Simulation And Prediction Of Non-Point Source Pollution In Yitong River Watershed

The Yitong River is an important tributary of the Songhua River. Affected by non-point source pollution in the basin, the water quality of the Yitong River is getting worse. The continuously deteriorating water quality has seriously constrained the sustainable development of the socio-economy in the Yitong River watershed and poses a great threat on the water quality of the Songhua River. Currently, there are few studies on the loss characteristic of the non-point source pollutants in the Yitong River watershed and on the simulation and prediction of output load. As a result, there are no enough decision bases for the prevention and management of non-point source pollution, greatly constraining the carrying out of work on the management of non-point source pollutions and the water environment in the Yitong River watershed.This dissertation taking the Yitong River watershed as the study area, in response to the prominent problems and actual needs in the prevention and treatment of non-point source pollutions in the area, the loss characteristics of non-point source pollutants under different land utilizations were studied in the Yitong River watershed. The source and migration patterns of non-point source pollutants in the basin were preliminary learned. Then the Soil and Water Assessment Tool (SWAT) was coupled with the modified export coefficient method in analog computation of the non-point source pollutant load. The spatial and temporal distribution characteristic of the non-point source pollution in the area were discussed. Taking into account the important effects of meteorological factors on non-point source pollution, the general circulation model (GCM), the statistical downscaling model (SDSM) and principal component analysis (PCA) were combined to predict the changing trends of precipitation and temperature in the Yitong River watershed under two future climate scenarios. Then, the predicted precipitation and temperature results for the study area were used as input to the SWAT model which was used to predict the effects of future climate changes on the output load of non-point source pollutants from the Yitong River watershed. The results are expected to provide scientific basis and decision support for the prevention and management of non-point source pollution in the Yitong River watershed and have important implications in improving the aquatic environment quality in the Yitong River even the Songhua River and in enhancing the sustainable development of the socio-economy in the watershed. The main contents include:1. Loss characteristic of the non-point source pollutants in the Yitong River watershedAccording to the unique physioclimate conditions and human activity modes in the Yitong River watershed, dryland, paddy field, grassland and woodland were chosen for the runoff plot experiment. The loss characteristics and trends of such non-point source pollutants as total N, total P, suspended solids and ammonia nitrogen were analyzed. The results showed that in terms of average export concentrations of non-point source pollutants, the different land-use types were in the following order: dryland>paddy field>grassland>woodland. The export concentrations of non-point source pollutants in dryland and paddy fields displayed a clear trend of increasing with the increasing runoff and decreasing with the decreasing runoff while no such a clear trend was found for woodland and grassland. During their loss with surface runoff, N was mainly in dissolved forms while P was mainly in particulate forms.2. Analog computation of the non-point source pollution load in the Yitong River watershed(1) Construction of the space and attribute databases for the Yitong River watershed"3S" (i.e.. GIS, RS. and GPS) were utilized to construct the spatial database and the attribute database for the Yitong River watershed. The spatial database mainly included the digital elevation model (DEM), the soil type map and the land use type map while the attribute database mainly included information for the meteorological stations and soil properties in the watershed. These databases would be the basis for the construction of a simulation model of non-point source pollution in the Yitong River watershed and a non-point source pollution model for the Changchun downtown.(2) The construction of the simulation model of non-point source pollution for the Yitong River watershedThe SWAT model was used to divide the whole study area into 30 sub-basins and 298 hydrologic responding units. Sensitivity analysis was carried out on the model parameters to find out the major sensitivity factors affecting the non-point source pollution of the basin. The monthly runoff data for 2006～2010 from the Nongan station and the monthly water quality data for 2009～2010 from the Kaoshan bridge section were used to calibrate and validate the model. A simulation model for the non-point source pollution in the Yitong River watershed was developed for analog computation of the non-point source pollution load of the Yitong River watershed (excluding the Changchun downtown).(3) The construction of a simulation model for the non-point source pollution in Changchun downtownTaking into account the important effects of Changchun downtown on the non-point source pollution in the whole study area and the applicability of the SWAT model, the improved export coefficient method was employed to develop the non-point source pollution simulation model for the Changchun downtown so as to analog compute the non-point source pollution load in Changchun downtown.(4) The analog computation of the non-point source pollution load of the Yitong River watershedCombining the SWAT model and the improved export coefficient model to analog compute the non-point source pollution load in the Yitong River watershed in 2010 and analyze the spatial and temporal distribution characteristics of the load. The results showed that the non-point source COD load was 15031.2 t, the sediment load was 1366.3×104 t, the ammonia nitrogen load was 1027.4 t, and total P load was 1369.6 t in the watershed at 2010. Affected by the non-uniform annual distribution of precipitation, the non-point source pollution load was relatively large in July and August, about 43%～54% of the total annual load of the watershed. Influenced by terrain, land use and surface coverage, the non-point source pollution load did not distribute uniformly throughout the basin. The key source regions of non-point source COD and ammonia nitrogen mainly concentrated in the areas surrounding Yitong County, Changchun City, and Nongan County which non-point source COD and ammonia nitrogen load were about 41% and 53%, respectively of the pollution load of the whole watershed. The key source area of non-point source sediment load and total P load mainly concentrated in such regions with relatively high intensity of agricultural activities as the Xinlicheng Reservoir, the Xinkai River and the Taiping Ditch which non-point source sediment load and total P load were 42% and 40%, respectively of the total loads of the whole watershed.3. Future precipitation and temperature prediction for the Yitong River watershed(1) Extraction of the principal components from the NCEP large scale weather forecast factorsPCA was employed to process the dataset of large scale weather forecast factors in the NCEP reanalysis data, from which the principal components of large scale weather forecast factors were extracted for using in forecasting regional precipitation and temperature. The original large scale weather forecast factors were replaced with these principle components as input to the SDSM. The input variables for precipitation prediction were reduced from 12 to 4, and those for temperature prediction were reduced from 8 to 2. The goal to downscale and compress the original dataset of large scale weather forecast factors was achieved.(2) Construction SDSMs between precipitation, temperature and the principal components of large scale weather forecast factorsOn the basis of the previous work, the meteorological factors of the NCEP reanalysis data were downscaled with the SDSM using the NCEP reanalysis data and the measured data from stations. SDSMs between the regional precipitation and temperature and the principal components of large scale weather forecast factors were developed. And the simulation results from the models were validated. The validation results showed that the precipitation and temperature results for the Yitong River watershed calculated with the SDSMs were reliable, and the model produced better results for temperature than for precipitation.(3) Prediction of future climatic changes in the Yitong River watershedBased on the A2 and B2 climate scenarios provide by the Hadley Centre Coupled Model Version 3 (HadCM3). a GCM, the SDSM was utilized to predict the future changing trends of precipitation and temperature in the Yitong River watershed under these two climate scenarios. The results showed that under the two scenarios, both the future precipitation and temperature display increasing trends in the Yitong River watershed, and the increase of precipitation and temperature in summer is the main reason to increase of the average annual precipitation and temperature in the study area. Under the A2 scenario, precipitation and temperature increase at 7.81 mm/per decade and 0.14℃/per decade, respectively in the watershed. In the future, September and July are the months with maximum precipitation and temperature increase, respectively for the watershed. Precipitation and temperature increase by 2.03 mm and 1.35℃, respectively compared with the baseline. Under the B2 scenario, precipitation and temperature increase at 5.67mm/per decade and 0.10℃/per decade, respectively in the watershed. In the future, September and August are the months with maximum precipitation and temperature increase, respectively for the watershed. Precipitation and temperature increase by 1.02 mm and 1.01℃, respectively compared with the baseline.4. Prediction of the effects of future climate changes on the non-point source pollution load in the Yiong River watershedThe precipitation and temperature prediction results for the future Yitong River watershed were used as input to the SWAT model which was employed to predict the effects of future climate changes on the streamflow and non-point source pollution load in the watershed and evaluate the changing trends and features of non-point source pollution load in the study area under climate changes. The results demonstrated that the flow of the basin will exhibit holistic rising trends under both scenarios. Future watershed flow will increase at 2.31 m3 s-1/per decade and 1.09 m3 s-1/per decade for the A2 and B2 scenarios, respectively. The increase of flow in summer is the main reason to the increase of average annual flow in the study area. The future sediment load and non-point source total P load will show increasing trends in the study area. The increases of sediment load and non-point source total P in summer are the mainly reasons for the increase of average annual pollution load under the two scenarios. Under the A2 scenario, future sediment load and non-point source total P will increase by 6.03×104 t/per decade and8.93 t/per decade, respectively in the watershed. In the future, August will be a month when sediment load and non-point source total P load increase fairly greatly, by 14.89×104 t and 140.8 t, respectively compared with the baseline. Under the B2 scenario, Future sediment load and non-point source total P load will increase by 3.78×104 t/per decade and 6.93 t/per decade in the watershed. In the future, September will be the month when sediment load and non-point source total P load increase fairly greatly, by 9.2×104 t and 101.4 t, respectively compared with the baseline. The non-point source ammonia nitrogen load will tend to reduce significantly in the future in the watershed, at 6.89 t/per decade and 10.97 t/per decade under the A2 and B2 scenarios, respectively. In the future, the reduction of non-point source ammonia nitrogen load in spring will be the main cause of the decrease of annual non-point source ammonia nitrogen load in the watershed.