DOI QR코드

DOI QR Code

A Study on Frost Occurrence Estimation Model in Main Production Areas of Vegetables

채소 주산지에 대한 서리발생예측 연구

  • Kim, Yongseok (Climate Change & Agroecology Division, Department of Agricultural Environment, National Institute of Agricultural Science) ;
  • Hur, Jina (Climate Change & Agroecology Division, Department of Agricultural Environment, National Institute of Agricultural Science) ;
  • Shim, Kyo-Moon (Climate Change & Agroecology Division, Department of Agricultural Environment, National Institute of Agricultural Science) ;
  • Kang, Kee-Kyung (Climate Change & Agroecology Division, Department of Agricultural Environment, National Institute of Agricultural Science)
  • 김용석 (국립농업과학원 농업환경부 기후변화생태과) ;
  • 허지나 (국립농업과학원 농업환경부 기후변화생태과) ;
  • 심교문 (국립농업과학원 농업환경부 기후변화생태과) ;
  • 강기경 (국립농업과학원 농업환경부 기후변화생태과)
  • Received : 2019.11.27
  • Accepted : 2019.12.30
  • Published : 2019.12.31

Abstract

In this study, to estimate the occurrence of frost that has a negative effect on th growth of crops, we constructed to the statistical model. We factored such various meteorological elements as the minimum temperature, temperature at 18:00, temperature at 21:00, temperature at 24:00, average wind speed, wind speed at 18:00, wind speed at 21:00, amount of cloud, amount of precipitation within 5 days, amount of precipitation within 3 days, relative humidity, dew point temperature, minimum grass temperature and ground temperature. Among the diverse variables, the several weather factors were selected for frost occurrence estimation model using statistical methods: T-test, Variable importance plot of Random Forest, Multicollinearity test, Akaike Informaiton Criteria, and Wilk's Lambda values. As a result, the selected meteorological factors were the amount of cloud, temperature at 24:00, dew point temperature, wind speed at 21:00. The accuracy of the frost occurrence estimation model using Random Forest was 70.6%. When it applied to the main production areas of vegetables, a estimation accuracy of the model was 65.2 and 78.6%.

채소작물과 과수작물의 생육에 악영향을 미치는 서리발생을 미리 예측하기 위해 모형을 구축하고 채소 주산지에 적용해 보았다. 서리 발생 전날에 관측되는 다양한 기상인자들(최저기온, 18시 기온, 21시 기온, 24시 기온, 평균풍속, 18시 풍속, 21시 풍속, 구름량, 5일간 강수량, 3일간 강수량, 상대습도, 이슬점온도, 초상최저기온, 지면온도)을 수집하고, 그 중에서 서리발생에 유의한 영향이 있다고 판단되는 변수들을 통계적 방법(T-test, Random Forest, Multicollinearity test, Akaike Informaiton Criteria, 그리고 Wilk's lambda values)을 통해 선택하였다. 여러 통계적 방법을 통해 선택된 유의한 기상 인자는 24시 기온, 구름량, 이슬점온도, 21시 풍속 이였으며, 이 기상인자를 기계학습법의 한 종류인 랜덤 포레스트에 적용하여 서리 발생 예측 모형을 구축하였다. 이렇게 구축 된 서리 발생예측 모형의 정확도는 70.6%로 나타났으며, 이 모형을 가을배추와 가을무의 주산지인 홍성과 서산에 적용하였을 때 65.2%와 78.6%로 나타났다.

Keywords

References

  1. Chung, U., Seo, H.C., and Yun, J.I., 2004, Site-specific frost warning based on topoclimatic estimation of daily minimum temperature. Korean Journal of Agricultural and Forest Meteorology 6(3), 164-169. (in Korean)
  2. Han, J.H., Choi, J.J., Chung. U., Cho, K.S., and Chun, J.P., 2009, Frostfall forecasting in the naju pear production area based on discriminant analysis of climatic data. Korean Journal of Agricultural and Forest Meteorology 11(4), 135-142. (in Korean) https://doi.org/10.5532/KJAFM.2009.11.4.135
  3. Kwon, Y.A., Lee. H.S., Kwon. W.T., and Boo, K.O. 2008, The weather characteristics of frost occurrence days for protecting crops against frost damage. Journal of the Korean Geographical Society 43(6), 824-842. (in Korean)
  4. Robinson, C. and Mort, N., 1996, A neural network solution to the problem of frost prediction. UKACC International Conference on Control. Control '96, UK, 136-139.
  5. Temeyer, B.R., W.A. Gallus Jr, K. A. Jungbluth, D. Burkheimer, and D. McCauley, 2003, Using an artificial neural network to predict parameters for frost deposition on Iowa bridgeways. Proceedings of the 2003 Mid-Continent Transportation Research Symposium, Iowa State University, Iowa, USA.