ISHS Acta Horticulturae 97: VI Symposium on Horticultural Economics
A MODEL OF POTATO DEVELOPMENT AND BULKING AS AN AID TO ECONOMIC DECISION-MAKING |
| Authors: | C. Hackett, Peter J. Sands |
| Abstract:
To aid a national potato crop reporting service for Australia (Hackett and Rattigan 1979), and to help with decision-making about the location and timing of potato production, a simple simulation model of the development and bulking of potatoes has been constructed. The two main components of the model are (a) a system for predicting the course and timing of tuber development and (b) a means of using this to predict tuber weight at any time. The model works well for a wide range of sites in Australia, and it is so simple that final yield can be estimated from commonly available weather data in 5 minutes using a pocket calculator and some look-up tables (Sands and Hackett 1979). Progressive yields can be estimated for the whole of bulking in 25 minutes using the same method. Reliable coefficients are available for the variety Sebago. Estimates for other varieties grown in Australia are being developed, and comparable work on varieties grown elsewhere will be welcomed. In the past, crop models have mostly been based solely on experimental data. Often, therefore, they have given poor estimates of commercial yields. The potato model has, however, been presented in two versions, one which is applicable to individual crops grown on a single occasion (Sands, Hackett and Nix 1979), the other being a version which estimates the long-term average performance likely from a population of commercial growers (Hackett, Sands, and Nix 1979a). Applications which have been illustrated in print (Hackett, Sands and Nix 1979b) include estimation of the benefits of modifying the timing of production in existing producing districts, identification or new producing districts in Australia, and examination of the prospects for seed potato production in cool parts of Australia for export to lowland tropics overseas. Each application took only about an hour on the desktop computer used, with very little preparation time. The model can be operated reliably by persons lacking formal training in biology or computing, and it has the additional advantage that its workings are transparent to the user. Thus, when errors are found, it is easy to tell whether it is the person or the model which is in error. | |
Download Adobe Acrobat Reader (free software to read PDF files) | |
