SPATIAL AND INTERANNUAL VARIATIONS IN THE PHOTOTHERMAL QUOTIENT - IMPLICATIONS FOR THE POTENTIAL KERNEL NUMBER OF WHEAT CROPS IN ARGENTINA

Analyses were performed to evaluate the hypothesis that, under conditions of good water availability, variations in yield in the Argentine wheat belt were related to variations in the photothermal quotient (Pq, the ratio of mean daily incident short-wave radiation to mean daily temperature in excess of 4.5 degrees C). Data relating to spring bread wheat yields collected between 1978 and 1985 at several sites of the national wheat nursery, for which good weather data were available, were filtered to eliminate trials which may have been exposed to drought and other adversities, such as disease and hail. There were significant (P = 0.05) differences in yield, kernel number (Nk, grains m(-2)) and, to some extent, kernel weight among crops grown at different sites under close to optimum levels of water and nutrient availability. Nk was the most important determinant of yield, variations in Nk accounting for 78% of the total variance in the yield. Cultivar effects, presumably cultivar-specific differences in kernel weight, accounted for a further 14% of this variance. For the filtered data set, the relationship between Nk and Pq for the 25 days prior to anthesis, accounted for 52% of the total variability in Nk. The Pq and cultivar together accounted for 82% of total variability. Associations between Nk and either radiation or temperature alone were weaker, and site effects not accounted for by Pg were very small. The Pq estimated from 12- to 15-year weather records for the currently accepted optimum date of anthesis differed significantly (P = 0.05) between each of three representative sites in the northern, central and southern parts of the wheat belt. Reductions in potential Nk associated with interannual variations in Pg at each site may be as much as 34% of maximum Nk for the site. The Pq for anthesis dates of up to 20 days before or after the currently accepted optimum at any of these sites did not differ with respect to the optimum date, thus no advantage in terms of greater Nk is to be expected from changes in anthesis dates.