General enquiries on this form should be made to:Defra, Procurements and Contracts Division (Science R&D Team)Telephone No. 0207 238 5734E-mail: research.competitions@defra.gsi.gov.uk

SID 5 Research Project Final Report

SID 5 (Rev. 07/10) Page 1 of 33

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 HH3509SFV

2. Project title

Targeting Phosphorus-Fertiliser Applications to Roots of Wide-Row Crops

3. Contractororganisation(s)

Warwick HRIUniversity of WarwickWellesbourneWarwickCV35 9EF     

54. Total Defra project costs £      (agreed fixed price)

5. Project: start date................ 01 April 2005

end date................. 31 March 2010

SID 5 (Rev. 07/10) Page 2 of 33

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.

Plants require phosphorus (P), which they acquire from the soil solution as phosphate. Agriculture relies on large inputs of P fertilisers to maintain crop yields and quality. However, excess P fertiliser may be leached from soils as soluble phosphates, or lost by erosion as P bound to soil particles. This can lead to diffuse pollution of surface waters, nutrient enrichment of adjacent environments, and a consequent loss of habitats and decline in biodiversity. Therefore, it makes financial and environmental sense to optimse P fertilisation of crops.

It has been observed that both P-inputs, and the P lost to the environment, can be reduced in some wide-row crops (crops planted in widely spaced rows) by placing a concentrated ‘starter’ P-fertiliser in the vicinity of the developing root system, rather than by broadcasting P-fertilisers. The placement of starter fertiliser close to the roots appears to improve the uniformity of a crop, as well as accelerating rooting and early plant growth. However, the agronomy of P-fertiliser placement has not been investigated systematically, and the physiological traits required of plants to optimise placed P-fertilisers are unknown. The purpose of this project was, therefore, to determine to what extent P inputs to wide-row crops can be reduced by placing starter P-fertiliser within the root zone. The question was addressed for wide-row crops generally, and then specifically for potato. The later work focused on elucidating root traits important for P acquisition by young potatoes.

This project delivered to SID1 WQ01 objectives, which seek to minimise the adverse impacts of UK agriculture on water quality. HH3509SFV delivers specifically to the WQ01 scientific objective, to mitigate transportation of pollutants into watercourses, notably the management of fertiliser nutrient additions to mitigate losses to water systems. HH3509SFV delivers to these by (1) assessing the potential for reducing inorganic phosphate fertiliser inputs to widely spaced vegetable and potato crops by placing the fertiliser into the rooting zone of the crop and (2) identifying potato root traits that improve capture of placed fertiliser, and develop a predictive model of the potato root system, to assist future breeding programmes in the development of varieties with improved nutrient capture by their roots. HH3509SFV also delivers to other Defra activities including the Water Quality Division’s interests in the Water Framework Directive and the activities of the Nutrient Management Unit.

Scientific Objectives

Objective 01. To assess the potential of P-fertiliser placement to reduce P inputs to wide-row crops (within 12 months)

SID 5 (Rev. 07/10) Page 3 of 33

Objective 02. To perform three field trials to determine the minimal P-fertiliser placement required to maintain potato yields (within 48 months).

Objective 03. To characterise the early growth, P acquisition and rooting patterns of potato genotypes in glasshouse experiments (within 54 months).

Objective 04. To identify potato genotypes that respond well to P-fertiliser placement in field trials (within 54 months).

Objective 05. To develop a mathematical model describing the development of the potato root system and its affect on P acquisition (60 months).

Initially field trials were conducted to determine the effect of P-fertiliser placement on the yield and mineral content of potato tubers, onions, carrot roots, cabbages, lettuces and maize cobs. A solid-fertiliser placement system was used to place triple-super-phosphate (TSP) fertiliser into the soil (7.62 cm (3”) below and 7.62 cm (3”) to the side of the seed/transplant) at half the rate recommended by RB209 fertiliser recommendations. Equivalent yields and mineral composition of produce were obtained by placing the fertiliser in comparison with broadcasting the recommended amount of TSP fertiliser.

Three years field trials were conducted to establish the optimal amount and placement of P-fertiliser for the potato crop. In these experiments, the amount, depth and distance of placement from the seed tuber of P-fertiliser were varied systematically. There was significant variability between years in terms of crop traits measured and no consistent effect of any of the treatments was observed. Consequently, there is little evidence to support the use of P fertiliser placement for the potato crop on sandy loam soil with a soil P index of 3 or greater. There may be benefits to using this system in other soil types with low soil P indices, but there is likely to be considerable year to year variation in the effectiveness of P fertiliser placement.

To investigate the genetic potential of potato plants to exploit placed P-fertilisers, the root systems of contrasting potato genotypes were studied in glasshouse experiments and the yield potential with placed P fertiliser was assessed in the field. Non-destructive imaging of the potato root systems in a rhizobox system was used to quantify these traits. Despite higher levels of replication, the strong environmental influence on root development resulted in no significant differences being observed between genotypes for root and shoot biomass, root area and root system growth rate. Significant differences were observed between genotypes for the root and shoot P concentrations, indicating some differences in the abilities of these genotypes to acquire P from the soil. No differences were observed between genotypes in their ability to exploit placed P fertiliser by expanding their root system around the placed fertiliser. However, in the field, significant differences in tuber yields between genotypes were observed.

The final aim of this project was to develop a mathematical model describing the development of the potato root system and its affect on P acquisition. An explicit algorithm for modelling water movement and solute transport in soil was been developed. The algorithm is robust and much simpler than the traditional algorithm, which is potentially very useful in modelling water and nutrient dynamics in the soil-plant system. A simple 2-D root growth model was then devised, parameterised and tested in modelling water and nitrogen dynamics for a wide range of vegetable and arable crops. A sophisticated 3-D model which simulates root growth and water and P uptake was also devised based on the work by Somma et al. (1999). It mechanistically simulates root growth, water and P uptake, and water movement and P transport in soil, and thus is a powerful tool to study the mechanism of water and P uptake by the entire root system. However, the model has dozens of parameters and some of them are difficult to determine. Further work is required to validate such a model.

A model to study the effect of P on potato growth PHOSMOD, based on Greenwood et al. (2001) and Zhang et al. (2007), was also updated. Results show that the model performs reasonably to predict the effect of P fertiliser on potato yield and crop %P.

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

SID 5 (Rev. 07/10) Page 4 of 33

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. 07/10) Page 5 of 33

8.1 Background

Plants require phosphorus (P), which they acquire from the soil solution as phosphate (Raghothama 1999; Hammond et al., 2004). Horticulture relies on large inputs of P fertilisers to maintain crop yields and quality. However, horticultural crops generally recover less than 10% of any broadcast P fertiliser in the year it is applied [Greenwood 1981; Laegreid et al., 1999). Unfortunately, some of the excess P may be leached from soils as soluble phosphates, or lost by erosion as P bound to soil particles (Sharpley 1995, Withers et al., 2001). This can lead to diffuse pollution of surface waters, nutrient enrichment of adjacent environments, and a consequent loss of habitats and decline in biodiversity. Therefore, it makes financial and environmental sense to optimise P fertilisation of horticultural crops.

It has been observed that both P-inputs, and the P lost to the environment, can be reduced in some wide-row crops by placing a concentrated ‘starter’ P-fertiliser in the vicinity of the developing root system, rather than by broadcasting P-fertilisers (Black 1992; Stone 2000ab; Dampney et al., 2002). The placement of starter fertiliser close to the roots appears to improve the uniformity of a crop, as well as accelerating rooting and early plant growth. Inefficient P fertilisation is a particular problem for the potato crop, which is grown on about 2.5% of arable land and consumes about 8% of the P-fertiliser applied in the UK, and land growing potatoes contributes significantly to P enrichment of surface waters (Dampney et al., 2002). Evidence in the literature suggests that placing P-fertilisers close to the roots of young potato plants can result in more efficient capture of fertiliser P and often results in a small increase in yield (Prummel, 1957; Verma and Grewal 1979; Kingston and Jones 1980; Lewis and Kettelwell 1992; Sparrow et al., 1992; Hegney et al., 1999; Anon et al., 2004). However, the agronomy of P-fertiliser placement has not been investigated systematically, and the physiological traits required of plants to optimise placed P-fertilisers are unknown. The purpose of this project was, therefore, to determine to what extent P inputs to wide-row crops can be reduced by placing starter P-fertiliser within the root zone. The question was addressed for wide-row crops generally, and then specifically for potato. The later work focused on elucidating root traits important for P acquisition by young potatoes.

The project has five objectives: (1) To identify wide-row horticultural crops that may benefit from P-fertiliser placement. This was achieved through a review of available literature complemented with new data from field trials assaying the effect of P-fertiliser placement on the yield of maize, cabbage, carrot, lettuce, maize, onion and potato. (2) To determine the minimal P-fertiliser placement required to maintain yields of potato. This was addressed by field experiments in which the P-fertiliser concentration and width of placement were systematically varied. (3) To characterise the genetic diversity and heritability of rooting, P acquisition and early growth traits of 18 potato varieties in the glasshouse. This provided data on the ability of different potato genotypes to exploit placed P-fertiliser. (4) To identify potato genotypes that respond well to P-fertiliser placement in the field. (5) To develop a mathematical model describing the development of the potato root system and its affect on P acquisition, plant growth and yield. It will be used both to identify root traits that may impact on the interception and uptake of placed P-fertiliser and to predict the effects of P-fertiliser placement strategies on plant growth and crop yield.

8.2 Policy targets

In 2005, the primary Defra policy objective underpinning HH3509SFV (HH35 ROAME A) was to support the horticultural and potato industry in reducing inorganic fertiliser inputs in order to protect and improve the environment, to preserve biodiversity, to promote a sustainable supply of high-value crops to the food chain, and to preserve natural resources. Subsequently, HH35 policy objectives have been superseded by SID1 WQ01 objectives, which seek to minimise the adverse impacts of UK agriculture on water quality. HH3509SFV delivers the WQ01 scientific objective, to mitigate transportation of pollutants into watercourses, notably the management of fertiliser nutrient additions to mitigate losses to water systems. HH3509SFV delivers to these by (1) assessing the potential for reducing inorganic phosphate fertiliser inputs to widely spaced vegetable and potato crops by placing the fertiliser into the rooting zone of the crop and (2) identifying potato root traits that improve capture of placed fertiliser, and develop a predictive model of the potato root system, to assist future breeding programmes in the development of varieties with improved nutrient capture by their roots. HH3509SFV also delivers to other Defra activities including the Water Quality Division’s interests in the Water Framework Directive and the activities of the Nutrient Management Unit.

6

8.3 Scientific Objectives

Objective 01. To assess the potential of P-fertiliser placement to reduce P inputs to wide-row crops (within 12 months)

Objective 02. To perform three field trials to determine the minimal P-fertiliser placement required to maintain potato yields (within 48 months).

Objective 03. To characterise the early growth, P acquisition and rooting patterns of potato genotypes in glasshouse experiments (within 54 months).

Objective 04. To identify potato genotypes that respond well to P-fertiliser placement in field trials (within 54 months).

Objective 05. To develop a mathematical model describing the development of the potato root system and its affect on P acquisition (60 months).

8.4 Objective 01. The potential of P-fertiliser placement to reduce P inputs to wide-row crops

8.4.1 Literature Review

A literature review was conducted to provide a general appraisal of the effectiveness of P-fertiliser placement for reducing P-fertiliser inputs to wide-row crops. Insights obtained from this review have been incorporated into recent publications (White et al., 2005, 2007 White and Hammond 2008). A brief summary of the review is presented here.

In general, placing compound (NPK) fertilisers adjacent to the root system of wide-row crops has been found to reduce their fertiliser requirements and improve their uniformity (Dampney et al., 2002; Black, 1992; Stone 2000ab). This practice also reduces the losses of N, P and K to the environment. These phenomena have been attributed to an acceleration of rooting and early plant growth. This research led to MAFF to suggest that the injection of high phosphate liquid ‘starter’ fertiliser 2-3 cm below the seed or around the roots of transplants will improve the growth and quality of crops such as bulb and salad onions, lettuce and leeks in their RB209 handbook (MAFF, 2000), with a caution that a maximum of 60 kg/ha phosphate should be applied in this manner to avoid ‘salt burn’, and editions of MAFF’s RB209 up to 1983, and current SAC/MLURI fertiliser recommendations, to advocate a reduction in the rates of P-fertilisation of potatoes by 20 to 25% if fertiliser is placed. The fertiliser industry and machinery manufacturers echo these recommendations in their literature, worldwide. Since the RB209 recommendations are the industry standard, it is important that they are adequately researched to ensure their credibility and acceptance by the agricultural and horticultural industries. However, a Defra-funded review by Dampney et al. (2002) concluded that there was an urgent need for new research to improve P-fertiliser recommendations for modern potato varieties cultivated on soils with a high P-Index (3 and above) using current agricultural practices such as fertiliser placement.

In contrast to the interest on placement of compound fertilisers, few studies have been undertaken to determine the effects of straight P-fertiliser placement on crop yield. In the arable sector, maize is commonly fertilised through banded or placed applications, and it has been observed that this agronomic practice is effective in increasing yields on soils with a low P-Index [e.g. Stone, 2000a; Smith et al., 1990; Sanchez et al., 1991; Lu et al., 1993; Okalebo et al., 1994; Dachler and Kochl 1995; Buerkert and Hiernaux, 1998; Withers et al., 2000; Buerkerrt et al., 2001; Roth et al., 2003). However, comparable studies are rare in horticultural crops. Nevertheless, several studies suggest that the agronomic efficiency of P-fertilisers is increased, the P lost to the environment is reduced, and a small increase in yield is obtained when P-fertiliser is targeted to the roots of wide-row crops, such as brassica, carrot, lettuce, onion and potato (Prummel, 1957; Ryan, 1962; Reith, 1972; Henriksen, 1987; Costigan and Heavyside, 1988; Smith et al., 1990; Brewster, 1991; McPharlin and Robertson 1997; Stone 1998, 2000ab; Dampney et al., 2002). It is expected that the targeting of a concentrated fertiliser close to the root system will benefit the establishment of most crops, but may be particularly useful for crops whose root systems are shallow or slow to exploit the inter-row space, such as lettuce, onion and potato.

Placing P-fertilisers close to the roots of young potato plants generally results in more efficient capture of fertiliser P, and often results in a small increase in yield (Prummel, 1957; Verma and Grewal 1979;

7

Harris, 1992; Lewis and Kettlewell, 1992; Sparrow et al., 1992; Hegney and McPharlin, 1999; Anon, 2004)., although care must be taken to avoid high salt concentrations in the rhizosphere, which inhibit root growth (Harris, 1992; Hegney and McPharlin, 1999). However, the response of potatoes to P-fertiliser placement has not been investigated systematically and, as observed by Dampney et al. (2002), there is an urgent need to determine the minimal P fertiliser placement required to maintain yields of commercial potato varieties using modern technology. This justifies the field trials of the current project in which the amount, depth and distance of P-fertiliser placement from the seed tuber will be varied systematically (see below).

8.4.2 Field Trials of Fertiliser Placement (2005-2006)

Field trials were conducted to determine the effect of P-fertiliser placement on the yield and mineral content of potato tubers, onions, carrot roots, cabbages, lettuces and maize cobs. A solid-fertiliser placement system was used for the reproducible placement of P-fertilisers. Triple-super-phosphate (TSP) was either placed into the soil (7.62 cm (3”) below and 7.62 cm (3”) to the side of the seed/transplant) at half the rate recommended by RB209 fertiliser recommendations or surface broadcast at the recommended rate. Data indicate that equivalent yields (Table 1) and mineral composition of produce (Table 2) were obtained by placing half the amount of P-fertiliser recommended in England and Wales close to the plant at planting as broadcasting the recommended amount at this time. For salad onions and carrots there was a clear yield advantage achieved through the placement of fertiliser into the soil at sowing. For potato and lettuce crops, equivalent yields were achieved, and for cabbage and maize, slight yield reductions were observed. Whilst these results represent potential financial and environmental benefits from reductions in the use of phosphate fertilisers, additional trials on different soil types and across multiple years are required to confirm these advantages. Soil type and weather conditions have previously been shown to affect the effectiveness of fertiliser placement (Prummel, 1957).

Table 1. Yield of potato tubers, onions, carrot roots, cabbages, lettuces and maize cobs obtained when either the amount of P-fertiliser recommended in MAFF’s RB209 was broadcast or half this amount was placed close to the seed or transplant at planting.

Placed Broadcast Yield RatioMean +/- SEM Mean +/- SEM Placed/Broadcast

Potato (kg/plot, tubers) 18.9 +/- 1.1 18.3 +/- 0.7 1.03Salad onion (g/plant) 18.7 +/- 1.2 16.3 +/- 1.2 1.14Carrot (g/plant, root) 69.9 +/- 6.2 56.6 +/- 5.5 1.23Cabbage (g/plant, heart) 889 +/- 11 924 +/- 11 0.96Lettuce (g/plant) 546 +/- 28 533 +/- 43 1.02Maize (kg/plot, cobs) 4.70 +/- 0.12 5.50 +/- 0.09 0.85

Table 2. Mineral concentrations in potato tubers, onions, carrot roots, cabbages, lettuces and maize cobs obtained when either the amount of P-fertiliser recommended in MAFF’s RB209 was broadcast or half this amount was placed close to the seed or transplant at planting.

N (% DM) P (% DM) K (% DM) Ca (% DM) Mg (% DM)

Mean +/- SE Mean +/- SE Mean +/- SE Mean +/- SE Mean +/- SEPotato tubers Placed 1.37 +/- 0.02 0.261

+/- 0.004 1.57 +/- 0.02 0.043 +/- 0.001 0.090 +/- 0.003

Broadcast 1.56 +/- 0.02 0.262 +/- 1.65 +/- 0.03 0.044 +/- 0.001 0.104 +/- 0.002

8

0.003Salad onion Placed 3.56 +/- 0.18 0.415

+/- 0.021 2.00 +/- 0.02 2.224 +/- 0.087 0.266 +/- 0.006

Broadcast 3.53 +/- 0.20 0.409+/- 0.023 2.02 +/- 0.19 2.297 +/- 0.088 0.278 +/- 0.009

Carrot roots Placed 1.09 +/- 0.10 0.345

+/- 0.023 1.63 +/- 0.07 0.421 +/- 0.006 0.200 +/- 0.002

Broadcast 1.17 +/- 0.11 0.343+/- 0.018 1.80 +/- 0.16 0.419 +/- 0.012 0.197 +/- 0.016

Cabbage Placed 3.91 +/- 0.13 0.356+/- 0.027 2.12 +/- 0.11 3.264 +/- 0.250 0.292 +/- 0.016

Broadcast 4.17 +/- 0.03 0.394+/- 0.009 2.33 +/- 0.07 3.046 +/- 0.213 0.282 +/- 0.005

Lettuce Placed 2.84 +/- 0.08 0.386+/- 0.010 3.21 +/- 0.11 0.962 +/- 0.031 0.546 +/- 0.022

Broadcast 2.61 +/- 0.10 0.387+/- 0.020 3.36 +/- 0.46 0.960 +/- 0.018 0.549 +/- 0.002

Maize Placed 1.34 +/- 0.04 0.243+/- 0.011 0.32 +/- 0.01 0.009 +/- 0.001 0.130 +/- 0.007

Broadcast 1.47 +/- 0.05 0.249+/- 0.010 0.33 +/- 0.01 0.008 +/- 0.000 0.134 +/- 0.007

8.5 Objective 02. Field trials to determine the minimal P-fertiliser placement required to maintain potato yields

8.5.1 Background

Three years of field trials were conducted to establish the optimal amount and placement of P-fertiliser for the potato crop. In these experiments, the amount, depth and distance of placement from the seed tuber of P-fertiliser were varied systematically (Table 3; Figure 1). All experiments were performed on Wharf Ground field, Wellesbourne, which had a soil P index of 3 (38 mg Olsen P L -1). Three replicate plots were grown for each fertiliser rate or placement position, using a randomised plot design. To soil that was K Index 1 (76 mg L-1) and had a Soil Nitrogen Supply (SNS) index 0, RB209 recommended rates for N and K for potatoes were applied.

Seed tubers were sown in early April and harvested in October in all three years. The crop was managed as a commercial crop, with the application of pesticides and irrigation applied when necessary. During the growth of the crop several parameters were recorded including, emergence rate, canopy development, canopy light interception and the nutrient status of diagnostic leaf samples. At commercial maturity, tubers were harvested, graded and analysed for their mineral nutrient status. For all samples analysed, tissue N, P, K, Ca, Mg, S, B, Cu, Fe, Mn, Na and Zn concentrations were determined.

Table 3. Potato mineral fertiliser applications incorporated prior to seed bed cultivations or placed at planting. Phosphate fertiliser rates for placement are measured as a percentage of the recommended broadcast rate.Crop Experiment Treatment N* K2O P2O5

9

kg ha-1 kg ha-1 Kg ha-1

Potato Optimal P fertiliser rate

Broadcast 185 350 130.0 Placed 100% 185 350 130.0Placed 50% 185 350 65.0Placed 25% 185 350 32.5Placed 10% 185 350 13.0Placed 5% 185 350 6.5Placed 0% 185 350 0.0

Potato Optimal P placement

Broadcast 185 350 1306” x 0” 185 350 656” x 3” 185 350 656” x 6” 185 350 653” x 0” 185 350 653” x 3” 185 350 653” x 6” 185 350 65

* this amount of N is determined by RB209, variety group 3, >120 days growing season

Figure 1. Statistical design for determination of optimal rates (2006 field trial shown as example), measured as a percentage of the recommended broadcast rate and placement locations treatments. P = Placed, B = Broadcast, s = distance placed to the side of seed, b = distance placed below the seed. Each plot is 3.04m wide (4 rows at 75cm centres plus furrow) and 4m long. There is a gap of 4m between each plot, therefore each bed = 52m long.

8.5.2 Results

Emergence rates of potato plants were monitored during crop establishment (Figure 2). Plants that can access nutrients earlier may emerge and establish quicker, resulting in an improved yield at harvest. Potato plants emerged between 17 and 30 days after planting. There was a significant (P=0.025) effect of treatment on emergence. This was most evident 20 days after planting, with plots supplied with 100% of the RB209 recommended amount of P fertilisers as a broadcast application having greater emergence than all other treatments. Within the rate experiment, placement of 25 and 50 % of the RB209 recommended amount of P fertilisers into the soil had higher emergence rates than placing 5, 10 or 100% of the RB209 recommended P fertiliser application (Figure 2A). However,

10

the no fertiliser application treatment had similar emergence to the placement of 25 and 50 % of the RB209 recommended P fertiliser application.

Days after planting

20 30 40

% e

mer

genc

e

0

20

40

60

80

100

100BC 0P 5P 10P 25P 50P 100P

Days after planting

20 30 40 50

100BC 3"s 0"b 3"s 3"b3"s 6"b6"s 0"b6"s 3"b6"s 6"b

A B

Figure 2. Emergence rates for (A) potato plants supplied with different rates (% of RB209 recommended phosphate fertiliser application) of inorganic triple super phosphate (TSP) fertiliser placed (3” to the side and 3” below the seed tuber) into the soil and for (B) potato plants supplied with 50% of the RB209 recommended rate of TSP at different locations relative to the seed tuber. P = Placed, BC = Broadcast, s = distance placed to the side of the seed tuber, b = distance placed below the seed tuber. Data points represent the REML means of four measurements per plot per time-point for three replicated plots over three years (n=36). Error bar represents the standard error of the difference.

Within the placement experiment, placing the fertiliser to the side of the seed tuber (0” below) resulted in the lowest emergence rates, as did placing the fertiliser 6” below and 6” to the side of the seed tuber (Figure 2B).

To monitor the continued development of the crop during the season, leaf area, intercepted radiation and elemental analyses of diagnostic leaves were measured. Over the three years of field trials there was no significant (P>0.05) effect of treatment observed for these measurements. There was a significant effect of year for all of these measurements, indicating year to year variation masked any effects of treatment. This is most notable in data for 2007, when the trial site was affected by flooding for part of the season.

Over the three years, the mean concentration of N, P and K in the diagnostic leaves declined over the growing season, whilst the concentrations of Ca and Mg increased, irrespective of treatment. There were significant positive correlations between leaf N, P and K concentrations and between Ca and Mg leaf concentrations over the growing season.

At commercial maturity, tubers were harvested and graded for size. Over the three years of field trials there was no significant (P>0.05) effect of treatment observed for tuber yield per plant (Figure 3). However, fertiliser treatment did have a significant effect on the weight of tubers in the >80mm category, with plots receiving 5 and 10% of the RB209 recommended P fertiliser application placed into the soil and plots where the fertiliser was placed 6" to the side of the seed tuber and either 3” or 6” below the tuber having greater tuber biomass in this grade.

11

Figure 3. Tuber yield per plant at commercial maturity for potato plants supplied with different rates (% of RB209 recommended phosphate fertiliser application) of inorganic triple super phosphate (TSP) fertiliser placed (3” to the side and 3” below the seed tuber) into the soil or supplied with 50% of the RB209 recommended rate of TSP at different locations relative to the seed tuber. P = Placed, BC = Broadcast, s = distance placed to the side of the seed tuber, b = distance placed below the seed tuber. Tubers were graded for size into <45mm (black), 45-65mm (light grey), 65-80mm (dark grey) and >80mm (white) and weighed. Bars represent the REML means of three replicated plots over three years (n=9). Error bar represents the standard error of the difference.

8.5.3 Summary

The purpose of Objective 02 was to optimise the placement of phosphate fertiliser to maintain or improve tuber yields in the potato crop, whilst reducing total P fertiliser inputs to the crop. However, there was significant variability between years in terms of the crop traits measured. No consistent effect of any of the treatments was observed. Consequently, there is little evidence to support the use of P fertiliser placement for the potato crop on sandy loam soil with a soil P index of 3 or greater. There may be benefits to using this system in other soil types with low soil P indices, but there is likely to be considerable year to year variation in the effectiveness of P fertiliser placement.

8.6 Objective 03. To characterise the early growth, P acquisition and root foraging patterns of potato genotypes

8.6.1 Background

To investigate the genetic potential of potato plants to exploit placed P-fertilisers, the root systems of contrasting potato genotypes were studied in glasshouse experiments to characterise and quantify genetic and environmental variation in these traits. For this study potato genotypes of heritage and commercial potato varieties available from commercial sources were used; Mayan Gold, Nadine, Kennebec, Maris Piper, Cara, Home Guard, Desiree, Lady Balfour, Pentland Dell, Estima, Red Duke Of York, Golden Wonder, King Edward, Marfona, Sutton Flourball, Irish Cobbler (American). These genotypes represent potato crops with different halum longevities, historic varieties (Sutton Flourball, Irish Cobbler) and varieties for which genetic information is available (Kennebec).

The developing root system was monitored non-destructively during growth using a Rhizobox system developed in association with Dr Andrew Thompson using information obtained in Defra project HH3615SPC (Objective 03/01). The Rhizobox was 400 mm (width) x 680 mm (length) x 60 mm (depth) and was sufficiently large to contain the roots of a young potato plant (Figure 4).

12

Treatment

100%

BC

0% P

5% P

10% P

25% P

50% P

100%

P3"s

0"b3"s

3"b3"s

6"b6"s

0"b6"s

3"b6"s

6"b

Tube

r yie

ld p

er p

lant

(kg)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Figure 4. The Rhizobox system used to grow potato plants and monitor the developing root system no destructively. The empty Rhizobox (A) is filled with compost and the developing root system can be imaged and analysed (B) before the whole root system is measured at the end of the experiment using a pin board to maintain some of the root system architecture (C).

The Rhizobox contained a glass front, and the unit was angled at 30° glass side down to encourage root growth down the glass to facilitate imaging of the root system (Figure 4). The whole Rhizobox was covered in a light-proof insulating jacket, to prevent the soil temperature from fluctuating during the day and having a negative impact on root growth. The glasshouse temperature was set to 18°C, with supplementary lighting. Rihzoboxes were watered automatically with deionised water via three drippers per box. Each Rhizobox was filled with 35 litres of John Innes No. 2 compost, which enabled the removal of soil at the end of the experiment to visualise the entire root system (Figure 5). The John Innes No. 2 compost was modified to contain black sand to improve automated detection of the root system. The Rhizoboxes were arranged in three rows, with six Rizoboxes in each row. Six experimental runs, were completed using a randomised block design, in which all varieties were grown in each run (n=6), and Kennebec was grown three times in each run (n=18).

Photographs of the developing root system were taken periodically and were analysed using a MatLab computer program developed by Dr Nick Parsons, (Warwick Medical School, University of Warwick) to estimate the growth rate, root density and morphology of the root system (Figure 5). Once the root systems of three of the control plants have reached the bottom of the Rhizobox system, plants were harvested. Pin boards were inserted into the Rhizobox and the root systems were washed and imaged (Figure 5C). The fresh and dry weights of the root and shoots and their tissue P concentrations are also determined.

8.6.2 Results

Root area and biomass data were collected for all 16 genotypes. Since root system development is strongly influenced by environment, six replicates for each genotype were analysed and 18 replicates for Kennebec. Despite this, no significant (P>0.05) effect of genotype was observed for root area, root system growth rate, root biomass or shoot biomass (Figure 6). There was a significant effect of genotype on root (P=0.009) and shoot (P=0.028) P concentrations, indicating differences between genotypes in their abilities to acquire P from the soil (Figure 6B). Interestingly, historic genotypes, Irish Cobbler and Suttons Flourball, had lower root biomass, root system growth rate and root area than more modern genotypes (Figure 6). There were also significant positive correlations between root and shoot biomass, and between shoot biomass and root area, root length and root system growth rate. The latter indicating the strong interaction between shoo and root development.

13

Figure 5. The root system of a potato plant growing in a Rhizobox under glasshouse conditions. Images of the root system (a) were taken periodically during growth. These images were filtered by the MatLab software (b) and processed to determine root length and density of the root system (c).

Bio

mas

s (g

DW

)

5

0

5

10

15

20Shoot

Root

Tiss

ue P

con

cent

ratio

n(%

DW

)

1.0

0.8

0.6

0.4

0.2

0.0

0.2

0.4

0.6

0.8

1.0Shoot

Root

Genotype

Irish C

obble

r

Sutton

s Flou

rball

Golden

won

der

Estima

Red D

uke o

f York

Homeg

uard

Cara

Marfon

a

Mayan

Gold

Nadine

King E

dward

Maris P

iper

Desire

e

Pentla

nd D

ell

Lady

Balf

our

Kenne

bec

Tota

l roo

t are

a (c

m2 )

0

200

400

600

Genotype

Irish C

obble

r

Sutton

s Flou

rball

Golden

won

der

Estima

Red D

uke o

f York

Homeg

uard

Cara

Marfon

a

Mayan

Gold

Nadine

King E

dward

Maris P

iper

Desire

e

Pentla

nd D

ell

Lady

Balf

our

Kenne

bec

Roo

t sys

tem

gro

wth

rate

(cm

2 day

-1)

0

2

4

6

A B

C D

Figure 6. Root and shoot biomass (A), root and shoot P concentration (B), total root area (C) and root system growth rate (D) for 16 potato genotypes grown in John Innes compost in glass fronted Rhizoboxes under glasshouse conditions. Plants were grown for between 30 and 45 days and imaged non-destructively during that period. At harvest, compost was washed from the root system and total root area, and root and shoot biomass recorded. Tissue P concentration was determined according to Hammond et al. (2009). Bars represent REML means from six biological replicates, except for Kennebec, which represents the REML mean from 18 replicates. Error bars represent standard errors of the difference.

14

Within the Rhizoboxes TSP fertiliser was placed 3’ below and to the side of the seed tuber, to investigate the interception and root development in and around the placed fertiliser. Images of the root system and root biomass for different regions of the root system were recorded to determine if the potatoes developed their root systems around the placed fertiliser. No significant differences between the genotypes were observed for root area or root biomass around the placed fertiliser in comparison to adjacent regions. It is known that plants expand their root systems in nutrient rich patches within the soil (Drew, 1975). Whilst the potato plants might also exhibit this trait, the sensitivity and resolution of the methods employed here were not sufficient to observe it.

8.6.3 Summary

The root systems of contrasting potato genotypes were studied in glasshouse experiments to characterise and quantify genetic and environmental variation in these traits and the ability of these genotypes to exploit placed P fertiliser. Non-destructive imaging of the potato root systems in a Rhizobox system was used to quantify these traits. Despite higher levels of replication, the strong environmental influence on root development resulted in no significant differences being observed between genotypes for root and shoot biomass, root area and root system growth rate. Significant differences were observed between genotypes for the root and shoot P concentrations, indicating some differences in the abilities of these genotypes to acquire P from the soil. No differences were observed between genotypes in their ability to exploit placed P fertiliser by expanding their root system around the placed fertiliser.

8.7 Objective 04. To identify potato genotypes that respond well to P-fertiliser placement in field trials

8.7.1 Background

To complement glasshouse trials in Objective 03, looking at differences in root system development between potato genotypes, field trials were established to investigate the field performance of a selection of these genotypes. The following genotypes were grown in the field Cara, Desiree, Homeguard, Kennebec, Lady Balfour, Maris Piper, Mayan Gold, Nadine, Pentland Dell. All experiments were performed on Wharf Ground field, Wellesbourne, which had a soil P index of 3 (38 mg Olsen P L-1). One plot was grown for each genotype in 2007 and 2008 using a randomised plot design. To soil that was K Index 1 (76 mg L-1) and had a Soil Nitrogen Supply (SNS) index 0, RB209 recommended rates for N and K for potatoes were applied.

Seed tubers were sown in early April and harvested in October in both years. The crop was managed as a commercial crop, with the application of pesticides and irrigation applied when necessary. During the growth of the crop several parameters were recorded including, emergence rate, canopy development, canopy light interception and the nutrient status of diagnostic leaf samples. At commercial maturity, tubers were harvested, graded and analysed for their mineral nutrient status. For all samples analysed, tissue N, P, K, Ca, Mg, S, B, Cu, Fe, Mn, Na and Zn concentrations were determined.

8.7.2 Results

Emergence rates of potato plants were monitored during crop establishment. Potato plants emerged between 20 and 24 days after planting. There were significant (P<0.001) differences between genotypes in 2007, but not in 2008. However, in both years Homeguard was one of the slowest to emerge and Desiree was one of the quickest to emerge.

To monitor the continued development of the crop during the season, leaf area, intercepted radiation and elemental analyses of diagnostic leaves were measured. No significant difference between genotypes was observed for leaf area development, intercepted radiation or elemental profiles of the diagnostic leaves. However, there was a significant (P<0.001) effect of year, mainly consequence of heavy rainfall during 2007.

At commercial maturity, tubers were harvested and graded for size. Over the two years of field trials there was a significant (P=0.008) effect of genotype on tuber yield per plant (Figure 7). Homeguard

15

produced the lowest tuber yields overall and Nadine produced the highest. There was also a significant effect of genotype on the <45mm and 45-65mm grades. Notably, Mayan Gold produced smaller tubers compared to all the other genotypes. There was also a significant (P=0.027) effect of genotype on the tuber P concentration, with Homeguard accumulating the least P and Kennebec accumulating the most, when the P fertiliser was supplied as a broadcast application (Figure 7).

Homeg

uard

Pentla

nd D

ell

Maris P

iper

Mayan

Gold

Desire

e

Kenne

bec P

Kenne

bec B

CCara

Lady

Balf

our

Nadine

Tube

r yie

ld (k

g pl

ant-1

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Homeg

uard

Pentla

nd D

ell

Maris P

iper

Mayan

Gold

Desire

e

Kenne

bec P

Kenne

bec B

CCara

Lady

Balf

our

Nadine

Tube

r P c

once

ntra

tion

(%P

DW

)0.00

0.05

0.10

0.15

0.20

0.25

0.30A B

Figure 7. Tuber yield per plant at commercial maturity (A) and tuber P concentration (B) for potato genotypes supplied with 50% of the RB209 recommended rate of triple super phosphate (TSP) fertiliser placed (3” to the side and 3” below the seed tuber) into the soil. P = Placed, BC = Broadcast. Tubers were graded for size into <45mm (black), 45-65mm (light grey), 65-80mm (dark grey) and >80mm (white) and weighed. Bars represent the REML means of two replicate years (n=2). Error bar represents the standard error of the difference.

8.7.3 Summary

There were significant differences in tuber yields between genotypes, with Homeguard having the lowest tuber yields and Nadine having the highest.

8.8 Objective 05. A mathematical model describing the development of the potato root system

8.8.1 BackgroundThe following modelling work has been carried out during the project:

Development of a hydrological model using a simple and highly accurate algorithm to increase accuracy of predictions of soil water movement and solute transport;

Development of a 2-D simple root growth model for a range of vegetable and arable crops; Development of a 3-D mechanistic model for root growth, water and P uptake, and water

movement and P transport in soil; Improvement of the model PHOSMOD to predict the effect of P fertiliser on potato yield.

8.8.2 Results

8.8.2.1 Hydrological model using a newly-developed algorithm

In 1-D situations, the differential equations for water and solute transfer within the soil profile in the soil-crop system are

(1)

(2)

16

where θ is the volumetric soil water content, h is the soil pressure head, Sw is the sink term, i.e. root water uptake, z is the vertical coordinate, t is time, K is the soil hydraulic conductivity, c is the solute concentration, vz is the water flux, f is the function for the zero- and first-order rate reactions for solute in the liquid and soil phases, Sc is the root solute uptake, Dz is the dispersion coefficient.

The soil hydraulic functions are defined according to van Genuchten (1980)

(3)

(4)

where is the relative saturation, and are the saturated and residual soil water contents, α and n are the shape parameters of the retention and conductivity functions, m=1-1/n, and Ks is the saturated hydraulic conductivity.

The dispersion coefficient in Eq. (2) is given by Bear (1972)

(5)

where Dd is the ionic molecular diffusion coefficient in free water, DL is the dispersivity, and is the tortuosity factor, which is defined:

(6)

Eqs. (1) and (2) are partial differential equations which normally requires complex numerical schemes such as the FE method to solve them (Šimůnek et al., 1992; Zhang et al., 2010b). This involves an iterative procedure to obtain the solution to the water flow equation (Eq. 1) by solving the system of linear algebraic equations and a solution to the transport equation (Eq. 2). To simplify the numerical procedure, we developed a simple and robust scheme to simulate water movement and solute transport in soil. The approach works with soil layers with uniform thickness of 5 cm. The proposed approach considers that water movement and solute transport in a soil layer is only influenced by the above and below layers in a small time step, allowing soil water flow and solute transport to be calculated on a layer basis.

Integrating Eqs. (1) (2) vertically over a soil layer leads to

(7)

(8)

where i is the soil layer number, is the time step, is the layer-average soil water content

change in layer i in , is the soil layer thickness, and , and represent water

flux and solute transport from the layer i+1 to i and from i to i-1, which are calculated

(9)

(10)

(11)

(12)

where , and , are the differences in soil pressure head and solute concentration between layers i+1 and i, and i and i-1, respectively.

17

Detailed steps of implementing the procedure can be seen elsewhere (Yang et al., 2009).

8.8.2.2 Hydrological model validation

To examine the performance of the proposed algorithm in hydrological simulations, a case of simulating water and nitrate movement in a soil column immediately after an application of 100 kg ha -1

nitrate-N (NO3-N) was assumed. The FE method was selected for comparison. The simulations were carried out on two contrasting soils: i.e. a coarse and a fine texture. The parameters describing water characteristics for both soils were set to those suggested by Wösten et al. (1999) (Table 4). The soil columns were assumed to have a depth of 100 cm, with an initial soil water content set at saturation throughout the column. The lower boundary condition was specified as free drainage, whereas no water flux was allowed at the surface. It was assumed that NO3-N was dissolved in the top 5 cm soil layer immediately after application. The calculated NO3-N concentrations were 0.513 and 0.390 mg cm-3 for the coarse and fine soils, respectively. For the proposed algorithm, the soil column was divided into 20 uniform 5 cm layers, with a simulation time step for both soils of 0.001 d. In the FE method, the soil column was divided into 50 soil layers with various thicknesses. Since numerical oscillations often occur in solving the transport equation, especially when convection is dominated in transport, the FE method with ‘upstream weighting and the artificial dispersion’ scheme was employed (Šimůnek et al., 1992), which was named as the ‘complex’ FE method in the subsequent sections. The FE method without ‘upstream weighting and the artificial dispersion’ scheme, named as the ‘ordinary’ FE method, was also used in the simulations for comparison.

Table 4. Soil hydraulic parameter values for the coarse and fine soils in the numerical experiments (Wösten et al., 1999)  θ s (cm3 cm-3) θ r (cm3 cm-3) θ (cm-1) n Ks (cm d-1)Coarse soil 0.40 0.03 0.0383 1.374 60.0

Fine soil 0.52 0.01 0.0367 1.101 24.8

Soil water content and NO3-N concentration distributions at various time intervals were simulated and compared using the proposed algorithm and the FE method for the fine soil (Figure 8). It is clear that the profiles predicted by the proposed algorithm agree well with those from the FE method. It was also found that the simulated results from both the ‘ordinary’ and ‘complex’ FE methods were virtually identical. This indicates that the ‘ordinary’ FE method may sufficiently be accurate in simulating solute transport in the fine soil, and the simple algorithm proposed in the study can achieve the same accuracy of the simulated results as those from the FE method in hydrological simulations for the fine soil.

The same simulations and comparisons were also carried out for the coarse soil (Figure 9). The simulated soil water content profiles at intervals using the proposed algorithm are in good agreement with those from the FE method, which confirms that the proposed algorithm is capable of simulating soil water movement in different soils accurately. While the simulated NO3-N concentration profiles at intervals using the proposed algorithm agree fairly well with those from the ‘complex’ FE method, large discrepancies were observed in the simulated results between the proposed algorithm and the ‘ordinary’ FE method. The ‘ordinary’ FE method severely underestimated NO3-N transport in the soil profile, resulting in much higher NO3-N concentration in the top 20 cm soil layer. This can be attributed to the steep NO3-N concentration front and the dominant convection in the simulated coarse soil. The differences in NO3-N concentration profiles simulated by the proposed algorithm and the ‘complex’ FE method may be attributed to the ‘upstream weighting and the artificial dispersion’ scheme. The scheme stabilizes the results by altering dispersion coefficient, but leads to the results deviated from the true values due to the artificial nature of the dispersion coefficient.

Figure 10 shows NO3-N concentration distributions down the soil profile after 30 day free drainage simulated using the proposed algorithm for the transport equation with and without the diffusion term. The diffusion term has a bigger effect on NO3-N transport in the fine soil than the coarse soil. This is due to the fact that in the coarse soil NO3-N transport is dominated by the convection term, i.e. NO3-N mainly moves with water flow. However, in the fine soil water flow is not as easy as that in the coarse soil due to narrow pores. As a result diffusion becomes an important process in NO 3-N transport in the

18

soil. This implies that in modelling NO3-N transport in the fine soil, the diffusion term has to be taken into consideration to enable the predictions to be reasonable.

It is evident, from the above, that the proposed algorithm presented in the study produces the results as accurately as those from the FE method in modelling soil water dynamics. The approach out-formed both the ‘ordinary’ and ‘complex’ FE methods in simulating NO3-N transport in the coarse soil. Given the simplicity, stability and the ability of the proposed algorithm to accurately simulate soil water and solute transport in different soils, it can be concluded that the proposed algorithm has a good potential to be used in agro-hydrological models for reliable hydrological simulations.

0

20

40

60

80

100

0.4 0.45 0.5 0.55Water content (cm3 cm-3)

Dept

h fro

m s

oil s

urfa

ce (c

m)

(a)

0

10

20

30

0 0.2 0.4 0.6

NO3-N conc (mg cm-3)

Dept

h fro

m s

oil s

urfa

ce (c

m)

(b)

01 0

2 03 0

0 0 .2 0 .4 0 .6

N O 3 c o n c e n tra tio n (m g c m -3 )

Initial after 5 d (FEM)after 5 d (presnet study) after 10 d (FEM)after 10 d (present study) after 30 d (FEM)after 30 d (present study)

Figure 8. Comparison of soil water content (a) and NO3-N concentration down the soil profile at intervals (b) for the fine soil.

19

0

20

40

60

80

100

0.15 0.25 0.35 0.45

Water content (cm3 cm-3)

Dept

h fro

m s

oil s

urfa

ce (c

m)

(a)

0

10

20

30

40

0 0.2 0.4 0.6NO3-N conc (mg cm-3)

Dept

h fro

m s

oil s

urfa

ce (c

m)

after 5 d

(b)

0

10

20

30

40

0 0.2 0.4 0.6NO3-N conc (mg cm-3)

Dept

h fro

m s

oil s

urfa

ce (c

m)

after 10 d

(c)

0

10

20

30

40

0 0.2 0.4 0.6NO3-N conc (mg cm-3)

Dept

h fro

m s

oil s

urfa

ce (c

m)

after 30 d

(d)

Legend for (a)0

2 0

4 06 0

8 01 0 0

0 0 .1 0 .2 0 .3 0 .4

W a te r c o n te n t (c m 3 c m -3 )Initial after 5d (FEM)after 5d (present study) after 10d (FEM)after 10d (present study) after 30d (FEM)after 30d (present study)

Legend for (b)(c)(d)

Figure 9. Comparison of soil water content (a) and NO3-N concentration down the soil profile after 5 days (b), 10 days (c) and 30 days (d) for the coarse soil.

20

01 0

2 03 0

4 0

Initial FEM (complex scheme)

FEM (ordinary scheme) Present study

0

10

20

30

40

50

0 0.1 0.2 0.3 0.4

NO3-N concentration (mg cm-3)

Dept

h fro

m s

oil s

urfa

ce (c

m)

Coarse soil (without diffusion)Coarse soil (with diffusion)Fine soil (without diffusion)Fine soil (with diffusion)

Figure 10. NO3-N concentration distributions down the soil profile after 30 days calculated with and without diffusion for the fine and coarse soils.

The test of the proposed simple and accurate algorithm for modelling water and nitrogen dynamics in a soil-wheat system has also been carried. The results were found to be encouraging and are being prepared for possible publication (Zhang 2010a).

8.8.2.3 Development of a 2-D root growth model

Root development simulation is a key part in modelling water and nutrient cycles in the soil-crop systems. Accurate modelling of root dynamics is extremely difficult since the development and proliferation of roots in soil are affected by a number of factors such as the supply of photosynthates from the shoot, the nutrient status of the plant, soil type and compaction, water potential at the root surface and availability and distribution of nutrients. Dynamic root architecture models including detailed modelling of individual roots have been developed. Such models are useful for basic research to understand the mechanisms of resource uptake by roots, but are generally not suitable for ago-hydrological models because of a lack of data to evaluate the models. In this study a generic 2-D dynamic root model has been developed (Pedersen et al., 2010). The model assumes root development is driven by temperature. Root penetration rates in both directions are dependent on degree days. The root biomass is assumed to be proportional to the above ground parts of the crop, and the proportionality decreases with increasing crop dry weight. The total root length is calculated using a specific root length density. A logarithmic function is employed to calculate root density. The root density distributions down the soil profile and in the horizontal direction are controlled by shape parameters.

The rooting depth and width are calculated based on the cumulative mean day temperature according to

(13)

where i = x, z stands for the coordinates in the horizontal and vertical directions, Ri (m) is the rooting width and depth, Ristart (m) is the starting rooting width and depth, (oC d) is the cumulative day degree, Tlag (oC d) is the threshold of cumulative day degree for root growth, Kri (m day-1 C-1) is the root growth rate in the corresponding direction. Rimax (m) is the maximum rooting depth and width restricted by physical barriers or the effective rooting width (= a half row width) for row crops.

21

Crop total root length is calculated as a product of root dry weight and a fixed specific root length. The increment in root dry weight is a function of the increment in crop dry weight , crop dry weight W, and a parameter defining root class:

(14)

where Rroot is the ratio of to , which declines with W and varies with the root class parameter as shown Zhang et al. (2009).

The root length declines logarithmically from the soil surface downwards, i.e.:

(15)

where L0 (m m-3) is the total root length, ax and az are the shape parameters controlling root distribution in x and z directions, respectively.

The proposed root model has been validated against root measurements from field experiments (Pedersen et al., 2010). In order to apply such a model for modeling water and nutrient cycles in the soil-crop system, the model was further parameterized for a range of crops (Zhang et al., 2009) and was incorporated into agro-hydrological models for water (Yang et al., 2009) and nitrogen (Zhang, 2010b; Zhang et al., 2010a) dynamics. It has been shown that with the proposal root model, the developed agronomic models were able to reproduce the measurements of water and nitrogen in the soil-crop system reasonably over a number of crops (Zhang et al., 2009a; Zhang, 2010; Zhang et al., 2010a). This suggests that the simple root model is accurate enough to describe root development for the purposes of water and nutrient management in crop production.

However, it should be pointed out that the model used in the study is rather simple. It does not take into consideration of other factors such as soil structure, soil water content and nutrients status. Although the model gave good descriptions of root development of crops observed in a number of studies in humid conditions, cautions need to be taken to apply the root model for crops grown under complex situations such as dry climates where low soil water content could be a limiting factor controlling root growth (Yang et al., 2009).

8.8.2.4 Model development for root architecture and water and P uptake

A root architecture and water and P uptake model has been developed. The model is largely based on the work by Somma et al. (1997). It simultaneously simulates soil water movement, P transport in soil, water and P uptake, and plant and root growth. The principles and main features of the model include (Somma et al., 1997).

Root apices are translocated in individual growth events as a function of current local soil conditions. A three-dimensional finite-element grid over the considered soil domain serves to define the spatial distribution of soil physical properties and as framework for modelling of transient water flow and solute transport, the latter including passive and active root solute uptake, as well as zero and first-order source/sink terms. Two water extraction functions have been embodied in the model, with the possibility of taking into account osmotic potential effects on water and passive solute uptake rate. Root age effects on root water and solute uptake activity have been included, as well as the influence of nutrient deficient or excessive concentration on root growth. A detailed description of the model is given in Somma et al. (1997) and Somma et al. (1998).

In order to test the model, the model is run for the soil domain of 10cmx4cmx60cm. The domain is divided into 4800 elements with each element having a volume of 1cmx1cmx1cm in simulating soil water movement and P transport using the finite element method. The main parameter values used in the simulations for root growth are determined according to Somma et al. (1997) and Somma et al. (1998). Initial soil water content is 0.4 cm3 cm-3 and the P concentration is 0.006 mmol.

22

The model is run for the following scenarios to assess the model’s predictions in the effects of different root architectures and P uptake rates on root growth, water and P uptake by roots, and water and P distributions in soil.

Scenario 1: Simulations of root growth, water and P uptake and soil water distribution

A medium soil texture is selected, and the van Genuchten soil hydraulic properties parameter values (Eqs. 3, 4) are given in Table 5.

Table 5. Soil hydraulic properties parameter values used in the simulations  θs (cm3 cm-3) θr (cm3 cm-3) α(cm-1) n Ks (cm h-1)

Medium soil 0.44 0 0.031 1.18 0.5

Figure 11 A and B shows the root distributions after 500 h and 1000 h growth. Different colours (white, green, red) of roots represent the different branching orders (0, 1st and 2nd orders). Figure 11 C and D show the soil water content distributions at 500 h and 1000 h, respectively. Low water content in the top region of the soil domain is a result of higher root length density. At 500 h, water uptake mainly occurs in the top 20cm, whereas at 1000 h, water uptake happens in the depth greater than 30 cm. Figure 12 illustrates the cumulative water and P uptake and shoot and root dry weight (DW) accumulation. It is clear that both shoot and root DW increase with time as the plant grows (Figure 12a), leading to the increases in water and P uptake by the entire root system (Figure 12b). Since in this scenario P uptake is treated as a passive process, i.e. P uptake as a result of water flow into the roots, the pattern of P uptake basically follows that for water uptake. It can be concluded, from this numerical experiment, that the model behaviour reasonably in predicting water and P uptake by the root system and water movement and P transport in soil.

(a) 500h (b) 1000h (c) 500h (d) 1000h

Figure 11. Simulated root distribution (a) (b) and soil water distribution (c) (d)

23

Figure 12. Simulated results of shoot and root dry matter accumulation (a) and cumulative water and P uptake (b)

Scenario 2: Simulation of the effects of different root systems on water and P uptake and soil water distribution

In this scenario the parameter values for root growth are different from those in S1, while the others remain identical. The purpose for running such a scenario is to test whether the simulated root growth responds to the parameter values and how this affects the plant growth and water and P uptake. Figure 13 a and b show the roots distributions simulated with a different set of root parameter values, which is evidently different from those shown in Figure 12. The approximately symmetric root distribution at 1000 h results in the soil water content distributed more symmetrically (Figure 13d) than that in S1 (Figure 11d). The differences in cumulative shoot and root DW and water and P uptake after 1000 h growth are not great. The simulated cumulative shoot and root DW and water and P uptake are 1.32 g, 0.44 g, 270 g and 1.62 mmol (Figure 14). Likewise, the values are 1.43 g, 0.47 g, 292 g and 1.75 mmol in S1 (Figure 12). The differences can be attributed to the different root architectures. A root system capable of occupying a greater volume of soil has a bigger capacity of capturing resources and thus resulting faster growth.

24

0

1

2

3

0 200 400 600 800 1000Time (h)

Dry

wei

ght (

g) ShootRoot

0

200

400

600

0 200 400 600 800 1000Time (h)

Wat

er u

ptak

e (g

)

0

1

2

3

P up

take

(mm

ol)Water uptake

P uptake

(a) 500h (b) 1000h (c) 500h (d) 1000h

Figure 13. Simulated root distribution (a,b) and soil water distribution (c,d).

Figure 14. Simulated results of cumulative water and P uptake (a, b), and shoot and root dry matter accumulation (c, d).

Scenario 3: Simulation of the effects of different active P uptake rates by root on plant P uptake and soil P distribution

In this scenario P uptake by roots is assumed to be an active process. Two cases are simulated with different maximum uptake rates of 1.2 x10 -4 mmol cm-2 h-1 (Case 1) and 0.2x10-4 mmol cm-2 h-1 (Case 2), respectively. The P concentration at which the uptake rate is halved is assumed to 0.002 mmol. Figure 15 shows the results of P concentration distributions in the soil after 1000 h growth and the cumulative P uptake by the entire root system. It can be seen that the P concentrations in the soil are both approximately symmetric (Figure 15ab), and the P concentration in the root zone in Fig. 8a is much lower than that in Figure 15b. This can be explained by the near symmetric root distribution (Figure 13b) and greater capacity of P uptake in Case 1.

25

(a) Case 1 (b) Case 2 (c) cumulative P uptake

Figure 15. Simulated P concentration distribution after 1000h growth in Case 1 (a) and in Case 2 (b), and the cumulative P uptake (c)

8.8.2.5 Improvement of the model PHOSMOD

The PHOSMOD was originally developed by Greenwood et al. (2001) to study the effect of P-fertiliser on crop growth over a wide range of crops. The model was, with the financial support from the DEFRA under the project of HH3507SFV, then modulised to serve as a component of the integrated NPK model for the reposes of combined NPK fertilizers to crop yield and mineral composition (Zhang et al., 2007).

The PHOSMOD model calculates the effects of extractable soil P, starter fertilizer P and granular fertilizer P on crop growth and P content. It assumes that roots absorb all of their P from the upper 30 cm of soil. The model calculates a daily increment in plant weight and root length from the %P in the plant, air temperature and the current plant weight. Root length is partitioned into segments divided between three regions of the soil: that enriched with either starter or granular fertilizer and the un-enriched remainder. The maximum amount of P that can diffuse daily to each of the previously formed segments is calculated. The actual uptake is calculated from this information and the %P in the plant, and the cycle of calculations repeated for the next day. There is a subroutine that updates soil water contents from Jan 1st until the day of the final harvest. There are other subroutines that calculate the interchange between extractable and non-extractable soil P and the diffusion coefficients in P-depleted zones of soil around each root segment and in the remainder of the soil. A detailed description of the PHOSMOD model is given in Greenwood et al. (2001) and Zhang et al. (2007).

During the project the PHOSMOD model was improved by replacing the very approximate algorithm for soil water movement with the newly developed simple but accurate algorithm.

26

0

1

2

3

0 200 400 600 800 1000

Time (h)

P up

take

(mm

ol) Uptake rate 1

Uptake rate 2

To validate and test the improved crop P model PHOSMOD for potato crop, the experimental data from the HRI field studies have been used (Table 5). The weather inputs are the daily values of rainfall (including irrigation), mean air temperature and potential evaporation from an open water surface, and were those experienced by the experiment. The other inputs are as described in Greenwood et al. (2001).

Table 5. Experiments used for the validation of the improved model PHOSMOD

Crop Planting date Harvest date Treatment P-Fertiliser rate (kg P ha-1)

Potato19/04/1971 16/08/1971 6

0, 49, 122, 196, 269, 34217/05/1972 06/09/1972 616/05/1973 12/09/1973 6

The simulated values of crop final DW and crop %P are shown in Figure 16, together with the measured values. It can be observed that the agreement between measurement and simulation is good for both crop final DW and crop %P. The simulated values are strongly correlated and almost proportional to the measured values (Figure 16), which indicates that the improved model PHOSMOD performs well in predicting the dependence of crop final DW and crop %P on P-fertiliser.

y = 1.1634x - 1.5054R2 = 0.9221

0

5

10

15

0 5 10 15Measured dry weight (t ha -1)

Sim

ulat

ed d

ry w

eigh

t (t h

a-1

)

y = 0.8022x + 0.0357R2 = 0.2916

0

0.2

0.4

0.6

0.8

0 0.2 0.4 0.6 0.8

Measured %P

Sim

ulat

ed %

P

(a) (b)Figure 16. Comparison of dry weight yield (a) and crop %P (b) under different P-fertiliser treatments for potatoes.

8.8.3 Conclusions

An explicit algorithm for modelling water movement and solute transport in soil has been developed. The algorithm is robust and much simpler than the traditional algorithm, which is potentially very useful in modelling water and nutrient dynamics in the soil-plant system.

A simple 2-D root growth model is devised, parameterised and tested in modelling water and nitrogen dynamics for a wide range of vegetable and arable crops. It has been shown that with the root model the agro-hydrological models are able to produce good agreement between measurement and simulation of water and nitrogen in the soil-plant system, indicating that the root model is parameterised reasonably.

A sophisticated 3-D model which simulates root growth and water and P uptake has been devised based on the work by Somma et al. (1999). It mechanistically simulates root growth, water and P uptake, and water movement and P transport in soil, and thus is a powerful tool to study the mechanism of water and P uptake by the entire root system. However, the model has dozens of parameters and some of them are difficult to determine. Further work is required to validate such a model.

A model to study the effect of P on potato growth PHOSMOD, based on Greenwood et al. (2001) and Zhang et al. (2007), was also updated. Results show that the model performs reasonably to predict the effect of P fertiliser on potato yield and crop %P.

27

8.9 Literature cited

Bear J. (1972). Dynamics of Fluid in Porous Media, Elsevier, New York, NY.Verma RS, Grewal JS (1979). Evaluation of levels and method of phosphatic fertilizer application to

potatoes in acid hill soils. Journal of the Indian Potato Association, 5, 76-82.Anon (2004). Growing your potential. Fertiliser placement for potatoes. Hydro Agri (UK)Black CA (1992). Soil Fertility Evaluation and Control. Lewis Publishers, Boca Raton, USA.Brewster JL, Rowse HR, Bosch AD (1991). The effects of sub-seed placement of liquid N and P

fertilizer on the growth and development of bulb onions over a range of plant densities using primed and non-primed seed. Journal of Horticultural Science, 66, 551-557.

Buerkert A, Bationo A, Piepho HP (2001). Efficient phosphorus application strategies for increased crop production in sub-Saharan West Africa. Field Crops Research, 72, 1-15.

Buerkert A, Hiernaux P (1998). Nutrients in the west African Sudano-Sahelian zone: losses, transfers and role of external inputs. Zeitschrift fur Pflanzenernahrung und Bodenkunde, 161, 365-383.

Costigan PA, Heavyside M (1988). The effects of starter fertilizer on the early growth and yield of transplanted crisp lettuce on fertile soils. Journal of Horticultural Science, 63, 247-253.

Dachler M, Kochl A (1995). Fertilizer placement to maize. Bodenkultur, 46, 119-124.Dampney P, Johnson P, Goodlass G, Dyer C, Sinclair A, Edwards T (2002). Review of the response

of potatoes to phosphate. Final Report on Defra Project PE0108.Drew (1975) Comparison of the effects of a localized supply of phosphate, nitrate, ammonium and

potassium on the growth of the seminal root system, and the shoot, in barley. New Phytol 75: 479–490.

Greenwood DJ, Karpinets TV, Stone D (2001). Dynamic model for the effects of soil P and fertilizer P on crop growth, P uptake and soil P in arable cropping: model description. Ann. Bot. 88: 279-291.

Greenwood DJ (1981). Fertilizer food production: world scene. Fertilizer Research, 2, 31-51.Hammond JP, Broadley MR, White PJ (2004). Genetic Responses to Phosphorus Deficiency. Annals

of Botany, 94, 323–332, 2004.Harris PM (1992). Mineral Nutrition. In: Harris PM (Ed.) The Potato Crop: The Scientific Basis for

Improvement, 2nd Edition, pp. 162-213. London: Chapman and Hall.Haygarth P et al. (2004). Reviewing the potential for reductions of nitrogen and phosphorus inputs in

current farm systems: A specification of project requirements. Final report on Defra Project ESO201.

Hegney MA, McPharlin IR (1999). Broadcasting phosphate fertiliser produces higher yields of potatoes than band placement on coastal sands. Australian Journal of Experimental Agriculture, 39, 495-503.

Henriksen K (1987). Effect of N- and P- fertilisation on yield and harvest time in bulb onions (Allium cepa L.). Acta Horticulturae, 208, 207-215.

Kingston BD, Jones RW (1980). Response of potatoes to phosphate rate and placement in the Texas Rolling Plains. Texas Agricultural Research Station Report 1980, College Station, Texas.

Laegreid M, Bøckman OC, Kaarstad O (1999). Agriculture, Fertilizers and the Environment. CABI Publishing, Wallingford, UK.

Lewis DJ, Kettlewell PS (1992). A comparison of broadcast granular fertiliser and placed liquid fertiliser for potatoes. Aspects of Applied Biology, 33, 29-35.

Lu S, Miller MH (1993). Determination of the most efficient phosphorus placement for field-grown maize (Zea mays L.) in early growth-stages. Canadian Journal of Soil Science, 73, 349-358.

McPharlin IR, Robertson WJ (1997). Response of spring-planted lettuce (Lactuca sativa L.) to freshly-applied and residual phosphorus and to phosphate fertiliser placement on a Karrakatta sand. Australian Journal of Experimental Agriculture, 37, 701-708.

Ministry of Agriculture, Fisheries and Food (2000). Fertiliser Recommendations for Agricultural and Horticultural Crops (RB209). Seventh edition. HMSO, Norwich.

Okalebo JR, Probert ME, Lekasi JK (1994). The effect of phosphate placement on maize in eastern Kenya. Tropical Agriculture, 4, 266-271.

Pedersen A., Zhang, K., Thorup-Kristensen K., Jensen L.S. (2010). Modelling diverse root density dynamics and deep nitrogen uptake – A simple approach. Plant Soil 326: 493–510.

Prummel J (1957). Fertilizer placement experiments. Plant and Soil, 8, 231-253.Raghothama KG (1999). Phosphate acquisition. Annual Review of Plant Physiology and Plant

Molecular Biology, 50, 665-693.

28

Reith WJS (1972). Soil properties limiting the efficiency of fertilizers. In: 7th Fertilizer World Congress, Zürich, C.I.E.C., pp. 275-278.

Roth GW, Beegle DB, Antle ME (2003). Evaluation of starter fertilizers for corn on soils testing high for phosphorus. Communications in Soil Science and Plant Analysis, 34, 1381-1392.

Ryan PF (1962). Fertiliser placement for kale. Irish Journal of Agricultural Research, 1, 231-236.Sanchez CA, Porter PS, Ulloa MF (1991). Relative efficiency of broadcast and banded phosphorus for

sweet corn produced on histosols. Soil Science Society of America Journal, 55, 871-875.Sharpley AN (1995). Identifying sites vulnerable to phosphorus loss in agricultural runoff. Journal of

Environmental Quality, 24, 947-951.Šimůnek J., Vogel T., Van Genuchten M.Th. (1992). The SWMS_2D code for simulating water flow

and solute transport in two-dimensional variably saturated media, v 1.1, Research Report No. 126, U. S. Salinity Lab, ARS USDA, Riverside.

Smith CB, Demchak KT, Ferretti PA (1990). Fertilizer placement effects on growth responses and nutrient uptake of sweet corn, snapbeans, tomatoes and cabbage. Communications in Soil Science and Plant Analysis, 21, 107-123.

Somma F., Clausnitzer V., Hopmans J.W. (1997). An algorithm for three-dimensional simultaneous modeling of root growth, transient soil water flow, and transport and uptake, V.2.1. Land, Air and Water Resources Paper No. 100034, Univ. of California, Davis.

Somma F., Hopmans J.W., Clausnitzer V. (1998). Transient three-dimensional modeling of soil water and solute transport with simultaneous root growth, root water and nutrient uptake. Plant Soil 202: 281-293.

Sparrow LA, Chapman KSR, Parsley D, Hardman PR, Cullen B (1992). Response of potatoes (cv Russset Burbank) to band placed and broadcast high cadmium phosphorous fertiliser on heavily cropped krasnozems in north-west Tasmania. Australian Journal of Experimental Agriculture, 32, 113-119.

Stone DA (1998). The effects of "starter" fertilizer injection on the growth and yield of drilled vegetable crops in relation to soils nutrient status. Journal of Horticultural Science and Biotechnology, 73, 441-451.

Stone DA (2000a). The effects of starter fertilizers on the growth and nitrogen use efficiency of onion and lettuce. Soil Use and Management, 16, 42-48.

Stone DA (2000b). Nitrogen requirement of wide-spaced row crops in the presence of starter fertilizer. Soil Use and Management, 16, 285-292.

Van Genuchten M.Th. (1980). A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44: 892-898.

Verma RS, Grewal JS (1979). Evaluation of levels and method of phosphatic fertilizer application to potatoes in acid hill soils. Journal of the Indian Potato Association, 5, 76-82.

White PJ, Broadley MR, Hammond JP, Thompson AJ (2005a) Optimising the potato root system for phosphorus and water acquisition in low-input growing systems. Aspects of Applied Biology 73, 111-118.

White PJ, Wheatley RE, Hammond JP, Zhang K (2007) Minerals, soils and roots. In: Potato biology and biotechnology: advances and perspectives, D Vreugdenhil (Ed). Submitted.

Withers PJA, Edwards AC, Foy RH (2001). Phosphorus cycling in UK agriculture and implications for phosphorus loss from soil. Soil Use Management, 17, 139-149.

Withers PJA, Peel S, Chalmers AG, Lane SJ, Kane R (2000). The response of manured forage maize to starter phosphorus fertilizer on chalkland soils in southern England. Grass and Forage Science, 55, 105-113.

Wösten J.H.M., Lilly A., Nemes A., Le Bas C. (1999). Development and use of a database of hydraulic properties of European soils. Geoderma 90: 169-185.

Yang D., Zhang T., Zhang K., Greenwood D.J., Hammond J., White P.J. (2009). An easily implemented agro-hydrological procedure with dynamic root simulation for water transfer in the crop-soil system: validation and application. J. Hydrol. 370: 177–190.

Zhang K. (2010a). An explicit hydrological algorithm for basic flow and transport equations and its application in agro-hydrological models for water and nitrogen dynamics. To be submitted.

Zhang K. (2010b). Evaluation of a generic agro-hydrological model for water and nitrogen dynamics (SMCR_N) in the soil-wheat system. Agric. Ecosyst. Environ. 137: 202-212.

Zhang K., Burns I.G., Greenwood D.J., Hammond J.P., White P.J. (2010b). Developing a reliable strategy to infer the effective soil hydraulic properties from field evaporation experiments for agro-hydrological models. Agric. Water Manage. 97: 399–409.

Zhang K., Greenwood D.J., Spracklen W.P., Rahn C.R., Hammond J.P., White P.J., Burns I.G. (2010a). A universal agro-hydrological model for water and nitrogen cycles in the soil-crop

29

system SMCR_N: critical update and further validation. Agric. Water Manage. doi:10.1016/j.agwat.2010.03.007.

Zhang K., Greenwood D.J., White P.J., Burns I.G. (2007). A dynamic model for the combined effects of N, P and K fertilizers on yield and mineral composition; description and experimental test. Plant Soil 298: 81–98.

Zhang K., Yang D., Greenwood D.J., Rahn C.R., Thorup-Kristensen K. (2009). Development and critical evaluation of a generic 2-D agro-hydrological model (SMCR_N) for the responses of crop yield and nitrogen composition to nitrogen fertilizer. Agric. Ecosyst. Environ. 132: 160–172.

30

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.

Publications and other outputs produced during the project that contain data, methodologies, or concepts developed, in part or in whole, in HH3509SFV.

Publications

Zhang K. (2010). An explicit hydrological algorithm for basic flow and transport equations and its application in agro-hydrological models for water and nitrogen dynamics. To be submitted.

Zhang K., Greenwood D.J., Spracklen W.P., Rahn C.R., Hammond J.P., White P.J., Burns I.G. (2010). A universal agro-hydrological model for water and nitrogen cycles in the soil-crop system SMCR_N: critical update and further validation. Agric. Water Manage. doi:10.1016/j.agwat.2010.03.007.

Zhang K. (2010). Evaluation of a generic agro-hydrological model for water and nitrogen dynamics (SMCR_N) in the soil-wheat system. Agric. Ecosyst. Environ. 137: 202-212.

Zhang K., Burns I.G., Greenwood D.J., Hammond J.P., White P.J. (2010). Developing a reliable strategy to infer the effective soil hydraulic properties from field evaporation experiments for agro-hydrological models. Agric. Water Manage. 97: 399–409.

Pedersen A., Zhang, K., Thorup-Kristensen K., Jensen L.S. (2010). Modelling diverse root density dynamics and deep nitrogen uptake – A simple approach. Plant Soil 326: 493–510.

Yang D., Zhang T., Zhang K., Greenwood D.J., Hammond J., White P.J. (2009). An easily implemented agro-hydrological procedure with dynamic root simulation for water transfer in the crop-soil system: validation and application. J. Hydrol. 370: 177–190.

Kloosterman B, De Koeyer D, Griffiths R, Flinn B, Steuernagel B, Scholz U, Sonnewald S, Sonnewald U, Bryan GJ, Prat S, Bánfalvi Z, Hammond JP, Geigenberger P, Nielsen KL, Visser RGF, Bachem CWB (2008). Genes driving potato tuber initiation and growth: identification based on transcriptional changes using the POCI array. Functional & Integrative Genomics, 8, 329-340.

Hammond JP, White PJ (2008). Sucrose transport in the phloem: integrating root responses to phosphorus starvation. Journal of Experimental Botany, 59, 93-109.

Hermans C, Hammond JP, White PJ, Verbruggen N (2006). How do plants respond to nutrient shortage by biomass allocation? Trends in Plant Science, 11, 610-617.

Amtmann A, Hammond JP, Armengaud P, White PJ (2006). Nutrient sensing and signalling in plants: potassium and phosphorus. Advances in Botanical Research, 43, 209-256.

White PJ, Broadley MR, Greenwood DJ, Hammond JP (2005) Proceedings of International Fertiliser Society 568. Genetic modifications to improve phosphorus acquisition by roots. IFS: York, UK. ISBN 0853102058.

White PJ, Broadley MR, Hammond JP, Thompson AJ (2005) Optimising the potato root system for phosphorus and water acquisition in low-input growing systems. Aspects of Applied Biology 73, 111-118.

Edited Contributions

George TS, Fransson A-M, Hammond JP, White PJ (2011) Phosphorus nutrition: Rhizosphere processes, plant response and adaptations. In: Bünemann E, Oberson A, Frossard E (eds), Phosphorus in Action. Springer, Dordrecht, Soil Biology, Volume 100, Part 2, pp 245-271.

Hammond JP, Broadley MR, Bowen HC, Hayden R, Spracklen WP, White PJ (2009) A Molecular Diagnostic for Phosphorus Deficiency in Potatoes. UC Davis: The Proceedings of the International Plant Nutrition Colloquium XVI. Retrieved from: http://escholarship.org/uc/item/9w97m6mz

31

Hammond JP, White PJ (2008) Diagnosing phosphorus deficiency in crops. In: Hammond JP, White PJ (eds), The Ecophysiology of Plant-Phosphorus Interactions. Springer, Dordrecht, pp 225-246.

White PJ, Hammond JP (2008) Phosphorus nutrition of terrestrial plants. In: Hammond JP, White PJ (eds), The Ecophysiology of Plant-Phosphorus Interactions. Springer, Dordrecht, pp 51-81.

Burns IG, Hammond JP, White PJ (2008) Precision placement of fertiliser for optimising the early nutrition of vegetable crops – Implications for yields, crop quality, and nutrient use efficiency. In: Proceedings of the IV International Symposium on Ecologically Sound Fertilization Strategies for Field Vegetable Production. Malmö, Sweden 22nd-25th September 2008. Acta Horticulturae.

Zhang K, Greenwood DJ, Hammond JP, White PJ, Burns IG (2008) A generic combined model for NPK fertilizer for vegetable and arable crops. In: Proceedings of the International Symposium on Plant Nutrition Management in Sustainable Agriculture, Nanchang, China, 13 th-16th October 2008, pp. 43-51.

White PJ, Wheatley RE, Hammond JP, Zhang K (2007) Minerals, soils and roots. In: Potato Biology and Biotechnology: Advances and Perspectives, D Vreugdenhil et al. (eds.) Elsevier Science.

Other Publications

Hammond JP, White PJ, Broadley MR (2009). Breeding for improved crop nutrient use efficiency. Horticulture Week (Grower), 27 February 2009, pp. 30-31.

Hammond JP, White PJ (2008) Sustainable approaches to plant nutrition. Horticulture Week (Grower), 07 August 2008, pp. 37-38.

Hammond JP, White PJ (2008) Reducing run-off for clean water. Horticulture Week (Grower), 26 June 2008, pp. 44-46.

Hammond JP, White PJ (2008) Sustainable future for fertilisers. Horticulture Week (Grower), 21 February 2008, pp. 33-34.

Hammond JP (2006) Halving the fertiliser burden. The Vegetable Farmer, February 2006, p. 30.

Meeting and Presentations

Hammond JP (2010) Predicting crop phosphorus requirements and developing new diagnostic tools. Nutrient Use Efficiency Symposium, University of Copenhagen, Denmark, 8 November 2010.

Hammond JP, Broadley MR, White PJ (2009) Assessing the suitability of struvite as a source of P for potato production. 2009 International Conference on Nutrient Recovery from Wastewater Streams, May 10-13, 2009, Vancouver, BC, Canada.

Hammond JP, Broadley MR, Bowen HC, Hayden R, Spracklen WP, White PJ (2009) P24. Chips and Ps: A diagnostic array for phosphorus (P) deficiency in potatoes. In: Abstracts of the Phoenix 2009 Symposium: Protein Complexes in Signalling and Development, University of Glasgow, 25-27 June 2009.

Hammond JP, Broadley MR, White PJ, King GJ, Bowen HC, Spracklen W, Hayden R, Meacham M, Overs T (2008) Phosphorus use efficiency in Brassica. Resource Capture by Crops, 10-12 September 2008, University of Nottingham, UK.

Hammond JP, Broadley MR, White PJ, King GJ, Bowen HC, Spracklen W, Hayden R, Meacham M, Overs T (2008) Phosphorus use efficiency in Brassica. Resource Capture by Crops, September 10-12 2008, University of Nottingham, UK.

Burns I, Hammond JP, White PJ (2008) Optimising the early nutrition of vegetable crops implications for yield, quality, and nutrient use efficiency. IV International Symposium on Ecologically Sound Fertilization Strategies for Field Vegetable Production, September 22-25, 2008, Malmö, Sweden.

Hammond JP, Broadley MR, White PJ (2008) Investigating crop nutrition through transcriptional studies. 2008 International Workshop on Chips, Computers and Crops, September 26-27, 2008, Zhejiang University, Hangzhou, China.

Hammond JP, Broadley MR, White PJ, King GJ, Bowen HC, Spracklen W, Hayden R, Meacham M, Overs

32

33