Intergovernmental Panel on Climate Change (IPCC) provides various prospects of future climate change under the Representative Concentration Pathways (RCP) scenarios using General Circulation Models (GCMs) of Coupled Model Intercomparison Project (CMIP). This paper describes a modified application of Ensemble Bayesian Model Averaging (EBMA) to produce daily mean temperature ensembles using 19 GCMs provided by CMIP. We proposed two types of approach: (1) monthly weighting scheme for a whole area (EBMA.v1) and (2) monthly weighting for each grid point (EBMA.v2), which can take into account the spatially heterogeneous pattern of GCM. For the training period of 1979- 2005 for East Asia, 9,855 sets of daily temperature ensembles (27 years × 365 days) were produced and compared to the ERA-Interim reanalysis data of European Centre for Medium-Range Weather Forecasts (ECMWF), which showed better validation statistics than the general mean and median ensembles. In particular, EBMA.v2 outperformed EBMA.v1 by diminishing the large errors of inland areas where the surface heterogeneity is larger than the ocean. The EBMA.v2 was able to handle the problem of spatial variability by employing monthly and spatially varying weighting scheme. We finally produced daily mean temperature ensembles for the period of 2006-2100 by using the EBMA.v2 under the RCP 6.0 scenario, which are going to be provided on the web.

Seasonal rainfall forecasts are one of the most important part of water resources management in minimizing climate-related risk. Recently, abnormal change in precipitation raised the attention of not only scientists it gets big interest in general public too. Seasonal climate forecasts are typically based on simulations from general circulation models (GCMs) that approximate the complex physical, chemical, and biological processes. But it has been known that General Circulation Models have considerable uncertainties. Recent studies suggested that Multi-Model Ensemble(MME) could reduce this uncertainties and give an improvement on the results. There have been used several MME estimation techniques that are simply averaging models and regression based techniques. This study aims to improve MME using Bayesian Model Averaging(BMA) technique which gives weights to the models based on each model performance to present observation. The result showed that BMA technique output is statistically more fitted to the observation than the other techniques and it is very important to further analysis such as downscaling and other simulation method that uses future precipitation as a main input data.

정상성 마코프 연쇄 모형은 일강우모의 모형으로 광범위하게 이용되고 있다. 하지만 정상성 마코프 연쇄 모형의 기본가정은 통계학적 특성이 시간에 따라 변화하지 않는 것으로, 일강우모의 시에 평균 또는 분산의 경향적 변화를 효과적으로 반영할 수 없다. 이러한 문제점을 인지하여 본 연구에서는 연주기 및 계절변화에 대하여 우수한 모의 능력을 나타내는 GCM의 모의결과를 입력자료로 이용하여 일강우량을 모의하기 위한 통계학적 상세화(downscaling) 기법인 비정상성 은닉 마코프 모형을 개발하였다. 개발된 모형을 낙동강 유역에 존재하는 영주지점, 문경지점 및 구미지점의 관측강우량에 적용한 결과, 일단위 및 계절단위의 강우량의 통계적 특성을 기존 모형에 비하여 개선된 결과를 도출할 수 있었으며, 또한 개발된 모형은 극치강수량 복원에 있어서도 관측값과 보다 유사한 결과를 보여 주었다. 이러한 점에서 정확성이 확보된 GCM 계절예측자료가 입력자료로 NHMM 모형에 활용된다면 예측기반의 일강수 상세화 모형으로 활용될 수 있을 것으로 판단된다. 이와 더불어, 기후변화 시나리오 입력자료가 사용된다면 기후변화 상세화 모형으로서도 적용될 수 있을 것으로 사료된다.