TL;DR
This paper introduces the WEP-karst model, a novel distributed hydrological model specifically designed for complex karst regions, significantly improving simulation accuracy. It reveals that while precipitation and infiltration are concentrated in the summer months, most water cycle fluxes (except green water) have decreased over the past 60 years, with blue water showing high sensitivity to climate change and a projected future decline.
What Problem Does This Paper Address?
Karst mountain regions, like Southwest China (KMRSC), are crucial water towers, supplying vast populations. However, their unique hydrogeology—characterized by thin soils, high infiltration capacity, and complex fissure networks—makes accurate hydrological simulation extremely challenging. Existing models often struggle to capture the spatial variability and dynamic behavior of water movement in these environments, leading to a poor understanding of how water resources are evolving and how they might be impacted by climate change. This lack of understanding hinders effective water resource management and planning for regions heavily reliant on these water sources.
Methods
The authors developed the WEP-karst model, an advancement of the existing WEP-L distributed hydrological model. The key innovation is the incorporation of an equivalent porous medium (EPM) approach to better represent the karst vadose zone. This approach divides the vadose zone into four distinct layers (soil, upper epikarst, lower epikarst, and transition layer) and treats them as a porous medium, allowing for a more nuanced simulation of water movement through the complex fissure and pore structures. The model also utilizes a stem-branch topological relationship for coding computation units and employs a one-dimensional kinematic wave approach for overland and riverine flow routing. Model calibration and validation were performed using observed discharge data from 18 hydrological stations, employing Nash-Sutcliffe Efficiency (NSE) and Relative Error (RE) as key performance metrics.
Datasets
The study utilized several datasets:
- Hydro-meteorological data: Daily records (1956-2015) from 69 national meteorological stations (precipitation, air temperature, sunshine hours, vapor pressure, relative humidity, wind speed).
- Topography data: Shuttle Radar Topography Mission (SRTM) with 90m spatial resolution.
- Soil data: 1km grid soil types and characteristic parameters from the National Second Soil Survey Data.
- River system data: River network map (1:250,000) and river cross-section data from hydrological yearbooks.
- Hydrogeology data: Hydrogeology unit division, permeability, and storage coefficient from the Distribution Map of Hydrogeology in China.
- Land-use data: National 1km grid land-use type data for seven periods (1980-2015).
- Vegetation data: Monthly vegetation information (leaf area index and area fraction) from MODIS satellite data (2001-2015).
- Model validation data: Naturalized monthly discharge data from 18 representative hydrological stations (1956-2000).
- Climate change projection data: Historical (1950-2005) and future (2006-2099) climate simulations from the RegCM4.0 regional climate model under the RCP4.5 scenario.
Key Insights and Ideas
- Karst complexity demands specialized models: The unique hydrogeology of karst regions cannot be adequately captured by standard hydrological models. The WEP-karst model's EPM approach is crucial for improving simulation accuracy.
- Epikarst zone is critical: The epikarst layer, often overlooked, plays a significant role in water storage and movement, and its inclusion dramatically improves hydrological simulations in karst areas.
- Water resources are declining: Over the past 60 years, most water cycle fluxes (precipitation, infiltration, evapotranspiration, blue water) have shown a statistically significant decreasing trend, highlighting a growing water scarcity challenge.
- Blue water is volatile: Blue water (river runoff) exhibits the largest interannual fluctuation and is the most sensitive to climate change, making regional water supply management particularly challenging.
- Seasonal concentration: Precipitation and infiltration are heavily concentrated in the May-August period, with blue water peaking in June, underscoring the importance of managing water during these wet months.
- Future climate impacts are concerning: Projections indicate a likely decrease in precipitation and blue water resources in the coming decades due to climate change, exacerbating existing water stress.
Key Results
- The WEP-karst model demonstrated good performance in simulating monthly discharge, with NSE values ranging from 0.71 to 0.94 and RE values from -9.8% to 8.3% during the validation period.
- Annual average precipitation in KMRSC is 1506 mm, approximately 2.4 times the national average.
- Annual average infiltration and evapotranspiration were 862 mm and 870 mm, respectively, each accounting for about 57% of precipitation.
- Annual average blue water reached 701 mm, with a per capita river runoff of about 3100 m³.
- Except for green water, other water cycle fluxes showed a statistically significant decreasing trend over 1956-2015.
- Climate change projections (RCP4.5) suggest a likely reduction in precipitation and water resources for KMRSC between 2021-2050.
Limitations and Future Work
- Surface-aquifer boundary inconsistency: The study acknowledges that the inconsistency between surface watershed boundaries and aquifer boundaries was not fully addressed, which could affect simulation accuracy, especially in small-scale watersheds. Future work should focus on better describing this inconsistency.
- Natural hydrological processes only: The current model simulates natural hydrological processes and does not account for human water use (reservoirs, withdrawal, irrigation, wastewater reuse). The authors plan to couple a social water cycle system into the WEP-karst model to improve its applicability in areas with significant human activity.
- Further regional analysis: While the study provides a broad overview, more detailed analysis of specific sub-regions and their unique challenges (e.g., water stressed areas in the northwest) is warranted.
- Green water's role: While the study notes green water's less significant trend, further exploration into its role in water resource management, especially in relation to vegetation capacity and reducing runoff, is suggested.
Summary updated