기상예보모델의 최적화를 통한 봄철 동아시아 황사 발생량 추정 및 예보 가이드라인 설계 = (The) Design of guideline in the presumption and forecast for east asia dust emission in the spring season through optimizing weather forecast model

국문 초록 (Abstract)

우리나라에 영향을 미치는 황사는 최근 들어 발생 빈도나 강도가 증가하고 있으며, 그것으로 인한 인적, 물적 피해도 늘어나고 있다. 황사 현상을 정확히 예측하기 위한 노력에도 불구하고 발원지 관측 자료의 부족과 예보 과정에 잔존하는 여러 불확실성으로 인해 황사 예보가 쉽지만은 않은 실정이다. 따라서 본 연구에서는 우리나라에 영향을 미치는 황사 현상을 정확하게 예측하기 위해 최적화된 기상모델을 이용해 봄철 동아시아 황사 발생량을 산정하고 예보 가이드라인을 설계하고자 하였다.
이를 위해 먼저 최신버전의 기상모델에 대해 우리나라의 지형과 봄철 기상특성을 정확히 반영할 수 있도록 최근 2년간의 기상자료를 이용한 다양한 조건의 수치모의를 통해 최적의 물리 모수화 방안과 예보 민감도 제고 방안을 강구하였다. 다음으로, 모델 결과와 경험식을 이용해 동아시아 황사 발생량을 추정하고, 황사 발생 농도별 기상요소를 분류하였으며, 이 결과를 통합해서 의사결정모형으로 구축하여 황사예보의 가이드라인을 제시하였다.

다국어 초록 (Multilingual Abstract)

The Asian dust (or yellow sand), which has influence upon our country from China and Mongolian deserts, is recently being increased in the frequency and intensity of its occurrence, and in the human and physical damage. In the face of an effort to correctly predict the phenomenon of yellow sand, it is the real situation that a forecast for yellow sand is not easy due to the lack of materials of observation on the source and to several uncertainties that are left in the process of a forecast. Accordingly, the purpose of this study was to calculate the Asian dust emission in the spring season and to design the meteorological guideline of a forecast, by using the optimized weather forecast models in order to accurately predict the phenomenon of yellow sands.
For this, it first devised the optimized physics parameterization scheme and the forecast-sensitivity strengthening method through numerical simulation in diverse conditions by using the weather-forecast materials for the recent 2 years so that the characteristics of our country's terrain and spring-season weather can be exactly reflected on the weather forecast model in the latest version. Next, the East Asia dust emission was assumed, and the meteorological elements by dust emission density were classified, by using the results of the model and the empirical formula. And, by integrating these results and then implementing it as the decision-oriented model, the guideline of a dust forecast was presented.
Given summarizing these results, those are as follows.
First, as a result of mutually comparing MM5 model and WRF model, both models were dropped the accuracy on complex terrain like Seoul in the simulation of surface wind fields, and the tendency of excessive simulation in wind velocity was indicated. This was considerably improved due to using the detailed land use category map, but the forecast error for the lowest temperature remained as the portion that will need to be reinforced. As a result of analyzing sensitivity in the wind fields and the temperature forecast by applying several physics parameterization schemes to MM5 model and WRF model, the optimized physics parameterization scheme in each model was presented. Overall, WRF model, which was recently released, shows performance that doesn't lag behind compared to MM5 model, and was judged to be more proper for the predictive and diagnostic model due to easy interface and to the support of diverse utilities available.
Second, as a result of calculating the dust emission by time through applying 5 empirical formulas as for the cases of dust emission in China and Mongolia, the aspect of regional distribution was indicated to be different little by little. And, the result of empirical formula in US EPA, which was considered the conditions such as vegetation, soil type and terrain, was better than the remaining four empirical formulas.
Third, regarding the cases of yellow sand by density in observation for the recent 2 years in our country, it performed the weather forecast model and extracted and spatially portrayed the meteorological element as for the grid domain of including our country from the source. Thus, the forecast guideline was designed as a guiding principle that can judge dust emission and transportation and our country's predicted density in observation through analyzing the pattern. As the guideline of yellow-sand forecast is a decision-oriented model, it will be able to function as a guideline that a weather forecaster can finally predict by comparing the measured and presumed meteorological elements with the criteria of decision.

목차 (Table of Contents)

• Ⅰ. 서론 = 1
• 1. 연구의 필요성 = 1
• 2. 연구목적 및 연구내용 = 3
• 3. 용어의 정의 = 3
• 4. 연구의 제한점 = 5
• Ⅱ. 이론적 배경 = 6
• 1. 선행 연구 = 6
• 2. 이론적 고찰 = 8
• Ⅲ. 연구 절차 및 방법 = 22
• 1. 연구 절차 = 22
• 2. 모델 초기조건 설정 = 25
• 3. MM5와 WRF 모델의 비교 연구 방안 = 27
• 4. MM5 및 WRF 모델 최적화 연구 방안 = 29
• 5. 황사 발원지 배출량 산정 방안 = 32
• 6. 황사 예보 가이드라인 설계 = 38
• Ⅳ. 연구 결과 = 39
• 1. MM5와 WRF 모델 비교 = 39
• 2. MM5 및 WRF 모델의 물리 스킴에 대한 최적화 방안 모색 = 55
• 3. 경험식을 이용한 황사 발원지 배출량 산정 = 72
• 4. 황사 예보 가이드라인 설계 = 79
• Ⅴ. 결론 및 제언 = 87
• 1. 결론 = 87
• 2. 제언 = 88
• 참고문헌 = 90
• ABSTRACT = 95
• 부록 = 98