Abstract: To study the change regulation of microclimate in sunlight greenhouse and establish the forecasting model of microclimate can provide a basis for rational development of greenhouse resources and microclimate control in sunlight greenhouse. We used hourly temperature and relative humidity to form microclimate observation instrument in sunlight greenhouse in Dongling of Shenyang, and calculated daily maximum temperature, daily minimum temperature, daily average relative humidity by the combination of winter, spring, autumn and sunny, cloudy, overcast data, to analyze the microclimate change regulation. And a series of forecasting models based on stepwise regression method were established by using data from automatic meteorological observation station. The results showed that the daily maximum temperature and daily minimum temperature in sunlight greenhouse of spring and autumn were close, and that of winter was significantly lower than that of spring and autumn, but the daily average relative humidity in winter was higher than that in spring and autumn. The daily maximum temperature reduced from sunny to cloudy to overcast days, characteristics of the daily maximum temperature were not obvious, but the daily average relative humidity increased from sunny to cloudy to overcast days. The stepwise regression forecasting models of microclimate in sunlight greenhouse all passed the significance test, the correlation coefficient was from 0.608 to 0.933, the relative error range was from 0.1% to 19.0%, and the average value of relative error was from 2.7% to 9.9%.