Compound vs. single droughts: different impacts on river water quality revealed by multi-decadal observations
Droughts impact both water quantity and quality. Research is needed on how water quality (e.g., total carbon, TC; total nitrogen, TN; and total phosphorus, TP) responds to different droughts (single or compound) and their characteristics (i.e., duration and severity). It also needs to be addressed w...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Ecological Indicators |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1470160X25008556 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Droughts impact both water quantity and quality. Research is needed on how water quality (e.g., total carbon, TC; total nitrogen, TN; and total phosphorus, TP) responds to different droughts (single or compound) and their characteristics (i.e., duration and severity). It also needs to be addressed whether these water quality indicators during different droughts can be simulated based on drought characteristics. We used 1979–2018 hydro-meteorological and water quality data from Harp Lake in south-central Ontario, Canada. First, we analyzed trends and annual distribution of relevant factors. Then, we identified meteorological, hydrological, and compound droughts using the Standardized Precipitation Index (SPI) and Standardized Streamflow Index (SSI), by applying run theory to extract the drought characteristics. Lastly, we established a response relationship of TC, TN, and TP to different drought characteristics, and used random forest method to construct simulation models for TC, TN, and TP during droughts. The simulation models were evaluated using correlation coefficient (CC) and Nash-Sutcliffe Efficiency (NSE). The results showed that: (i) In the HP4 watershed, TC showed a significant upward trend (p < 0.01) over 40 years of long-term data, while TN and TP showed a significant downward trend (p < 0.01). (ii) The concentration of TC did not show significant changes in drought, while TN concentration increases, and TP concentration decreases. (iii) The random forest model can effectively simulate the cumulative concentrations of TC, TN, and TP during drought periods (CC > 0.90, NSE > 0.60). However, the simulation results for compound droughts are poorer than for other drought types, highlighting the greater complexity of compound drought impact on river water quality relative to single droughts. These findings may aid in water quality management during droughts and may have broader applications in environmental monitoring and water resource management. |
|---|---|
| ISSN: | 1470-160X |