新西兰果树浇水

在新西兰

两篇有价值的基础性文章：1，http://www.fao.org/docrep/004/Y3655E/y3655e10.htm

Scheduling deficit irrigation of fruit trees for optimizing water use efficiency

I. Goodwin and A.-M. Boland,
Department of Natural Resources and Environment,
Institute of Sustainable Irrigated Agriculture, Tatura, Australia

Summary

Regulated deficit irrigation (RDI) of fruit trees in the Goulburn Valley of southeastern Australia has increased water use efficiency by approximately 60 percent with no loss in yield or substantial reductions in vegetative vigour. Original techniques to schedule RDI were based on a 12.5 percent (peach) and 20 percent (pear) replacement of US Class A pan evaporation. Subsequent research into soil moisture measurement led to a recommended soil suction of 400 kPa to trigger irrigation. To extend the application of RDI to other environments and fruit crops, practical scheduling steps have been developed. Firstly, fruit growth is measured to determine when to apply RDI. Secondly, an irrigation plan is developed to estimate irrigation run time and interval based on soil type, root distribution, wetting pattern and average daily water use. Thirdly, soil moisture sensors are installed and irrigation is applied when soil suction reaches 200 kPa. Irrigation run time is adjusted by measuring soil moisture immediately following irrigation. Finally, US Class A pan evaporation is measured or reference crop evapotranspiration is calculated to estimate irrigation interval for scheduling in later years.

Regulated deficit irrigation (RDI) was developed to improve control of vegetative vigour in high-density orchards in order to optimize fruit size, fruitfulness and fruit quality. RDI is usually applied during the period of slow fruit growth when shoot growth is rapid. However, it can also be applied after harvest in early-maturing varieties. Furthermore, RDI can generate considerable water savings. Thus, it is useful for reducing excessive vegetative vigour, and also for minimizing irrigation and nutrient loss through leaching.
Increasingly, orchards are being planted with compact, closely spaced trees. Higher density improves profitability as trees bear earlier, yields are higher, and production costs are lower (Chalmers, 1986). While the benefits of high-density orchards are well known, excessive vegetative vigour in badly managed high-density orchards can lead to shading and associated barrenness (Chalmers et al., 1981). Fruitlet retention, fruit size and fruit colour can be reduced in the current season while fruit-bud formation in the following season can be inhibited (Purohit, 1989). Therefore, when full canopy cover is reached, it is critical that excessive vegetative growth minimized.
Techniques for controlling vegetative vigour include branch manipulation, mechanical shoot and root pruning, the application of chemical growth regulators, manipulating crop load, fertilizer management, and RDI (Chalmers et al., 1984). Of these, RDI is arguably the most economical, as less water is applied with no loss in fruit size or total yield. Genetic control methods such as the use of dwarfing rootstocks will control vegetative vigour for the life of an orchard and are widely used in apple production. However, vigour management based on cultural practices ensures that trees remain inherently vigorous and are capable of rapidly filling their allotted space and producing high early yields (Chalmers et al).
Extensive research means that the effects of regulated water deficits on tree growth and development are well understood. Most studies have shown that mild water stress applied during the period of slow fruit growth controlled excessive vegetative growth while maintaining or even increasing yields. These included studies on peach (Prunus persica) (Li et al., 1989; Williamson and Coston, 1990), European pear (Pyrus communis) (Brun et al., 1985a, 1985b; Chalmers et al., 1986; Mitchell et al., 1984, 1986, 1989), Asian pear (Pyrus serotina) (Caspari et al., 1994) and apple (Malus domestica) (Irving and Drost, 1987). In addition, water stress applied after harvest reduced vegetative growth of early-maturing peach trees (Larson et al., 1988; Johnson et al., 1992). RDI applied to olives over a ten-week period following pit hardening had no adverse effect on oil production (Alegra et al., 1999). Moderate levels of water stress applied to prunes (Prunus domestica), by withholding irrigation in a deep soil during stage II of fruit growth, increased return fruit bloom, crop load, and total fruit dry matter yield (Lampinen et al., 1995).
The application of RDI improves water use efficiency (WUE). Mitchell and Chalmers (1982) found WUE, expressed as yield per unit irrigation, increased from 4.9 to 8.0 t/Ml under RDI in canning peaches that yielded 48 t/ha. Similarly, Mitchell et al. (1989) found WUE increased from 12.5 to 22 t/Ml under RDI in WBC pears that yielded approximately 90 t/ha. In the Goulburn Valley in southeastern Australia these improvements in WUE would lead to water savings of 3 Ml/ha and 2 Ml/ha for peaches and pear, respectively. Even larger water savings have been reported for peaches in China (Goodwin et al., 1998). In this case, total irrigation applied was reduced from 3.0 Ml/ha to 1.4 Ml/ha without any effect on yield. Goldhamer (1999) reported water savings of 25 percent for RDI applied to olives in California, United States of America. with no yield reduction.
Increased WUE under RDI is due largely to reductions in transpiration, which might be as much as 50 percent (Boland et al., 1993b). Reduced transpiration appears attributable to partial stomatal closure. Despite reduced transpiration, measured increases in fruit osmotic potential (Jerie et al., 1989) indicate that fruit dry weight accumulation is not impaired. This also holds for Asian pear (Behboudian et al., 1994), grapefruit (Cohen and Goell, 1988) and apple (Failla et al., 1992), and is thought to be a mechanism of adaptation to water stress (Mitchell et al., 1994).
Both the timing and level of water stress are critical to the success of RDI. These factors need to be considered in relation to what is understood of the growth and development of the species in question. In addition, it is necessary to adopt modern techniques for scheduling irrigation that allow adequate assessment of water stress in any environment. This paper describes how to determine the timing and frequency of RDI, and it presents practical scheduling techniques for estimating water application rates.
Timing of RDIThe development of RDI was not possible without first understanding patterns of tree and fruit growth. Initially, RDI experiments focused on peach and pear, and a comparison of the development of these fruits illustrates the importance of the timing of RDI application. Although patterns of growth and development may vary in other horticultural crops, the basic principle of applying RDI when fruit growth is minimal remains the same.
The growth curve of peach is double-sigmoidal with two periods of increasing growth rate. Three phases are commonly attributed to fruit growth. Stages I and III are separated by a phase of decreasing growth rate (Stage II) known as the lag phase (Chalmers and van den Ende, 1975, 1977). Changes in the relative sink strengths of the seed and pericarp govern development. Only 25 percent of total fruit growth occurs when vegetative parts are growing rapidly; the majority of fruit growth occurs in the final 6-8 weeks before harvest when vegetative growth is almost complete (Chalmers et al., 1975, 1984) (Figure 1a). This asynchronous growth of fruit and shoots reduces competition for resources at critical stages, and provides a sound basis for the application of the RDI, which relies on water stress during Stage II having a small effect on fruit growth but a significant effect on vegetative growth.
The growth of pear fruit is curvilinear with less than 20 percent occurring by midway from bloom to harvest (Mitchell, 1986). The majority of shoot growth occurs during this period of slow fruit growth (Mitchell et al., 1986). Thus, RDI is applied for the first 70-80 days after bloom. The majority of fruit growth occurs in the remaining 6-8 weeks to harvest (Figure 1b).

Figure 1
Typical shoot and fruit growth pattern for (a) peach and (b) European pear

The above generic descriptions of fruit and shoot growth of peach and pear are useful for explaining the theoretical basis for RDI and the general timing of RDI. However, to implement RDI for a particular variety requires a more accurate description of the growth periods. Stages of fruit growth for different fruit varieties can be readily determined by tagging several fruit and shoots on a tree and making weekly determinations of their circumference (or diameter) and length with a tape measure. Fruit circumference can be converted to relative volume by cubing.
Scheduling RDI -- historyUnderstanding of when and how to apply RDI has improved substantially over the past 20 years. Scheduling has evolved from the initial recommendations based on US Class A pan evaporation (Epan) toward measuring both soil moisture and tree responses before making management decisions. Although the original simple recommendations may still work for many orchards, the emphasis on measuring soil moisture to estimate orchard water use and tree water stress allows more precise control over vegetative vigour and fruit growth.
Under trickle irrigation, the original recommendation for scheduling RDI was to irrigate daily and calculate irrigation amount from a percent replacement of Epan. The formula used to calculate irrigation run time was:

Replacement amounts were derived from the original RDI experiments at Tatura (Mitchell et al., 1989). For peaches, the recommended replacement was 12.5 percent from flowering until the start of rapid fruit growth. From the start of rapid fruit growth to harvest, the recommended replacement was 100 percent. The start of rapid fruit growth was based on a date for different varieties, e.g. Golden Queen was mid-January. With William Bon Chretien (WBC) pears the strategy was slightly different, consisting of a period of withholding irrigation during spring until attaining a cumulative deficit of 100-125 mm of evaporation from 1 October. After this, a replacement of 20 percent Epan was used until mid-December to calculate required irrigation application. From mid-December to harvest, the recommended replacement was 100-120 percent for pears.
Adapting these recommendations to fit other irrigation systems concentrated on altering the interval between irrigations. During the RDI period, the recommended intervals were 7 days for microjets (40 litres/h/tree in 3x5 m planting) and 21 days for sprinklers (120 litres/h/tree in 6x6 m planting) (Goodwin, 1995). Applying RDI using flood irrigation was based on increasing the interval between irrigations or irrigating every second row.
The next improvement was to estimate irrigation interval for systems other than trickle. Estimates were based on the volume of water in the rootzone and average daily water use, and utilized the measurement of soil moisture to adjust the interval. Calculation of run time was essentially unchanged, although soilmoisture measurements following irrigation were recommended to adjust run time. Mitchell and Goodwin (1996) recommended a formula to calculate interval based on average daily pan evaporation:

Where:
volume of water in rootzone (litres) = width of wetted strip (m) x tree spacing (m) x 0.3 m wetting depth (m) x soil type factor ranging from 60 (sandy soils) to 80 (loams and clays) average daily water use (litres/day) = row spacing (m) x tree spacing (m) x replacement factor x average daily Epan (mm).
This method of scheduling remains well suited to the Goulburn Valley. However, it is not applicable to other soil types and climates. RDI experiments in China on peaches, with root systems up to 2.5 m deep, emphasized the need to measure soil moisture over the entire rootzone depth to trigger the initial irrigation in spring or early summer (Goodwin et al., 1998).
In conjunction with the above formulae to estimate run time and interval based on pan evaporation, recommendations to measure soil moisture were developed to ensure soil dryness was sufficient but not excessive. Measurements of rootzone soil moisture were included in the scheduling of RDI to adjust irrigation interval and run time. Recommendations were based on intensive soil suction monitoring with gypsum blocks in an RDI experiment on pears at Tatura (Goodwin et al., 1992). Under trickle irrigation, soil suction of 400 kPa at 0.1-0.25 m depth, 0.15 m from the emitter, was recommended to trigger irrigations with irrigation run time based on the above formula. Soil moisture measurements after irrigation at 0.6 m from the tree line were recommended to adjust irrigation run time.
Work undertaken on RDI of wine grapes across a range of climates and soil types (Goodwin and Jerie, 1992) highlighted the need for adjustments in soil moisture values to trigger irrigation depending on rootzone depth, soil texture and climate. Recommendations for wine grapes were as follows. In sandy soils with shallow rootzones (<0.4 m) and hot climates (e.g. average January daily evaporation >8 mm), soil suction under RDI should not exceed 100 kPa. In loam soil with intermediate rootzones (0.4-0.8 m) and mild climates (e.g. average January daily evaporation 5-8 mm), soil suction under RDI should not exceed 200 kPa. In clay soil with deep rootzones (>0.8 m) and cool climates (e.g. average January daily evaporation <5 mm), soil suction under RDI should not exceed 400 kPa.
Scheduling RDI -- current recommendationThe following is a list of necessary steps implementing RDI successfully:

• Measure fruit and shoot growth to determine the RDI period for fruit species/varieties in an orchard.
• Dig up a tree to determine the rootzone distribution -- width and depth (80 percent of total).
• Determine the wetting pattern of the irrigation system and estimate wetted rootzone.
• Develop a season irrigation plan for run time and interval based on soil type and average Epan or reference crop evapotranspiration (ETo).
• Install soil moisture sensors (preferred measure is soil suction using gypsum blocks)
  
  
  - at 0.3 m and bottom of rootzone in shallow soil,
  - at 0.3 m, 0.6 m and bottom of rootzone in deep soil.
  

During RDI period

• Measure and record soil suction and irrigate when the entire rootzone dries out to a minimum of 200 kPa.
• Irrigate to wet the top 0.3 m of the root zone.
• Measure and record soil moisture 6-12 h after irrigation and, where necessary, adjust the amount applied in previous irrigations.
• Irrigate when the wetted rootzone soil at 0.3 m depth dries out to 200 kPa.
• Measure evaporation (or ETo) interval between irrigations -- irrigate in future years based on this evaporation interval.
• Repeat steps 3-6.
  

During rapid fruit growth

• Irrigate to wet at least the top 0.6 m of rootzone.
• Measure and record soil suction 6-12 h after irrigation, and, if the soil is dryer than 30 kPa (sandy soil) or 50 kPa (clay soil) at 0.6 m, apply more irrigation.
• Irrigate when the wetted rootzone soil suction at 0.3 m depth dries out to 30 or 50 kPa.
• Measure evaporation (or ETo) interval between irrigations -- irrigate in future years based on this evaporation interval.
• Repeat steps 2-5.
  

Measuring shoot and fruit growthAn understanding of the changes in fruit and shoot growth for different varieties is critical for the timing of RDI. Water stress should be applied only during the vegetative growth period when fruit is growing slowly. Water stress must be avoided or minimized (where water is limited) during rapid fruit growth. The stages of fruit growth for a given variety can be determined by tagging several fruit and shoots and weekly measuring their circumference and length with a tape measure. Converting fruit circumference to volume [volume = 0.02 x (circumference)3] gives a true indication of fruit weight. This technique is simple and the measurements are useful for adjusting irrigations, especially where shoot growth continues despite high soil water deficits.
Root distributionRoot distribution is an important component for RDI scheduling because of thepotential store of available moisture in the soil. The best method for determining root distribution is to dig a pit next to an orchard tree and estimate the amount of roots in 0.2-m depth increments until the bottom of the rootzone (80 percent of roots). Root depth is important for determining the volume of water in the rootzone when the profile is wet from rainfall, and for deciding where to site soil moisture sensors.
Wetted root zoneIt is critical to determine the volume of the wetted rootzone. This can be estimated from the root distribution and the wetted volume of soil. To determine the wetting volume, it is necessary to observe the wetted surface area and depth following an irrigation event.
A hole is dug to observe wetting at depth. The wetted rootzone is then estimated from the volume of roots that are wet following irrigation. The calculation in the following irrigation plan assumes that the wetting pattern is a continuous strip of soil with a wetting depth of 0.3 m. This wetted strip pattern will occur with closely spaced microjets or drippers where the wetting pattern overlaps. For other irrigation systems where the wetting patterns are separate, the wetted rootzone is calculated assuming the shape of a cylinder.
Irrigation planThe aim of setting out a season irrigation plan for the approximate interval and run time is to provide a theoretical basis for irrigation scheduling and water budgeting. For each month of a growing season, the interval between irrigations is calculated based on the equation:

At the start of the season, the interval between irrigations is equivalent to the withholding irrigation period where the volume of water in the rootzone (i.e. stored soil moisture) can be calculated by substituting the wetted volume with the root volume:
Volume of water in rootzone (litres/tree) = Lateral root distribution width (m) x Tree spacing (m) x Root depth (m) x Deficit available water ranging from 9 percent (sandy soils) to 13 percent (loams and clays) x 1 000.
Once irrigation commences, the volume of water in the root zone is equivalent to the irrigation amount to be applied:

Volume of water in rootzone (i.e. irrigation amount) (litres/tree) = Width of wetted strip (m) x Tree spacing (m) x 0.3 m wetting depth (m) x Deficit available water ranging from 9 percent (sandy soils) to 13 percent (loams and clays) x 1 000

Run time calculations use the emitter rate per tree and the system irrigation efficiency:

To estimate average daily water use, the plan uses local long-term average USA Class A pan evaporation data and appropriate crop factors for RDI (Mitchell and Goodwin, 1996). Alternatively, it is possible to use ETo and crop coefficients (Kc) (Allen et al., 1998) and appropriate percent replacements for RDI to estimate daily water use.
Soil moistureRDI scheduling requires measurements of soil moisture. In shallow rootzones, soil moisture is measured at two depths (Figure 2). In deep rootzones (>0.6 m), soil moisture is measured at three depths. The aim is to dry out the soil throughout the rootzone to a minimum suction of 200 kPa by withholding irrigation (positions A, B and C). If there is no rain, the soil in the upper rootzone (positions A and B) will become much drier than the soil towards the bottom of the rootzone (position C). If the entire rootzone becomes drier than 200 kPa, stress levels on the tree will cause loss in productivity. Irrigation is necessary.

Figure 2
Position of soil moisture sensors in (a) shallow and (b) deep rootzone soils

Once irrigation commences, the objective is to maintain a moderate level of stress on the trees. This is best achieved by irrigating with less water than the usual full recommendation. Irrigations should aim to wet to 0.3 m depth (position A).
It is necessary to measure soil moisture 6-12 h after irrigation to adjust the amount of water applied in proceeding irrigations. If the soil in the top rootzone (position A) remains dry then the irrigation amount must be increased. If the soil in the mid-rootzone (position B) becomes wet immediately following irrigation then irrigation amount must be cut back.
The gypsum block is preferred over other methods of determining moisture because it measures soil water suction, which relates to the level of water stress on the trees. It is the only instrument capable of measuring soil suction in the range suitable for RDI. It is relatively inexpensive, robust to handle, and simple to install. It requires a portable hand-held meter to measure the resistance between the two electrodes embedded in the block of gypsum. The electronics in the meter convert the resistance automatically to suction. The measurement is simple: requiring the connection of the two wires to the meter and a button to be pushed to directly measure soil suction.
Alternatively, soil samples may be collected with an auger and the moisture content assessed. This is much less accurate than the gypsum block method, but may be useful to assess wetted depth and moisture below the top 0.05 m depth.
RDI in practiceAs part of an extension programme in the Goulburn Valley, sites were established on growers properties to demonstrate RDI. Growers were interested in controlling vegetative vigour in high-density orchards and saving water. One site consisted of 6-year-old Golden Queen peach trees on Tatura Trellis (van den Ende et al., 1987) irrigated with 45 litres/h microjets (one every second tree). Thirty trees (three rows each of ten trees) received normal irrigation and 30 (also three rows of ten) received the deficit irrigation. Measurements recorded to indicate WUE and vigour control included water applied, soil moisture (tensiometers and gypsum blocks), butt diameter and fruit growth (mm).
RDI was applied from the first week of November to the last week of December, to provide approximately 40 percent of evaporation; control trees received full irrigation. Soil suction was maintained between 0 and 65 kPa on the control treatment and between 0 and 200 kPa on the RDI treatment. For the remainder of the season, soil suction was maintained between 0 and 50 kPa on all of the trees.
Fruit growth was measured over the season (four fruits per tree, 120 per treatment). There was no apparent difference in fruit size between the RDI-treated trees and the controls. Tree butt size was used as an indicator of vigour. The 30 trees irrigated under the RDI strategy exhibited an overall reduction in measured butt diameter at the end of the season. The grower also noted a reduction in tree vigour, with more fruiting wood established. There was a reduction in the water applied under RDI management with a saving of 2.3 Ml/ha: total irrigation for the control was 7.9 Ml/ha, whereas that for the RDI treatment was 5.6 Ml/ha.
The demonstration site showed that RDI can generate considerable savings. Fruit size and yield were maintained, and vegetative vigour appeared to be reduced.
Other issues related to RDI -- root volume and salinityIt is evident that root volume is an important factor in the tree growth response to RDI. Some studies have suggested that the success of RDI in controlling vigour and maintaining yield arises from both an adaptation to moderate water stress developed in a shallow soil volume (Jerie et al., 1989) and/or restricted wetted root volume (Richards and Rowe, 1977a, 1977b).
To further explore this effect, an experiment was established to determine the interaction of RDI and root volume on Golden Queen peaches (Boland et al., 1994, 2000a, 2000b). This study demonstrated that the effect of root volume was independent of the RDI water stress response. However, there are important implications for the practical application of RDI under various conditions. In the Goulburn Valley, shallow root volume assists the development of water stress under RDI. In a deep soil with an unrestricted root system, it takes considerably longer to develop water stress; under these conditions it may be necessary to physically restrict the volume of roots.
Therefore, the control of vegetative growth and establishment of RDI depends on the interaction between rainfall/evaporation, available soil volume for root exploration and the readily available water (time taken to develop water stress).
The application of RDI in a saline environment presents potential advantages and disadvantages. Management of orchards irrigated with saline water has traditionally relied on leaching to prevent accumulation of salts, in order to maintain a soil volume that will permit root development. Leaching is regarded as the key to salinity control (Hoffman and van Genutchen, 1983). Although, RDI does not provide the same degree of leaching, it does have the potential to improve salinity management, firstly by a reduction in the importation of salt, and secondly by control of the rising water table (Shalhavet, 1994).
An experiment that assessed the impact of saline irrigation when applying RDI (Boland et al., 1993a) demonstrated significant adverse effects on the productivity of peach trees, with similar results expected on other fruit trees that are generally sensitive to salinity. Therefore, while RDI may lessen the volume of drainage and applied salts, the detrimental effects on productivity would generally outweigh these benefits. Where RDI is applied in a saline environment to either save water or control vegetative vigour, it is necessary to adopt specific management strategies: strategic leaching irrigations (e.g. every five to seven irrigations), and careful monitoring of soil salinity.
ConclusionAlthough the control of vegetative vigour in high-density orchards was the original objective of RDI, increased WUE has become a critical issue in areas where water scarcity is a problem. RDI is an ideal water saving technique. Its application and adaptation in various environments have led to improved understanding of the process, the benefits, and the requirements for adoption. Scheduling has evolved to include weather and soil-based monitoring. As a consequence, this wealth of knowledge has enabled the implementation of a practical and achievable programme for grower adoption of RDI.
References

Alegra, S., Girona, J., Marsal, J., Arbones, A., Mata, M., Montagut, D., Teixido, F., Motilva, M.J. & Romero, M.P. 1999. Regulated deficit irrigation in olive trees. Acta Horticulturae 474: 373-375.

Behboudian, M.H., Lawes, G.S. & Griffiths, K.M. 1994. The influence of water deficit on water relations, photosynthesis and fruit growth in Asian pear (Pyrus serotina Rehd.). Scientia Horticulturae 60: 89-99.

Boland, A.-M., Jerie, P.H., Mitchell, P.D., Goodwin, I. & Connor, D.J. 2000a. Long-term effects of restricted root volume and regulated deficit irrigation on peach I. Growth and mineral nutrition. Journal of the American Society of Horticultural Science 125: 135-142.

Boland, A.-M., Jerie, P.H., Mitchell, P.D., Goodwin, I. & Connor, D.J. 2000b. Long-term effects of restricted root volume and regulated deficit irrigation on peach II. Productivity and water use. Journal of the American Society of Horticultural Science 125: 143-148.

Boland, A.-M., Mitchell, P.D., Goodwin, I. & Jerie, P.H. 1994. The effect of soil volume on young peach tree growth and water use. Journal of the American Society of Horticultural Science 119: 1157-62.

Boland, A.-M., Mitchell, P.D. & Jerie, P.H. 1993a. Effect of saline water combined with restricted irrigation on peach tree growth and water use. Australian Journal of Agricultural Research 44: 799-816.

Boland, A.-M., Mitchell, P.D., Jerie, P.H. & Goodwin, I. 1993b. The effect of regulated deficit irrigation on tree water use and growth of peach. Journal of Horticultural Science 68: 261-274.

Brun, C., Raese, J.T. & Stahly, E.A. 1985a. Seasonal responses of `Anjou' pear trees to different irrigation regimes. I. Soil moisture, water relations, tree and fruit growth. Journal of the American Society of Horticultural Science 110: 830-834.

Brun, C., Raese, J.T. & Stahly, E.A. 1985b. Seasonal responses of `Anjou' pear trees to different irrigation regimes. I. Mineral composition of fruit and leaves, fruit disorders and fruit set. Journal of the American Society of Horticultural Science 110: 835-840.

Caspari, H.W., Behboudian M.H. & Chalmers, D.J. 1994. Water use, growth and fruit yield of `Hosui' Asian pears under deficit irrigation. Journal of the American Society of Horticultural Science 119: 383-388.

Chalmers, D.J. 1986. Research and progress in cultural systems and management in temperate fruit orchards. Acta Horticultura 175: 215-225.

Chalmers, D.J., Burge, G., Jerie, P.H. & Mitchell, P.D. 1986. The mechanism of regulation of `Bartlett' pear fruit and vegetative growth by irrigation withholding and regulated deficit irrigation. Journal of the American Society of Horticultural Science 111: 904-907.

Chalmers, D.J., Canterford, R.L., Jerie, P.H., Jones, T.R. & Ugalde, T.D. 1975. Photosynthesis in relation to growth and distribution of fruit in peach trees. Australian Journal of Plant Physiology 2: 635-645.

Chalmers, D.J. & van den Ende, B. 1977. The relationship between seed and fruit development in peach. Annals of Botany 41: 707-714.

Chalmers, D.J. & van den Ende, B. 1975. A reappraisal of the growth and development of peach fruit. Australian Journal of Plant Physiology 2: 623-634.

Chalmers, D.J., Mitchell, P.D. & van Heek, L. 1981. Control of peach tree growth and productivity by regulated water supply, tree density and summer pruning. Journal of the American Society of Horticultural Science 106: 307-12.

Chalmers, D.J., Mitchell, P.D. & Jerie, P.H. 1985. The relationship between irrigation, growth and productivity of peach trees. Acta Horticulturae 173: 283-288.

Chalmers, D.J., Mitchell, P.D. & Jerie, P.H. 1984. The physiology of growth control of peach and pear trees using reduced irrigation. Acta Horticulturae 146: 143-149.

Cohen, A. & Goell, A. 1988. Fruit growth and dry matter accumulation in grapefruit during periods of water with holding and after reirrigation. Australian Journal of Plant. Physiology 15: 633-639.

van den Ende, B., Chalmers, D.J. & Jerie, P.H. 1987. Latest developments in training and management of fruit crops on Tatura Trellis. Hortscience 22: 561-568.

Failla, O., Zocchi, G., Treccani, C. & Cocucci, S. 1992. Growth, development and mineral content of apple fruit in different water status conditions. Journal of Horticultural Science 67: 265-271.

FAO. 1998. Crop evapotranspiration - guidelines for computing crop water requirements by R.G. Allen, L.S. Pereira, D. Raes and M. Smith. FAO Irrigation and Drainage Paper 56. Rome.

Goldhamer, D.A. 1999. Regulated deficit irrigation for California canning olives. Acta Horticulturae 474: 373-375.

Goodwin, I. 1995. Irrigation scheduling for regulated deficit irrigation (RDI). Agricultural Notes #AG0299. Melbourne, Australia. Department of Natural Resources and Environment.

Goodwin, I. & Jerie, P.H. 1992. Regulated deficit irrigation: From concept to practice. Aust. and N.Z. Wine Industry Journal 4: 258-61.

Goodwin, I., Jerie, P.H. & Boland, A.-M. 1998. Water saving techniques for orchards in Northern China. In: Water is Gold, Irrigation Association of Australia 1998 National Conference and Exhibition, Brisbane, 19-21 May 1998. Brisbane, Australia, Department of Natural Resources.

Goodwin, I., Mitchell, P.D. & Jerie, P.H. 1992. Measurement of soil matric potential under William Bon Chretien pear comparing regulated deficit with normal irrigation. Australian Journal of Experimental Agriculture 32: 487-92.

Hoffman, G.J. & van Genuchten, M.Th. 1983. Soil properties and efficient water use: water management for salinity control. In: Limitations to efficient water use in crop production, H.M. Taylor, W.R. Jordan & T.R. Sinclair eds. Madison, Wisconsin, United States of America, American Society of Agronomy.

Irving, D.E. & Drost, J.H. 1987. Effects of water deficit on vegetative growth, fruit growth and fruit quality in Cox's Orange Pippin Apple. Journal of Horticultural Science 62: 427-432.

Jerie, P.H., Mitchell, P.D. & Goodwin, I. 1989. Growth of William Bon Chretien pear fruit under Regulated Deficit Irrigation. Acta Horticulturae 240: 271-274.

Johnson, R.S., Handley, D.F. & DeJong, T.M. 1992. Long-term response of early maturing peach trees to postharvest water deficits. Journal of the American Society of Horticultural Science 117: 881-886.

Lampinen, B.D., Shackel, K.A., Southwick, S.M., Olson, B., Yeager, J.T. & Goldhamer, D. 1995. Sensitivity of yield and fruit quality of French prune to water deprivation at different fruit growth stages. Journal of the American Society of Horticultural Science 120: 139-147.

Larson, K.D., DeJong, T.M. & Johnson, R.S. 1988. Physiological and growth responses of mature peach trees to postharvest water stress. Journal of the American Society of Horticultural Science 113: 296-300.

Li, S-H., Huguet, J-G., Schoch, P.G. & Orlando, P. 1989. Response of peach tree growth and cropping to soil water deficit at various phenological stages of fruit development. Journal of Horticultural Science 64: 541-552.

Mitchell, P.D. 1986. Pear fruit growth and the use of diameter to estimate fruit volume and weight. Hortscience. 21: 1003-1005.

Mitchell, P.D. & Chalmers, D.J. 1982. The effect of reduced water supply on peach tree growth and yields. Journal of the American Society of Horticultural Science 107: 853-56.

Mitchell, P.D., Chalmers, D.J., Jerie, P.H. & Burge, G. 1986. The use of initial withholding of irrigation and tree spacing to enhance the effect of regulated deficit irrigation on pear trees. Journal of the American Society of Horticultural Science 111: 858-861.

Mitchell, P.D., van den Ende, B., Jerie, P.H. & Chalmers, D.J. 1989. Response of "Bartlett" pear to withholding irrigation, regulated deficit irrigation, and tree spacing. Journal of the American Society of Horticultural Science 114: 15-19.

Mitchell, P.D. & Goodwin, I. 1996. Micro-irrigation of vines and fruit trees. Melbourne, Australia, Agmedia.

Mitchell, P.D., Goodwin, I. & Jerie, P.H. 1994. Pear and quince. In: B. Schaffer and P. C. Anderson eds. Handbook of environmental physiology of fruit crops Volume 1 Temperate crops. Boca Raton, Florida, United States of America, CRC Press, Inc.

Mitchell, P. D., Jerie, P. H. & Chalmers, D. J. 1984. The effect of regulated water deficits on pear tree growth, flowering, fruit growth and yield. Journal of the American Society of Horticultural Science 109: 15-19.

Purohit, A.G. 1989. High density planting of fruit tree - a review II. Control of vegetative growth and tree size and deciding critical space. Journal of Maharashtra Agricultural University 14: 133-136.

Richards, D. & Rowe, R.N. 1977a. Root-shoot interactions in peach: The function of the root. Annals of Botany 41: 1211-1216.

Richards, D. & Rowe, R.N. 1977b. Effects of the root restriction, root pruning and 6-benzylamino-purine on the growth of peach seedlings. Annals of Botany 41: 729-740.

Shalhavet, J. 1994. Using water of marginal quality for crop production: major issues. Agricultural Water Management 25: 233-269.

Williamson, J.G. & Coston, D.C. 1990. Planting method and irrigation rate influence vegetative and reproductive growth of peach planted at high density. Journal of the American Society of Horticultural Science 115: 207-212.

评论
2， http://hortsci.ashspublications.org/content/35/6/1048.full.pdf

Using Water Stress to Control Vegetative Growth and
Productivity of Temperate Fruit TreesR. Scott Johnson
1
and Dale F. Handley
Department of Pomology, University of California, Kearney Agricultural Center, 9240 S. Riverbend Avenue,
Parlier, CA 93648

评论
这是果树浇水吗，简直就是果树上太空了，博士论文都出来了

评论
这两篇文章讲的是果树进入结果期后的浇水问题。摘录供有兴趣者参阅。

评论

这是新西兰本地讨论果树浇水基本上都会涉及的基础性文章。是基础性问题。

评论

第二篇文章没法整体copy，以下是文字部分：

Water stress almost always reduces yield in agricultural crops.
Numerous studies have demonstrated the negative impact of water
stress on various plant processes (Bradford and Hsiao, 1982). A close
correlation exists between water use or evapotranspiration and final
yield in annual crops (Hanks, 1983), so stress should be avoided at all
stages of development. However, for mature fruit trees, this relation-
ship may not hold in many situations, thus providing the opportunity
for saving water without reducing production.
The perennial nature of fruit trees helps provide this opportunity.
Because a mature tree already has grown a supporting structure of
scaffolds and branches, it can rapidly grow a full canopy of leaves in
the spring to maximize light interception and provide renewal fruiting
wood for the following year. Further vegetative growth does not
substantially increase total light interception and is generally undesir-
able growth that must be pruned out later. Therefore, imposing
moderate water stress to reduce vegetative growth may have no
negative effect on total photosynthate production and continued produc-
tivity. In fact, even beneficial results may be obtained because of less
shading of lower and interior fruiting wood, better light distribution
throughout the tree, and reduced need for summer and dormant pruning.
Reduced canopy development (shoot growth and leaf expansion)
is one of the earliest responses to water stress, occurring before
stomatal closure and reduction of photosynthesis (Bradford and Hsiao,
1982). Theoretically, this should allow for greater availability of
carbohydrates for the growth of other organs. Indeed, increased root
growth has been demonstrated under moderate water stress (Sharp and
Davies, 1975). In fruit trees the goal is to divert these carbohydrates
into fruit growth. Since fruit growth tends to dominate over vegetative
growth (Higgs and Jones, 1991; Lenz, 1967), we have hypothesized
that moderate water stress will not reduce fruit growth and may even
promote it (Fig. 1). The idea of purposely imposing moderate stress to
achieve certain beneficial results has generally been termed regulated
deficit irrigation (RDI) (Behboudian and Mills, 1997). Of course,
moderate, beneficial stress can quickly turn into severe, harmful stress
(point A in Fig. 1) and may depend on many factors, such as variety,
environmental conditions, and timing of the stress. This emphasizes
the need to accurately measure stress and to thoroughly understand the
processes occurring in the tree as moderate stress develops.
Timing of RDI is critical, especially in temperate fruit trees in
which growth of the various organs tends to occur at different times of
the season (Chalmers et al., 1985). We have hypothesized that temper-
ate fruit trees can tolerate or benefit from moderate water stress during
two periods. First, the period after harvest should have potential
because no fruit is present and the concern is with proper development
of fruit buds for next year’s crop. With early maturing varieties, this
period can be quite long, so the potential for water savings is substan-
tial. The second period is during the lag phase (stage II) of peach
[
Prunus persica
(L.) Batsch] fruit growth (Lilleland, 1933). Stone fruit
typically follow a double sigmoid pattern of growth. Rapid periods of
growth after bloom and before harvest are separated by a slower lag
phase of growth. Since fruit diameter is increasing more slowly, water
stress should have only a minor effect on fruit growth but a major
impact on vegetative growth. During the subsequent period of rapid
fruit growth (stage III) carbohydrates should be diverted from vegeta-
tive growth into fruit growth.


POSTHARVEST WATER STRESS

Larson et al. (1988), working with a June-harvested peach cultivar
in California reported no loss of production or fruit size when RDI
treatments were imposed between mid-June and mid-October in a
flood-irrigated orchard. Continuing this same experiment for 4 years,
Johnson et al. (1992) demonstrated continued productivity, reduced
vegetative growth, substantial savings of water and no indication of
decreased tree health or vigor. The main drawback to the RDI treat-
ment was an increase in fruit doubles (Fig. 2A). A single postharvest
irrigation in early August effectively reduced the level of this disorder
but still provided substantial water savings. Subsequent work with a
May-harvested peach cultivar using microsprinkler irrigation docu-
mented another fruit quality disorder, termed “deep suture” (Fig. 2B),
associated with late-season water stress (Handley, 1991; Handley and
Johnson, 2000). Alleviation of water stress by irrigating between early
August and early September reduced the incidence of both doubles and
deep sutures to the level of the fully irrigated control (Table 1).
Microscopic examination of developing fruit buds showed that carpel
differentiation was occurring in late August and early September.
Therefore, the practice of reducing or eliminating irrigation after
harvest of early maturing peach cultivars appears to be a feasible
means of saving water without reducing production, as long as stress
is alleviated during the period of carpel differentiation.
Similar treatments were imposed on an early-maturing plum (
P.
salicina
L.) cultivar irrigated with foggers (Johnson et al., 1994). Even
under extensive water stress, no double fruit or deep sutures were
induced. However, completely cutting off irrigation led to partial
defoliation within a few weeks and subsequent loss of yield. In trees
that were irrigated daily, but at half the rate of the fully irrigated
control, no reduction in yield or fruit quality occurred over a 3-year
period.
Other researchers have reported negative effects of postharvest
water stress on apricot (
P. armeniaca
L.) fruit (Brown, 1953; Uriu,
1964). This could be due to the severity of stress but may also reflect
differences in species. Additional research is needed to determine how
well different species under different conditions can tolerate posthar-
vest stress

WATER STRESS DURING FRUIT GROWTH

It has been hypothesized that water stress imposed during the lag
phase of stone fruit growth will have a much greater effect on reducing
vegetative growth than reproductive growth. If full irrigation (or
greater) is restored during the final, rapid phase of fruit growth,
reduced competition from vegetative growth should allow final fruit
size to be equal to or greater than that of fully irrigated controls. Studies
on peach trees in Australia during the early 1980s supported this
hypothesis (Chalmers et al., 1981, 1984; Mitchell and Chalmers,
1982). Increases in fruit size and yields of up to 30% were reported.
The same researchers subsequently carried out similar experiments on
pear (
Pyrus communis
L.), although pears do not exhibit a lag phase
(Chalmers et al., 1985), and reported similar results (Mitchell et al.,
1984, 1986, 1989). Again, RDI produced yields equal to or 20%
greater than that of the controls. The authors proposed that fruit
osmoregulation accounted for the stimulation of fruit growth upon
resuming irrigation (Chalmers et al., 1986).
Other researchers in several countries have tried to replicate this
technique, but have generally been less successful. Some have re-
ported negative effects of RDI on fruit size in apple (
Malus domestica
Borkh.) (Ebel et al., 1993) and peach (Girona, 1989); others have
reported no significant reduction in fruit size of apple (Ebel et al., 1995;
Irving and Drost, 1987), Asian pear (Behboudian and Lawes, 1994;
Caspari et al., 1994), and peach (Li et al., 1989; Strabbioli, 1992), but
none have shown the substantial increase in fruit size reported from
Australia. After 5 years of RDI treatments on peach in California, we
observed an increase in fruit weight (8%) in only the fifth year of the
experiment (R.S. Johnson, unpublished data). Even though substantial
increases in yield and/or fruit size were not observed in these experi-
ments, the effects can still be considered beneficial, given the substan-
tial savings of water with no significant loss in productivity.
Why does RDI appear to work in Australia but not as well
elsewhere in the world? First, stress may need to be applied before the
lag phase of fruit growth. Li et al. (1989) imposed stress treatments
during different periods of peach fruit growth. They found no change
in fruit size by stressing during the lag phase of growth, but a
significant increase by imposing stress during the first phase of rapid
fruit growth. Furthermore, the Australian workers often withheld
water during early fruit growth, which may have contributed to the
success of their experiments. Perhaps the emphasis should be on
imposing stress early enough in the season so vegetative growth can be
substantially reduced.
Second, the degree and duration of stress may also be important
factors. The experiments in Australia were conducted on a shallow
soil. Theoretically, trees in this condition could be put into and brought
out of stress more quickly than those in a deeper soil. In a deeper soil
there may not be sufficient time to develop the stress necessary to
substantially inhibit vegetative growth early in the season. Because
water stress during the last rapid phase of fruit growth invariably
reduces fruit size (Behboudian and Lawes, 1994; Caspari et al., 1994),
the stress must be alleviated quickly. Again, this can be a problem on
deeper soils, especially if penetration of water into the soil is slow
(Girona et al., 1993). Further research is needed to fine-tune the timing,
duration and degree of stress needed to obtain consistent results from
RDI treatments.

FUTURE RESEARCH

Many studies have shown that moderate water stress can provide
beneficial effects in temperate fruit trees. The key to obtaining consis-
tent and predictable results will be a better understanding of how water
stress affects physiological processes in the plant. For example,
Chalmers et al. (1986) hypothesized that RDI causes osmotic adjust-
ment in the fruit, which can then increase growth rates. What are the
conditions that maximize osmotic adjustment in the fruit? The timing
of stress is obviously important (Behboudian and Lawes, 1994; Mills
et al., 1996) but the duration, degree and rate of stress development
could also be determining factors. Furthermore, there may be signifi-
cant interactions with other factors, such as crop load (Berman and
DeJong, 1996).
Chalmers et al. (1984) suggested that root growth is the variable
controlling vegetative vigor and fruitfulness under RDI treatments.
What is the underlying physiological mechanism for this? Does a tree
maintain a constant root-to-shoot ratio (Richards and Rowe, 1977),
and how is this ratio affected by stress, planting density, irrigation
type, pruning, soil type, etc.?
The better we understand how stress affects various physiological
processes within the tree, the more likely we will be able to exploit its
beneficial effects under various environmental, soil and cultural
conditions. Much has still to be learned about how moderate water
stress affects vegetative growth and productivity in fruit trees. How-
ever, over the last two decades research has clearly shown that stress
can be imposed in ways to reduce vegetative growth, yet maintain
productivity, while saving significant amounts of water.



评论
所以，结论是？        

评论

可能看第二篇文章更方便些，
结论：：The better we understand how stress affects various physiological
processes within the tree, the more likely we will be able to exploit its
beneficial effects under various environmental, soil and cultural
conditions. Much has still to be learned about how moderate water
stress affects vegetative growth and productivity in fruit trees. How-
ever, over the last two decades research has clearly shown that stress
can be imposed in ways to reduce vegetative growth, yet maintain
productivity, while saving significant amounts of water.