Bustan, A., Kennedy-Leigh Centre for Horticultural Research, Faculty of Agriculture, Food and Environmental Quality Sciences, Hebrew University of Jerusalem, Rehovot 76100, Israel
Goldschmidt, E.E., Kennedy-Leigh Centre for Horticultural Research, Faculty of Agriculture, Food and Environmental Quality Sciences, Hebrew University of Jerusalem, Rehovot 76100, Israel
Erner, Y., Agricultural Research Organisation, Institute of Horticulture, Bet Dagan 50250, Israel

The foundations of the Citrus productivity model 'CITROS' were laid by Goldschmidt and Monselise (1977). The model is based on the assumption that where all other needs (water, fertilizer, etc.) are managed at, or close to the optimum, dry matter production and allocation become the fundamental processes that limit tree productivity. The purpose of the model is to identify and quantify productivity problems, predict yields and devise means for optimization of Citrus crop production. Data from the literature supplemented by new field photosynthesis measurements were used to obtain reasonable estimates of daily and annual photoassimilate production by a Citrus tree canopy. The potential relative growth rate (RGR) of reproductive organs was followed through the growing season and the cost of fruit production was calculated. A model for an effective heat unit (EHU) was constructed to improve and refine the expression of temperature effects on fruit growth. Respiratory losses, including temperature effects, were also evaluated. This quantitative information was incorporated into a series of mathematical functions that provide estimates of carbohydrate demand and consumption by the reproductive organs (flowers and fruit) within a citrus tree. The carbon consumption of other tree organs (particularly roots) and the management of carbohydrate reserves should also be incorporated into the model. Matching the daily available amount of photoassimilates with reproductive organ demand for carbon, the model predicted that a shortage of carbohydrates might already occur during bloom. While early fruit development appears to be sink-limited, at least in certain cultivars, all subsequent stages of fruit growth and maturation are source-limited. Yield predictions by the model fall within a reasonable range, but the problem of trade-off between fruit number and fruit size (during the fruit set and abscission period) requires further investigation. Actual differences in the performance of grapefruit (Citrus paradisi Macf., cv 'Marsh seedless') orchards from different climate regions were successfully predicted by the model. The model, which is already a powerful diagnostic and predictive tool, may gain further strength by incorporation of the effects of management practices (e.g., drip irrigation) and frequently encountered environmental stresses.