During this period immediately following flowering, care needs to be exercised to avoid water stress particularly in stonefruit. For example, in peaches there is an initial rapid fruit growth for approximately 4 weeks following flowering when the soil should not be allowed to dry out beyond 40 kPa in sandy soil and 60 kPa in clay loam soils.

In most seasons in the Goulburn Valley, crops are not irrigated until reference crop evapotranspiration (ET_o) exceeds rainfall by 125 mm. Generally, this is in late October but could be as late as mid-November in a wet spring. However, in recent years there has been insufficient winter and early spring rain to wet up the root zone. Root zone soil moisture must be measured to avoid water stress. Similarly, in environments dissimilar from the Goulburn Valley (for example, trees growing in lighter soil types) measurements of soil moisture will avoid the root zone drying out excessively.

Period 2 commences approximately four to five weeks after flowering and continues until six weeks before harvest for early-maturing fruits (that is, before mid-January), and eight weeks before harvest for later maturing fruits. Trees are irrigated with greatly reduced volumes of water compared to that which would normally be applied. Irrigation to replace 30 % of orchard water use capability is recommended. Soil moisture in the middle of the wetted fibrous root zone should not exceed 100 kPa in sand or 400 kPa in clay loams.

In this period six to eight weeks before harvest, the fruit is growing rapidly, and the tree now needs ample water to reach maximum fruit size. Water stress must not occur during this final period of fruit growth. Irrigation to replace 100 % of orchard water use capability is recommended. Soil moisture in the middle of the wetted fibrous root zone should not exceed 40 kPa in sand or 60 kPa in clay loams.

After harvest a similar strategy as during period 2 can be implemented. In early maturing varieties and species (for example, cherries and apricots) there is considerable shoot growth after harvest which should be kept in check to maintain fruitfulness and even cropping within the canopy. Irrigation to replace 30 % of orchard water use capability is recommended. Soil moisture in the middle of the wetted fibrous root zone should not exceed 100 kPa in sand or 400 kPa in clay loams.

Values for daily and weekly ET_o (mm) can be obtained from the Bureau of Meteorology. Data shown in the table are typical for the Goulburn Valley. To use the table, you merely have to replace the figures given in the example by those that you have collected in the previous week.

Orchard effective area of shade (EAS, %) is a simple and practical estimate of tree size and hence the actively transpiring leaf area in an orchard. EAS is determined from measurements of the percent shade cast by the trees at three key times a day (3½ h before solar noon, at solar noon and 3½ h after solar noon). Taking three measurements per day accounts for differences in foliage extent (i.e. training system and tree size), planting arrangement (i.e. row orientation and tree spacing) and leaf area density (i.e. pruning management). EAS is calculated from the average of the three measurements. The percent shade can be estimated visually or measured using a light bar known as a ceptometer.

The understorey coefficient (K_e) is a factor to convert ET_o to understory water use; the combination of soil evaporation and cover crop water use. For modern high-density orchards, micro-irrigation is designed to deliver water requirements to individual trees and minimize the contribution of irrigation to understorey water use. Hence under drip irrigation K_e can be set to 0.1 and under microjet set to 0.2. Whereas under sprinkler irrigation there is substantial understorey water use and K_e can be set to 1-EAS. For example, if EAS = 20 % then K_e = 1-20 % = 0.8.

The stress coefficient (K_s) is a factor used for setting the amount of stress deliberately imposed on the orchard. A value of 1.0 is no stress. For example, during period 2 under RDI it is recommended to replace 30 % of orchard water use capability, hence K_s = 0.3.

Weekly irrigation for a peach orchard (I) is calculated from weekly ET_o, effective area of shade (EAS), the understorey coefficient (K_e) and the stress coefficient (K_s) using the following formula:

I = K_s x ET­­₀ x [(1.1xEAS)+K_e]

The area of planting square (m²) is calculated from the distance between rows multiplied by the in-row distance between trees. Different spacings between trees are given for each form of irrigation in the example in Table 1.

Weekly irrigation requirement per tree is calculated from the the weekly orchard irrigation multiplied by the area of planting square.

The interval between irrigations (day) is also important with RDI, and recommended intervals are given in Table 1. For drip irrigation, the rationale behind these recommendations relates to the size of the wetted root zone. In period 2, frequent irrigation (that is, daily) wets a small volume of soil regularly. In contrast, using a two-day interval in period 1 (and daily interval in period 3) enables a much greater volume of water to wet a larger root zone. This manipulation in wetting the root zone could be responsible for the observed improved growth of fruit in period 3 and higher yields on RDI-managed trees. If, with drip irrigation, the system must be run for more than 24 hours every second day to provide the required quantity of water, serious thought should be given to upgrading the system to a higher rate of discharge.

The longer interval between irrigations in period 2, than in period 1 and 3, for both microjet and sprinkler irrigation is necessary to allow enough water to wet the soil to a reasonable depth.

In period 2, with microjet and sprinkler irrigation, an interval of seven and 21 days respectively is recommended. If the combined effects of evaporation, spacing of trees and rate of application result in less than two- and eight-hour irrigation times respectively for microjet and sprinkler irrigations, the interval will need to be extended until such figures are reached. For these long intervals, irrigation is based on the accumulated evaporation since the previous irrigation.

The quantity of water required at each irrigation is multiplied by the interval between irrigations in days and divided by 7 (that is, by the number of days in the week). For example, if the weekly irrigation requirement is 52 litre but the interval is only two days, then approximately 15 litre of water is applied every 2^nd day (52 x 2 ¸ 7 = approximately 15).

Application rate (litre/hour/tree) is the amount of irrigation applied to each tree per hour. This is calculated from the emitter discharge rate multiplied by the number of emitters per tree. If not known, this should be measured.

Run time (hour) is calculated by dividing the number of litres per tree required at each irrigation by the application rate.

To target high quality fruit, strategic irrigation management practices, geared to specific fruit growth stages are currently being investigated in a design field experiment, at DEDJTR Tatura, at the Stone fruit field laboratory. The effect of water stress on stonefruit production varies with the severity of the water deficit and the duration and timing of that water stress. Typically fruit size is reduced under water deficits with the accumulation of soluble solids. The effect of other important fruit quality attributes like maturity firmness and skin colour are less well understood. So water stress during Stage 1 for example of fruit development is known to restrict cell division, reduce fruit size and decrease yield. Water stress during Stage 2 reduces the vegetative vigour with minimal impact on yield and fruit size. Consequently Stage 2 represents that traditional RDI regulated deficit irrigation period for reduced irrigation supply.

In Stage 2 of fruit growth, fruit growth rate is relatively low relative to the vegetative growth. Water stress however in stage three, reduces cell expansion of the fruit and is likely to restrict the remobilization of assimulate in the following seasons crop. Stage 3 water stress therefore impacts negatively on yield and fruit size. So at a Tatura the irrigation experiment aims to identify combinations of irrigation levels and timing of water deficits that will enable a late season nectarine variety of September Bright to achieve maximum uniformity in yield and fruit quality. The irrigation level's percentage of tree evaporation of the control being applied using the drip irrigation are, zero, 20 percent and 40 percent of the control. Zero treatment deficit irrigation treatment is imposed to give extreme high levels of water stress. The 20 percent deficit is to impose severe levels of water stress and a 40 percent deficit rirrigation treatment is to impose moderate levels of water stress. The control, 100 percent irrigation delivery of water is crop water requirement and in this stage to maximize yield and fruit size. There's different stages and timings of these water deficits, including stage one of fruit development, stage two of development and stage 3 of fruit development.

So when we observed this graph, we can see stage one, stage 2 and stage 3 of fruit growth. In stage 1, the period from September to early November, we can see the fruit growth rate in this curve here. In to stage 2, from November to January and Stage 3 is where the rapid fruit growth occurs,post January up until harvest in March. The relative growth of Shoots is also shown in the green line over those three periods, stage 1, stage 2 and stage 3 of fruit growth. So I've detailed here the 12 irrigation treatments we're using where we have different water deficits at different growth stages throughout the irrigation season.

The control, as mentioned, receives 100 per cent of water requirements all year across all fruit growth stages and post harvest stage four. As an example, in stage one we have zero percent irrigation. We have a 20 percent irrigation treatment and in stage one we also have a 40 percent irrigation treatment. In stage two, we have similar levels. Only in stage two the traditional RDI period has 0% treatment, 20 percent stage two treatment and a 40 percent stage two treatment.

In stage 3 of fruit growth, we have actually split stage three in half, that for a month between January and February, we call that three A and in the February - March window we call that a three B phase of fruit growth where we're also apply deficit irrigation treatments.

The results from last irrigation season, season 2016-17 season, we can see, we look at the year data, and we can compare each treatment, compared to the control, and typically we had no effect in Stage 2 with the green bars here, which is the traditional RDI period due to the deficit. They tend to yield no effect in Stage 3 A (let's call it the early phase of rapid growth). Major effect in yield production in Phase 3 B, and in the early phase 1 deficit irrigation treatment there appeared to be a couple of differences which we'll cover in a minute. So when we look at stage one. This is a histogram of the distribution of fruit weight compared to the control. So the black line here represents the control irrigated treatment, the well watered treatment compared to stage 1 deficits of 0 percent water, 20 percent water and 40 percent water on fruit size.

And what we can see as we have deficit irrigated in stage one we have a reduction in fruit overall fruit size, and a shift to the left in these distribution curves.  So we may compare the same data for fruit sweetness, the same treatments, we see a very similar response in overall fruit sweetness. However we did note that there was fruit size reduction.

Moving onto the stage 2 treatments, compared to the control, we have very similar profiles for fruit size and very similar profiles in fruit sweetness, suggesting that fruit size and yield was not impacted due to stage 2 deficit irrigation, which is the traditional RDI period, and fruit outcomes, fruit sweetness outcomes were quite similar as well to the control.

When we go to stage three, we can see major changes in reductions in our fruit size profiles compared to the control, and extreme changes in both fruit size for the late season, stage three, part two (let's call it the second phase, the last phase, the last month prior to harvest) where we deficit irrigated trees that were either 0 percent or 20 percent of water. We saw a huge shift to the left in smaller fruit compared to the control, and consequently we saw an increase in their fruit size reduced and a major change in the profiles of sweetness and those two deficit irrigated treatments just prior to harvest.

There was also, when you look at the early stage 3 deficit treatments, the 3.1 Zero %, the 3.1 twenty % and 3.1 forty % treatment , we also had fruit size reductions, but we didn't see any changes in their fruit sweeteners profiles.

The irrigation experiment has shown in the first season, that we need to well water our fruit trees in stage one of fruit growth to avoid that fruit size penalty. To maximize fruit size, we need to have well watered trees in phase one, and phase 2 (the RDI period), we saw no effect on fruit size or fruit sweetness, suggesting that there is potential for water savings using RDI in stage two of fruit growth. In stage 3, we found big differences in fruit size and fruit sweetness when we set deficit irrigation close to harvest, and so withholding water did increase fruit sweetness, but it also penalized fruit size. So further work needs to be done to refine that to optimize for equality in Phase 3 of fruit growth.