Estimating Vegetation Greening Influences on Runoff Signatures Using a Log‐Based Weighted Ensemble Method. Issue 12 (8th December 2022)

• Record Type: Journal Article
• Authors: Huang, Qi; Zhang, Yongqiang; Ma, Ning; Post, David
• Is Part Of: Water resources research. Volume 58:Issue 12(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-12-08
• Subjects: vegetation greening -- log‐based weighted ensemble method -- runoff signature -- leaf area index change -- the Yellow River basin; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2022WR032492
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: Vegetation greening profoundly impacts the water cycle, and recent concerns about greening impacts have focused on various hydrological cycle components. However, the impacts of greening on catchment runoff signatures reflecting magnitude, low/high flow frequency, low/high flow duration and flow dynamics remain poorly understood. To properly simulate these runoff signatures, we use five modified hydrological models incorporating vegetation dynamics and further derive three ensemble approaches to obtain eight runoff time series outputs in a major tributary of the Yellow River Basin. Multiple validations suggest that the log‐based weighted ensemble (LWE) approach is robust for depicting the impact of greening on selected runoff signatures. This is especially true for the low flow part of the runoff time series and the overall performance of the selected signatures since LWE explicitly reduces the low flow bias. With this approach, five experiments were designed to isolate the impact of vegetation greening on runoff signatures, and the comparisons among the experiments indicate that greening noticeably decreases runoff magnitude, increases low flow frequency/duration and decreases high flow frequency/duration signatures. However, greening has little influence on runoff dynamic signatures. Each percent increase in leaf area index results in (a) changes of −0.2 ± 0.1% for magnitude signatures; (b) changes of −0.34 ± 0.30% and 0.56 ± 0.28% with wide ranges for annual high flow days and annual low flow days, respectively; and (c) marginal change on flow dynamic signatures. This study provides new insights by disentangling greening impacts on various runoff signatures using a trade‐off ensemble method. Key Points: Isolate vegetation greening impact on runoff signatures by comparing modeling experiments applying a log‐based weighted ensemble approach Increasing leaf area index decreased flow magnitude, increased low flow duration, and resulted in little change in flow dynamic signatures Flow frequency and duration signatures showed a greater response than magnitude signatures