Spatiotemporal aerosol prediction model based on fusion of machine learning and spatial analysis

이권호,표성훈,Man Sing Wong 한국대기환경학회 2024 Asian Journal of Atmospheric Environment (AJAE) Vol.18 No.1

This study examined long-term aerosol optical thickness (AOT) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) to quantify aerosol conditions on the Korean Peninsula. Time-series machine learning (ML) techniques and spatial interpolation methods were used to predict future aerosol trends. This investigation utilized AOT data from Terra MODIS and meteorological data from Automatic Weather System (AWS) in eight selected cities in Korea (Gangneung, Seoul, Busan, Wonju, Naju, Jeonju, Jeju, and Baengyeong) to assess atmospheric aerosols from 2000 to 2021. A machine-learning-based AOT prediction model was developed to forecast future AOT using long-term observations. The accuracy analysis of the AOT prediction results revealed mean absolute error of 0.152 ± 0.15, mean squared error of 0.048 ± 0.016, bias of 0.002 ± 0.011, and root mean squared error of 0.216 ± 0.038, which are deemed satisfactory. By employing spatial interpolation, gridded AOT values within the observation area were generated based on the ML prediction results. This study effectively integrated the ML model with point-measured data and spatial interpolation for an extensive analysis of regional AOT across the Korean Peninsula. These findings have substantial implications for regional air pollution policies because they provide spatiotemporal AOT predictions. This study examined long-term aerosol optical thickness (AOT) data from the Moderate Resolution Imaging Spectroradiometer (MODIS) to quantify aerosol conditions on the Korean Peninsula. Time-series machine learning (ML) techniques and spatial interpolation methods were used to predict future aerosol trends. This investigation utilized AOT data from Terra MODIS and meteorological data from Automatic Weather System (AWS) in eight selected cities in Korea (Gangneung, Seoul, Busan, Wonju, Naju, Jeonju, Jeju, and Baengyeong) to assess atmospheric aerosols from 2000 to 2021. A machine-learning-based AOT prediction model was developed to forecast future AOT using long-term observations. The accuracy analysis of the AOT prediction results revealed mean absolute error of 0.152 ± 0.15, mean squared error of 0.048 ± 0.016, bias of 0.002 ± 0.011, and root mean squared error of 0.216 ± 0.038, which are deemed satisfactory. By employing spatial interpolation, gridded AOT values within the observation area were generated based on the ML prediction results. This study effectively integrated the ML model with pointmeasured data and spatial interpolation for an extensive analysis of regional AOT across the Korean Peninsula. These findings have substantial implications for regional air pollution policies because they provide spatiotemporal AOT predictions.

여기가 북핵 3자대화 고비 ; 중국은 북핵해결 결정적 중재자?

이권호 평화문제연구소 2003 통일한국 Vol.234 No.-

정지궤도위성 관측 기반의 대기변수 산출

이권호,김희섭,하종성 한국대기환경학회 2023 한국대기환경학회 학술대회논문집 Vol.2023 No.10

First Retrieval of Data regarding Spatial Distribution of Asian Dust Aerosol from the Geostationary Ocean Color Imager

이권호,유주형,안재현,김영준 한국해양과학기술원 2012 Ocean science journal Vol.47 No.4

Aerosol optical thickness (AOT) was retrieved from the Geostationary Ocean Color Imager (GOCI) on board the Communication, Ocean, and Meteorological Satellite (COMS) for the first time. AOT values were retrieved over the ocean at a spatial scale of 0.5 × 0.5 km2 by using the look-up table (LUT)- based separation technique. The radiative transfer model (RTM)was used for different models of atmosphere–ocean environmental conditions, taking into account the realistic variability of scattering and absorption. Ocean surface properties affected by whitecaps and pigment content were also taken into account. The results show that the radiance observed by the GOCI amounts to only 5% of the radiation that penetrated the ocean and, consequently, 95% of the radiation is scattered in the atmosphere or reflected at the ocean surface in the visible wavelengths longer than 0.6 ìm. Within these wavelengths, radiance variations at the top of atmosphere (TOA) due to pigment variations are within 10%, while the radiance variation due to wind speed is considerably higher. For verification of GOCI-retrieved AOTs, comparison between GOCI and ground-based sunphotometer measurement at Gosan, Korea (126.10°E, 33.23°N)) showed good correlation (r = 0.99). The GOCI observations obtained by using the proposed technique showed promising results for the daily monitoring of atmospheric aerosol loading as well as being useful for environmental supervisory authorities.

정지궤도 인공위성을 이용한 대기 에어로솔 및 수증기 산출

이권호,염종민,하종성 한국대기환경학회 2021 한국대기환경학회 학술대회논문집 Vol.2021 No.10

지구환경변화로 인한 대기 환경 프로세스가 가속화됨에 따라, 대기 환경요소에 대한 시공간적 범위에 대한 정보를 제공할 수 있는 정밀 모니터링이 필요하게 되었다. 따라서, 본 연구는 현재 운용중인 정지궤도위성탑재체(GK2A/AMI)가 관측하는 다중 분광 관측 자료를 이용하여 고해상도의 대기중의 에어로솔 입자와 수증기량을 산출하였다. 기본적으로 대기 에어로솔은 위성의 가시채널 반사도에서 대기조성과 지표에 의한 영향을 제거하는 방법론을 사용한다. 수증기는 적외선 채널의 차등 분광 흡수도를 이용한 수치해석적 방법으로 산출하게 된다. 동아시아 지역의 사례를 대상으로 타 위성 산출물과 비교한 결과는 오차범위 내에서 만족할 만한 수준의 산출결과를 얻을 수 있었다. 이러한 정지궤도 위성 관측을 통한 대기환경정보는 대기환경 감시뿐 아니라, 더욱 고해상도의 위성영상정보(예: 다목적 실용위성)의 활용에 사용될 수 있을 것이다.

중적외선 영역의 원격탐사 기술 고찰

이권호,김희섭,Lee, Kwon-Ho,Kim, Heeseob 대한원격탐사학회 2022 大韓遠隔探査學會誌 Vol.38 No.6

Throughout the wavelengths used by remote sensing sensors, the mid-wave infrared (MWIR) has wide applicability in our life such as land, environment, disasters, and military defense. However, due to the unique characteristics of the MWIR, remote sensing techniques in this wavelength region has not been greatly developed. Recently, the applied remote sensing techniques for the MWIR sensors are presenting in a new research field, and various research results are being reported domestically and internationally. In this study, research results on the MWIR remote sensing techniques developed were investigated and related literature records were statistically analyzed. Furthermore, based on the literature review of the current and past MWIR remote sensing techniques and major research results, current status and research trends of MWIR remote sensing are presented.

Overexpression of Ref-1 Inhibits Lead-induced Endothelial Cell Death via the Upregulation of Catalase

이권호,이상기,김효신,조은정,주희경,이은지,이지영,박명수,장석종,조정현,박진봉,전병화 대한약리학회 2009 The Korean Journal of Physiology & Pharmacology Vol.13 No.6

The role of apurinic/apyrimidinic endonuclease1/redox factor-1 (Ref-1) on the lead (Pb)-induced cellular response was investigated in the cultured endothelial cells. Pb caused progressive cellular death in endothelial cells, which occurred in a concentration- and time-dependent manner. However, Ref-1 overexpression with AdRef-1 significantly inhibited Pb-induced cell death in the endothelial cells. Also the overexpression of Ref-1 significantly suppressed Pb-induced superoxide and hydrogen peroxide elevation in the endothelial cells. Pb exposure induced the downregulation of catalase, it was inhibited by the Ref-1 overexpression in the endothelial cells. Taken together, our data suggests that the overexpression of Ref-1 inhibited Pb-induced cell death via the upregulation of catalase in the cultured endothelial cells.

Role of the Scattering Phase Function in the Remote Sensing of Non-spherical Dust Aerosols

이권호 한국물리학회 2013 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.62 No.6

The effects of differences between the angular scattering structures of the spherical and the spheroidal scattering phase functions (SPFs) were studied and shown to be significant in radiative transfer calculations and for remote sensing for Asian dust aerosols. We used a discrete ordinate radiative transfer (DISORT) code to estimate the effects of different SPFs with a radiance identical to that for satellite retrieval of dust aerosols. Estimated errors due to the use of SPFs of spherical particles and the use of Henyey-Greenstein (H-G) approximation functions for dust aerosols are substantial. The variability of the resulting remote sensing reflectance was in the range 10 to 70% for the backward-scattering region, suggesting that dust aerosols cannot be treated as spheres because the dust reflectance is strongly affected by the scattering angle. In the case study of moderate resolution imaging-spectroradiometer (MODIS)-retrieved aerosol optical thickness (AOT) during the Asian dust storm day of May 1, 2011, the use of an SPF for spheroidal particles with 256 Legendre polynomial expansion (LPE) terms produced the smallest error whereas the largest errors occurred for the cases of a spherical-particle SPF with 32 LPE terms. These results imply that nonspherical dust aerosols have a large impact on the radiative transfer and that accurate information about the SPF may help reduce errors in both radiative transfer calculations and satellite remote sensing of Asian dust.

바이오매스 연소활동이 지역 대기환경에 미치는 영향

이권호 한국대기환경학회 2012 한국대기환경학회 학술대회논문집 Vol.2012 No.5

국제 : 집중분석 ; 중공 16차 전대, 자본가 입당 명문화

이권호 평화문제연구소 2002 통일한국 Vol.228 No.-