A crop-irrigation simulation model for individual farmer use

Abstract

Agricultural research has shown that it is possible to achieve excellent economic results by the addition of irrigation to those farms situated in the low rainfall areas of the South Island. However, many farmers who have irrigation are not achieving results which place them in a better financial position than their counterparts without irrigation, because of a lack of understanding of all the factors involved in managing an irrigation property. It is argued that if prospective and practising irrigation farmers were supplied with a model of their farm, they could experiment with the model and quickly learn by experience, those management techniques associated with efficient irrigation farm management. This would lead directly to the second use of the model, as an aid for individual farm planning.

Such a model would be constructed in such a way that it would embrace only the basic logic of the crop-irrigation system of the farm, and call upon modules, or sub-programmes as they are required. Such a module would contain specific details relevant to individual crops (e.g. lucerne growth). The farmer would add to this basic outline his own farm details (e.g., paddock areas and crops to be grown) and in so doing, would construct a model of his own farm.

In order to investigate the feasibility of building such a model, a two crop, crop irrigation model was constructed, based on lucerne and wheat. This model maintained a daily soil moisture budget and related this to plant growth. The soil moisture budget was the result of monitoring daily moisture input (rainfall and irrigation) into the soil and daily output (actual evapo-transpiration) from the soil. With lucerne growth, if the soil moisture available was sufficient for the crop's requirements for that day, the potential yield increase expected (a function of stage of growth and time of year) for that day, was added to the total accumulation of dry-matter at the end of the previous day. When a moisture stress occurred, the increase in yield for that day was modified to allow for the effect of the moisture stress. With the wheat model, if the soil moisture available was sufficient for the crop's requirements for that day the expected maximum yield was maintained. If the plant was suffering a moisture stress, then the expected maximum yield was reduced, depending on the stage of growth and intensity of the moisture stress for that day. Following validation, it was concluded that the model was capable of predicting crop yields with certain limitations. Validation was carried out by comparing actual field observations of crop yields with model predictions, given the same climatic and management criteria.