이상기상 조건에서 AutoML 논벼수확 모델의 강건성 평가
Evaluation of the Robustness of an AutoML Model for Paddy Rice Yields under Extreme Weather Conditions
4
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Climate change can make a significant impact on the growth of paddy rice in Korea, leading to increased uncertainty in predicting yields. While previous studies mainly focused on the concept of extreme weather effects on paddy rice yields, this study presented a quantitative analysis of the robustness of an Artificial Intelligence (AI) yield model under several extreme weather conditions such as heavy rain, low temperature, heatwave, and drought. Our nationwide experiment with cross-validation for the period 2000-2019 in Korea showed that the paddy rice yields under extreme weather conditions did not show a significant difference from the regular period, with a Mean Absolute Error (MAE) of 0.22 to 0.23 ton/ha (3-4% of the average yield). Our findings demonstrate that our AI model for paddy rice yields produces stable predictions even under extreme weather conditions, suggesting that the model can be further enhanced for future use.
Abstract
Climate change can make a significant impact on the growth of paddy rice in Korea, leading to increased uncertainty in predicting yields. While previous studies mainly focused on the concept of extreme weather effects on paddy rice yields, this study presented a quantitative analysis of the robustness of an Artificial Intelligence (AI) yield model under several extreme weather conditions such as heavy rain, low temperature, heatwave, and drought. Our nationwide experiment with cross-validation for the period 2000-2019 in Korea showed that the paddy rice yields under extreme weather conditions did not show a significant difference from the regular period, with a Mean Absolute Error (MAE) of 0.22 to 0.23 ton/ha (3-4% of the average yield). Our findings demonstrate that our AI model for paddy rice yields produces stable predictions even under extreme weather conditions, suggesting that the model can be further enhanced for future use.
참고문헌 (14)
- 안중배, "수치예보모형을 이용한 역학적 규모축소 기법을 통한 농업기후지수 모사", 한국농림기상학회지, 12(1), pp.1~10 (2010), 한국농림기상학회
- 조수빈, "우리나라 시군단위 벼 수확량 예측을 위한 다종 기상자료의 비교평가", 대한원격탐사학회지, 37(2), pp.337~357 (2021), 대한원격탐사학회[DOI]
- He, X., "AutoML: a survey of the state-of-the-art", Knowledge-Based Systems, 212, pp.106622~ (2021)
- The Impacts of Climate Change on Food Production: A 2020 Perspective, "Hisas, L.", Universal Ecological Fund, ~ (2011)
- 홍석영, "MODIS NDVI와 기상자료를 이용한 우리나라 벼 수량 추정", 대한원격탐사학회지, 28(5), pp.509~520 (2012), 대한원격탐사학회[DOI]
- Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, "IPCC", IPCC, ~ (2014)
- 김나리, "위성영상과 기후자료를 이용한 미중서부 옥수수와 콩의 수확량 추정", 기후연구, 9(4), pp.315~329 (2014), 기후연구소[DOI]
- Krishnan, P., "Impact of elevated CO2 and temperature on rice yield and methods of adaptation as evaluated by crop simulation studies", Agriculture, Ecosystems & Environment, 122(2), pp.233~242 (2007)
- 이승호, "기후변화가 농업생태에 미치는 영향 -나주지역을 사례로-", 대한지리학회지, 43(1), pp.20~35 (2008), 대한지리학회
- 마종원, "SSAE 알고리즘을 통한 2003-2016년 남한 전역 쌀 생산량 추정", 대한원격탐사학회지, 33(5), pp.631~640 (2017), 대한원격탐사학회
- 나상일, "시계열 마스크 맵이 논벼 NDVI와 단수와의 관계에 미치는 영향", 대한원격탐사학회지, 36(5), pp.725~733 (2020), 대한원격탐사학회[DOI]
- Schierhorn, F., "Machine learning reveals complex effects of climatic means and weather extremes on wheat yields during different plant developmental stages", Climatic Change, 169, pp.39~ (2021)
- 서명철, "기후변화에 따른 국내 벼 품종과 재배기술의 적응성에 관한 고찰", 한국작물학회지, 65(4), pp.327~338 (2020), 한국작물학회