Scholarlink

<p>Water use efficiency (WUE) is a key indicator of the coupling relationship between water and carbon in terrestrial ecosystems and is crucial for understanding water use characteristics and their responses to environmental changes across different ecosystem types. However, substantial uncertainty remains regarding the magnitude and dominant controls of WUE in various ecosystems. In this study, we synthesized daily-scale eddy covariance flux data, meteorological observations, and remote sensing datasets from 21 ChinaFLUX sites, covering forest, cropland, grassland, and wetland ecosystems, to systematically examine the spatiotemporal patterns of WUE. A random forest (RF) algorithm was employed to quantify the relative importance of environmental drivers. The results show that WUE significantly varied among ecosystems, with forests and croplands exhibiting the highest values, and grasslands the lowest (<i>P</i> &lt; 0.05). Air temperature (<i>T</i><sub><i>a</i></sub>) was identified as the primary determinant of WUE (average contribution: 25.62%), followed by leaf area index (<i>LAI</i>, 19.41%), net radiation (<i>R</i><sub><i>n</i></sub>, 16.92%), soil water content (<i>SWC</i>, 13.07%), and vapor pressure deficit (<i>VPD</i>, 11.99%). Wind speed (<i>WS</i>) and precipitation (<i>P</i>) contributed relatively less (&lt; 10%). Furthermore, the dominant environmental controls of WUE varied across ecosystems: <i>T</i><sub><i>a</i></sub> primarily influenced forest WUE, <i>LAI</i> was the key factor for croplands, while <i>SWC</i> and <i>VPD</i> had stronger effects on wetlands and grasslands. WUE also exhibited significant spatial heterogeneity across climatic zones: in arid and semi-humid regions, WUE was mainly regulated by <i>T</i><sub><i>a</i></sub>, <i>LAI</i>, and <i>SWC</i>, whereas in humid regions, <i>VPD</i> and <i>P</i> played more important roles. These findings enhance our understanding of the mechanisms underlying ecosystem-scale water-carbon coupling and provide scientific guidance for ecosystem management, water resource optimization, and climate adaptation strategies.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Quantifying the Environmental Determinants of Water use Efficiency in Terrestrial Ecosystems: A Machine Learning Perspective

  • 期刊论文

  • 发表日期：

    2026年2月1日

查看全文
  • Shengqi Jian,
  • Jun Gao,
  • Ying Zhang,
  • Jiayi Wang,
  • Shentang Dou,
  • Xueli Zhang

摘要

Water use efficiency (WUE) is a key indicator of the coupling relationship between water and carbon in terrestrial ecosystems and is crucial for understanding water use characteristics and their responses to environmental changes across different ecosystem types. However, substantial uncertainty remains regarding the magnitude and dominant controls of WUE in various ecosystems. In this study, we synthesized daily-scale eddy covariance flux data, meteorological observations, and remote sensing datasets from 21 ChinaFLUX sites, covering forest, cropland, grassland, and wetland ecosystems, to systematically examine the spatiotemporal patterns of WUE. A random forest (RF) algorithm was employed to quantify the relative importance of environmental drivers. The results show that WUE significantly varied among ecosystems, with forests and croplands exhibiting the highest values, and grasslands the lowest (P < 0.05). Air temperature (Ta) was identified as the primary determinant of WUE (average contribution: 25.62%), followed by leaf area index (LAI, 19.41%), net radiation (Rn, 16.92%), soil water content (SWC, 13.07%), and vapor pressure deficit (VPD, 11.99%). Wind speed (WS) and precipitation (P) contributed relatively less (< 10%). Furthermore, the dominant environmental controls of WUE varied across ecosystems: Ta primarily influenced forest WUE, LAI was the key factor for croplands, while SWC and VPD had stronger effects on wetlands and grasslands. WUE also exhibited significant spatial heterogeneity across climatic zones: in arid and semi-humid regions, WUE was mainly regulated by Ta, LAI, and SWC, whereas in humid regions, VPD and P played more important roles. These findings enhance our understanding of the mechanisms underlying ecosystem-scale water-carbon coupling and provide scientific guidance for ecosystem management, water resource optimization, and climate adaptation strategies.

Graphical Abstract