시계열 위성영상과 머신러닝 기법을 이용한 산림 바이오매스 및 배출기준선 추정 = Machine-learning Approaches with Multi-temporal Remotely Sensed Data for Estimation of Forest Biomass and Forest Reference Emission Levels

다국어 초록 (Multilingual Abstract)

The study aims were to evaluate a machine-learning, algorithm-based, forest biomass-estimation model to estimate subnational forest biomass and to comparatively analyze REDD+ forest reference emission levels. Time-series Landsat satellite imagery and ESA Biomass Climate Change Initiative information were used to build a machine-learning- based biomass estimation model. The k-nearest neighbors algorithm (kNN), which is a non-parametric learning model, and the tree-based random forest (RF) model were applied to the machine-learning algorithm, and the estimated biomasses were compared with the forest reference emission levels (FREL) data, which was provided by the Paraguayan government.
The root mean square error (RMSE), which was the optimum parameter of the kNN model, was 35.9, and the RMSE of the RF model was lower at 34.41, showing that the RF model was superior. As a result of separately using the FREL, kNN, and RF methods to set the reference emission levels, the gradient was set to approximately -33,000 tons, -253,000 tons, and -92,000 tons, respectively. These results showed that the machine learning-based estimation model was more suitable than the existing methods for setting reference emission levels.

국문 초록 (Abstract)

본 연구는 다중시기 위성영상과 머신러닝 알고리즘을 이용하여 준국가수준의 시계열 산림바이오매스량을 추정하였으며, 이를 바탕으로 산림배출기준선 설정하여 비교·분석하였다. 머신러닝기반의 산림바이오매스 추정 모델을 구축하기위하여 Landsat TM 위성영상과 유럽항공우주국에서 제공하는 Biomass Climate Change Initiative 정보를 이용하였으며, 머신러닝 알고리즘은 비모수 학습모델인 k-Nearest Neighbor(kNN)과 의사결정나무 기반의 Random Forest(RF)를 적용하였다. 또한, 추정된 산림바이오매스량은 Forest reference emission levels(FREL) 자료와 비교하였다. 머신러닝 알고리즘 별산림바이오매스 추정 모델을 비교해보면, 최적의 kNN 모델과 RF 모델의 Root Mean Square Error (RMSE)는 각각 35.9와34.41였으며, RF모델이 kNN모델보다 상대적으로 우수하였다. 또한, FREL, kNN, RF 모델 별 산림배출기준선의 기울기는각각 약 –33천ton, -253천ton, -92천ton으로 설정되었다.

참고문헌 (Reference)

• 1 유소미 ; 송철호 ; 홍민아 ; 김휘진 ; 김지원 ; 고영진 ; 이우균, "해외 조림사업 방향성 모색을 위한 인도의 A/R CDM 사업 현황분석 및 평가" 한국기후변화학회 12 (12): 443-460, 2021
• 2 임종수 ; 한원성 ; 황주호 ; 정상영 ; 조현국 ; 신만용, "위성영상자료 및 국가 산림자원조사 자료를 이용한 산림 바이오매스 추정" 대한원격탐사학회 25 (25): 311-320, 2009
• 3 정재훈 ; 허준 ; 유수홍 ; 김경민 ; 이정빈, "kNN 알고리즘과 계절별 Landsat TM 위성영상을 이용한 단양군 지역의 지상부 바이오매스 탄소저장량 추정" 대한공간정보학회 18 (18): 119-129, 2010
• 4 서환석 ; 이정수 ; 박동환 ; 임종수, "k-NN기법을 이용한 산림바이오매스 자원량 평가 −강원대학교 학술림을 대상으로 −" 한국산림과학회 101 (101): 547-557, 2012
• 5 VCS, "VCS MODULE VMD0007 Redd Methodological Module: Estimation of Baseline Carbon Stock Changes and Greenhouse Gas Emissions from Unplanned Deforestation (BL-UP)"
• 6 Kuyper, J., "The Evolution of the UNFCCC" 43 : 343-368, 2018
• 7 Lee, W. K., "REDD+ Introduction"
• 8 Lee, W. K., "REDD+ Implementation Guidelines"
• 9 Ministry of Environment and Sustainable Development, "Nivel de Referencia de las Emisiones Forestales por Deforestación en la República del Paraguay para pago por resultados de REDD+ bajo la CMNUCC"
• 10 Meng, S., "Mapping aboveground biomass using texture indices from aerial photos in a temperate forest of Northeastern China" 8 (8): 230-243, 2016
• 11 Nicodemus Nyamari, "Impact of land cover changes on carbon stock trends in Kenya for spatial implementation of REDD+ policy" Elsevier BV 133 : 102479-, 2021
• 12 Mutanga, O., "High density biomass estimation for wetland vegetation using WorldView-2 imagery and random forest regression algorithm" 18 : 399-406, 2012
• 13 Li, Y., "Forest aboveground biomass estimation using Landsat 8 and Sentinel-1A data with machine learning algorithms" 10 : 1-12, 2020
• 14 Korea Forest Service, "Forest Master Plans 6th" Korea Forest Service 55-, 2018
• 15 Zhao, P., "Examining spectral reflectance saturation in Landsat imagery and corresponding solutions to improve forest aboveground biomass estimation" 8 (8): 469-495, 2016
• 16 Pan, Y., "Estimation of real-driving emissions for buses fueled with liquefied natural gas based on gradient boosted regression trees" 660 : 741-750, 2019
• 17 Wu, C., "Comparison of machine-learning methods for above-ground biomass estimation based on Landsat imagery" 10 (10): 035010-, 2016
• 18 Vahedi, A. A, "Artificial neural network application in comparison with modeling allometric equations for predicting above-ground biomass in the Hyrcanian mixedbeech forests of Iran" 88 : 66-76, 2016
• 19 Goussanou, C. A., "Application of site-specific biomass models to quantify spatial distribution of stocks and historical emissions from deforestation in a tropical forest ecosystem" 29 (29): 205-213, 2017
• 20 Ma, J., "Analyzing the leading causes of traffic fatalities using XGBoost and grid-based analysis : a city management perspective" 7 : 148059-148072, 2019
• 21 2050 Carbon Neutrality and Green Growth Commission, "2030 Nationally Determined Contribution (NDC) Upgrade Plan"