|
|
|
| Authors: | N.W. Barba, J.P. De Melo-Abreu |
| Keywords: | chilling, dormancy, algorithm, thermal time, heat units, Prunus persica |
Abstract:
Peach trees are grown in many regions subjected to mild winter temperatures, including some regions of Portugal.
The aim of this study was to determine whether the existing models to simulate dormancy break predict this occurrence as determined by the forcing method applied to floral buds, to evaluate a computer algorithm to predict dormancy break, and to yield data on chilling requirements of two varieties, ‘Starcrest’ and ‘Baby Gold 9’. In 1999 and 2000, cuttings of these two varieties were taken periodically in two orchards, located in two regions with distinct thermal regimens, and forced at 20 °C. An optimization computer algorithm was constructed to find the accumulation of chilling that minimizes the standard error of the estimates of the days of the beginning of flowering, and computes the average chilling units accumulated for the ‘chill hours’, ‘Utah chill-units’, ‘Utah low-chilling’, ‘North Carolina’ and ‘dynamic’ models.
Data of eight dates of flowering of the var. ‘Baby Gold 9’, in different year/locations, were used for these computations.
Results from the forcing trials show that the accumulation of Utah low-chilling was a more constant quantity (‘Starcrest’: 1257 ± 39.5 chill-units; ‘Baby Gold 6’: 1428± 50.0 chill units), but the performance of the other models was only slightly worse.
The values of accumulated chilling necessary for dormancy break obtained by the algorithm for the variety ‘Baby Gold 9’, using merely the dates of flowering, were not significantly different from the measured values.
However, there is probably room for substantial improvement of these models.
|
Download Adobe Acrobat Reader (free software to read PDF files) |
|