Temperature as a factor in nectarine production in the Western Cape.
Phenological and climatic data were collected from nectarine orchards containing the cultivars Sunlite, Flavortop and Fantasia, growing in six climatically divergent areas in the Western Cape, South Africa. Shoots were forced at 25⁰C and bud volumes measured in order to develop an index of when the endodormancy requirement for each cultivar in each area had been satisfied. Flower bud break gave the only consistent indication of when the chilling requirement had been satisfied and 10% flower bud break after 14 days at 25⁰C was adopted as the best index of end of endodormancy. Unlike vegetative bud break, flower bud break was not influenced by forcing period or gibberellic acid (GA3) application. Comparison of winter chilling models using the temperature data for the endodormancy period, showed that the 'dynamic' model (DP) was more accurate than the currently used 'Utah' model (UCU), for estimating the rest requirement of the nectarine cultivars Sunlite, Flavortop and Fantasia. The DP model failed, however, to identify orchards displaying 'delayed foliation' symptoms as a result of insufficient winter chilling. The 'dynamic' model not only takes into account the positive effect of cool temperatures and negative effect of high temperatures on rest completion that the 'Utah' model does, but also the positive effect of moderate temperatures and the fact that chilling negation is subject to time constraints. The nectarine cultivars tested required about 33 ± 5; 46 ± 8 and 46 ± 8 'dynamic' portions of chilling for Sunlite, Flavortop and Fantasia, respectively. Modification of the UCU model, using principles from the DP model, resulted in a substantial increase in its accuracy. The modification assumes that the negating effect of high temperatures is confined to the diurnal cycle and that there is not a carry-over effect of chilling negation from one day to the next. Coefficients of variation for the estimated chilling requirements of 'Sunlite', 'Flavortop' and 'Fantasia' were reduced from 24.9%, 21.7% and 23.8% to 19.7%, 17.9% and 20.1 % respectively. This modification, now called the 'daily positive Utah chill unit model (PCU) , is currently used in the Western Cape Province to determine winter chilling efficiency in deciduous fruit orchards. The PCU requirements for 'Sunlite', Flavortop' and 'Fantasia' were 566± 111, 807± 145 and 817 ± 165 respectively. Correlations were then made between the heat component during endodormancy (degree hours > 10⁰C [D H]) and the apparent chilling requirement. The subsequent regressions gave almost parallel slopes regardless of cultivar or chilling requirement, indicating that in milder winter areas, heat accumulation could also play a vital role in endodormancy. As an approximation, an elevation in mean air temperature during the main endodormancy months of May and June, resulted in an apparent reduction in the chilling requirement of 20% for every 10⁰C increase within the range of 11-14°C. The post-endodormancy heat requirement for bud development in spring was highest in the milder winter areas and lower in areas receiving more winter chilling than their endodormancy requirement. Although there are already 65 automatic weather stations operating in South Africa's Western Province deciduous fruit industry, the variation in microclimate is much too large to make accurate extrapolations and alternative methods of estimation of chilling are required. Tables for the expeditious determination of daily chill unit accumulation from minimum and maximum temperatures were developed, using the assumption that air temperature follows either a straight line between extremes or that it follows a sine curve during the heating cycle followed by a logarithmic cooling phase. Both tables gave an average accuracy of 95 % for. determining monthly or total winter chilling in seven divergent locations over three seasons. The sine- logarithmic table appeared to be slightly more accurate under mild winter conditions. Daily positive Utah chill units are now used extensively by the South African deciduous fruit industry and colour maps depicting average PCU chilling values for the region were in the process of being plotted at the time of going to press.