International Journal of Applied Earth Observations and Geoinformation (Aug 2021)
Using remote sensing information to enhance the understanding of the coupling of terrestrial ecosystem evapotranspiration and photosynthesis on a global scale
Abstract
Understanding the coupling of terrestrial ecosystem evapotranspiration (ET) and photosynthesis (gross primary productivity, GPP) is limited by inherent difficulties to provide accurate approximations of transpiration (T) and leaf-to-air vapor pressure difference (D) that is a key variable needed to define the stomata behaviors in generic methods. To address the issues and better characterize the ET-GPP relationship, we developed a novel remote sensing (RS)-driven approach (RCEEP) based on the underlying water use efficiency (uWUE) method (termed uWUE-Model). RCEEP partitions T from ET using the RS-derived fraction (rsFt) of T in ET and then links T and GPP via RS-derived canopy conductance (rsGc) instead of D. RCEEP, the original uWUE-Model, and two other uWUE-based versions (RT or RG) that only incorporate rsFt or rsGc were adjusted using the calibration data, and then inter-compared in terms of their performances (Nash-Sutcliffe efficiency, NSE) in estimating GPP from ET on a daily and monthly scale for both calibration and validation datasets — two subsets of data from 177 flux sites covering 11 biome types over the globe. Results revealed better performances of RT and RG as compared to uWUE-Model over most biomes, implying remarkable contributions of rsFt and rsGc to a more meaningful relationship between ET and GPP. RCEEP yielded the best performances over all biome types except for evergreen forest with reasonable mean NSE values of 0.67 – 0.68 (0.75) on a daily (monthly) scale. Further comparisons between RCEEP and two existing approaches concerning estimating GPP from ET revealed consistently better performances of RCEEP and thus, positive implications of introducing rsFt and rsGc in bridging ecosystem ET and GPP. Besides, rsFt should be used combined with rsGc to avoid degraded effectiveness for specific biome types (Savannah and Woody Savannah). These results are promising in view of improving or developing algorithms on coupled estimates of ecosystem ET and GPP and understanding the GPP dynamics concerning ET on a global scale.
