정지 궤도 기상 위성을 이용한 기계 학습 기반 강우 강도 추정: 한반도 여름철을 대상으로 = Rainfall Intensity Estimation Using Geostationary Satellite Data Based on Machine Learning: A Case Study in the Korean Peninsula in Summer

국문 초록 (Abstract)

강우 현상은 물 순환과 에너지 순환의 주요 요소 중 하나이며 강우량 추정은 수자원 확보와 수재해 예측 및 피해 감축에 매우 중요한 역할을 한다. 위성 기반 강우량 추정은 시공간적으로 고해상도인 자료를 통하여 넓은 지역을 연속적으로 감시할 수 있다는 장점이 있다. 본 연구에서는 Himawari-8 Advanced Himawari Imager (AHI) 수증기 채널(6.7 μm), 적외 채널(10.8 μm)과 기상 레이더 Column Max (CMAX) 합성장을 이용하여 기계학습 기반 정량적 강우량 추정 모델을 개발하였다. 기계학습 기법으로는 랜덤 포레스트(Random Forest, RF)를 사용하였으며 기상 레이더 반사도(dBZ)와 Z-R식으로 변환한 강우강도(mm/hr)를 타겟으로 하는 모델을 구축하여 비교하였다. 레이더 강우강도를 통해 검증하였을 때 임계성공지수(Critical Success Index, CSI)는 0.34, Mean-Absolute-Error (MAE) 4.82 mm/hr였다. GeoKompsat-2(GK-2A) 강우강도 산출물, Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN)-Cloud Classification System (CCS) 산출물과 비교하였을 때 강우 유무 분류에서 CSI 21.73%, 10.81%, 강우강도 정량적 평가에서MAE 31.33%, 23.49% 높은 성능을 보였다. 강우량 산출물을 지도화 한 결과, 실제 강우강도 분포와 유사한 분포를 모의하여 기존 산출물 대비 높은 정확도의 강우량을 추정했다.

다국어 초록 (Multilingual Abstract)

Precipitation is one of the main factors that affect water and energy cycles, and its estimation plays a very important role in securing water resources and timely responding to water disasters. Satellite-based quantitative precipitation estimation (QPE) has the advantage of covering large areas at high spatiotemporal resolution. In this study, machine learning-based rainfall intensity models were developed using Himawari-8 Advanced Himawari Imager (AHI) water vapor channel (6.7 μm), infrared channel (10.8 μm), and weather radar Column Max (CMAX) composite data based on random forest (RF). The target variables were weather radar reflectivity (dBZ) and rainfall intensity (mm/hr) converted by the ZR relationship. The results showed that the model which learned CMAX reflectivity produced the Critical Success Index (CSI) of 0.34 and the Mean-Absolute-Error (MAE) of 4.82 mm/hr. When compared to the GeoKompsat-2 and Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks (PERSIANN)-Cloud Classification System (CCS) rainfall intensity products, the accuracies improved by 21.73% and 10.81% for CSI, and 31.33% and 23.49% for MAE, respectively. The spatial distribution of the estimated rainfall intensity was much more similar to the radar data than the existing products.

참고문헌 (Reference)

1 안숙희, "한반도 기상재해의 원인별 발생 및 피해 특성" 한국방재학회 15 (15): 133-144, 2015

2 최학림, "기상위성 휘도온도와 기상레이더 반사도 자료를 이용한 한반도 영역의 강우강도 추정 비선형 관계식 개선" 한국지구과학회 39 (39): 131-138, 2018

3 유철희, "기계학습 기반 상세화를 통한 위성 지표면온도와 환경부 토지피복도를 이용한 열환경 분석: 대구광역시를 중심으로" 대한원격탐사학회 33 (33): 1101-1118, 2017

4 Marcos, C, "Validation report for “convective rainfall rate”(CRR-PGE05 v4.0). NWCCDOP2-GEO-AEMET-SCI-ATBD-Precipitation_v1.1"

5 Song, H. -J., "Two heavy rainfall types over the Korean peninsula in the humid East Asian summer environment: A satellite observation study" 143 (143): 363-382, 2015

6 Vicente, G. A., "The operational GOES infrared rainfall estimation technique" 79 (79): 1883-1898, 1998

7 Hou, A. Y., "The global precipitation measurement mission" 95 (95): 701-722, 2014

8 Huffman, G. J., "The global precipitation climatology project(GPCP)combined precipitation dataset" 78 (78): 5-20, 1997

9 Sakolnakhon, K., "The estimation rainfall using infrared(IR)band of Himawari-8 satellite over Thailand" 39 : 236-248, 2016

10 Marshall, J. S., "The distribution of raindrops with size" 76 (76): 165-166, 1948

11 Sukmin Yoon, "Statistical modeling of extreme rainfall events and applications of radar rainfall estimates for reducing flood risk in Gyeongnam area" 경상대학교 대학원 2013

12 Meisner, B. N., "Spatial and annual variations in the diurnal cycle of large-scale tropical convective cloudiness and precipitation" 115 (115): 2009-2032, 1987

13 Breiman, L., "Random forests" 45 (45): 5-32, 2001

14 Griffith, C. G., "Rain estimation from geosynchronous satellite imagery–Visible and infrared studies" 106 (106): 1153-1171, 1978

15 Germann, U., "Radar precipitation measurement in a mountainous region, Quarterly Journal of the Royal Meteorological Society : A Journal of the Atmospheric Sciences" 132 (132): 1669-1692, 2006

16 Prigent, C., "Precipitation retrieval from space : An overview" 342 (342): 380-389, 2010

17 Joss, J., "Precipitation measurement and hydrology" 1990 : 577-606, 1990

18 Hsu, K. -L., "Precipitation estimation from remotely sensed information using artificial neural networks" 36 (36): 1176-1190, 1997

19 Hong, Y., "Precipitation estimation from remotely sensed imagery using an artificial neural network cloud classification system" 43 (43): 1834-1853, 2004

20 Sadeghi, M., "PERSIANN-CNN : Precipitation estimation from remotely sensed information using artificial neural networks-convolutional neural networks" 20 (20): 2273-2289, 2019

21 Prat, "PERSIANN-CDR: Daily precipitation climate data record from multisatellite observations for hydrological and climate studies" 96 (96): 69-83, 2015

22 정성래, "Meteorological Products of Geo-KOMPSAT 2A (GK2A) Satellite" 한국기상학회 56 (56): 185-185, 2020

23 Cho, D., "Improvement of spatial interpolation accuracy of daily maximum air temperature in urban areas using a stacking ensemble technique" 57 (57): 633-649, 2020

24 Jang, E., "Improvement of SMAP sea surface salinity in river-dominated oceans using machine learning approaches" 58 (58): 138-160, 2021

25 Kidd, C., "Global precipitation measurement" 18 (18): 334-353, 2011

26 Miao, C., "Evaluation of the PERSIANN-CDR daily rainfall estimates in capturing the behavior of extreme precipitation events over China" 16 (16): 1387-1396, 2015

27 Hong, Y., "Evaluation of PERSIANNCCS rainfall measurement using the NAME event rain gauge network" 8 (8): 469-482, 2007

28 Sorooshian, S., "Evaluation of PERSIANN system satellite-based estimates of tropical rainfall" 81 (81): 2035-2046, 2000

29 Kim, M., "Deep learning-based monitoring of overshooting cloud tops from geostationary satellite data" 55 (55): 763-792, 2018

30 Mishra, K. V., "Deep learning framework for precipitation retrievals from communication satellites" 1-9, 2018

31 Tsay, J. -D., "Deep learning for satellite rainfall retrieval using Himawari-8 multiple spectral channels"

32 Yu, P. -S., "Comparison of random forests and support vector machine for real-time radarderived rainfall forecasting" 552 : 92-104, 2017

33 Shin, J. -Y., "Assessing the Applicability of Random Forest, Stochastic Gradient Boosted Model, and Extreme Learning Machine Methods to the Quantitative Precipitation Estimation of the Radar Data : A Case Study to Gwangdeoksan Radar, South Korea, in 2018" 2019

34 Saltikoff, E., "An overview of using weather radar for climatological studies : Successes, challenges, and potential" 100 (100): 1739-1752, 2019

35 Tao, Y., "A two-stage deep neural network framework for precipitation estimation from bispectral satellite information" 19 (19): 393-408, 2018

36 Wu, H., "A spatiotemporal deep fusion model for merging satellite and gauge precipitation in China" 584 : 124664-, 2020

37 Sun, Q., "A review of global precipitation data sets : Data sources, estimation, and intercomparisons" 56 (56): 79-107, 2018

38 Han, D., "A novel framework of detecting convective initiation combining automated sampling, machine learning, and repeated model tuning from geostationary satellite data" 11 (11): 1454-, 2019