• 한국습지학회지
• /
• 제15권1호
• /
• pp.115-125
• /
• 2013

본 연구에서는 강수의 공간적 편차가 큰 산악지역에서 축소기법을 적용하기 위한 방법론을 마련하고 이를 이용하여 미래 강수특성의 변화를 추정하고자 하였다. 이를 위하여 한반도내 산악지역이라고 할 수 있는 남한강유역을 대상유역으로 선정하였고 일반적인 축소기법 중의 하나인 신경망과 고도자료를 부가자료로 활용하여 유역의 지형적 특성을 반영할 수 있는 SKlm 기법을 연계하여 신경망-SKlm 모형(ANN-SKlm : Artificial Neural Network - Simple Kriging with varying local means)을 구축하였다. 유역내 6개의 기상관측소 지점의 월강수량을 이용하여 신경망-SKlm 기법과 기존 강수량의 공간분포 방법인 Thiessen 및 Ordinary Kriging 을 적용하여 비교 평가하였다. 유역내에 보다 밀도있게 구성되어 있는 25개 강우관측소 지점을 대상으로 각 기법을 평가한 결과 고도자료를 부가자료로 사용하는 SKlm 기법이 가장 우수한 결과를 나타내었다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2009년도 학술발표회 초록집
• /
• pp.112-115
• /
• 2009

The research of climate change impact in hydrometeorology often relies on climate change information. In this paper, neural networks models such as support vector machine neural networks model (SVM-NNM) and multilayer perceptron neural networks model (MLP-NNM) are proposed statistical downscaling of the monthly precipitation. The input nodes of neural networks models consist of the atmospheric meteorology and the atmospheric pressure data for 2 grid points including $127.5^{\circ}E/35^{\circ}N$ and $125^{\circ}E/35^{\circ}N$, which produced the best results from the previous study. The output node of neural networks models consist of the monthly precipitation data for Seoul station. For the performances of the neural networks models, they are composed of training and test performances, respectively. From this research, we evaluate the impact of SVM-NNM and MLP-NNM performances for the downscaling of the monthly precipitation data. We should, therefore, construct the credible monthly precipitation data for Seoul station using statistical downscaling method. The proposed methods can be applied to future climate prediction/projection using the various climate change scenarios such as GCMs and RCMs.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2009년도 학술발표회 초록집
• /
• pp.1142-1146
• /
• 2009

A rainfall simulation and forecasting technique that can generate daily rainfall sequences conditional on multi-model ensemble GCMs is developed and applied to data in Korea for the major rainy season. The GCM forecasts are provided by APEC climate center. A Weather State Based Downscaling Model (WSDM) is used to map teleconnections from ocean-atmosphere data or key state variables from numerical integrations of Ocean-Atmosphere General Circulation Models to simulate daily sequences at multiple rain gauges. The method presented is general and is applied to the wet season which is JJA(June-July-August) data in Korea. The sequences of weather states identified by the EM algorithm are shown to correspond to dominant synoptic-scale features of rainfall generating mechanisms. Application of the methodology to seasonal rainfall forecasts using empirical teleconnections and GCM derived climate forecast are discussed.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2005년도 학술발표회(1)
• /
• pp.65-65
• /
• 2005

• 한국기후변화학회지
• /
• 제9권1호
• /
• pp.47-58
• /
• 2018

In this study, we evaluated the uncertainty in the process of selecting GCM and downscaling method for assessing the impact of climate change, and influence of user-centered climate change information on reproducibility of Chungju Dam inflow was analyzed. First, we selected the top 16 GCMs through the evaluation of spatio-temporal reproducibility of 29 raw GCMs using 30-year average of 10-day precipitation without any bias-correction. The climate extreme indices including annual total precipitation and annual maximum 1-day precipitation were selected as the relevant indices to the dam inflow. The Simple Quantile Mapping (SQM) downscaling method was selected through the evaluation of reproducibility of selected indices and spatial correlation among weather stations. SWAT simulation results for the past 30 years period by considering limitations in weather input showed the satisfactory results with monthly model efficiency of 0.92. The error in average dam inflow according to selection of GCMs and downscaling method showed the bests result when 16 GCMs selected raw GCM analysi were used. It was found that selection of downscaling method rather than selection of GCM is more is important in overall uncertainties. The average inflow for the future period increased in all RCP scenarios as time goes on from near-future to far-future periods. Also, it was predicted that the inflow volume will be higher in the RCP 8.5 scenario than in the RCP 4.5 scenario in all future periods. Maximum daily inflow, which is important for flood control, showed a high changing rate more than twice as much as the average inflow amount. It is also important to understand the seasonal fluctuation of the inflow for the dam management purpose. Both average inflow and maximum inflow showed a tendency to increase mainly in July and August during near-future period while average and maximum inflows increased through the whole period of months in both mid-future and far-future periods.

• 대한의용생체공학회:의공학회지
• /
• 제26권4호
• /
• pp.185-191
• /
• 2005

In this paper, we proposed a downscaled 3D object technique using medical images for telediagnostic use. The proposed system consisted of downscaling/thresholding processes for building a downscaled 3D object and a process for obtaining 2D images at specific angles for diagnosis support. We used 80 slices of Digital Imaging and Communication in Medicine(DICOM) CT images as sample images and the platform-independent Java language for the experiment. We confirmed that the total image set size and transmission time of the original DICOM image set using a down-scaled 3D object decreased approximately $99\%\;and\;98.41\%,$ respectively. With additional studies, the proposed technique obtained from these results will become useful in supporting diagnosis for home and hospital care.

• 한국농공학회논문집
• /
• 제65권1호
• /
• pp.1-13
• /
• 2023

Various techniques for bias correction and statistical downscaling have been developed to overcome the limitations related to the spatial and temporal resolution and error of climate change scenario data required in various applied research fields including agriculture and water resources. In this study, the characteristics of three different statistical dowscaling methods (i.e., SQM, SDQDM, and BCSA) provided by AIMS were summarized, and climate change scenarios produced by applying each method were comparatively evaluated. In order to compare the average rainfall characteristics of the past period, an index representing the average rainfall characteristics was used, and the reproducibility of extreme weather conditions was evaluated through the abnormal climate-related index. The reproducibility comparison of spatial distribution and variability was compared through variogram and pattern identification of spatial distribution using the average value of the index of the past period. For temporal reproducibility comparison, the raw data and each detailing technique were compared using the transition probability. The results of the study are presented by quantitatively evaluating the strengths and weaknesses of each method. Through comparison of statistical techniques, we expect that the strengths and weaknesses of each detailing technique can be represented, and the most appropriate statistical detailing technique can be advised for the relevant research.

• 한국농공학회논문집
• /
• 제52권4호
• /
• pp.83-91
• /
• 2010

Generally, the GCM (General Circulation Model) data by IPCC climate change scenarios are used for future weather prediction. IPCC GCM models predict well for the continental scale, but is not good for the regional scale. This paper tried to generate future temperature and precipitation of 8 scattered meteorological stations in South Korea by using the MIROC3.2 hires GCM data and applying LARS-WG downscaling method. The MIROC3.2 A1B scenario data were adopted because it has the similar pattern comparing with the observed data (1977-2006) among the scenarios. The results showed that both the future precipitation and temperature increased. The 2080s annual temperature increased $3.8{\sim}5.0^{\circ}C$. Especially the future temperature increased up to $4.5{\sim}7.8^{\circ}C$ in winter period (December-February). The future annual precipitation of 2020s, 2050s, and 2080s increased 17.5 %, 27.5 %, and 39.0 % respectively. From the trend analysis for the future projected results, the above middle region of South Korea showed a statistical significance for winter precipitation and south region for summer rainfall.

• 한국기후변화학회지
• /
• 제9권1호
• /
• pp.31-45
• /
• 2018

This study implemented the prediction of drought properties (number of drought events, intensity, duration) using the user-oriented systematical procedures of downscaling climate change scenarios based the multiple global climate models (GCMs), AIMS (APCC Integrated Modeling Solution) program. The drought properties were defined and estimated with Effective Drought Index (EDI). The optimal 10 models among 29 GCMs were selected, by the estimation of the spatial and temporal reproducibility about the five climate change indices related with precipitation. In addition, Simple Quantile Mapping (SQM) as the downscaling technique is much better in describing the observed precipitation events than Spatial Disaggregation Quantile Delta Mapping (SDQDM). Even though the procedure was systematically applied, there are still limitations in describing the observed spatial precipitation properties well due to the offset of spatial variability in multi-model ensemble (MME) analysis. As a result, the farther into the future, the duration and the number of drought generation will be decreased, while the intensity of drought will be increased. Regionally, the drought at the central regions of the Korean Peninsula is expected to be mitigated, while that at the southern regions are expected to be severe.

• 한국컴퓨터정보학회논문지
• /
• 제26권4호
• /
• pp.47-53
• /
• 2021

본 논문에서는 초해상도(Super-Resolution, SR)을 계산하는데 필요한 물리 기반 시뮬레이션 데이터를 효율적으로 분류하고 분할하여 빠르게 SR연산을 가능하게 하는 쿼드트리 기반 최적화 기법을 제안한다. 제안하는 방법은 입력 데이터로 사용하는 연기 시뮬레이션 데이터를 다운스케일링(Downscaling)하여 쿼드트리 연산 소요 시간을 대폭 감소시킨다. 이 과정에서 연기의 밀도를 이진화함으로써, 다운스케일링 과정에서 밀도가 수치 손실되는 문제를 완화하며 쿼드트리를 구축한다. 학습에 사용된 데이터는 COCO 2017 데이터 셋이며, 인공신경망은 VGG19 기반 네트워크를 사용한다. 컨볼루션 계층을 거칠 때 데이터의 손실을 막기 위해 잔차(Residual) 보완 방식과 유사하게 이전 계층의 출력 값을 더해주며 학습을 진행한다. 실험결과가 연기의 경우 제안된 방법은 이전 접근법에 비해 약 15~18배 정도의 속도향상을 얻었다.