Abstract:

In order to reveal the influence of meteorological factors on the quality of Shanghang tobacco, based on the daily meteorological data of Shanghang from 2009 to 2024 and the tobacco quality data of Shanghang tobacco in the same period, a variety of mathematical statistical methods were used to analyze the change characteristics of Shanghang tobacco quality, to reveal the key meteorological factors and sensitive periods affecting the quality of tobacco leaves, and then build a response model and validate the model. The results showed that: (1) from 2009 to 2024, the nicotine content of Shanghang tobacco leaves in Fujian was 1.7%−3.1%, the total sugar content was 22.4%−38.7%, the total nitrogen content was 1.5%−2.4%, the potassium content was 1.9%−3.0%, except the total sugar and reducing sugar contents were slightly higher, the other contents were all within the suitable range. (2) From 2009 to 2024, the total sugar content of Shanghang tobacco leaves showed a significant upward trend (P<0.01), the total nitrogen content showed a significant downward trend (P<0.01), and the other quality changes were not significant. (3) In terms of the transplanting period, the average temperature on the transplanting day and the extreme low temperature after transplanting for 40d were the key meteorological factors that affected nicotine. The relative humidity after transplanting for 15d was the key meteorological factor affecting total sugar and total nitrogen. Precipitation after transplanting for 15d and after transplanting for 50d were respectively the key meteorological factors affecting reducing sugar and potassium. For the picking period, the average temperature in the first 5d of picking was the key meteorological factor affecting nicotine. Extreme low temperature on the picking day was the key meteorological factor affecting total sugar and reduction of sugar. Total nitrogen were comprehensively affected by the average temperature 20d before picking, the extreme high temperature 20d before picking, the extreme low temperature on the picking day and the relative humidity 15d before picking. Extreme low temperature 40d before picking was the key meteorological factor affecting potassium. (4) The prediction model of tobacco leaf quality during the transplanting period and picking period was verified by the meteorological factor and tobacco leaf quality data from 2022 to 2024. It showed that the root mean square error between the simulated values and the actual values of nicotine, total nitrogen and potassium contents was between 0.29% and 0.71%, and the mean absolute error (MAE) was between 0.28% and 0.71%. The simulation model effect was good. It indicates that the model can guide the growth of characteristic tobacco leaves, respond to meteorological disasters during the production process, and predict the quality of tobacco leaves.

Key words: Key meteorological factors, Quality of to bacco leaf, Correlation coefficient, Regression model