The current watershed-scale quantification method for non-point source (NPS) pollution does not reflect the annual or seasonal variation of precipitation as well as spatial characteristics of soil, land use, and topography. It is necessary to develop a method for quantifying spatially and temporally changing NPS pollution loads in order to utilize the seasonal forecast data that predicts the change of precipitation up to 6 months in advance. In addition, it is possible to consider the application of nonstructural management measures within the time scale of several months as a part of proactive countermeasures against NPS pollution. Especially in agricultural lands including paddy fields, it is possible to apply various management methods such as adjustment of sowing and irrigation time and adjustment of fertilizer and irrigation amount as a part of best management practices (BMPs). Therefore, the objective of this study is to develop a method that quantifies and predicts the reduction of NPS loads at the watershed scale according to applied various farming methods by considering temporal variations based on seasonal prediction information as well as spatial variability of soil, land use, and topography.

The study consists of three detailed sub-items. First, the Observation-based Moving Window Regression (MWR-Obs) method using reanalysis data was added to the existing system. Second, the applicability of the APEX-Paddy model, which is known to be able to consider the characteristics of paddy fields in Korea, was evaluated. Finally, spatial and temporal variations of NPS pollutants was considered in a watershed with multiple land use characteristics.

Analyzing the monitoring data from the previous study on plot-scale paddy field showed that precipitation amount during the growing season acted as an important factor for the non-point source (NPS) pollutant load. The APEX-Paddy model applied to quantify the NPS pollutant loading according to various farming practices in the plot-scale paddy field showed similar tendency of annual variations in total nitrogen (TN) and total phosphorus (TP) loads but quantitatively underestimated the loads. NPS load for TN and TP at the Jeonju-A TMDL (Total Maximum Daily Load) water quality station was evaluated for different land covers including paddy areas. TN and TP loads by overall NPS pollutants corresponded to 43.5% and 42.9% of total pollutant loads, respectively, showing slightly smaller loads than that of point source pollutants. In addition, the proportion occupied by NPS pollutants during wet years tended to be higher compared to dry or normal years. Considering different land covers, TN and TP loads contributed by paddy area alone corresponded to 6.9% and 8.8% of total pollutant loads, respectively. The forecasted TN and TP loads at the Jeonju-A TMDL station using 6-month (May to October) seasonal forecast information for 2011 and 2012 showed underestimations of 26.5% and 13.2%,

respectively, compared to observed values. As a result, we presented the possibility of the utilization of seasonal forecast information in the decision making process for quantifying NPS pollutant loads and selecting appropriate farming practices as BMPs for reducing the NPS loads from agricultural areas.

Keyboards: Seasonal Forecast, Long-term Prediction, Water Quality, Downscaling, Climate Index, Multi-Model Ensemble (MME), Teleconnection, Non-point Source Pollution, Best Management Practices

1. 서론

2. 연구 방법

2.1 대상 유역 및 모니터링 사이트

2.1.1 새만금 논 비점오염 모니터링 사이트

2.1.2 만경강 유역

2.2 기후예측 예측성 정보

2.2.1 실시간 계절예측 시스템

2.2.2 예측성 평가

2.3 농업비점오염 부하량 정량화 모델링

2.3.1 APEX-Paddy 모형

2.3.2 유역규모 비점오염 부하량 정량화

3. 연구결과

3.1 기후 변동에 따른 비점오염 부하량 변동 특성 분석

3.2 기후예측 예측성 평가

3.3 필지규모 농업비점오염 부하량 정량화

3.4 유역규모 비점오염 부하량 정량화

3.5 유역규모 비점오염 부하량 예측 및 최적관리방안 제시

4. 결론

REFERENCE

부록 1. 계절예측 상세화 프로그램 사용자 매뉴얼