A wavelet coherence approach to detecting ecosystem services trade-off response to land use change

Clarifying the complex land use impacts on ecosystem services (ESs) trade-off will be beneficial to watershed sustainable development, especially through scientific land use management and decision making. Dongting Lake Basin (DLB) is not only one of the most significant ecological barriers for the Yangtze River Economic Belt, but also an important grain production base of China. The trade-off between the grain production (GP) and water purification (WP) has become increasingly prominent. Here, we chose DLB as a case study area, applied spatial continuous wavelet transform and wavelet coherence analysis, characterized the ES trade-off intensity by wavelet coherence coefficient, and explored the influence of land use type, conflict and intensity on the trade-off between GP and WP. The results showed that the trade-off intensity between GP and WP in the DLB in 2015 had alleviated compared with 2005, and the coherence coefficient had increased while maitaining the negative value. The trade-off intensity was the strongest in farmland and forest land, and weaker in grassland and water body. The impact of land use conflict mainly depended on the specific types of land use conversions. For the transects where land use conversions mainly appeared between farmland and forest land, the intensification of land use conflict would increase the trade-off intensity (2005: R² = 0.3862, p < 0.05; 2015: R² = 0.2543, p < 0.05), while for the transects dominated by conversions to water body and grassland, stronger land use conflicts would reduce the trade-off intensity (2005: R² = 0.3438, p < 0.05; 2015: R² = 0.2668, p < 0.05). The impact of land use intensity was also realized through the land use type, with lower interpretation ratio. In addition, the wavelet coherence analysis showed that the scale about 10.51 km was the most suitable for exploring the trade-off between GP and WP, which was equivalent to the scale of the secondary watershed in the study area.