고해상도 원격탐사 자료와 기계학습을 이용한 한국 산림의탄소 저장량 산정 (Estimation of Forest Carbon Stock in South Korea Using Machine Learning with High-Resolution Remote Sensing Data)

서지정보

· 발행기관 : 한국기상학회
· 수록지 정보 : 대기 / 33권 / 1호 / 61 ~ 72페이지
· 저자명 : 신재원, 정수종, 장동영

초록

Accurate estimation of forest carbon stocks is important in establishing greenhousegas reduction plans. In this study, we estimate the spatial distribution of forest carbon stocksusing machine learning techniques based on high-resolution remote sensing data and detailedfield survey data. The high-resolution remote sensing data used in this study are Landsat indices(EVI, NDVI, NDII) for monitoring vegetation vitality and Shuttle Radar Topography Mission(SRTM) data for describing topography. We also used the forest growing stock data from theNational Forest Inventory (NFI) for estimating forest biomass. Based on these data, we built amodel based on machine learning methods and optimized for Korean forest types to calculatethe forest carbon stocks per grid unit. With the newly developed estimation model, we createdforest carbon stocks maps and estimated the forest carbon stocks in South Korea. As a result,forest carbon stock in South Korea was estimated to be 432,214,520 tC in 2020. Furthermore,we estimated the loss of forest carbon stocks due to the Donghae-Uljin forest fire in 2022 usingthe forest carbon stock map in this study. The surrounding forest destroyed around the fire areawas estimated to be about 24,835 ha and the loss of forest carbon stocks was estimated to be1,396,457 tC. Our model serves as a tool to estimate spatially distributed local forest carbonstocks and facilitates accounting of real-time changes in the carbon balance as well as managingthe LULUCF part of greenhouse gas inventories.

영어초록

Accurate estimation of forest carbon stocks is important in establishing greenhousegas reduction plans. In this study, we estimate the spatial distribution of forest carbon stocksusing machine learning techniques based on high-resolution remote sensing data and detailedfield survey data. The high-resolution remote sensing data used in this study are Landsat indices(EVI, NDVI, NDII) for monitoring vegetation vitality and Shuttle Radar Topography Mission(SRTM) data for describing topography. We also used the forest growing stock data from theNational Forest Inventory (NFI) for estimating forest biomass. Based on these data, we built amodel based on machine learning methods and optimized for Korean forest types to calculatethe forest carbon stocks per grid unit. With the newly developed estimation model, we createdforest carbon stocks maps and estimated the forest carbon stocks in South Korea. As a result,forest carbon stock in South Korea was estimated to be 432,214,520 tC in 2020. Furthermore,we estimated the loss of forest carbon stocks due to the Donghae-Uljin forest fire in 2022 usingthe forest carbon stock map in this study. The surrounding forest destroyed around the fire areawas estimated to be about 24,835 ha and the loss of forest carbon stocks was estimated to be1,396,457 tC. Our model serves as a tool to estimate spatially distributed local forest carbonstocks and facilitates accounting of real-time changes in the carbon balance as well as managingthe LULUCF part of greenhouse gas inventories.

참고자료

· 없음