초록 ▼

기후변화 적응대책 마련을 위해 미래 기후변화 상세정보에 대한 지방자치단체 및 산업체의 요구가 증대되고 있다. 고해상도 전지구 모델링은 계산 부담이 커서 현실적인 어려움이따르므로 지역기후모형을 이용한 역학적 규모축소법의 활용이 늘고 있다. 그동안 역학적규모축소법을 이용하여 우리나라 및 주변 영역의 미래 상세변화 전망에 관한 연구가 있어왔으나 주로 대기 변화를 모사하는데 그쳤고, 우리나라 주변해역의 미래 상세변화 전망에관한 연구는 매우 제한적이었다. 한반도 주변해의 미래상세변화를 전망하는 데는 해양과대기의 상호작용을 적절히 모사하는 것

기후변화 적응대책 마련을 위해 미래 기후변화 상세정보에 대한 지방자치단체 및 산업체의 요구가 증대되고 있다. 고해상도 전지구 모델링은 계산 부담이 커서 현실적인 어려움이따르므로 지역기후모형을 이용한 역학적 규모축소법의 활용이 늘고 있다. 그동안 역학적규모축소법을 이용하여 우리나라 및 주변 영역의 미래 상세변화 전망에 관한 연구가 있어왔으나 주로 대기 변화를 모사하는데 그쳤고, 우리나라 주변해역의 미래 상세변화 전망에관한 연구는 매우 제한적이었다. 한반도 주변해의 미래상세변화를 전망하는 데는 해양과대기의 상호작용을 적절히 모사하는 것이 중요하다.이 연구과제에서는 지역기후 해양-대기 접합모형을 수립하고, 수립된 접합모형을 이용하여 한반도 주변해역에 대한 미래 상세변화 전망을 시범적으로 제시하였다. 수립된 접합모형은 해양 성분모형 ROMS와 대기 성분모형 WRF로 구성되었고, 접합자 MCT를 이용하여 매 시간마다 양방향 변수 교환이 이뤄진다. 한반도 주변해를 포함하는 북서태평양 접합모형의 해양 성분모형은 동쪽으로는 쿠로시오 이안 위치를 고려하고 북쪽으로는 동해가 모두 포함되도록 위도 10°N∼55°N, 경도 105°E∼175°E로 설정하였고, 해상도는 수평으로는 1/12도(약 10km) 이며, 연직으로는 30개 층을 두었다. 이 해역에 대해 100년 스핀업 실험을 수행하였고, 수심 보정을 통해 대한해협과 쓰가루해협을 통과하는 수송량을 개선하였다. 접합모형의 대기 성분모형은 서쪽으로 편서풍 사행을 고려하여 티벳고원을 포함하고, 남쪽으로 강수관련 중규모 에디의 북향전파를 고려하여 저위도 강수대를 포함하며, 수평 해상도는 50km 이고 연직으로 27개 층을 두였다. 파장 1000km 이상의 장파에 대해 분광너징 기법을적용함으로써 바람 장을 좀 더 현실적으로 개선하였다.수립된 지역기후 접합모형을 이용한 미래변화 상세전망 실험은 유사지구온난화 방법을이용하여 수행하였고, 미래기후 전망을 위해 CMIP5 전구모형 중에서 동아시아 몬순 기후모사성능이 비교적 우수한 CanESM2의 Historical 및 RCP4.5 시나리오 모사자료를 이용하였다. 이 접합모형으로 역학적 규모축소를 통해 시범적으로 제시한 한반도 주변해 미래변화는 황해보다 동해에서 수온 증가가 클 것으로 전망되며, 이 결과는 지금까지 관측된 경향성과 일치한다.이 과제를 통해 국내 최초로 수립한 지역기후 해양-대기 접합모형은 지속적인 개선이 필요함에도 불구하고 한반도 주변해 미래변화 상세전망 자료를 생산하는 데 활용될 수 있으며, 기후변화에 따른 한반도 연안기후 및 수산자원 전망 등에도 활용될 수 있다. 아울러,이 과제를 통해 습득한 지역모형 접합기술은 앞으로 해양생태계 등 다양한 권역의 성분모형을추가하여 지역기후 통합모형을 개발하는 데 유용하게 쓰일 것으로 사료된다.

Abstract ▼

Ⅳ. Results
Table 1 Road map of the project
Chapter 1. Implementation of a regional climate coupled model
1. Implementation of an ocean/atmosphere component model
∘ The ocean component of the coupled model is a regional ocean circulation model, ROMS(Regional Ocean Modeling System) that is

Ⅳ. Results
Table 1 Road map of the project
Chapter 1. Implementation of a regional climate coupled model
1. Implementation of an ocean/atmosphere component model
∘ The ocean component of the coupled model is a regional ocean circulation model, ROMS(Regional Ocean Modeling System) that is widely used. The K-profile method is chosen for vertical turbulence mixing.
∘ To set up a coupled model for the East Sea, we set the spatial domain of the ocean component as 33°N∼52°N in latitude and 127°E∼142°E in longitude with 1/6 degree (about 20km) interval and 30 vertical layers.We have completed 10-year spin-up for this ocean domain. The ocean model was stabilized for seawater temperature and sea surface level after 5-year spin-up simulation.
∘ To set up a coupled model for the Northwest Pacific, we set the spatial domain of the ocean component as 10°N∼55°N in latitude and 105°E∼175°E in longitude with 1/12 degree (about 10km) interval and 30 vertical layers. We have completed 100-year spin-up simulations for this ocean domain.
∘ The Northwest Pacific model underestimated the transports in the Korea and Tsugaru Strait compared with observational data and we improved the underestimation. This improvement was achieved by replacing the ETOPO1 bathymetry with the ETOPO5 and by calibrating the bottom topography of the Tsugaru Strait in the ocean model closer to the observation. This calibration resulted in an increase in the transport through the Korea Strait and in turn, this increment also increased the transport through the Tsugaru Strait.
∘ To set up a coupled model for the North Pacific, we set the spatial domain as 20°S∼65°N in latitude excluding sea-ice area and 100°E∼ 80°W in longitude. Through these setting up processes, we haveestablished a three-domain nested ocean model system for the East Sea,Northwest and North Pacific Ocean. The East Sea model, however, revealed a boundary problem caused by its small domain and the Pacific model loaded a computational burden by simulating on a fine resolution over its large domain. Hence, we focused our project on producing a testbed of regional coupled model simulation for the projection of detailed climate changes in the Northwest Pacific.
∘ The atmospheric component of the coupled model for this project is a WRF(Weather Research and Forecasting model) that is widely used for regional climate research as well as numerical weather forecasting. Kain-Fritsch scheme has been selected for cumulus process parameterization in this project.
∘ We used NCEP/DOE (National Centers for Environmental Prediction/Department of Energy) reanalysis data as the initial and boundary condition of the atmospheric model and conducted a dynamical downscaling experiment using the model for East Asia. The downscaled results showed southward surface winds in summer while typically northward surface winds are observed in the region during summer.
∘ To remove such an error, we applied a spectral nudging technique in WRF and obtained a realistic summer wind for both of its direction and speed. A regional model is incapable of simulating long waves on planetary scale due to its limited domain size and to overcome this limitation regional modeling requires a spectral nudging technique which restores the long waves in global data to prognostic variables in the regional model. Such a technique has been widely used in dynamical downscaling with a regional atmospheric model. The regional coupled model established through this project restores the long waves with a wavelength longer than 1000km for horizontal wind, air temperature and geopotential height above 500hPa to the top of the atmospheric model(50hPa) every one hour.
2. Implementation of a regional coupled model
∘ Through MCT(Model Coupling Toolkit), our coupled model calculates surface wind stress and heat flux with 10m wind, surface air pressure and humidity output from its atmospheric component WRF and inputs them to its oceanic component ROMS and also inputs sea surface temperature produced by ROMS to WRF. The ocean-atmosphere variable exchanges through MCT occurs every one hour in the coupled model.
∘ We have optimized the MCT source codes and shorten the model integration time by the optimization.
∘ We have upgraded the atmospheric component of the coupled model, WRF from its version 3.1 to 3.4 which was the latest released version as of 2013.
∘ We set the spatial domain of the coupled model for the Northwest Pacific with 10km resolution for its ocean component and 50km resolution for its atmospheric component, considering that the domain can include a westerly meandering in the atmosphere from the Tibetan Plateau, mesoscale eddies transported from the Tropical rainbands and Kuroshio extension to the east, and exclude sea-ice area to the north.
Chapter 2. A testbed simulation for detailed climate change projection
1. A testbed simulation with a coupled model and the model evaluation
∘ Our coupled model established through this project has been evaluated with an idealized experiment for typhoon. This experiment was conducted to investigate the upper ocean responses to typhoon when the typhoon approaches to coastal sea from open ocean. Vertical mixing and upwelling caused by cyclonic wind cooled the upper ocean and a maximum of the cooling occurred at the sea surface on the right-hand side of the typhoon trajectory.
∘ We have simulated the East Sea using our coupled model from 2000 to 2004 and the simulated results were similar to observational data as the model integration continued.
∘ An evaluation of our regional coupled model for the Northwest Pacific was made in comparison with a global model, CanESM2 which was used as the initial and boundary condition of the regional model to project detailed climate changes. The regional coupled model showed a better spatial pattern of sea surface temperature in the East and Yellow Sea similar to the OISST observational data than the CanESM2 temperature, especially for the pattern near coastal line with details.
2. Preliminary projection results with detailed climate changes
∘ The experiments for future change projection were conducted with each of the ocean and the atmosphere component model prior to an experiment with their coupled model. Such experiment with a component model only was designed to isolate a coupling effect from its future change projection. We adopted a pseudo global warming method in our dynamical downscaling experiments for future projection. This pseudo global warming method is a dynamical downscaling method that can reduce a systematic error in the global model used as the initial and boundary condition of the regional model and also the computational burden.
∘ As the initial and boundary condition of our coupled model, we used ECMWF-interim reanalysis data for its ocean component and NCEP/DOE reanalysis data for its atmospheric component to simulate the present climate and CanESM2 global model projection data, one of the CMIP5 global models to project a future change.
∘ In the future changes projected solely by the ocean component model, the surface temperature of the seas around Korea showed a greater increase in summer than in winter.
∘ In the future changes projected solely by the atmospheric componen model, precipitation showed an increase over the Tropical Pacific Ocean and over the Northwest Pacific and a partial land area of China and Japan. Changes in large-scale atmospheric circulation and water vapor transport played a crucial role in determining such precipitation changes in East Asia including the Korean Peninsula.
∘ In the future changes projected by the regional coupled model, the sea surface temperature in the East Sea showed a greater increase than in the Yellow Sea as global warming continued. This regional difference in the linear trend of sea surface temperature is consistent with the trend based on observation.

목차 Contents

• 표지 ... 1
• 제출문 ... 2
• 보고서 초록 ... 3
• 요약문 ... 5
• SUMMARY 및 KEYWORDS ... 12
• 목차 ... 21
• CONTENTS ... 23
• List of Figures ... 24
• List of Table ... 29
• 제1장 서론 ... 30
• 제1절 연구개발의 필요성 ... 30
• 1. 기술적 측면 ... 30
• 2. 경제·산업적 측면 ... 30
• 3. 사회·문화적 측면 ... 31
• 4. 기술원 고유기능 발전과의 연관성 ... 31
• 제2절 연구개발목표 및 내용 ... 31
• 1. 연구개발의 목표 ... 31
• 2. 연차별 연구개발 세부목표 및 내용 ... 31
• 3. 연구 추진계획 및 수행 방법 ... 33
• 제2장 국내외 기술개발 현황 ... 35
• 제1절 국내 연구동향 ... 35
• 제2절. 국외 연구동향 ... 36
• 제3장 연구개발수행 내용 및 결과 ... 39
• 제1절 지역기후접합모형 수립 ... 39
• 1. 해양/대기 성분모형 수립 ... 39
• 2. 접합모형 수립 ... 55
• 제2절 미래변화 상세도 시범 제시 ... 61
• 1. 접합모형 초기 실험 및 성능 검증 ... 61
• 2. 미래변화 상세도 시범 제시 ... 70
• 제4장 연구개발목표 달성도 및 대외기여도 ... 87
• 제1절 연구개발목표 달성도 ... 87
• 1. 3단계 연구기간 내 연구내용 대비 달성률(%) ... 87
• 2. 정량적 목표 달성도 (부록 참조) ... 88
• 제2절 대외 기여도 ... 89
• 제5장 연구개발결과의 활용계획 ... 92
• 제6장 참고문헌 ... 93
• 제7장 부록 ... 94
• 끝페이지 ... 113