Predicting Black Locust Flowering and Honey Production Using Meteorological Indices in Korea

Black locust ( Robinia pseudoacacia L.) accounts for 70% of Korea’s honey production(200 billion KRW annually), but climate change has caused extreme yield variability(4.3-43.8 kg/colony) and unpredictable flowering (±15 days). This study developed an integrated prediction system combining flowering phenology and honey production models. Using 7-year data(2019-2025) from five regions, we applied a Single Triangle Degree-Day(STDD) model with region-specific heat requirements (185-220°C day). We constructed an Acacia Flowering Index(AFI) integrating five meteorological factors(temperature, precipitation, wind speed, solar radiation, diurnal range) with weighted coefficients. The STDD model predicted flowering dates with high accuracy(RMSE= 3.9 days, R² = 0.74), representing 30% improvement over existing models. Based on Colony Activity Index (CI) and stage-specific AFI values, the honey production model explained 76% of yield variability (R² = 0.76, MAPE=9.67%). Flowering-period weather conditions (CIflowering) emerged as the dominant predictor (β= 0.929, p<0.001). Cross-validation confirmed robust performance (R² = 0.71-0.77). This system enables flowering prediction 2-3 weeks in advance and provides decision-support tools for optimizing migratory beekeeping, colony management, and production forecasting under climate variability.