ANFIS를 활용한 GloSea5 앙상블 기상전망기법 개선 = An enhancement of GloSea5 ensemble weather forecast based on ANFIS

다국어 초록 (Multilingual Abstract)

ANFIS-based methodology for improving GloSea5 ensemble weather forecast is developed and evaluated in this study. The proposed method consists of two steps: pre & post processing. For ensemble prediction of GloSea5, weights are assigned to the ensemble members based on Optimal Weighting Method (OWM) in the pre-processing. Then, the bias of the results of pre-processed is corrected based on Model Output Statistics (MOS) method in the post-processing. The watershed of the Chungju multi-purpose dam in South Korea is selected as a study area. The results of evaluation indicated that the pre-processing step (CASE1), the post-processing step (CASE2), pre & post processing step (CASE3) results were significantly improved than the original GloSea5 bias correction (BC_GS5). Correction performance is better the order of CASE3, CASE1, CASE2. Also, the accuracy of pre-processing was improved during the season with high variability of precipitation. The post-processing step reduced the error that could not be smoothed by pre-processing step. It could be concluded that this methodology improved the ability of GloSea5 ensemble weather forecast by using ANFIS, especially, for the summer season with high variability of precipitation when applied both pre- and post-processing steps.

국문 초록 (Abstract)

본 연구에서는 ANFIS 기반 GloSea5 앙상블 기상전망 개선 기법을 개발하고 평가하였다. 대상유역은 국내 주요 다목적댐인 충주댐 유역을 선정하였으며, 개선 기법은 ANFIS 기반의 전·후처리기법으로 구성된다. 전처리 기법에서 GloSea5의 앙상블 멤버에 가중치를 부여하며(OWM), 후처리과정에서는 전처리결과를 편의보정 한다(MOS). 평가결과 편의보정된 GloSea5에 비해 예측성능이 개선되었으며, CASE3, CASE1, CASE2 순으로 모의성능이 우수하였다. 전처리 기법은 강수의 변동성이 큰 계절에 개선효과가 우수하였으며, 후처리 기법은 전처리로 개선하지 못한 오차를 줄일 수 있는 것으로 나타났다. 따라서 본 연구에서 개발한 ANFIS 기반 GloSea5 앙상블 기상전망 개선 기법은 전·후처리 기법을 함께 사용하는 것이 가장 좋으며, 특히 여름철과 같이 강수의 변동성이 큰 계절에 활용성이 높을 것으로 판단된다.

참고문헌 (Reference)

1 Krasnopolsky, V. M., "neural network nonlinear multimodel ensemble to improve precipitation forecasts over continental US" 2012 : 2012

2 Tebaldi, C., "The use of the multi-model ensemble in probabilistic climate projections" 365 (365): 2053-2075, 2007

3 Lorenz, E. N, "The predictability of a flow which possesses many scales of motion" 21 (21): 289-307, 1969

4 Best, M. J., "The Joint UK Land Environment Simulator (JULES), model description - Part 1: Energy and water fluxes" 4 : 595-640, 2011

5 Buizza, R., "Stochastic representation of model uncertainties in the ECMWF ensemble prediction system" 125 (125): 2887-2908, 1999

6 Shiri, J., "Short-term and long-term streamflow forecasting using a wavelet and neuro-fuzzy conjunction model" 394 (394): 486-493, 2010

7 Min, S. K., "Probabilistic climate change predictions applying Bayesian model averaging" 365 (365): 2103-2116, 2007

8 Nair, A., "Prediction of Monthly Summer Monsoon Rainfall Using Global Climate Models Through Artificial Neural Network Technique" 175 (175): 403-419, 2018

9 Maraun, D., "Precipitation downscaling under climate change: recent developments to bridge the gap between dynamical models and the end user" 48 (48): 2010

10 Korea Meteorological Administration, "Operating system and 2014 verification of the high resolution joint seasonal forecast system between KMA and Met Office" KMA 1-7, 2015

11 Xu, J., "Online multitask learning framework for ensemble forecasting" 29 (29): 1268-1280, 2017

12 Madec, G., "Note du Pole de Modélisation" Institut Pierre-Simon Laplace (IPSL) 2008

13 Sun, W., "Multiple model combination methods for annual maximum water level prediction during river ice breakup" 32 (32): 421-435, 2018

14 Mitchell, T. M., "Machine Learning" WCB/McGraw-Hill 108-112, 1997

15 Jang, J. S. R, "Input selection for ANFIS learning" 2 : 1493-1499, 1996

16 Awan, J. A., "Improving ANFIS based model for long-term dam inflow prediction by incorporating monthly rainfall forecasts" 28 (28): 1185-1199, 2014

17 Wood, A. W., "Hydrologic implications of dynamical and statistical approaches to downscaling climate model outputs" 62 : 189-216, 2004

18 World Meterological Organization, "Guidelines on ensemble prediction systems and forecasting" World Meterological Organization 1-17, 2012

19 Wu, M. C., "Effective Use of Ensemble Numerical Weather Predictions in Taiwan by Means of a SOM-Based Cluster Analysis Technique" 9 (9): 836-, 2017

20 Chau, K. W., "Comparison of several flood forecasting models in Yangtze River" 10 (10): 485-491, 2005

21 Hunke, E. C., "CICE: the Los Alamos Sea Ice Model Documentation and Software User’s Manual Version 4.1" T-3 Fluid Dynamics Group, Los Alamos National Laboratory 675-, 2010

22 Raftery, A. E., "Bayesian model averaging to calibrate forecast ensembles" 133 : 1155-1174, 2005

23 Doycheva, K., "Assessment and weighting of meteorological ensemble forecast members based on supervised machine learning with application to runoff simulations and flood warning" 33 : 427-439, 2017

24 Yaseen, Z. M., "Artificial intelligence based models for stream-flow forecasting: 2000-2015" 530 : 829-844, 2015

25 Sarraf, B. S., "Analysis of post-processing method for dynamic models output using network data for the drought in North West of Iran" 181 : 2017

26 Jang, J. S. R, "ANFIS: Adaptive-network-based fuzzy inference system" 23 (23): 665-685, 1993

27 Cuo, L., "A review of quantitative precipitation forecasts and their use in short-to mediumrange streamflow forecasting" 12 (12): 713-728, 2011

28 Davies, T., "A new dynamical core for the Met Office's global and regional modelling of the atmosphere" 131 (131): 1759-1782, 2005

29 Cheng, L., "A methodology for deriving ensemble response from multimodel simulations" 522 : 49-57, 2015