Frost damage significantly restrains global wheat production. Frost development is well documented at the landscape scale but inadequately studied at the sub-paddock scale. Particularly at sites with flat topography, which are where frost field trials are conducted. In these trials, wheat varieties are compared for frost resistance with researchers assuming minimum air temperature (frost severity) is relatively uniform. But previous research has suggested that this assumption leads to falsely identifying frost resistance. Here, we demonstrate how fibre-optic Distributed temperature sensing DTS can be used to measure nighttime temperatures in a wheat crop, to better understand frost development in field trials. DTS uses the Raman Effect and the scattering of laser light to measure temperature continuously across a fibre-optic cable providing temperature data with high spatial and temporal resolution. We demonstrate that DTS can be used to record nighttime temperature in a frost field trial with an average accuracy of 0.105 °C across 3487 m by constructing a fibre-optic fence with eight rungs, spaced at 100 mm increments from ground level, through seven blocks of wheat with different sowing times. Our research shows that even in mild frost events, vertical temperature gradients of 0.24 °C per 100 mm develop in wheat crops, with the coldest temperatures occurring ∼100 to 200 mm below the top of the ear. We also show that cold temperature development during frost is non-uniform but spatially organised in a sowing block of two varieties (Wyalkatchem and Elmore). In a sowing block there was up to a 1.3 °C range in minimum air temperature and a 5.0 °C hr (degree hours below 0 °C) variation in cold. Whereas, across the site, there was a 0.5 °C range in minimum temperature, and a 3.7 °C hr range between sowing blocks.The larger variation in minium temperature within than between sowing blocks suggests that trial design may have a greater impact on the development of cold temperature than topographic or soil differences across flat sites. There is also a varietal impact on cold development with Wyalkatchem recording more degree hours below 0 °C than Elmore, and this is suggested as driven by morphometric differences (height, canopy density and closure).Our results provide an improved understanding of cold temperature development in field trials that will aid in the search for frost resistance. They may also help to better understand cold temperature-yield relationships so that the economic impact of frost to growers can potentially be predicted to enable effective post-event management decisions.