Is Soil Temperature Better than Air Temperature for Predicting Winter Wheat Phenology?

In predicting wheat (Triticum aestivum L.) phenology, logic suggests that basing thermal unit accumulation on near-surface soil temperature should give a better representation of shoot apex thermal unit accumulation than air temperature until internode elongation raises the apex above the soil surface. A study was undertaken to determine if predictions of winter wheat phenology are improved when based on measured near-surface soil temperature rather than air temperature. Air temperature 1.5 m above the soil surface and soil temperature at crown depth (the position of the shoot apex before stem elongation) were collected for 23 site-years across the U.S. Central Great Plains representing a range of cultivars, soils, management practices, and climates. Seven site-years from different sites were randomly selected to calculate the mean thermal units from both seeding and 1 January to specific gowth stages based on both air and soil temperature. These means were used to predict occurrence of gowth stages for the remaining 16 site-years. In no instance did soil temperature significantly improve prediction of winter wheat phenology. From these results, we conclude that the additional effort and expense of using soil temperature in predicting winter wheat phenology are not justified. 0 NE RECURRING QUEST in wheat phenological research in recent decades has been to better understand the role of temperature and to improve the thermal time concept (reviewed by McMaster, 1997). Most wheat phenological modeling is based on the concept of thermal time and its many modifications and variations, which incorporate factors such as photoperiod, water stress, nutrient stress, upper temperature thresholds, and varying the base temperature (e.g., McMaster et al., 1992; Rickman et al., 1996; Ritchie and Otter, 1985; Weir et al., 1984). Although introduction of other factors can improve the prediction of phenology, temperature remains the primary factor driving wheat development (e.g., Frank and Bauer, 1995; Jamieson et al., 1995; Wilhelm and McMaster, 1995). Purvis (1961) was one of the first to indicate that the wheat shoot apex directly perceives temperature. Further, the long history of experiments in which roots and shoots are maintained at different temperatures also indicates that the grass shoot apex perceives temperature. See, for example, Kleinendorst and Brouwer, 1970, Watts, 1972, and Bollero et al., 1996, for corn (Zea mays L.); Ong, 1983, for pearl millet [Pennisetum glaucum (L.) R.Br.; syn. P. typhoides (Burm. f.) Stapf & C.E. Hubb.]; Peacock, 1975, for perennial ryegrass (Loliumperenne L.); and Sato and Ito, 1968, for orchardgrass (Dactylis glomerata L.) and perennial ryegrass. Thermal time, often expressed in growing degreedays (GDD), is commonly calculated from air temperature, and a constant relationship between air temperature and shoot apex temperature is assumed. The wheat shoot apex is located in the crown of the plant until internode elongation raises the apex above the soil surface. It seems logical that while the shoot apex is in the crown, soil temperature at crown depth might be a better indicator of shoot apex temperature than air temperature. When the shoot apex is elevated above the soil G.S. McMaster, USDA-ARS, Great Plains Systems Res., P.O. Box E, Fort Collins, CO 80522 and W.W. Wilhelm, USDA-ARS, Soil and air temperature may be a better measure of Water Conservation Res., Univ. of Nebraska, East Campus, 119 Keim apex temperature than soil temperature. Following this Hall, Lincoln, NE 68583-0934. *Corresponding author (greg@gpsr. colostate.edu). Abbreviations: GDD, growing degree-days; LAI, leaf area index; RMSE, root mean square error; SARES, sum of absolute residuals; Published in Agron. J. 90:602407 (1998). SRES, sum of residuals. MCMASTER AND WILHELM: SOIL VS. AIR TEMPERATURE FOR PREDICTING WINTER WHEAT PHENOLOGY 603 Table 1. Site descriptions, with soil type, wheat cultivar, rotation, and site-year seeding dates.