Statistical Prediction of Growing Season in Vegetation over South Korea Using Climate Factors

초록

This study defines growing season in vegetation over South Korea and examines the impact of climatic factors on the growing season. Also, a statistical model for predicting growing season was developed using stepwise regression based on climate factors. Using satellite-observed data, start of the growing season (SOS), end of the growing season (EOS), and length of the growing season (LOS) dates were estimated and compared with observed leaf-out and flowering dates from arboretums. Results indicated that between 2001 and 2020, SOS is advancing at a rate of  0.490 days year1, EOS is delaying at 0.229 days year1 and LOS is extending at 0.771 days year1. Climatic factors affecting the growing season, such as temperature and precipitation from 1, 2, 3, and 4 months before SOS and EOS dates, winter duration, preseason temperature, growing season temperature, and preseason rainfall were analyzed and some of these factors were used for the statistical prediction model. Results using leave-one-out cross validation showed that the prediction model for SOS, EOS, and LOS demonstrated good predictive performance. Our study provides information for understanding the growing season in vegetation in response to climate change over South Korea. This suggests the potential applicability of statistical models for forecasting vegetative growing season under the future climate change.

키워드