랜덤포레스트를 이용한 낙엽송과 편백의 적지적수도 제작: 경상남도를 대상으로
Mapping Species-Specific Optimal Plantation Sites Using Random Forest in Gyeongsangnam-do Province, South Korea
0
조회
0
다운로드
6
인용
0
참고문헌
키워드
초록
본 연구의 목적은 적지적수 판단에 있어 최근 분류 예측에 활용되고 있는 랜덤포레스트 기법의 적용가능성을 살펴보는데 있다. 즉, 수종별 조림 적지 판단에 있어 랜덤포레스트 기법을 소개하고 적지적수도를 작성하여 적용성을 판단하고자 한다. 그 결과 랜덤포레스트 기법의 예측 정확도는 낙엽송89.29%, 편백 73.89%로 높은 편으로 나타났다. 변수 중요도는 두 개의 수종 모두 표고, 경사, 방위의순으로 영향력이 높은 것으로 나타났으며 지형, 토성, 토양형이 낮은 영향력을 보였다. 적지적수도 작성 결과, 낙엽송은 경상남도 중부를 제외한 대부분 지역이 가능지와 적지로 나타났으며, 편백은 경상남도의 북동부 지역이 적지로 나타났다. 랜덤포레스트 기법은 적지적수도 작성뿐만 아니라 산림 분야에서적용되어 왔던 다양한 형태의 분류 및 예측 연구에서 활용 가능성이 높을 것으로 사료된다.
Abstract
The purpose of this study is to examine the applicability of the random forest technique, which is recently used for classification prediction. In other words, we will introduce the random forest technique and make a map the species-specific optimal plantation sites. As a result, the accuracy of prediction was 89.29% for Larix kaempferi and 73.89% for Chamaecyparis obtusa. In terms of variable importance, both species showed high influence in the order of DEM, slope, and aspect, and showed low influence in the order of topography, soil texture and soil type. According to the results of the mapping species-specific optimal plantation sites, most regions except for the central part of Gyeongsangnam-do province were judged to be high and moderate suitability of L. kaempferi. The northeastern part of Gyeongsangnam-do province was judged to be high suitability of C. obtusa. The random forest is expected to be highly applicable to various types of classification and prediction studies that have been applied in the forest field as well as the mapping species-specific optimal plantation sites.
참고문헌 (29)
- Breiman L, "Random Forests", Machine Learning, 45(1), pp.5~32 (2001)
- Carvalho, MC, "Modeling ecological niche of tree species in brazilian tropical area", J. of Plant Ecology, 11(6), pp.803~814 (2017)
- Ceh M, "Estimating the performance of random forest versus multiple regression for predicting prices of the apartments", International J. of Geo-Information, 7(5), pp.168~ (2018)
- Choi JH, "Decision Trees and Its Applications", J. of The Korean Official Statistics, 4(1), pp.61~83 (1999)
- Hastie T, "The elements of statistical learning: Data mining, inference, and prediction", pp.745~ (2009), Springer
- 강성표, "삼나무와 편백의 지위지수 추정 및 적지판정 - 제주특별자치도와 남부지역을 중심으로 -", 한국도서연구, 27(3), pp.125~144 (2015), 한국도서(섬)학회
- 김동현, "기후변화를 고려한 낙엽송 지위지수 추정", 한국산림과학회지, 101(1), pp.53~61 (2012), 한국산림과학회
- 김의경, "주요 조림수종의 투자효율성 분석 및 적지 판정을 위한 GIS 응용모델의 개발", 한국지리정보학회지, 9(2), pp.1~10 (2006), 한국지리정보학회
- Kim HH, "Neural network applications to determining suitable tree species for site-specific conditions", J. of Korean Forest Society, 90(4), pp.437~444 (2001)
- "Korea Forestry Promotion Institute" (2019)[URL]
- Establishment of Softwood Monitoring Base for Management of Larix kaempferi Plantation Area in Taebaeksan National Park, "Korea National Park Research Institute", 202~ (2017)
- Lee DS, "Development of hazard rating system to pine wilt disease based on environmental factors using a random forest model", pp.2018
- Lee JM, "A Study on Statistical Regression Analysis", pp.2002
- 이영근, "기후변화가 신갈나무의 적지분포에 미치는 영향", 한국산림과학회지, 103(4), pp.605~612 (2014), 한국산림과학회[DOI]
- Li H, "Substituting random forest for multiple linear regression improves binding affinity prediction of scoring functions : Cyscore as a case study", BMC Bioinformatics, 15, pp.291~ (2014)
- Louppe G, "Understanding Random Forests", pp.2014
- 문가현, "환경요인을 이용한 남원시의 적지적수도 제작", 한국농림기상학회지, 17(2), pp.126~135 (2015), 한국농림기상학회[DOI]
- Moon SY, "Performance Comparison of Classification methods based on the random forest in class imbalanced data", pp.2018
- History and Utilized Research of Species-Specific Optimal Plantation Sites, "National Institute of Forest Science", 128~ (2009)
- National Institute of Forest Science, "Forest Location Soil Map (1:5,000) Production standard Manual", 143~ (2011)
- Economic species Hinoki Cypress, "National Institute of Forest Science", 155~ (2012)
- Economic species Japanese Larch, "National Institute of Forest Science", 180~ (2012)
- Predicting the changes of productive areas for major tree species under climate change in Korea, "National Institute of Forest Science", 240~ (2014)
- "Right Tree on Right Site" (2019)[URL]
- Park YS, "Hazard ratings of pine forests to a pine wilt disease at two spatial scales (individual trees and stands) using self-organizing map and random forest", Ecological Informatics, 13, pp.40~46 (2013)
- Robert LM, "Site Index Curves for White Fir in the Southwestern United States Developed Using a Guide Curve Method", Western journal of applied Forestry, 21(2), pp.87~93 (2006)
- "right tree on right site" (2019)[URL]
- 신만용, "기후대별 입지환경 인자에 의한 소나무류의 지위지수 추정식 및 적지 구명", 한국농림기상학회지, 9(3), pp.179~187 (2007), 한국농림기상학회
- United States Department of Agriculture, "Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys", pp.871~ (1999), United States Department of Agriculture