High tunnel soil health is crucial for successful and sustainable crop production within protected environments. Soil microbial activity is highly temperature-dependent, and soils that are slightly warmer will foster increased metabolic rates within soil communities enhancing microbial diversity and enzymatic activity, promoting nutrient availability. However, little is understood about the potential for microbial activity during colder seasons in norther latitudes when high tunnels are taken out of production and soils are left fallow. Temperature variation in high tunnels could also create variation in microbial community activity, creating spatial nutrient variation with impacts on production the following season. To analyze soil temperature fluxes, we buried an array of 27 soil sensors four inches deep within the soil in a newly built, 30-foot-wideby 96-foot-long tunnel located in Brookings, South Dakota. The high tunnel was oriented east to west and soil was bare. Soil temperatures were recorded at 30-minute intervals from December 22 to March 15, (2023 – 2024). Air temperature and light (lux) data was also collected inside of the high tunnel as well as external weather data from a nearby (>1km) Mesonet weather station. We used multiple linear regression to model the relationship between average internal soil temperature and internal light and temperature data. We also compared sensor location (latitude, longitude, and Euclidean distance from the center of the high tunnel) on soil temperature within the high tunnel using an ANOVA and multiple linear regression to examine how sensor location was related to soil temperature. Our top model of internal soil temperature showed light, internal temperature, and the interaction between light and internal temperature explained a large amount of high tunnel soil temperature variation (R2 = 0.87, p < 0.0001). There was also significant variation in soil temperature throughout the high tunnel, with the daily mean difference of 3.12 degrees Celsius (p < 0.0001) observed between our sensor at the center of the high tunnel and our sensor near the northwest corner of the high tunnel. Our top model showed that latitude, the quadratic of longitude, and the Euclidean distance from the center of the high tunnel explained a moderate amount of high tunnel soil temperature variation (R2 = 0.44, p < 0.0001). This analysis demonstrates a need to further investigate how microbial communities react to temperature variation within high tunnels when they are not in production.
