Abstract: Establishing an early warning model for the maximum or minimum temperature and humidity could provide timely forecast information for vegetable farmers to prevent heat injury, chilling damage, and high humidity diseases in extreme weather, when the greenhouse film is being uncovered or covered. In this study, environmental data for the year 2020 was collected using temperature and humidity sensors, and the daily regression analysis method was used along with statistical methods such as the F-value test to explore the characteristics and variation of temperature and humidity in unequal height greenhouse in Dongting Lake area. A quadratic equation was used to fit temperature and humidity data. Using the outside temperature and relative humidity values, the temperature and humidity at any given time inside the greenhouse could be estimated, and maximum or minimum warning could be provided accordingly. The results showed that there were significant differences in temperature and humidity changes in different months in unequal height greenhouses in Dongting Lake area, with stable temperature and humidity changes throughout the year. The higher temperature in the greenhouse appeared from 12:00 to 14:00 between June to September. The lower temperature in the greenhouse appeared from 4:00 to 6:00 in December and January. The higher humidity in the greenhouse appeared from 4:00 to 6:00 between March and April and between November to December. The F-value of the early warning models of the highest and lowest temperature and humidity in the greenhouse passed the 0.01 extremely significant level with a high fitting degree of R~2＞0.8, reflecting the diurnal variation of temperature and humidity in the greenhouse.