome and climate, with the models producing the best results for wet to moderately wet environments. Our findings show the suitability of individual models for a number of combinations of land cover types, biomes, and climates. At the same time, no model outperformed the others for all conditions, which emphasizes the need for adapting individual algorithms to take into account intrinsic characteristics of climates and ecosystems in South America.

Highlights

Land evaporation, or evapotranspiraEtion (ET ), is the phenomenon by which water is converted from a liquid into its vapor phase over land
In geEneral, ET can be reasonably well predicted by all four models, despite an overall tendency of overestimation by Priestley–Taylor Jet Propulsion Laboratory (PT-JPL) and Penman–Monteith Mu model (PM-MOD), and underestimation by Global Land Evaporation Amsterdam Model (GLEAM)
Our analysis emphasizes the need of improving mEodel ET partitioning, the link between flEawed ET partitioning and poor model skill is not evident based on our results

Summary

Introduction

EvapotranspiraEtion (ET ), is the phenomenon by which water is converted from a liquid into its vapor phase over land. Over the past three decades, eddy covariance (EC) systems have become the stateof-the-art and standard in situ method to quantify land surface energy and mass fluxes for different types of ecosystems (Campos et al, 2019; Restrepo-Coupe et al, 2013; Rodrigues et al, 2016; Wang et al, 2020) These techniques estimate fluxes for areas of relatively limited spatial dimensEionEs ( 1 km2 ) depending on the heterogeneity of the landscape), and they are affected by specific local conditions, such as the occurrence of advection across sharp contrasts in vegetation and/or irrigation conditions, and those caused by topographic features, such as cold air drainage for sloping terrain (Allen et al, 2011; Mauder et al, 2020; Mutti et al, 2019; Rahimzadegan & Janani, 2019; Rwasoka et al, 2011)

Results

Discussion

Conclusion