| 要約: | <p>The growing demand for water presents a significant sustainability challenge. Understanding vegetation changes is crucial, as plants significantly influence water exchange through transpiration. However, global climate models show considerable uncertainty in predicting future vegetation trends ranging from −0.007 to 0.083 m<sup>2</sup> m⁻<sup>2</sup> decade⁻<sup>1</sup>, impacting water management. Here, we apply an emergent constraint method to reduce uncertainty in global vegetation projections for the period 2015–2100 from a climate model ensemble (Coupled Model Intercomparison Project Phase 6 [CMIP6]), focusing on the leaf area index (LAI). Our approach reduces uncertainty in global LAI projections by 37.7%–53.1%. We find that this uncertainty is primarily due to incomplete representations of the CO<sub>2</sub> fertilization effect. Our results also show that models underestimate future LAI increases by 28.2%–32.1%, leading to underestimated water loss from increased transpiration. These findings improve predictions of future vegetation and transpiration, providing valuable insights for policymakers to adjust water management strategies and better prepare for water-related challenges.</p>
|
|---|