Loading…
Improved statistical models for the relationship between riparian vegetation and river flow in arid environments: Implications for flow management
Riparian vegetation (RV) provides critical ecosystem services but has been degraded worldwide due to river flow change. Quantitative relationships between RV and river flow are essential for understanding RV developments and managing flow to conserve RV. Based on the improved statistical model frame...
Saved in:
Published in: | The Science of the total environment 2023-05, Vol.874, p.162487-162487, Article 162487 |
---|---|
Main Author: | |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Riparian vegetation (RV) provides critical ecosystem services but has been degraded worldwide due to river flow change. Quantitative relationships between RV and river flow are essential for understanding RV developments and managing flow to conserve RV. Based on the improved statistical model framework that incorporates previous RV conditions into explanatory variables to estimate later RV conditions, this study quantified the RV-flow relationships on the annual scale in the arid Ejina Delta through regression analysis coupled with the normalized difference vegetation index (NDVI) and hydrological data during 2002–2020. The median of NDVIs over the April–October growing season (SMN) was used to indicate annual vegetation conditions, and annual RV cover was derived using a dynamic SMN threshold (0.077–0.084) based on its better vegetation conditions than surrounding deserts. The water year was determined as September–August based on the defoliation time and lag time of the groundwater response to river flow. The results showed that (1) the RV cover approximately expanded from 1619 to 2914 km2, and the total SMN of RV cover increased from 3711 to 7880; (2) the spatial pattern of SMN declining away from rivers was well described by an exponential function with two physically meaningful parameters (R2 = 0.99); (3) the water-year runoff ranged from 4.0 × 108 to 10.6 × 108 m3 with an increasing trend; and (4) the annual RV condition, including both the total SMN and the spatial pattern of SMN, was well estimated by the multiple linear models incorporating a previous RV condition with a coefficient |
---|---|
ISSN: | 0048-9697 1879-1026 |
DOI: | 10.1016/j.scitotenv.2023.162487 |