Measurable sediment discharge from a karst underground river in southwestern China: temporal variabilities and controlling factors
Severe soil erosion is a critical issue in karst areas. Due to a lack of surface streams, soils often discharge from karst catchments via groundwater. Quantifying sediment discharge (SD) from karst groundwater will help managers to develop effective methods of soil conservation. The time series of t...
Gespeichert in:
Veröffentlicht in: | Environmental earth sciences 2020-02, Vol.79 (4), Article 90 |
---|---|
Hauptverfasser: | , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Severe soil erosion is a critical issue in karst areas. Due to a lack of surface streams, soils often discharge from karst catchments via groundwater. Quantifying sediment discharge (SD) from karst groundwater will help managers to develop effective methods of soil conservation. The time series of the monthly SD and controlling factors (water discharge, precipitation, potential evapotranspiration, and normalized difference vegetation index) of the Nandong Underground River System (NURS), a typical karst underground river catchment in southwest China, from 1998 to 2015 were analyzed. To investigate the changing seasonal characteristics of monthly sediment discharge and controlling factors and predict the variations in monthly sediment discharge, an analysis of variance (ANOVA), seasonal Mann–Kendall test, and seasonal decomposition of time series by loess (STL) were conducted to identify changes in the seasonal characteristics of the SD and the controlling factors on a monthly scale. The results of these analyses indicated that the SD and its controlling factors varied considerably from month to month and the annual soil loss mainly occurred from June to September. The SD gradually decreased during April and May due to the decreasing of precipitation in March and April during the 1998–2015 period. The decrease of rainfall not only reduces the intensity of soil erosion on the surface, but also reduces the flow and velocity of underground rivers, reducing the transport capacity of suspended matter. Our study showed that the bivariate state-space model had the lowest Akaike’s information criterion (AIC) score (− 7.594 and − 7.686) and root mean square error (RMSE) (0.022 × 10
6
m
3
and 0.020 × 10
6
m
3
) values and the highest
R
2
values (0.983 and 0.984) for the calibration and validation periods, and was the best state-space model to describe the temporal distribution of the monthly sediment discharge in the NURS. A method that allows for the correct estimation and evaluation of soil erosion and the determination of the regional soil and water conservation can be useful for better karst catchment management in the NURS. |
---|---|
ISSN: | 1866-6280 1866-6299 |
DOI: | 10.1007/s12665-020-8826-7 |