Abstract:
Land use change is one of the key factors that directly dominate the hydrological process of the watershed. Human activities have changed land-use patterns, thereby indirectly posing an impact on surface infiltration, evapotranspiration, and soil water in the water cycle process. The water resources were also evolved for the water balance in the basin. This study aims to further explore the impact of land-use evolution on the spatiotemporal changes of blue and green water in the Weihe River Basin. A Cellular Automata-Markov (CA-Markov) model was selected to predict the land-use pattern in 2035 using land-use data in 1995 and 2015. Firstly, a standardized preprocessing was performed on the constraint data using a FUZZY module, including elevation, slope, highway, and railway. Secondly, the Markov module was superimposed onto the land use data to obtain the land transfer probability matrix from 1995 to 2015. Thirdly, the multi-criteria evaluation (MCE) module was used to make land-use maps suitability. The CA-Markov model was used to predict land-use structure in 2035. A variety of land-use change scenarios were set to quantitatively analyze the temporal and spatial responses of land-use change and blue-green water using the soil and water assessment tool (SWAT). The results showed that: 1) The CA-Markov model was effective in simulating the land-use pattern in 2015, where the Kappa coefficient was 0.89, suitable for the prediction. 2) During 1995-2035, the cropland was mainly transferred to building land and grassland, while the grassland was mainly transferred to woodland and cropland. Meanwhile, the cropland showed a decreasing trend, with the single attitude of -0.27%, and the building land had the largest growth rate, with the dynamic attitude of 3.75%. 3) The model calibration showed that the evaluation indicators of five hydrological stations all met the standards of R2>0.6 and ENS>0.5, indicating that the SWAT model presented good applicability in each basin of the Weihe River Basin. 4) The blue water increased by 2.38 mm/a, and green water decreased by 18.74 mm/a from 1995 to 2035. By contrast, the blue water increased by 14.82 mm/a, and the green water decreased by 15.23 mm/a from 2015 to 2035. As such, the evapotranspiration was reduced in the basin, but the runoff was increased. 5) Returning cropland to woodlands and grasslands contributed to reducing the blue and green water, where the blue water volume decreased by 9.27, 11.37 mm/a, while the green water volume decreased by 32.94, 21.13 mm/a, respectively, under the extreme land-use scenarios. The main reason was that the cropland increased the runoff and evapotranspiration in the basin. The amount of blue and green water decreased, after a large area of cropland was converted into woodland and grassland. Therefore, it is necessary to reasonably plan the cropland, woodland, and grassland for the control of soil erosion, thereby promoting the improvement of the ecological environment in the study area. These findings can be expected to provide a sound scientific reference for land-use planning and water resources management in the Weihe River Basin.