Search

Currently, non-point source (NPS) pollution is the primary contributor to water pollution in river basins. Therefore, assessing the benefits of rural non-point source pollution management measures and optimizing their allocation are of great significance for improving water quality. In this study, a SWAT model was firstly constructed for the NPS pollution simulation and optimization based on the four-period land use data, which was more suitable for the Guishui River Basin. And then it was calibrated and verified by using the monitoring data. On the basis of spatiotemporal evolution of total nitrogen (TN) and total phosphorus (TP) in the Guishui River Basin from 2007 to 2019, and the priority control areas for TP and TN were identified based on the RUSLE and additive models. Next, seven management measures scenarios were designed in the priority control areas from the perspectives of hydrological response units (HRUs), sub-basins, and basin scales. Finally, the cost-effectiveness of each scenario was quantitatively evaluated, and the optimal allocation scheme of pollution control management measures was determined. Results showed that the amounts of TP in the Guishui River basin was lower than that of TN, showing a fluctuating downward trend, and the spatial distribution of TP and TN was high in the central and low in the north and southeast. Compared the effect of management measures, it was found that process interception measures were more effective than source control and landscape management measures in reducing the TP and TN. Among them, vegetation filter strips (FS) and grassed waterways (GW) had more environmental benefits and better cost-effectiveness for the pollution reduction. As for the source control measures, 20% chemical fertilizer reduction (FRT20) had relative more effects for the pollutants control than the 10% chemical fertilizer reduction (FRT10) and no-tillage (NTIL). When it refers to landscape management, the reduction effect of return farmland to forest (CFTF) on TP and TN was not much different from that of GW. Considering the costs and benefits of optimized allocation, the combination of FRT20, FS and CFTF had better environmental benefits, while the combination of FRT20 and FS had higher cost-effectiveness. Therefore, the government should control the source emission, strengthen the process interception and pay attention to the end-of-pipe treatment in the downstream. These conclusions can provide a scientific basis for the prevention, control and management of NPS pollution in the Guishui River Basin.

This study analyzed the process of vegetation restoration in the eastern margin ecotone of Qinghai-Tibet Plateau (EMEQTP) from 2001 to 2019 under different restoration modes based on the vegetation net primary productivity (NPP) which was estimated by the Carnegie-Ames-Stanford approach (CASA) model. Then the potential NPP was calculated and the durations of NPP restoration were predicted based on the Autoregressive Integrated Moving Average (ARIMA) model. Finally, the relative importance of main parameters in the CASA model as well as global sensitivity and uncertainty were discussed. The results showed that the NPP under a natural restoration mode had the highest increase with 20.1%, and its spatial change showed an overall trend of restoration. The NPP under the comprehensive restoration mode achieved an increase of 11.5%, with the southwestern area experiencing an insignificant trend of degradation, whereas the northeastern area experiencing a significant restoration trend. As for the NPP under the artificial restoration mode, it experienced the lowest increase with 10.5%, and the northern part experiencing a significant restoration trend, whereas the southeastern part experiencing an insignificant degradation trend. In addition, the highest potential NPP of vegetation restoration would be appeared in the area under a comprehensive restoration mode. The area under an artificial restoration mode would have the lowest potential NPP of vegetation restoration, but the duration of restoration would be longest under this mode, reaching up to 64 years. Although the natural restoration mode achieved the best effect of vegetation restoration, considering the limited scope of the promotion of this mode, further attention should be placed on restoration measures and the effects of the artificial and comprehensive restoration modes. Finally, APAR, GST and PET were relatively important for the CASA model, and the parameters including Topt, GST, and PET were sensitive. Those results can offer scientific references for the management of ecological environment in the EMEQTP.