Productivity and water relations of field-grown cashew:: a comparison of sprinkler and drip irrigation

Cashew is an emerging crop in the seasonally 'wet-dry' tropical regions of northern Australia. In North Queensland flowering and fruiting of cashew coincides with the dry season (May-November). During this period growers sprinkler irrigate at 500 L/tree.week. A 3-year (1996-98) experiment compared this strategy with alternatives, including no irrigation or drip irrigation in which 115 or 230 L/tree.week was applied by drippers placed near the tree trunk and near the canopy drip line throughout the dry season. Measurements of soil water to 1.3 m, leaf gas exchange, chlorophyll fluorescence, tree sap flow and yield were made. Data collected in the first 2 years showed that the water requirement of the trees increased progressively as the crop load and evaporative demand increased during the dry season. During the final year of the study, additional sprinkler and drip treatments, in which water applications were progressively increased during the dry season, were introduced. The productivity of cashew in this experiment was strongly influenced by irrigation treatments, ranging (over all years) from 42 to 160 g nut/m(2) canopy surface area. Depletion of plant-available water in the root zone was associated with a reduction in photosynthesis mediated by partial stomatal closure. These effects of soil drying were evident in all irrigated treatments during the mid and late stages of the dry season but were more severe in treatments receiving the least water. When irrigation was withheld until the mid-stage of the dry season the trees had similar yields to those that were irrigated throughout, emphasising the importance of providing adequate irrigation between nut set and harvest. When rainfall from January to September in each year of the study was taken into account, there was a strong linear relationship between nut yield and water applied (rainfall + irrigation), with each extra kilolitre of water applied resulting in about 6 extra g nut/m(2) canopy surface area. This linear relationship was based on water application in the range 25-50 kL per season. It is possible that if the seasonal water application had exceeded 50 kL the marginal response to extra water may have diminished. Using drippers was slightly more efficient than sprinklers, with drip-irrigated trees requiring about 5% less water applied to achieve a given nut yield. In years when rainfall is average, and subject to other economic factors, growers in North Queensland should aim to irrigate about 500 L/tree.week. In years of low rainfall between January and September it is likely that yield will be improved by applying more irrigation water; high rainfall during these months of the year may reduce the irrigation requirement. In all cases growers should be careful to accurately monitor water applications, particularly when the total (from rainfall + irrigation) exceeds 40 kL/tree for the season.