Frost response in wheat and early detection using proximal sensors

Abstract

Frost can significantly reduce production of field crops grown in Mediterranean-type environments, where significant economic losses for Australian wheat occur annually. If non-destructive sensors could make rapid, spatial assessment of frost damage, this could limit economic losses through timely management decisions. This paper reports on a methodology for imposing frost treatments to wheat under field conditions and the utility of canopy reflectance data for detecting early crop damage. Purpose-built chambers using stepped additions of dry ice allowed for a range of frost scenarios to be tested when applied at wheat heading and anthesis. For frost treatments applied at anthesis, grain number and yield were reduced by 8.8% and 7.2%, respectively, for every degree Celsius below zero (down to −4°C). This effect was additive over two consecutive nights. For cold load equivalent, there was a 2.2% and 1.9% reduction in grain number and yield, respectively, per °C hr (below 0°C). For wheat, spectral indices PRI and NDVI (reflectance) and FRF_G and SFR_G (fluorescence) showed significant relationships, with cold load applied for heading treatments. Next steps include targeting frost intensities equivalent to cold loads between 20 and 80°C hr and testing the utility of these proposed indices in a commercial paddock setting.