Land surface models significantly underestimate the impact of land-use changes on global evapotranspiration. (3rd December 2021)

• Record Type: Journal Article
• Authors: Wang, Qilin; Wang, Yingping; Zhang, Lu; Qin, Shujing; Zhang, Quan; Liu, Pan; Liu, Liu; Zou, Kaijie; Cheng, Shujie; Cheng, Lei
• Is Part Of: Environmental research letters. Volume 16:Number 12(2021)
• Publication Date: 2021-12-03
• Subjects: land-use changes -- terrestrial evapotranspiration -- land surface models -- Budyko framework -- land-use change data interpretation -- model structure; Environmental sciences -- Periodicals; Human ecology -- Research -- Periodicals; Environmental health -- Periodicals; 333.7
• DOI: 10.1088/1748-9326/ac38db
• Languages: English
• ISSNs: 1748-9326

Abstract

Abstract : Despite numerous assessments of the impact of land-use change (LUC) on terrestrial evapotranspiration (ET) that have been conducted using land surface models (LSMs), no attempts have been made to evaluate their performance in this regard globally. Errors in simulating LUC impacts on ET largely stem from LUC data interpretation (LI, i.e. mapping of gridded LUC data into annual plant function types) and model structure (MS, i.e. parameterization of land-surface processes). The objective of this study was to benchmark ET estimates from four LSMs using the Zhang-curve, a prototype of the Budyko framework that has been validated against global hydrological observations and used widely to quantify the impacts of LUC on ET. A framework was further proposed to quantify and attribute errors in estimated ET changes induced by LI or MS. Results showed that all LSMs underestimated ET changes by about 55%–78%, and 37%–48% of the error was attributable to LI, but only 11%–32% of the error was attributable to MS across the four LSMs. From a hydrological perspective, our analysis provided insights about the errors in estimated impacts of LUC on ET by LSMs. The results demonstrated that LUC data interpretation accounted for a larger fraction of errors than LSM structure. Therefore, there is an urgent need for the defining and development of consistent protocols for interpreting global LUC data for future assessments.