General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: research.competitions@defra.gsi.gov.uk

SID 5 Research Project Final Report

SID 5 (Rev. 3/06) Page 1 of 29

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code HH3609STX

2. Project title

Partial rootzone drying: delivering water saving and sustained high quality yield into horticulture

3. Contractororganisation(s)

Lancaster UniversityEast Malling ResearchDundee University               

54. Total Defra project costs £ 1,117,553.00(agreed fixed price)

5. Project: start date................ 01 April 2004

end date................. 31 March 2009

SID 5 (Rev. 3/06) Page 2 of 29

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.

Against a background of increasingly limited water resources, there is a need to develop efficient horticultural production systems in which water and nutrients are utilised effectively in line with principles of sustainable development. Defra Policy Area WU01 addresses the need to optimise water use by UK agriculture and food production industries. A major objective of this programme is to identify opportunities for water saving in agriculture and horticulture. Genetic improvement programmes aimed at identifying Quantitative Trait Loci (QTL) for improved drought tolerance will eventually facilitate the marker-assisted selection of new lines with improved water use efficiency (e.g. HH3608TX, WU0107). However, UK horticulture is unlikely to benefit from these advances for at least a decade. In the meantime, irrigation management techniques that improve the efficiency of water use in existing crops grown in areas where water resources are most threatened are an attractive option.

These techniques, which include Partial Rootzone Drying (PRD), Deficit Irrigation (DI) and Regulated Deficit Irrigation (RDI), involve applying slightly less water than the plant needs so that mild soil water deficits develop. Roots exposed to the drying soil produce chemical signals that are transported to the shoots where they influence both vegetative and reproductive growth. High profile successes in the vineyards of South Australia prompted many growers worldwide to try these irrigation management techniques, often with only limited success. Experiments were conducted with seven major crop species (strawberry, raspberry, runner bean, potato, tomato, lettuce and poinsettia) over 5 years. Container- and soil-grown plants were treated with PRD, DI or RDI at various percentages of full irrigation. Full irrigation was determined on the basis of potential evapotranspiration, actual evapotranspiration (in container-grown plants) or soil moisture monitoring. Physiological measurements were made throughout the cropping cycle to assist in irrigation scheduling, and crop yield, quality and shelf-life potential assessed at harvest.

Physiological responses (increased water use efficiency [WUE], decreased yield) depended on the severity of the deficit irrigation imposed. In strawberry and raspberry, 20-25% less irrigation did not decrease marketable yields or berry quality indicating considerable scope for water savings. More severe deficit irrigation (50% less irrigation than fully irrigated crops) incurred, on average, a 20% yield penalty but increased WUE by 20% in strawberry, runner bean and lettuce, and approximately 50% in raspberry, potato and tomato. Generally, there were no detectable yield differences between PRD, DI or RDI plants supplied with the same irrigation volumes, although in runner beans PRD out yielded DI crops by over 20% when soil-

SID 5 (Rev. 3/06) Page 3 of 29

grown in polytunnels, while in substrate-grown strawberry, RDI plants out yielded PRD plants. RDI had positive impacts on berry and fruit crops such as increased colour, flavour and

altered chemical (e.g. antioxidant) composition. RDI was also used successfully to limit poinsettia stem height so that retailers specifications were met, despite a 90% reduction in the use of environmentally unsustainable chemical growth retardants. RDI also improved poinsettia quality and shelf-life potential. However, even mild deficit irrigation (20% less irrigation) resulted in unmarketable crops of lettuce (bitter taste) and potato (unacceptable skin finish).

Deficit irrigation (and especially alternating wet and dry parts of the rootzone in PRD) stimulated biomass allocation towards the roots (which may limit the severity of leaf water deficit) and fruits. Limitation of crop photosynthesis likely accounted for the yield and quality penalties sustained.

Plant long-distance signalling was quantified by measuring leaf water relations, and collecting xylem sap to assay its chemical constituents. Irrespective of crop or sap sampling methodology, deficit irrigation increased concentrations of abscisic acid, (ABA - a hormone that regulates water use by closing leaf pores known as stomata) and decreased concentrations of cytokinins (hormones that regulate shoot growth).

The substrates in which crops were grown generally didn’t influence yield, but affected long-distance signalling. Simulating leaf xylem sap ABA concentration in response to a standardised PRD treatment showed that ABA-mediated restriction of whole plant transpiration may be more readily achieved in mineral substrates. Alternating wet and dry parts of the rootzone in PRD in highly organic substrates (e.g. peat) caused transient leaf water deficits as it was difficult to re-wet the dry substrate.

Theoretical analysis showed that the optimum time to alternate the wet and dry sides was when the dry side has dried to a level where root ABA output started to diminish. Direct ABA measurement is impractical thus alternative indirect approaches including measurement of soil moisture and plant responses were evaluated. Soil moisture sensing offers the advantage of ready automation while plant-based measurements can be time-consuming: practicalities (ease of grower use; equipment expense or availability) are likely to dominate the choice of an appropriate scheduling technique.

Project progress, the development of new proposals and plans for technology transfer were discussed regularly with our industry advisory group to help ensure that our research continued to meet the needs of the industry. Several trials were conducted on commercial holdings to facilitate direct technology transfer. Consortium members made numerous presentations at grower days and contributed articles to the trade press. 35 scientific papers have already been published (see Section 9) and additional manuscripts are currently being prepared for submission to international scientific journals. Strong industry support for our work has meant that a HortLINK project (HL0187) and an HDC-funded project (SF 107) have been developed directly as a result of this Defra project. Further development of PRD and RDI in UK horticulture is being delivered via Defra projects WU0110 and WU0118, and the EU-funded SIRRIMED project (commencing October 2009). Other project proposals are currently being prepared.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

SID 5 (Rev. 3/06) Page 4 of 29

IntroductionApproximately 70% of world-wide water use is committed to agriculture. Despite this, water shortages limit food production in many regions. If crop production is to be sustained and even increased in a changing environment then water must be used more efficiently. In the UK, irrigation is required in most regions in most years to ensure reasonable crop yields of the required quality. The recent Defra study on water use in agriculture (WU0102) suggests that climate change will further increase demand for irrigation in these areas by about 20% by 2020 and 30% by 2050. Consequently, UK growers, Defra and the Environment Agency are becoming increasingly concerned about the future availability of

SID 5 (Rev. 3/06) Page 5 of 29

abstracted water for irrigation. New legislation designed to safeguard these resources (The Water Act 2003) will limit future water use and growers will have to demonstrate efficient use of available water before time-limited abstraction licences are renewed. Mains supplies will also be limited and expensive.

These unavoidable issues have recently focussed attention on irrigation techniques that allow more efficient use of water. Increasingly, higher value horticultural crops are grown in the UK under protected cropping. This increases yields and quality but also provides opportunities both to save water and nutrient resources and to modify supplies of these variables to regulate growth and development. This ‘natural’ growth regulation has the potential to replace the widespread use of growth regulating chemicals, an expensive, undesirable and increasingly unsustainable component of crop production.

Partial Rootzone Drying (PRD) is a deficit irrigation technique designed to enhance crop water use efficiency (WUE) by exploiting the plant’s long-distance signalling mechanisms that modify plant growth, development and functioning as the soil dries. The novel science behind these mechanisms has been revealed largely by the members of this consortium (Davies and Zhang 1991; Davies et al. 2005; Dodd 2005, Dodd and Beveridge 2006; Jia and Davies 2007). Exploitation of this science has been actively pursued by the South Australian wine industry and an EU-funded consortium (IRRISPLIT) co-ordinated by Lancaster. Although these projects have delivered substantial water savings and significantly added value in terms of increased yield quality, labour saving and crop scheduling in Mediterranean (climatic) environments, when this project was conceived there had been little development of PRD (or deficit irrigation more generally) in the UK, representing a potential “missed opportunity” for UK growers, retailers and consumers.

The main project objective was to develop the potential to deliver substantial savings of irrigation water in different horticultural sectors while maintaining or improving crop quality by optimising PRD technology. While many PRD trials have delivered very positive results, this is not always the case and we sought to understand why this happens. Since crop responses to deficit irrigation are a cumulative response to dynamic changes in plant long-distance signalling, much more information is needed on the nature and origin of these signals and how they impact on plants grown under different environmental conditions and rooting media. Furthermore, signalling impacts that may be desirable in some crops (e.g. limitation of vegetative growth in over-vigorous grape vines may save pruning costs) may be undesirable in others (e.g. lettuce where vegetative growth is essential) thus it seems likely that different regimes / approaches need to be developed for different crops under different production systems. A key aspect on which the success (or otherwise) of deficit irrigation may be judged is consumer perceptions of product quality: increased quality may be highly sought after (as evidence by consumer willingness to pay a premium) in certain crops (e.g. strawberry) yet attract little market share in others (e.g. lettuce). For these reasons, the responses of a wide variety of crops (strawberry, raspberry, runner bean, potato, tomato, lettuce, poinsettia) to deficit irrigation were investigated.

Scientific objectives1. To quantify potential water saving under deficit irrigation and the impact of these treatments on

yield and WUE 2. To determine and quantify signalling mechanisms under deficit irrigation 3. To determine the effects of PRD on resource partitioning and root and leaf functioning 4. To determine the environmental conditions and substrates that maximise the benefits of PRD 5. To determine the optimal scheduling of deficit irrigation in the field 6. To determine the 'quality' of deficit-grown produce7. To ensure effective technology transfer

Progress in relation to stated objectivesObjective 1 was fully met. Experiments with a number of crops showed it was possible to design appropriate deficit irrigation strategies to save water without compromising yield. Generally, there were no detectable yield differences between PRD and DI plants, although RDI-grown strawberries outyielded PRD-grown strawberries while PRD-grown runner beans outyielded DI-grown runner beans.

Published outputs: See Reference 5 in Section 9A compilation paper from the entire consortium is in preparation

Objective 2 was fully met. Various hydraulic (leaf water potential -leaf) and chemical signals (xylem sap pH, and the plant hormones abscisic acid (ABA), indole-3-acetic acid (IAA) and various cytokinins (CKs)) were quantified. Experiments with poinsettia and strawberry showed that deficit irrigation increased delivery rates of ABA, but decreased delivery rates of CKs, from the root system. PRD

SID 5 (Rev. 3/06) Page 6 of 29

experiments showed similar changes in hormone concentrations in xylem sap collected from tomato leaves, and that alternating the wet and dry parts of the root system increased xylem sap ABA concentration. Modelling leaf xylem sap ABA concentration as a function of ABA contributions from the wet and dry parts of the root system showed that total soil water availability determined whether DI or PRD elicited a higher ABA concentration.

Published outputs: See References 9, 17, 19, 20, 21 in Section 9

Objective 3 was fully met. Deficit irrigated crops partitioned more of their resources into the root system, negatively impacting on yield in lettuce but with no apparent effect in tomato (presumably since fruit growth was maintained at the expense of vegetative growth).

Published outputs: See Reference 4 in Section 9

Objective 4 was fully met. Extensive data sets in several crops showed that stomatal responses to PRD were more pronounced at high atmospheric vapour pressure deficit. Substrate did not affect yield or stomatal responses to PRD in runner bean, but had different effects on raspberry yield in two consecutive years. Each substrate generated unique relationships between soil matric potential and the fraction of sap flow from roots of PRD plants in drying soil, and root ABA concentration. Incorporation of these relationships into a model showed that the increase in shoot xylem ABA concentration in response to a defined PRD treatment varied with substrate.

Published outputs: See References 27, 28, 33, 34 in Section 9

Objective 5 was fully met. Adoption of automatic soil moisture based irrigation scheduling in RDI-grown poinsettia successfully produced high quality plants while obviating the need for undesirable sprays of chemical growth retardants. Various methods of plant-based irrigation scheduling (leaf water potential, porometric and thermal camera measurements of stomatal conductance) were compared in a number of crops. Empirical assessment of the effects of different frequencies of alternating wet and dry sides of PRD plants was made on runner bean stomatal responses and yield.

Published outputs: See References 7, 8, 24, 25, 26, 30, 31, 32 in Section 9

Objective 6 was fully met. Detailed studies of crop quality were made in experiments with strawberry (including biochemical analysis of flavour volatile production and antioxidant concentrations) and poinsettia (including plant height specifications and post-production quality and shelf-life tests). Visual assessments of crop quality were made in potato (tuber size distribution and skin quality), runner bean (pod curling assessments) and lettuce (colour and taste).

Published outputs: See Reference 23 in Section 9

Objective 7 was fully met. Throughout this project, we held several Advisory Group meetings with industry representatives working with all crop species. Several of our trials were conducted on commercial holdings. A number of follow-on studies (funded by Defra, HortLINK and HDC) have been developed from this project. A full list of technology transfer activities is included as Appendix 1.

Methodology Conventionally, irrigation scheduling has aimed to meet full crop evapotranspiration (ET), determined from direct measurements or micrometeorology-based estimations of ET, since the relationship between ET and crop yield is near-linear at suboptimal water supply. In contrast, deficit irrigation (DI) applies less water than full crop ET. This project compared two forms of DI, which varied the placement of irrigation within the rootzone. In conventional DI, less water was applied to the entire rootzone. In PRD, only one side of the crop row (or one soil compartment in containerised plants) was irrigated at a time and the other was allowed to dry the soil. Maintenance of the same wet and dry parts of the rootzone throughout the growing season (fixed PRD) contrasts with studies where the wet and dry parts of the rootzone are regularly alternated (alternate PRD). Unless stated otherwise, all mentions of PRD in this report relate to alternate PRD, although the duration of switching between wet and dry sides varied with the different crops and production systems.

An alternative irrigation management technique is Regulated Deficit Irrigation (RDI), which aims

SID 5 (Rev. 3/06) Page 7 of 29

to exploit the differential sensitivity of yield-determining processes to water deficits. RDI has been used extensively in tree crops and vines, where water deficits have been imposed at a particular stage(s) of crop development, rather than throughout the growing season. When crop phenology is sensitive to water deficit (e.g. during fruit cell division), 100% ET is supplied but when crop phenology is less sensitive to water deficit (e.g. during vegetative growth prior to fruit development and late fruit expansion) less irrigation is supplied. Since many of our experiments compared irrigation placement (PRD versus DI) rather than timing, the term RDI is used in this report only in relation to work with strawberry and poinsettia.

Experiments were conducted with seven major crop species grown with conventional deficit irrigation and PRD, with various percentages of full irrigation. Several approaches were adopted by the three partner institutions (Dundee, East Malling Research [EMR], Lancaster) to determine full irrigation:

- potential evapotranspiration (ETp) measured using a Skye Evaposensor (EMR)- direct weighing of pot-grown plants to determine actual evapotranspiration (Lancaster and

EMR)- soil moisture monitoring to maintain soil moisture at ≥ 95% of field capacity (Dundee) or at pot

capacity (EMR)- automatic scheduling according to daily radiation load (Lancaster) or substrate moisture content

EMR)Various physiological measurements (leaf elongation to assay vegetative growth, stomatal

conductance to determine restriction of water loss, pre-dawn and midday leaf water potential to determine plant water status, soil water status) were made to determine effects of irrigation treatments during crop growth. Assessments of yield and quality parameters (see Objective 6) relevant to the different crops were made at harvest / market date.

Although experiments were replicated over multiple years (a minimum of 2 years for each species), substrates (3 different substrates for containerised raspberry and runner beans) and varieties (June bearers and ever-bearers in strawberry; a traditional variety and an F1 hybrid in tomato), each crop was trialled under conditions approaching those used commercially by only one partner. Similarly, each partner used a different type of irrigation scheduling (as discussed above) thus it is not possible to separate scheduling from species effects. However, adequate replication within each experiment allows great confidence within the limitations of each experimental system.A more detailed description of the methodology adopted for each crop can be found in Appendix 2

ResultsObjective 1 To quantify potential water saving under deficit irrigation and the impact of these

treatments on yield and WUE IntroductionFollowing the high profile successes in Southern Australia with grapevine (Dry et al., 1996), PRD has been trialled in many countries and has been shown to improve water use efficiency (WUE) in many crops in both in glasshouse studies and field trials. A meta-analysis of these data will shortly be published (Dodd 2009). In 2004, Defra commissioned this research project to assess the potential of PRD to deliver water saving into several UK horticulture sectors. The potential benefits associated with PRD were compared with those of another widely used irrigation management technique, Regulated Deficit Irrigation (RDI) to determine the most appropriate water saving strategy for the seven different horticultural sectors.

In assessing the yield responses of plants to deficit irrigation generally, or PRD in particular, two questions were framed: (1) What is the relationship between yield and irrigation volume ? (this determines the potential water savings and/or yield penalties – Figure 1.1-1.3) (2) Does PRD differentially influence yield compared to DI or RDI ? (Figure 1.4). For six of the seven crops assessed (it was not appropriate to quantify yield in poinsettia as quality attributes are key to market value – see Objective 6), relative yield data were summarised (Figure 1.1-1.3) to gain an overview of the responses of a range of crops.

Results and DiscussionThe yield penalty sustained was dependent on the severity of deficit irrigation imposed (see Figure 1.1). For example, in container-grown lettuce, decreasing irrigation by 15% and 30% significantly decreased shoot fresh weight (=yield) by 13% and 29% respectively. Such close correspondence between yield and crop evapotranspiration suggests that very little water can be saved without incurring a yield penalty. For the container-grown strawberry crop in 2004, decreasing irrigation (PRD

SID 5 (Rev. 3/06) Page 8 of 29

and RDI data combined) by 20% and 40% decreased class I berry yield by 16% and 38% respectively, even though the yield reductions were not statistically significant. In contrast, for the strawberry crops grown in 2005 and 2006, the link between yield and crop evapotranspiration was not so apparent, with a 40% decrease in irrigation incurring a 25% yield penalty (Figure 1.1). Of more importance was the finding that a 20% decrease in irrigation only incurred a significant yield penalty in one (2005) of three years, suggesting a potential water saving under deficit irrigation with minimal impact on yield; this work is discussed further below. Crops grown with unrestricted root systems in polytunnels (raspberries in 2005 and 2006, potatoes in 2005) suffered no significant yield penalty despite a 25% decrease in irrigation. Thus, in several crops a potential water saving of 20-25% can be made without significant negative impacts on crop yield, with maintenance of crop quality in some (raspberry, strawberry) but not all (potato) crops – see Objective 6. There does not seem to be any potential for water saving in lettuce.

Since the original concept of PRD envisaged watering only half the rootzone at each irrigation, several projects including ours (e.g. IRRISPLIT, the South Australian winegrape project) have imposed a 50% decrease in irrigation in some trials. Even though the analysis above indicates this will likely incur some yield penalty, it was informative to compare results obtained under UK conditions. Decreasing irrigation by 50% decreased yield (relative to fully irrigated crops) by an average of 20% (Figure 1.2). A notable exception was lettuce, where leaf growth (responsible for the harvested crop portion) was especially sensitive to soil drying,

such that a 29% yield penalty was incurred when 30% less irrigation was applied. In crops that were grown with both restricted and unrestricted root systems (raspberry and runner bean), there was apparently greater scope for saving water in crops grown with unrestricted root systems. When 50% less irrigation was applied, the yield penalty incurred was

less for crops grown with unrestricted, rather than restricted, root systems (Figure 1.2), presumably since crops with unrestricted root systems could access water deeper in the soil profile, thus decreasing their reliance on surface-supplied irrigation.

SID 5 (Rev. 3/06) Page 9 of 29

Figure 1.1. Crop yield ratio of deficit irrigation to full irrigation (a ratio of 1 indicates that yield with both techniques is equivalent) for different crops irrigated with various percentages of full irrigation. Each point is the mean response of 16 (lettuce) and 12 (strawberry) plants. Significant (P < 0.05) differences between deficit irrigation and full irrigation are arrowed. The dotted line indicates the 1:1 relationship for irrigation percentages < 100%, and no yield limitation at percentages > 100%.

Figure 1.2. Crop yield ratio of deficit irrigation (mean of PRD and DI) to full irrigation (a ratio of 1 indicates that yield with both techniques is equivalent). The fraction of full irrigation was 50% except for crops 1 (60%), 3-unrestricted (66%) and 6 (75%). Columns are means ± SE of the number of experiments / seasons given in parentheses at the base of each column. Shaded columns denote pot experiments where the root system was unrestricted. Significant (P < 0.05) differences between deficit irrigation and full irrigation are indicated with an asterisk (*). Studies are numbered thus: (1) Strawberry - Fragaria ananassa cv. Elsanta (2) Raspberry - Rubus idaeus cv. Glen Ample (3) Runner Bean - Phaseolus coccineus cv. Emergo (4) Potato - Solanum tuberosum cv. Maris Piper (5) Tomato - Solanum lycopersicum cv. Ailsa Craig (6) Lettuce - Lactuca sativa cv. Rex.

Water use efficiency can be defined as the ratio of crop yield to irrigation applied (applied water use efficiency) or water used (intrinsic water use efficiency). In our experiments, the latter could only be determined in containerised crops that were weighed directly. Thus applied water use efficiency was calculated, and was increased by deficit irrigation in all crops (Figure 1.3). Although the magnitude of the increase was roughly 20% in strawberry, runner bean and lettuce, it was approximately 50% in raspberry, potato and tomato. In runner beans, the increase in WUE was independent of whether the root systems were constrained or not; but in raspberries grown in unconstrained soil, the WUE almost doubled for the deficit irrigation treatments. While increased WUE was achieved in all crops, these gains must be tempered against yield losses in some crops (cf. Figure 1.2)

Generally, there were no detectable yield differences between PRD and DI plants (Figure 1.4). There were two exceptions: yield of PRD-grown strawberries was 12% less than RDI-grown strawberries while PRD-grown runner beans outyielded DI-grown runner beans by over 20% when grown in polytunnels with unrestricted roots.

SID 5 (Rev. 3/06) Page 10 of 29

Figure 1.4: Crop yield ratio (and WUE since both techniques received the same irrigation volumes) of PRD to DI. A ratio of 1 indicates that yield or WUE with both techniques is equivalent. The fraction of full irrigation was 50% except for crops 1 (60%), 3-unrestricted (66%) and 6 (75%). Columns are means ± SE of the number of experiments / seasons given in parentheses at the base of each column. Shaded columns denote pot experiments where the root system was unrestricted. Significant (P < 0.05) differences between deficit irrigation and full irrigation are indicated with an asterisk (*). Studies are numbered thus: (1) Strawberry - Fragaria ananassa cv. Elsanta (2) Raspberry - Rubus idaeus cv. Glen Ample (3) Runner Bean - Phaseolus coccineus cv. Emergo (4) Potato - Solanum tuberosum cv. Maris Piper (5) Tomato - Solanum lycopersicum cv. Ailsa Craig (6) Lettuce - Lactuca sativa cv. Rex.

Figure 1.3. Water use efficiency ratio of deficit irrigation (mean of PRD and DI) to full irrigation (a ratio of 1 indicates that water use efficiency with both techniques is equivalent). The fraction of full irrigation was 50% except for crops 1 (60%), 3-unrestricted (66%) and 6 (75%). Columns are means ± SE of the number of experiments / seasons given in parentheses at the base of each column. Shaded columns denote pot experiments where the root system was unrestricted. Studies are numbered thus: (1) Strawberry - Fragaria ananassa cv. Elsanta (2) Raspberry - Rubus idaeus cv. Glen Ample (3) Runner Bean - Phaseolus coccineus cv. Emergo (4) Potato - Solanum tuberosum cv. Maris Piper (5) Tomato - Solanum lycopersicum cv. Ailsa Craig (6) Lettuce - Lactuca sativa cv. Rex.

PRD versus RDI - Strawberry (EMR)Although it is often possible to improve WUE with deficit regimes, in crops like strawberry marketable yields must be maintained if the techniques are to be commercially viable. In this type of work, the most relevant expression of WUE is the weight of class 1 fruit produced per unit volume of irrigation water applied. Perhaps not surprisingly, the success of deficit regimes can vary (Pudney and McCarthy, 2004); in studies on strawberry, deficit irrigation reduced fruit size and number, resulting in losses of marketable yield (Kirnak et al., 2003; Yuan et al., 2004). We conducted polytunnel experiments with the ‘June-bearer’ ‘Elsanta’ and the ever bearer ‘Flamenco’ to determine if PRD or RDI regimes could be developed that delivered substantial water savings while maintaining or improving yields and quality of class 1 fruit.

ResultsIrrigation scheduling and water savingsA novel method of scheduling irrigation to substrate-grown strawberries was developed in this project. Daily plant water use was estimated using an Evaposensor (Figure 1.5) and Evapometer and cultivar-specific calibration factors were calculated weekly to account for increases in canopy area over each season. Using this system, run-off from well-watered plants following irrigation events was avoided because plant demand for water was matched exactly with supply. Deficit irrigation treatments were applied as a percentage of potential evapotranspiration (ETp) at 50% full bloom in each season to try to ensure that berry numbers were not reduced in deficit-grown plants.

SID 5 (Rev. 3/06) Page 11 of 29

Table 1.1: Water use efficiency (kg class 1 fruit produced per cubic metre of irrigation water applied) for ‘Elsanta’ strawberry plants under deficit irrigation regimes for two seasons. Estimated values for current commercial practice in substrate-grown strawberry production are presented for comparison.

Irrigation regime (% of ETp)

WUE (kg class 1 fruit per m3 H2O)60-day Main season Two seasons

WW 120 5.5 11.1 8.4PRD 100 5.8 12.3 9.1PRD 80 6.2 17 11.7PRD 60 7.3 25.4 16.5RDI 100 5.8 12.3 9.1RDI 80 7.5 20.3 14.0RDI 60 8.4 24.7 16.7SED 0.41 2.27 1.35Commercial crop* 3.4-4.8 10.7-13.6 7.1 - 9.2

* Assuming average class 1 yields of 250 g and 550 g per plant and daily irrigation volumes of 500 ml and 700 ml per plant in 60-day and main season plants, respectively

0 20 40 600

10

20

30

Days after treatments applied

0 20 40 60 80

Cum

ulat

ive

irrig

atio

n vo

lum

e (L

)

0

10

20

30

40

Commercia regimeWell-watered 120%PRD & RDI 100%PRD & RDI 80%PRD & RDI 60%

60-day season 2006

Main season 2007

Figure 1.6. Irrigation volumes applied to 60-day and to main season plants (WW = 120%; PRD and RDI regimes of 100%, 80%, 60% of ETp). The solid line represents the current recommended value for substrate-grown strawberries.

WW PRD100

PRD80

PRD60

RDI100

RDI80

RDI60

0

50

100

150

200

Irrigation regime

***

A) 60-day crop

B) main season crop

Ave

rgae

yie

ld o

f cla

ss 1

frui

t per

pla

nt (g

)

0

200

400

600

B) main season crop

Figure 1.7. Effects of irrigation regimes on yields of class 1 fruit per plant in A) 60-day and B) main season plants. Results are means of 12 replicate plants; asterisks indicate statistically significant differences (P ≤ 0.05) from WW values.

Cumulative irrigation volumes applied to each plant under the seven regimes were calculated for both 60-day and main season plants and compared to current industry ‘best practice’ recommendations (Figure 1.6). Substantial water savings were achieved by irrigation scheduling, a total of 12 L water per plant over both seasons. This would equate to a saving of 700 m3 of water per hectare of substrate-grown strawberries in commercial production. Further water savings were delivered by the different severities of the PRD and RDI regimes; in the most severe regimes (60% ETp) only 57% of the volume of water that would have been used in a commercial crop was applied.

Yields of class 1 fruitYields of class 1 fruit from well-watered plants receiving 120% of daily ETp averaged 230 g and 500 g per plant in 60-day and main season, respectively (Figure 1.7). Yields were significantly reduced by the PRD 80% and 60% regimes and by the RDI 60% regime in the first year but there were no significant treatment differences in the main cropping year. Over three seasons; marketable yields were consistently reduced by the most severe PRD and RDI regimes (60 % ETp) but yield penalties were also incurred with 100% and 80% PRD regimes during the hot summer in 2006. Reductions in yield were always due to treatment effects on berry size, not berry number.

Water use efficiency Water use efficiency can be expressed as the yield (mass) of class 1 fruit produced per unit volume of irrigation water and a higher value indicates a greater WUE. The irrigation volumes used to produce the yields of class 1 fruit recorded for 60-day (2006) and main season (2007) were combined to give an estimate of the overall WUE over both cropping seasons (Table 1.1). The 80% and 60%.PRD and RDI regimes improved WUE significantly, compared to WW values, in both cropping seasons. However, as noted above, yields were reduced by some of these treatments. Values of WUE for the conservative 120% ETp WW regime were similar to those calculated for a typical commercial crop.

DiscussionCurrent guidelines for irrigating

substrate-grown strawberries recommend volumes of between 0.5 and 0.7 L of water per plant per day (ADAS, 2003). Values of WUE for WW plants were similar to those estimated for a typical commercial crop since we used a rather conservative approach in which we supplied 120% of estimated daily ETp

SID 5 (Rev. 3/06) Page 12 of 29

to account for leaks, run-off and evaporation from the substrate surface. We have since shown in WU0110 that 100% of ETp is adequate to maintain yields of WW plants and that effective scheduling can improve WUE by a further 25% (WU0110 SID 4 2009).

We have shown that if RDI is applied judiciously, substantial water savings can be made, without compromising yields of class 1 fruit. WUE, expressed in terms of kg fruit produced per cubic metre of water applied, was increased from 8.4 in well-watered plants to 14 in RDI-treated plants (Table 1.1). RDI also improved several components of fruit quality (see Objective 6). The potential to use irrigation scheduling and RDI to improve WUE, nutrient use efficiency and fruit quality in substrate-grown strawberry production will be developed further in our new HDC-funded project (SF 107).

Although increases in WUE were obtained with PRD, compared to well-watered values, reductions in yield of class 1 fruit occurred with most of the PRD regimes tested. Our attempts to overcome this shortcoming by modifying the ways in which the irrigation was switched were unsuccessful. The loss in productivity was presumably due to plant water deficits that developed during the irrigation switching events inherent in all PRD regimes. This problem is likely to be exacerbated in hot weather and therefore PRD may be a risky strategy to use in commercial strawberry production. However, loss of turgor during switching may only occur in containerised plants where root growth, and therefore access to water, is restricted. We are currently testing the potential of PRD to deliver water savings while maintaining marketable yields and improving fruit quality and shelf-life potential in field-grown strawberries (HL0187).

Our preliminary results also suggested that if PRD and RDI regimes applied to 60-day plants were continued during the flower initiation phase (Sept-Oct), the number of trusses produced in the second cropping year (2006) was increased. Despite the greater number of fruit, berry size was not reduced and so marketable yields were increased by up to 40% in the second cropping year. Although the increases in yields were not always statistically significant (e.g. Figure 1.7B), our data suggest that deficit irrigation offers the potential not only to improve WUE and berry quality, but also to increase yields of class 1 fruit. Clearly, this result has very important economic implications and, to our knowledge, has not been reported before. This aspect of the work is currently being progressed in WU0110.

We also conducted experiments with the ever-bearer ‘Flamenco’ over two cropping seasons (HH3609TX SID 4 2005-2006). Our results suggested that although vegetative growth could be limited by both PRD and RDI, yields of class 1 fruit were reduced by both techniques. This may be due to the detrimental effects of the deficit regimes on flower initiation which occurs continuously throughout the season in ever-bearers. More applied work is needed to determine whether irrigation scheduling and deficit regimes can be applied during specific stages of development to restrict vegetative growth without compromising fruit size or number. We are currently developing such a concept note for submission to the HDC Soft Fruit panel later this year.

Objective 2 To determine and quantify signalling mechanisms under deficit irrigationIntroductionPartial Rootzone Drying is a deficit irrigation technique designed to enhance crop WUE by exploiting the plant’s long-distance signalling mechanisms that modify plant growth, development and functioning as the soil dries. It is an adaptation of laboratory split-root experiments (Blackman and Davies 1985; Gowing et al. 1990; Sobeih et al. 2004) that utilises plant root-to-shoot chemical signals to influence shoot physiology. It can be operated in drip- or furrow-irrigated crops where each side of the row is watered independently. When the crop is irrigated, only one side of the row receives water and the other is allowed to dry the soil. The root system senses soil drying and produces chemical signals that are transmitted to the shoots to close the stomata (decreasing water loss) and limit vegetative growth, thus improving WUE (Gowing et al. 1990; Dry et al. 1996; Davies et al. 2000). Although there is evidence that the plant hormone ABA is one of the components involved in the control of stomatal conductance as the soil dries (reviewed in Davies and Zhang 1991; Dodd 2005; Dodd et al. 2008a), more information is needed on how PRD affects the outputs of root-derived signals (in integrating chemical information from both wet and dry parts of the rootzone) and whether other chemical signals fluctuate during deficit irrigation treatments.

Experiments on Strawberry (EMR)We have shown that deficit regimes can be used to limit transpirational water loss, slow leaf extension

SID 5 (Rev. 3/06) Page 13 of 29

Days after start of treatment0 1 2 3 4 5 6

0

200

400

600

WW PRD 80% ETp80 RDI 80% ETp

Sto

mat

al c

ondu

ctan

ce (m

mol

H20

m-2

s-1

)

**

***

Figure 2.1. Effects of PRD and RDI on stomatal conductances measured on a fully expanded strawberry leaf in a GroDome experiment. Results are means of six replicate plants; asterisks indicate statistically significant differences (P ≤ 0.05).

0.1 0.2 0.3 0.4 0.5

Xyl

em s

ap h

orm

one

deliv

ery

rate

(ng

h-1)

0

10

20

30

40

50

0.1 0.2 0.3 0.4 0.56.2

6.4

6.6

6.8

7.0

7.2

Substrate moisture content (m3m-3)

0.1 0.2 0.3 0.4 0.50

100

200

300

0.1 0.2 0.3 0.4 0.50

10

20

30

ABA pH

ZR Z

PRD 80% ETp RDI 80% ETp

WW

Figure 2.2. Effects of substrate drying during PRD and RDI regimes on the delivery of abscisic acid (ABA), pH, zeatin riboside (ZR) and zeatin (Z) from pressurised detopped strawberry root systems. Delivery rates were calculated from multiplying sap flow rates by hormone concentrations quantified by GC-MS-SIM. The horizontal dashed line in each figure indicates the deliveries at which significant reductions in stomatal conductances first occurred.

and reduce canopy area; other benefits include improved flavour volatile production and enhanced bioactive content (see Objective 6). These beneficial effects are thought to be controlled by chemical signals (e.g. plant hormones) synthesised in response to drying soil (Davies et al., 2000). However, the signals produced in response to deficit irrigation that regulate shoot responses in strawberry or any other member of the Rosaceae are not known, although ABA is presumed to be involved (Stoll et al., 2000; Dodd et al. 2006)

The aim of this work was to identify the signals in strawberry that initiate leaf responses to help minimise water loss from leaves of deficit-grown plants. To help establish cause from effect, it was important to be able to detect any minor changes in hormone flux from the roots immediately before the leaf responses were initiated. The precise control over environmental conditions afforded by the GroDome enabled us to predict when the leaf responses to PRD and RDI first occurred and we were able to target sap collection times accordingly. The hormone profile in these many sap samples was then established using definitive quantification by GC-MS-SIM. Hormone delivery rates were calculated by multiplying concentrations by sap flow rates to ensure that any changes in the output of root-sourced chemical signals were not simply artefacts of reduced transpirational water flow (Else et al., 1995). Transpiration bioassays were used to try to establish whether changes in hormone output were physiologically significant.

ResultsTranspiration rates and stomatal conductances of PRD- and RDI-treated plants began to diverge from WW values three days after the beginning of the experiment and were significantly lower by the afternoon of day 4 (Figure 2.1). Leaf elongation rate (LER) was not affected by the deficit regimes in these short-term experiments. Leaf water potentials remained unchanged in the PRD- and RDI-treated plants suggesting that stomatal closure was regulated by chemical rather than hydraulic signals. ABA delivery rates from PRD- and RDI-treated roots increased as the substrate dried (Figure 2.2) and were significantly higher in RDI-treated plants on days 4 and 6; similar increases were detected in PRD-treated plants although the results were not statistically significant due to variability between plants. No consistent effects on xylem sap pH were detected over the six-day experiment but the deliveries of zeatin (Z) and zeatin riboside (ZR) decreased as the substrate dried (Figure 2.2) and were significantly lower in PRD-treated plants on day 4. Export of dihydrozeatin and dihydrozeatin riboside were low and were not affected by deficit regimes. The output of indole-3-acetic acid (IAA) from PRD- and RDI-treated roots was reduced within two days and remained lower than WW values thereafter.

To test the causal status of the increased ABA delivery, the concentrations of ABA detected in xylem sap from WW, PRD- and RDI-treated plants were used in bioassays where detached leaves were fed with a range of synthetic ABA solutions. These tests suggested that the changes in ABA output from PRD- or RDI-treated roots were not sufficient to initiate stomatal closure (Figure 2.3). Although this may indicate that other signals are involved, differences in the sensitivity of the transpiration bioassay to ABA compared to intact leaves may also be a factor.

Discussion

SID 5 (Rev. 3/06) Page 14 of 29

Clock time

09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00Tran

spira

tion

rate

(mm

ol H

2O m

-2s-1

)

0.0

1.0

1.5

2.0

*** *

Transfer

0 mol m-3 ABA10 mol m-3 ABA

20 mol m-3 ABA40 mol m-3 ABA

Figure 2.3. Effects of different concentrations of synthetic ABA on rates of water loss from detached leaves. All leaves were transferred from buffer to experimental solutions at 11:00. Results are means of six replicate leaves; asterisks indicate statistically significant differences (P ≤ 0.05).

09:10 11:12 13:15 15:05 16:45

Sto

mat

al c

ondu

ctan

ce (m

mol

H 2O m

-2 s

-1)

0

100

200

300

400

*

** *

*

*

Clock time

CC WW RDI

Figure 2.4. Effects of RDI on stomatal conductances measured on a fully expanded poinsettia leaf compared to values from commercial control (CC) and well-watered (WW)) plants. Results are means of 12 replicate plants; asterisks indicate statistically significant differences (P ≤ 0.05) from CC values.

The changes in root-sourced chemical signals produced in response to deficit irrigation have been quantified in strawberry. Increased output of ABA and decreased delivery of Z and ZR from roots exposed to drying soil coincided temporally with the onset of stomatal closure in RDI-treated plants. Delivery of ABA was also increased from PRD-treated roots at the time stomata began to close although the change was not statistically significant in this experiment. In other experiments, increased output of ABA from PRD-treated roots was followed within hours by the onset of stomatal closure. The reduction in cytokinin export from roots exposed to PRD and to RDI may also have increased stomatal sensitivity to ABA.

Xylem sap ABA concentrations in planta at the time stomata first began to close were approximately 30 µmol m-3 but concentrations above 40 µmol m-3

were needed to prompt stomatal closure in detached leaf transpiration bioassays. A lower sensitivity of stomata to ABA in these bioassays is often observed and on-going work in our HortLINK project (HL0187) suggests that increased ethylene production in detached leaves may reduce the extent of ABA-induced stomatal closure (see Tanaka et al., 2005).

The lack of a leaf growth response to PRD and RDI was always observed during these short-term GroDome experiments. This was unexpected since a slowing of LER is usually one of the first detectable responses to soil drying. In our polytunnel experiments, LER didn’t begin to diverge from WW values until 18 days after the beginning of the deficit irrigation treatments. This implies that either different signals regulate stomatal and leaf growth responses or that the responses are triggered by different intensities of the same signal.

These studies have greatly improved our understanding of how the plant’s internal signalling system changes in the first few days of deficit irrigation. Although this work is a very important first step, for a crop like strawberry where the aim is to maintain marketable yields and improve fruit quality, more work needs to be done on optimising the delivery of these root-sourced chemicals over the cropping season. A lowered photosynthetic rate imposed by continuous and intensive signalling from roots will also limit production of important precursors to key flavour components such as sugars and flavour volatiles. In strawberry, partial rather than complete stomatal closure could help to maintain photosynthesis and the production of essential sugar and flavour volatile precursors while limiting water loss. Optimising the production and transport of the chemical signals that interact to control water loss and photosynthetic rate over the cropping season is a key component of HL0187.

Very little is known about the signals that regulate the production of antioxidants in fruit. Fruit ripening is a developmental event where oxidative processes take place and a variety of reactive oxygen species (ROS) accumulates (Jimenez et al., 2002). However, in addition to this developmental control, mechanisms must exist that are able to stimulate further antioxidant production. Our results with PRD support the notion that the perception of stress can lead to an elevated ascorbic acid (AsA) and ellagic acid (EA) content (see Objective 6); however, the response was not consistent across all deficit treatments. The mechanism behind stress-induced increases in antioxidants is unclear, although methyl jasmonate (MeJA) may be involved (Wang 1999). To increase fruit antioxidant content reliably and consistently, we need to understand more about the signals that regulate fruit AsA and EA production. This work is currently being carried out in WU0110.

Experiments on Poinsettia (EMR)Both PRD and RDI were imposed on a commercial poinsettia crop to try to control plant height without reliance on undesirable and costly plant growth regulators (see Objective 6). Both techniques effectively limited height but RDI was deemed to be the easier technique to implement on a commercial scale using

SID 5 (Rev. 3/06) Page 15 of 29

26/09 27/10 25/11

Hor

mon

e de

liver

y ra

te (n

g h

-1)

0

20

40

60

80

100

120

140

160

*

26/09 27/10 25/110.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Date of measurement26/09 27/10 25/11

0

1

2

3

*

*

ABA IPAR ZRCC WW RDI-GP1

Figure 2.5. Effects of RDI on deliveries of ABA and the cytokinins isopentenyl-adenosine (IPAR) and zeatin riboside (ZR) in xylem sap collected from pressurised detopped poinsettia root systems. Delivery rates were calculated from multiplying sap flow rates by hormone concentrations quantified by GC-MS-SIM. Results are means of sap samples collected from six replicate plants; asterisks indicate statistically significant differences (P ≤ 0.05) from CC values.

flood-and-drain irrigation systems that many of the larger growers use. Therefore, work from 2007 onwards concentrated on developing RDI regimes that could be used on a commercial scale. The effects of RDI on root-to-shoot signalling were investigated to try to identify the signals that regulated

poinsettia stem and shoot responses to deficit irrigation.

ResultsStomatal conductance was significantly reduced within seven days of imposing RDI and remained lower than well-watered controls throughout the five-week deficit regime. During this period, diurnal measurements showed that stomatal conductances were similar in all treatments at the beginning of the photoperiod (08:00 in late October) but values were reduced in RDI-treated plants during the middle of the day and remained lower than in well-watered controls for the rest of the photoperiod (Figure 2.4). When sampled in the same week as the diurnal data were collected, xylem sap delivery of ABA from detopped, pressurised roots was increased 2.5-fold by RDI (Figure 2.5). This increased output of ABA presumably triggered and maintained stomatal closure, but reduced cytokinin deliveries may also have contributed (Figure 2.5). After the RDI regime was ended, output of ABA returned to pre-stress levels and was similar in all treatments by simulated market date (25 November). Deliveries of root-sourced IAA were unaffected by the RDI regime.

DiscussionPoinsettia stem height was effectively reduced by RDI (see Objective 6) but the ways in which deficit irrigation limits stem extension in poinsettia, or in other ornamental crops, is not yet known. Although ABA was traditionally viewed as a growth inhibitor, more recent research suggests that it can help to maintain shoot growth in plants exposed to drying soil by limiting the growth-inhibiting effects of ethylene. However, foliar ethylene production was not increased in RDI-treated plants so the reduced internode lengths in plants exposed to mild soil drying must be regulated in other ways. A reduction in gibberellin (GA) biosynthesis and/or transport, increased metabolism of active GAs to inactive metabolites or altered signal transduction are likely lo be involved. The regulation of GA biosynthesis and transduction in deficit-grown plants will be investigated in future work.

Plants previously exposed to mild soil drying by imposing RDI during the period of rapid shoot extension have an improved shelf-life potential and tolerance to chilling temperatures when tested several months later (see Objective 6). These results imply that exposure to a mild form of stress (e.g. soil drying) can pre-condition plants so that they are more tolerant of stressful conditions encountered later. Our analyses showed that root-to-shoot signalling had returned to pre-stress levels by market date in plants previously exposed to RDI and it remained unchanged at the end of the shelf-life tests. The mechanisms that bestow this enhanced stress tolerance are not yet known but may arise from an

SID 5 (Rev. 3/06) Page 16 of 29

Figure 2.7. Whole pot soil water content (a) leaf water potential (b), and xylem ABA concentration (c) of WW (), PRD-F () and PRD-A (▼) plants for which irrigation was alternated at 1000 h. Each point represents a single plant and regression lines were fitted to each treatment. Further details of these experiments published in Dodd, Theobald, Bacon & Davies 2006. Functional Plant Biology 33, 1081-1089.

ABA-mediated increase in the production of reactive oxygen species (ROS) (e.g. hydrogen peroxide) and associated changes in antioxidant capacity during RDI. When stress is next encountered, the plant’s antioxidant system may then be able to scavenge ROS more efficiently than previously unstressed plants. Experiments designed to test this hypothesis form part of a concept note that we recently submitted to the HDC Protected Crops panel and to the HortLINK programme.

Experiments on Tomato (Lancaster)Experiments aimed to determine signals in xylem sap collected from detached leaves, thus avoiding concerns that hormone concentrations are altered between exiting the root system and arrival at sites of action in the leaves. Various hydraulic (leaf water potential -leaf) and chemical signals were assayed (xylem sap pH, and the plant hormones ABA, IAA, and various CKs) based on their perceived physiological relevance. The effects of alternating wet and dry sides of PRD plants was also investigated, to assist in determining when such alternation events should occur (see also Objective 5).

Results and DiscussionDeficit irrigation of tomato (50% of ET applied as PRD) decreased soil water potential (soil) by 0.4 MPa, leaf by 0.08 MPa, whole plant transpiration rate by 22%, whole plant leaf area by 25%, and increased leaf xylem ABA concentration ( [X-ABA]leaf ) 2.5-fold. Although PRD caused no detectable change in xylem CK concentration, it decreased the CK concentration of fully expanded leaves by 46% (Figure 2.6). Maintenance of xylem CK concentrations (rather than an increase) while transpiration decreased suggests that CK loading into the xylem also decreased as the soil dried. Since leaf CK concentration did not decline proportionally with CK delivery, other mechanisms such as CK

metabolism may also influence leaf CK status of PRD plants (Kudoyarova et al. 2007). For deficit irrigated tomato grown as above (50% of ET applied as PRD), alternation of wet and dry sides (PRD-A) increased [X-ABA]leaf up to 2-fold above that of plants where the wet and dry sides were fixed (PRD-F). Thus alternation further decreased stomatal conductance (Dodd et al. 2006). Differences in [X-ABA]leaf were detected within an hour of alternation but did not persist beyond the photoperiod of alternation (Figure 2.7c). [X-ABA]leaf

increased linearly as whole pot soil water content (pot) and leaf water potential (leaf) declined, but the difference in [X-ABA]leaf

between the two sets of PRD plants was not due to differences in either pot or leaf

SID 5 (Rev. 3/06) Page 17 of 29

Figure 2.6. Changes in xylem sap (a) and bulk leaf CK concentration (b) on 3 measurement occasions (Days 2, 4 and 5 after initiating PRD) for the compounds zeatin (hollow bars), zeatin riboside (cross-hatched bars) and zeatin nucleotide (filled bars). Xylem sap data are means of 8 replicates (comprising 2 samples per treatment per day from all four experiments) while bulk leaf data are means of 3 replicates. Further details of these experiments are published in Kudoyarova, Vysotskaya, Cherkozyanova & Dodd 2007. Journal of Experimental Botany 58, 161-168.

Figure 2.9. Simulated changes in soil water content surrounding (a), fractions of sap flow from (b) and root xylem ABA concentration of (c) wet () and dry () parts of the root system of PRD and DI plants. Leaf xylem ABA concentration (d) of DI plants equalled root xylem ABA concentration, and of PRD plants is modelled Further details published in Dodd, Egea & Davies 2008. Plant, Cell and Environment 31, 1263-1274.

(Figure 2.7a, b). In PRD-F plants, the unwatered part of the root system contributes proportionally less to the transpiration stream as the soil progressively dries (see Figure 2.8 below). In PRD-A plants, we hypothesise that re-watering the dry part of the root system allows these roots to contribute proportionally more to total sap flux, thus liberating a pulse of ABA to the transpiration stream as the root ABA pool accumulated during soil drying is depleted. Since the enhancement of [X-ABA] leaf caused by PRD-A increased as pot and leaf declined, an optimal frequency of alternation to maximise the cumulative physiological effects of this ABA pulse must consider possible negative impacts of leaf water deficit as soil water status declines.

Experiments on “Two root, one shoot” grafted plants (Lancaster)Further work with pot-grown tomatoes showed that fixed PRD either increased or decreased [X-ABA]leaf

compared to deficit irrigated plants depending on total soil water availability (Dodd 2007). In attempting to explain these divergent responses, it was considered necessary to assess the contribution of different parts of the root system to [X-ABA] leaf. This was achieved by developing a novel grafting procedure where a single shoot was grafted onto the root systems of two plants, and xylem sap collected from both detached leaves and each individual root system. Since particular roots can supply particular leaves in the canopy in so-called “sectorial” plants, approach grafting was avoided. Some work used sunflower as its cylindrical stems allowed good contact between the stem and sap flow sensors (necessary to assess the sensitivity of sap flow to soil drying). We hypothesised that decreasing sap flow from roots in drying soil would limit ABA export to the shoot during PRD, such that soil moisture heterogeneity would influence the relationship between leaf xylem ABA concentration and

total soil water availability.Results and DiscussionInitial work used a peat-based substrate with high water holding capacity to ensure that the irrigated root system of PRD plants was adequately supplied with water (soil > - 1 kPa). During PRD, once decreased below a threshold, the fraction of sap flow from drying roots decreased linearly with soil water content (Figure 2.8b). Root xylem ABA concentration increased in both DI and PRD plants as declined. Although [X-ABA]leaf

increased in DI plants as declined, in PRD plants [X-ABA]leaf actually decreased within a certain range (Dodd et al. 2008b). A simple model that weighted the ABA contributions of wet and dry root systems to [X-ABA]leaf according to the sap flow from each, better predicted [X-ABA]leaf of PRD plants than either the root xylem ABA concentrations derived from wet or dry root

SID 5 (Rev. 3/06) Page 18 of 29

Figure 2.8. Fractions of initial sap flow from wet (blue) and dry (red) parts of the root system during a typical PRD experiment (a). Arrows are when the wet side was watered. Relationship between soil water content and fraction of initial sap flow from the dry part of the root system (b). Each point in (b) is derived from a single experiment as in (a). Further details published in Dodd, Egea & Davies 2008. Plant, Cell and Environment 31, 1263-1274.

systems, or their mean. This model revealed that for the same whole pot soil water content, simulated [X-ABA]leaf was higher in PRD plants than DI plants with moderate soil drying, but continued soil drying (such that sap flow from roots in drying soil ceased) resulted in the opposite effect (Figure 2.9). That total soil water availability determined whether DI or PRD elicited a greater [X-ABA] leaf provides a possible physiological explanation for differences in plant yield between DI and PRD plants supplied with the same irrigation volumes (Figure 1.4).

All modelling thus described shows if the wet part of the root system is adequately watered, there is an optimal dry part soil water content to maximise ABA signalling from the entire root system (Dodd 2008; Dodd et al. 2008b; c). If the soil dries beyond this threshold, the wet and dry parts of the rootzone should be switched. However, this approach has yet to be trialled in the field, and in many field experiments with PRD, partial drying of the irrigated roots occurs if irrigation is infrequent (Kirda et al., 2004) and it is important to assess the implications for ABA signalling. Soil water status of the irrigated pot affected the relationship between the fraction of sap flow through the dry part of the root system and soil water content: although soil of the irrigated pot determined the threshold soil at which sap flow from roots in drying soil decreased, the slope of this decrease was independent of the wet pot soil (Dodd et al. 2008c). Further modelling predicted that the specific dry to maximise ABA signalling from the entire root system will vary according to wet (Dodd et al. 2008c), suggesting flexibility in dry

set points if the water status of the irrigated pot varies. Irrigation scheduling based on ABA modelling of PRD plants may inform and complement other plant- or soil-based methods of scheduling irrigation, but further work is required to validate this approach (Dodd et al. 2009).

Objective 3 To determine the effects of PRD on resource partitioning and root and leaf functioning

Experiments at Lancaster with glasshouse-grown tomato plants with roots split between 2 x 5 L pots revealed that total biomass did not differ between PRD and DI plants after four cycles of PRD, but PRD increased root biomass by 55% as resources were partitioned away from all shoot organs. When the crop was allowed to fruit on six trusses and ripe tomatoes were harvested over an eight week period, fruit yield was statistically similar between PRD and control plants, suggesting that biomass allocation within the shoot altered to maintain fruit yield at the expense of stem and leaf biomass (Mingo et al. 2004).

Further experiments showed that after two cycles of PRD, the promotion of root biomass in PRD plants was associated with the alternation of wet and dry compartments, with increased root biomass occurring in the re-watered compartment after previous exposure to soil drying (Figure 3.1). Promotion of root biomass in field-grown PRD plants may allow the root system to access water and nutrients that would otherwise be unavailable to control plants. This may contribute to the ability of PRD plants to maintain a similar leaf

water potential to conventionally irrigated plants, even when smaller irrigation volumes are supplied. While these deficit irrigation-induced changes in biomass allocation (leaf growth inhibition and

increased biomass allocation to the roots and fruits) may be advantageous in some crops, this would be a major disadvantage in lettuce where the shoot is harvested. When containerised (5 L pots) lettuce was subjected to four irrigation treatments (115%, 100%, 85% & 70% of measured ET in the greenhouse, increased root biomass correlated with decreased irrigation. Despite this, marketable heads (> 150g within an eight week growing period) were also achieved with the 85% irrigation treatment.

Decreased vegetative growth (biomass allocation to the leaves) was advantageous in other crops. In poinsettia, decreased vegetative growth (with no effect on reproductive growth) resulted in

SID 5 (Rev. 3/06) Page 19 of 29

Figure 3.1. Root biomass of DI and PRD plants after one (a) and two (b) drying cycles. Black bars indicate soil drying. Data are means ± S.E. of 5-6 replicate plants, with different letters above the bars indicating significant differences. Further details published in Mingo, Theobald, Bacon, Davies & Dodd 2004. Functional Plant Biology 31, 971-978.

more compact and robust marketable plants (see Objective 6). In strawberry, decreased vegetative growth allowed easier crop harvesting and increased the penetration of light to developing fruit. Deficit irrigation during the flower initiation phase (Sept-Oct) increased the number of trusses produced in the second cropping year. Despite the greater number of fruit, berry size was not reduced and so marketable yields were increased by up to 40% in the second cropping year.

To summarise, deficit irrigation (and especially alternating wet and dry parts of the rootzone in PRD) stimulated biomass allocation towards the roots (which may limit the severity of leaf water deficit) and fruits. The latter may have quality implications, as discussed below.

Objective 4 To determine the environmental conditions and substrates that maximise the benefits of PRD

Since the action of long distance signals of soil drying (Objective 2) is more effective at high evaporative demand (Loveys et al. 2004), the low evaporative demand commonly experienced under UK conditions may limit the effectiveness of PRD. An extensive data set on grapevine water use made available to us by the Adelaide group shows that PRD only provides tangible benefits (in terms of water saving) at atmospheric VPDs typically > 2 kPa. In contrast, the high humidity (RH 60 – 80%) and moderate temperatures (16-25C) maintained in commercial tomato glasshouses in the UK sets the range of achievable VPD between 0.4 and 1.3 kPa. Thus benefits of PRD are likely to occur only at limited times of the year when cloud cover is low and radiation load is high, so permitting the internal glasshouse environment to equilibrate more closely with the external when glasshouse vents are opened fully. Thus analysis of stomatal responses in our experiment with tomato grown on rockwool blocks showed that significant effects of PRD on stomatal conductance were typically restricted to periods of higher VPD.

The effect of imposing deficit irrigation (both PRD and DI) in a range of different substrates (a peat-based medium, a 3:1 sand: peat mix and a coir-based medium) on the physiology and yield of pot-grown raspberry and runner beans was investigated over two seasons with each crop. Irrigation treatments were based on maintaining the media of control plants at or near pot capacity, with DI and PRD receiving 50% of control irrigation. No significant differences between media were found for yield (or any other traits) in the runner bean experiments. In contrast, raspberry yield varied with substrate in each year. Averaged across irrigation treatments (since there was no interaction between irrigation treatment and substrate), raspberry yield was 42% lower in coir in 2005, and 39% lower in sand in 2006. The inconsistent effects of different substrates on raspberry yield from year-to-year negated attempts at explaining these responses in terms of either substrate moisture holding capacity and/or moisture release curves.

The influence of different substrates on the contribution of different parts of the root system to total sap flow and root xylem ABA concentration (see discussion in Objective 2) was assessed using “two root, one shoot” grafted plants as described above. During PRD, one pot (“wet”) was watered and other (“dry”) was not. While each substrate gave a unique relationship between soil matric potential of the dry pot (dry) and the fraction of photoperiod sap flow from roots in drying soil (Fdry) (Figure 4.1a) , Fdry decreased with root water potential (root) similarly in all substrates. Likewise, each substrate gave a unique relationship between soil matric potential (soil) and root xylem ABA concentration ( [ABA]root ) (Figure 4.1b) but [ABA]root increased similarly with decreasing root in all substrates. Across all substrates, whole plant transpiration was most closely correlated with the mean soil of both pots, and then with [X-ABA]leaf. A model of leaf xylem ABA concentration, which weighted ABA contributions of each root system according to the sap flow from each, showed that in some substrates the increase in [ABA]root from roots in drying soil was partially offset by a decrease in sap flow from roots in drying soil, such that in response to defined PRD treatment, the increase in shoot xylem ABA concentration ( [ABA]shoot ) varied from 1.3-3.8-fold according to the substrate. Thus a desired outcome of

SID 5 (Rev. 3/06) Page 20 of 29Figure 4.1 Relationship between substrate matric potential and fraction of initial sap flow from the dry part of the root system (a) and root xylem ABA concentration (b).

PRD (ABA-mediated restriction of whole plant transpiration) may be more readily achieved in certain substrates.

To summarise, choice of substrate in containerised plants will determine the precision with which desired deficit irrigation treatments can be imposed.

Objective 5 To determine the optimal scheduling of deficit irrigation in the fieldThe majority of PRD trials conducted to date have used fairly arbitrary decisions as to when to alternate the side of the root system that is irrigated. For example, in this project, in pot-grown runner beans, the impact of alternating every week versus every two weeks was compared. Although there was no difference in the number or quality of pods produced or in vegetative growth, stomatal conductance of PRD plants alternated every week was 30% lower than PRD plants alternated every two weeks. This suggests that root signal output decreased after the first week of root drying, indicating additional scope for saving water if the frequency of alternation is optimised.

A theoretical analysis (now substantiated by results described in Objective 2 – see Figure 2.9) suggests that the optimum time would be when the dry side has dried to a level where root signal output starts to diminish. Direct measurement of this signal (if practical) could provide a good indicator for scheduling the switch. As this is likely impractical, alternative approaches need to be considered. These include direct measurement of soil moisture (ideally with a prior calibration of soil moisture against signal intensity) and through the detection of other plant responses such as leaf wilting or stomatal closure (reviewed by Jones 2004a; b). These plant responses potentially provide a good indirect measure of the intensity of the drought signal although product quality may already be jeopardised if wilting has occurred.

Soil moisture sensing offers the advantage of ready automation, and once sensors are installed, data can be downloaded continuously by wireless data-link by the irrigation manager. Although such sensors are increasingly used in production systems, expense usually limits the number of sensors that can be installed in a given crop thus selection of “average” plants becomes important if appropriate management decisions are to be made. Work at EMR with RDI-grown poinsettia used a GP1 data logger (Delta-T Devices, Burwell, UK) to trigger irrigation based on soil moisture sensors (e.g. SM 200 probes) to provide a series of dry down cycles of various intensities. The soil moisture threshold at which irrigation was triggered was adjusted throughout the growing season to regulate stem extension, allowing the RDI-treated plants to remain within commercial height specifications.

A consistent relative response of stomatal conductance to PRD is frequently observed (Figure 5.1). However, a difficulty with using stomatal conductance to schedule irrigation, at least under many UK environments, is that stomatal aperture is sensitive not only to soil drying and root-shoot signalling but also to environmental conditions; and large day-to-day variation in mean conductances can be attributed to differences in environmental conditions. It follows that irrigation scheduling using stomatal measurements is most likely to be successful where one has a well-irrigated control as a reference.

Direct measurement of stomatal conductance can also be time-consuming (even with a transient-time porometer which can make a reading every 30 seconds) if measurements are required over a large spatial scale. An alternative is to use thermal imaging to measure leaf or canopy

SID 5 (Rev. 3/06) Page 21 of 29

Figure 5.1. Temporal changes of stomatal conductance of glasshouse-grown Phaseolus coccineus cv. Emergo in response to different irrigation treatments. Fluctuations in mean conductance on different days were primarily related to differing environmental conditions, with low conductance on days of very low humidity/high temperatures.

WW120

PRD100

PRD80

PRD60

RDI100

RDI80

Treatment

Con

cent

ratio

n (n

g g

-1 FW

)

0

20

40

60

80

Ethyl hexanoate

0

10

20

30

Ethyl butanoate

RDI60

WW120

PRD100

PRD80

PRD60

RDI100

RDI80

Treatment

RDI60

Figure 6.1. Effects of PRD and RDI on the production of two important strawberry flavour volatiles.

temperature which is directly related to stomatal conductance (higher temperatures indicate stomatal closure). Approaches to estimate stomatal conductance using canopy temperature combined with the use of wet and/or dry reference surfaces or measurements of environmental variables (humidity, air temperature, radiation and windspeed) have been summarised by Leinonen et al. (2006) and Guilioni et al. (2008).

To summarise, results from this study show several successful methodologies to schedule irrigation (when to water, and when to switch) although practicalities (ease of grower use; equipment expense or availability) are likely to dominate over physiology in choice of an appropriate technique. These methods are currently being used to schedule irrigation to field-grown strawberries (HL0187).

Objective 6 To determine the 'quality' of deficit-grown produceIntroductionAlthough irrigation management strategies such as PRD or RDI can deliver large water savings, these techniques will only be taken up by the UK horticultural industry if yields and product quality are maintained or improved. The potential to use these water saving strategies without incurring yield penalties is summarised under Objective 1 for six of the seven crops studied. In addition to yields, product quality is also crucial because producers must continue to meet the expectations of retailers and consumers with regard to product quality and shelf-life.

At the project outset (2004), both the Adelaide experiments and IRRISPLIT work with grapes and tomato had shown that if PRD was managed carefully, fruit quality could be improved resulting in substantial added value. Unfortunately, effects of PRD on aspects of crop quality are either not reported or are assessed only cursorily and so rigorous scientific analysis of quality attributes in deficit-grown plants is sparse. Having established that PRD or RDI regimes could be imposed in some crops without reducing yields, we quantified the effects of these techniques on various components of product quality, including those most commonly used by the industry and retailers.

Experiments on Strawberry (EMR)Retailers and consumers expect high quality fruit of good shape, colour and size, with a good aroma, sugar acid balance and shelf-life. These quality criteria can be difficult to meet in some seasons and reduced consumer confidence in UK-produced berries can impact on return sales and limit the overall profitability of the sector. The potential of PRD and RDI to deliver consistent improvements in berry quality was determined over three cropping seasons.

ResultsAppearance and textureBoth PRD and RDI regimes affected fruit colour; the brightness was reduced and the chroma (or intensity) was increased. These changes are indicative of a deeper, more-intensely red, coloured berry which may be more appealing to consumers. Fruit firmness was increased by the 60% PRD and RDI regimes and by the 80% PRD regime but this response was not consistent over the three cropping seasons.

Flavour and bioactive compoundsBerry soluble solids content (SSC – [BRIX]) was unaffected by PRD and RDI treatments and concentrations of sugars (sucrose, fructose, glucose) and organic acids (citric, malic, oxalic and ascorbic) were also maintained. However, RDI increased the production of several important flavour volatiles but some aroma concentrations were reduced by PRD (Figure 6.1).

Berry concentration of several antioxidants (e.g. phenols, ASA, EA) were increased by some PRD and RDI regimes and these differences were not due simply to less dilution caused by a reduced water content of the PRD- and RDI-grown fruit. Berry total antioxidant capacity was also increased by the 60% PRD regime.

DiscussionTexture and colour are important components of fruit quality that influence consumer appeal and SSC is used by retailers as a proxy measure of berry sweetness. Each of these attributes was either

SID 5 (Rev. 3/06) Page 22 of 29

Date

25/07/06 22/08/06 19/09/06 17/10/06 14/11/06

Pla

nt h

eigh

t (cm

)

0

5

10

15

20

25

30

Irrigation regimesapplied

Irrigation regimesended

4039383736 41 42 43 44 45 4635 47Week

StaplehurstControlWWRDI

Figure 6.2. RDI effectively controlled stem extension so that all plants were within height specifications at market date, despite receiving 90% fewer sprays of PGR’s than the controls. Results are means of 12 replicate plants with associated standard errors.

maintained or enhanced by PRD and RDI and the all-important ratio of sugars to acids was also unaffected. A recent report that strawberry quality could be significantly improved by RDI should be viewed with caution; berry size was reduced by 35% due to the severity of water deficits imposed (Terry et al., 2007) and so any differences were likely to be artefacts due to dilution effects.

Our work showed that the production of important flavour volatiles was significantly increased by RDI but these volatiles were either maintained or reduced by PRD. This was not due simply to a greater supply of precursors (carbohydrates) resulting from reduced vegetative growth since canopy areas were reduced to a similar extent by PRD and RDI. Many flavour volatiles are derived from products of photosynthesis and partial stomatal closure may limit their availability. The increase in berry antioxidant concentration in deficit-grown fruit has important implications both for human health and for improved berry shelf-life. We are currently trying to exploit the differential responses to PRD and RDI by developing a new deficit irrigation technique to promote both berry flavour and bioactive content without incurring yield penalties (WU0110). The potential to use deficit irrigation to improve fruit quality in field-grown strawberry production is being determined in our on-going HortLINK project (HL0187).

Experiments on Poinsettia (EMR)Unlike the industry sectors listed under Objective 1, delivering water savings into the protected ornamentals sector is not a high priority since water use by this industry is generally very low compared to others. Some growers use efficient flood-and-drain systems that recycle irrigation water and so their WUE is already high. However, growers using drip irrigation and mains water would benefit from irrigation management strategies that reduce water inputs and losses. All growers add fertiliser to the irrigation water (fertigation) and although exempt from NVZ regulations, the reduction in fertiliser inputs associated with effective scheduling and deficit regimes would help to offset the rising costs of fertilisers.

Control of plant height is a major issue for growers of protected potted ornamental plants. This is currently achieved by frequent applications of plant growth regulators (PGR’s) but these are costly and environmentally undesirable and are at risk of being withdrawn under EU legislation. Alternative, non-chemical methods of height control are needed to counter these legislative changes. Therefore, the aim of these experiments was to use deficit irrigation techniques to try to control plant height in poinsettia and reduce the reliance on PGR’s. The potentially beneficial effects of RDI on plant quality, and shelf- and home-life potential were also determined. Conditions during distribution and retailing were quantified and the potential of RDI to improve tolerance to stresses encountered during the distribution chain was also determined. Investigations into the chemical signals that impact on plant growth and shelf-life in deficit-grown poinsettia plants are described under Objective 2.

ResultsPlant quality at simulated market dateExperiments over three seasons confirmed that a six-week RDI regime imposed during the period of rapid growth successfully reduced stem extension so that plant heights were well within specifications at simulated market date (mid-late November) (Figure 6.2). RDI-treated plants received only one PGR spray shortly after pinching, compared to the nine or ten sprays needed to control height in the commercial controls (CC) and well-watered (WW) plants. Plant widths were unaffected by the treatments but the stage of cyathia (flower) development at simulated market date was delayed in RDI-treated plants. This delayed anthesis may have contributed to the improved overall quality of these plants during home-life tests (see below). In each season, overall plant quality at market date was maintained, and sometimes improved slightly compared to commercial controls, by the irrigation scheduling and RDI regimes (Figure 6.3).

SID 5 (Rev. 3/06) Page 23 of 29

03/12/07 10/12/07 17/12/07 24/12/07 31/12/07 07/01/08 14/01/08

Pla

nt q

ualit

y sc

ore

(1-5

)

0

1

2

3

4

5 CCWWRDIRDI-GP1

Date of measurement

10/12/07 17/12/07 24/12/07 31/12/07 07/01/08 14/01/08

Cum

mul

ativ

e le

af d

rop

per p

lant

0

10

20

30

40

A)

B)

Figure 6.3. Effects of the irrigation treatments on A) overall plant quality and B) cumulative leaf drop during a six-week shelf life test at EMR. In A) a score of 5 represents excellent quality, 3 is marketable while 1 represents very poor quality. Results are means of six replicate plants with associated standard errors.

Shelf-life potentialThe deterioration in plant quality during the six-week home-life test was slowed by both RDI regimes, compared to CC and WW plants (Figure 6.4A). RDI-treated plants remained in good condition throughout l the home-life tests which were ended in mid January. Greatly reduced rates of leaf abscission (Figure 6.4B), bract abscission and leaf and bract paling contributed to the higher quality of RDI-treated plants. Effects of distribution on shelf-life potentialPoinsettia are notoriously prone to chilling injury and temperatures below 10 ºC can lead to severe

defoliation and leaf and bract necrosis, which greatly reduces plant quality. Therefore, care is taken during distribution to ensure that exposure to chilling temperatures is minimised. Heated lorries are used for dispatch and the temperature in each lorry is checked immediately on arrival at depot. Ambient temperatures encountered by our experimental plants fluctuated at the different stages during distribution but were maintained above 12 ºC until plants arrived in-store. Despite clear labelling on the boxes (‘store at ambient temperature’), our experimental plants were placed in a cold-store (4 ºC) until they were retrieved by EMR staff later that morning. During the subsequent shelf-life test at EMR, leaf abscission was reduced by 33% in RDI-treated plants compared to CC plants. Our attempts to repeat this work in 2008 were unsuccessful because the experimental plants were sold in error when they arrived in store.

Effects of ethylene on shelf-life potentialAnecdotal evidence suggests that poinsettia are very susceptible to ethylene accumulation and that exposure to low concentrations during distribution can limit plant quality during shelf- and home-life. Our measurements in 2006 suggested that whole-plant ethylene production rates both at simulated market date and during shelf-life tests were very low (0.005 parts per million - ppm). Although these concentrations are far too low to impact on plant quality, it is likely that plants could encounter very high ethylene concentrations in depot and during transport to store that may be sufficient to trigger leaf drop and reduce quality.

In 2007, we were granted access to the Asda Erith depot to determine whether accumulations of ethylene during distribution subsequently limited shelf-life potential and reduced quality. Ambient ethylene concentrations were fairly low (< 1 ppm) at Staplehurst Nursery before dispatch, and in-store. However, ambient ethylene concentrations reached 5 ppm in the depot, presumably due to the presence of climacteric fruit (e.g. bananas, apples) and other ethylene- producing goods such as Christmas trees. But in dose-response tests at EMR, exposure to concentrations of up to 20 ppm for 12 h failed to alter any aspect of plant quality.

DiscussionOur experiments demonstrated unequivocally that plant height could be effectively controlled by RDI imposed during the period of rapid stem extension. Importantly, stringent plant specifications imposed by retailers at market date were met consistently, despite a 90% reduction in PGR use. Overall quality of RDI-treated plants at market date was maintained or improved and cyathia development was delayed slightly, compared to CC and WW plants. Similar results were obtained over three seasons (2006-2008) using the flood-and-drain benches at Staplehurst Nursery and irrigation management strategies such as RDI now need to be developed for a range of crops to help reduce this industry’s reliance on PGR’s (see ‘Future work’).

SID 5 (Rev. 3/06) Page 24 of 29

The shelf- and home-life of RDI-treated plants was also consistently improved. The improved shelf-life potential of RDI-treated plants was presumably bestowed by changes in the metabolism of deficit-grown plants. Since the deficit irrigation treatment and shelf-life tests were three months apart, RDI must have somehow ‘conditioned’ the plants so that leaf and bract abscission was reduced. Our work with RDI on other crops such as strawberry indicates that antioxidant levels can be improved in deficit-grown plants. We will determine whether the total antioxidant capacity is increased in RDI-treated poinsettia and whether this increased capacity to neutralise ROS generated during exposure to chilling temperatures underlies the improved shelf-life potential.

Ambient ethylene concentrations of up to 5 ppm were detected in the depot en route to the store. However, in tests, ethylene concentrations four times higher than this failed to impact on plant quality.

Other crops Potatoes were grown in a field on a commercial holding next to a traditionally managed crop. PRD and RDI treatments received 50% of the amount of water to maintain soil moisture at ≥ 95% of field capacity. Although the commercially managed crop received approximately two-fold more water (including rainfall), yield of all experimental treatments was only about 10% less, with no differences between PRD and RDI. However tuber quality of the conventional crop was high, while the experimental crop had a high incidence of common scab and poor skin finish. This (micro)biological response limits the applicability of deficit irrigation to potato, since the soil moisture content required to preserve tuber quality was far greater than that which would trigger a physiological drought response. However, we anticipate that substantial water savings can still be delivered by improving irrigation scheduling and targeting delivery of water more efficiently to help maintain tuber quality and reduce water and nutrient losses to the environment. This work is being carried out in field trials at EMR (WU0118).

Experiments with pot-grown (5L) lettuce imposed four irrigation treatments (nominally 115%, 100%, 85% & 70% of measured ET), with the latter two treatments decreasing ET by an average of 11% and 23%. There was no difference in ET between the two most generous watering regimes despite differences in soil water content at harvest, indicating wasteful water application in the highest treatment. The 100% irrigation treatment maximised marketable shoot fresh weight, but marketable heads were also achieved with the 85% irrigation treatment. The severest irrigation treatment resulted in dark-green, unmarketable heads, due to chlorophyll accumulation and bitter taste.

To summarise, while deficit irrigation had positive (or neutral) impacts on berry and fruit crops, significant risk of unmarketable crops exists in lettuce and potato subjected to deficit irrigation.

Objective 7 To ensure effective technology transfer To enable grower interaction and facilitate technology transfer, several trials were conducted on commercial holdings: poinsettia throughout the project (Staplehurst Nurseries, Kent), potatoes in 2006 (Hilton of Fern, Angus) and tomatoes in 2005 (Flavourfresh, Southport, Lancashire). The research at the first two enterprises is discussed in Objective 6, while that at Flavourfresh is briefly discussed below.

A trial was conducted with rockwool-grown tomatoes in a commercial greenhouse near Lancaster. Transplants were placed atop paired rockwool slabs, allowing equal root establishment between both slabs. Significant effects of PRD on stomatal conductance (up to 30%) and total sap flow (22% at mid-day) were typically restricted to periods of higher VPD when cloud cover was low and radiation load high. Although PRD did not significantly affect individual fruit fresh weight and mean weekly yield through the season, fruit SSC values increased by 16% and dry matter allocation to the fruit increased by 20%. Despite these advantages, higher setup costs likely will prevent adoption of PRD by the industry.

We have already transferred the technology and knowledge from this project to the UK soft fruit industry (HL0187, SF 107) and we have continued to interact closely with key representatives of the protected potted ornamental sector during the development of our HortLINK concept note. While industry representatives have been informed of our results with other crops, the outcomes of these trials does not, at present, seem to justify rapid technology transfer to industry.

A complete list of technology transfer activities can be found in Appendix 1

SID 5 (Rev. 3/06) Page 25 of 29

Future workSix new projects have been developed, based on the successful delivery of the project objectives:

HortLINK HL0187: Improving water use efficiency and fruit quality in field-grown strawberry, 2007-2012 (EMR)

WU0110: Developing novel water-saving irrigation strategies to produce fruit with more consistent flavour and quality and an improved shelf-life, 2007-2012 (EMR)

WU0118: Improving water use efficiency and tuber quality in potato production by optimising irrigation scheduling, 2008-2011 (EMR)

SF 107: Managing water, nitrogen and calcium inputs to optimise flavour and shelf-life in soil-less strawberry production. 2009-2012 (EMR)

EU-funded SIRRIMED project comprising deficit irrigation of tomatoes and potatoes (Lancaster - commencing October 2009)

A concept note on improving energy, water, and nutrient use efficiencies and reducing pesticide use in protected pot plant production has recently been submitted to the HortLINK programme (EMR)

Main implications of the findingsDuring the course of this project, we sought to identify opportunities to deliver substantial water savings while maintaining marketable yields and improving product quality in several sectors of UK horticulture. Five new research projects have already been developed as a direct result of the progress made in this project. These strategic and applied projects will assist Defra in meeting the objectives in Policy Area WU01 and will also deliver into WQ01 since precise and targeted irrigation will also reduce fertiliser inputs that should, in turn, help to reduce diffuse pollution. These approaches can be expected to deliver significant economic gains and more efficient production systems that optimise the use of valuable resources such as water, fertiliser and labour and reduce waste during production. Novel irrigation strategies developed in this project are also transferable to many other irrigated crops and new proposals on potted ornamental plants, raspberry and potted living herbs are currently being developed.

In terms of the potential to deliver water savings, PRD, DI and RDI are all very similar. However, our results indicate that specific irrigation strategies should be used with some cropping systems to ensure that yields and product quality are maintained or improved. For example, although PRD may lead to yield losses in substrate-grown plants compared to DI or RDI, both techniques may be suitable for field-grown crops. The severity of the deficit irrigation regimes is also critical to success; although the more severe regimes may use less water, crop yields and quality are nearly always reduced. Our work with strawberry has shown that substantial water savings can be made simply by scheduling irrigation effectively. This represents a low risk strategy for growers with none of the inherent risks associated with managing deficit irrigation. Tools for effective irrigation scheduling have been developed and tested in this project and, combined with efficient and targeted delivery of irrigation water and fertiliser, these offer immediate opportunities for improving WUE and nutrient use efficiency (NUE) in many growing systems (e.g. potato).

Of all of the crops studied, potato is the most important in terms of its water use. The potato sector uses at least 56% all water abstracted for irrigation in England, and 25% of all water used in agriculture; about half of this water is used to control ‘common scab’ (Defra WU0101). However, potatoes are largely grown in eastern regions where water resources are most threatened (Environment Agency, 2007) and abstraction rates are predicted to rise by a further 30% by 2050 (Defra WU0102). Although our results suggest that deficit irrigation can reduce potato tuber quality (skin finish) leading to losses in marketable yield, there are other opportunities for delivering water savings. Genetic improvement programmes are already underway in potato (e.g. HH3615SPC, HP0218) and can be expected to produce new commercial varieties with improved water use efficiency (WUE) within 15-20 years. This approach will eventually help to lessen the impact of the predicted climate change scenarios of dry summers, especially in Southern England (UKCIP08, Jenkins et al., 2007). In the short-term, improved irrigation scheduling and high-precision delivery methods will help to secure more immediate water savings without impacting on yields and tuber quality. These approaches are curremtly being developed in WU0118.

SID 5 (Rev. 3/06) Page 26 of 29

The large number of technology transfer activities undertaken by the project consortium has also helped to draw attention to issues of water availability and water use efficiency among growers in many different sectors, especially in regions such as the south and east where resources are most threatened. In addition, our work on deficit irrigation has highlighted the potential benefits on crop or plant quality that can be achieved with effective irrigation management strategies.References

ADAS (2003) Irrigation Best Practice Grower Guide – Top and Soft Fruit;Blackman PG, Davies WJ (1985) Journal of Experimental Botany 36, 39-48. Davies WJ, Bacon MA, Thompson DS, Sobeih W, Gonzalez Rodriguez L (2000) Journal of Experimental Botany 51, 1617-1637.Davies WJ, Kudoyarova G, Hartung W (2005) Journal of Plant Growth Regulation 24, 285-295. Davies WJ, Zhang J (1991) Annual Review of Plant Physiology and Plant Molecular Biology 42, 55-76.Dodd IC (2005). Plant and Soil 274, 251-270.Dodd IC (2007) Functional Plant Biology 34, 439–448.Dodd IC (2008) Acta Horticulturae 792, 225-231. Dodd IC (2009) Journal of Experimental Botany in pressDodd IC, Beveridge CA (2006) Journal of Experimental Botany 57, 1-4Dodd IC, Davies WJ, Safronova VI, Belimov AA (2008a) Acta Horticulturae 792, 233-239. Dodd IC, Egea G, Davies WJ (2008b) Plant, Cell and Environment 31, 1263-1274.Dodd IC, Egea G, Davies WJ (2008c) Journal of Experimental Botany 59, 4083-4093.Dodd IC, Egea G, Davies WJ (2009) Acta Horticulturae in pressDodd IC, Theobald JC, Bacon MA, Davies WJ (2006) Functional Plant Biology 33, 1081-1089.Dry PR, Loveys BR, Botting D, During H (1996) Proceedings of the 9th Australian Wine Industry Technical Conference, 126-131 Else MA, Hall KC, Arnold GM, Davies WJ, Jackson MB (1995). Plant Physiology 107, 377-384.Gowing DJG, Davies WJ, Jones HG (1990) Journal of Experimental Botany 41, 1535–1540.Guilioni L, Jones HG, Leinonen I, Lhomme JP (2008) Agricultural and Forest Meteorology 148, 1908-1912Jia W, Davies WJ (2007) Modification of leaf apoplastic pH in relation to stomatal sensitivity to root sourced ABA signals. Plant Physiology 143, 68-77. Jimenez A, Creissen G, Kular B, Firmin J, Robinson S, Verhoeyen M, Mullineaux P (2002). Planta

214, 751-758.Jones HG (2004a) Journal of Experimental Botany 55, 2427-2436.Jones HG (2004b) Advances in Botanical Research Incorporating Advances in Plant Pathology 41, 107-163. Kirnak H, Kaya C, Higgs D, Bolat I, Simsek M, Ikinci A (2003). Australian Journal of Experimental

Agriculture 43, 105-111. Kudoyarova GR, Vysotskaya LB, Cherkozyanova A, Dodd IC (2007) Journal of Experimental Botany 58, 161-168Leinonen I, Jones, HG (2004) Journal of Experimental Botany 55, 1423-1431. Loveys BR, Stoll M, Davies WJ (2004) In ‘Water use efficiency in plant biology’ (Ed MA Bacon) pp. 113-141. (Blackwell: Oxford, UK)Mingo DM, Theobald JC, Bacon MA, Davies WJ, Dodd IC (2004) Functional Plant Biology 31, 971-978.Pudney S, McCarthy MG. (2004). Acta Horticulturae 664, 567-573.Sobeih WY, Dodd IC, Bacon MA, Grierson D, Davies WJ (2004) Journal of Experimental Botany 55, 2353-2363. Stoll M, Loveys B, Dry P (2000) Journal of Experimental Botany 51, 1627-1634.Tanaka Y, Sano T, Tamaoki M, Nakajima N, Kondo N, Hasezawa S (2005). Plant Physiology 138,

2337-2343.Wang SY (1999). Journal of Plant Growth and Regulation 18, 127-134.Yuan BZ, Sun J, Nishiyama S (2004). Biosystems Engineering 87, 237-245.

SID 5 (Rev. 3/06) Page 27 of 29

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

SID 5 (Rev. 3/06) Page 28 of 29

1. Bacon MA (2004) Water use efficiency in plant biology Blackwell Publishing, Sheffield, pp. 1-22. ISBN 1-4051-1434-7.

2. Loveys BR, Stoll M, Davies WJ (2004) In Bacon, M.A. (Ed) Water Use Efficiency of Plants. Blackwell. Oxford. pp113-141. ISBN 1-4051-1434-7.

3. Davies WJ, Hartung W (2004) In, New directions for a diverse planet: Proceedings of the 4th International Crop Science Congress Brisbane, Australia, www.cropscience.org.au.

4. Mingo D, Theobald JC, Bacon MA, Davies WJ, Dodd IC (2004) Functional Plant Biology 31, 971-978

5. Grant OM, Stoll M, Jones HG (2004) Journal of Horticultural Science and Biotechnology 79, 125-130.

6. Jones HG (2004) In Water use efficiency in plant biology. (ed. M.A.Bacon), Blackwell Publishing, Sheffield, pp. 27-41. ISBN 1-4051-1434-7.

7. Leinonen I, Jones, HG (2004) Journal of Experimental Botany 55, 1423-1431. 8. Jones HG (2004) Journal of Experimental Botany 55, 2427-2436.9. Sobeih W, Dodd IC, Grierson D, Bacon MA, Davies, WJ (2004) Journal of Experimental Botany

55, 2365-2384. 10. Jones HG (2004) Advances in Botanical Research Incorporating Advances in Plant Pathology 41,

107-163. 11. Davies WJ, Kudoyarova G, Hartung W (2005) Journal of Plant Growth Regulation 24, 285-295. 12. Dodd IC, Davies WJ (2005) In: Plant Hormones: Biosynthesis, Signal Transduction, Action! P.J.

Davies ed. Kluwer Academic Publishers, Dordrecht, The Netherlands. pp 493-51213. Dodd IC (2005) Plant and Soil (Root physiology : from gene to function Special Issue) 274, 257-

275.14. Pospisilova J, Dodd IC (2005) In Handbook of Photosynthesis 3rd edition (Ed: M Pessarakli)

Marcell Dekker, pp 811-825.15. Dodd IC, Beveridge CA (2006) Journal of Experimental Botany 57, 1-416. Davies WJ (2006) In Plant Growth and Climate Change. (Eds. JIL Morison and M Morecroft)

Blackwells, Oxford. pp 96-117. 17. Dodd IC, Theobald JC, Bacon MA, Davies WJ (2006) Functional Plant Biology 33, 1081-1089.18. Grant O, Chaves MM, Jones HG (2006) Physiologia Plantarum 127, 507-518.19. Dodd IC (2007) Functional Plant Biology 34, 439–448.20. Kudoyarova GR, Vysotskaya LB, Cherkozyanova A, Dodd IC (2007) Journal of Experimental

Botany 58, 161-16821. Jia W, Davies WJ (2007) Plant Physiology 143, 68-77. 22. Morison, JIL, Baker NR, Mullineaux PM, Davies WJ (2007) Philosophical Transactions of the

Royal Society 363, 639-658. 23. Dodds PAA, Taylor JM, Else MA, Atkinson CJ, Davies WJ (2007) Acta Horticulturae 744, 295-

302. 24. Jones HG (2007) Journal of Experimental Botany 58, 119-130. 25. Grant OM, Tronina L, Jones HG, Chaves MM (2007) Journal of Experimental Botany 58, 815-825. 26. Stoll M, Jones HG (2007) Journal International des Sciences de la Vigne et du Vin 41, 77-84 27. Dodd IC, Egea G, Davies WJ (2008) Plant, Cell and Environment 31, 1263-1274.28. Dodd IC (2008) Acta Horticulturae 792, 225-231. 29. Dodd IC, Davies WJ, Safronova VI, Belimov AA (2008) Acta Horticulturae 792, 233-239. 30. Jones HG (2008) Acta Horticulturae 792, 391-40331. Loveys BR, Jones HG, Theobald JC, McCarthy MG (2008) Acta Horticulturae 792, 421-427.32. Guilioni L, Jones HG, Leinonen I, Lhomme JP (2008) Agricultural and Forest Meteorology 148,

1908-191233. Dodd IC, Egea G, Davies WJ (2008) Journal of Experimental Botany 59, 4083-4093.34. Dodd IC, Egea G, Davies WJ (2009) Acta Horticulturae in press35. Dodd IC (2009) Journal of Experimental Botany in press

SID 5 (Rev. 3/06) Page 29 of 29