title / The Coates Farm Study II: Nitrogen Flows in a Changing Mixed Farming System
/ DEFRA
project code / NT1853
Department for Environment, Food and Rural Affairs CSG 15
Research and Development
Final Project Report
(Not to be used for LINK projects)
Two hard copies of this form should be returned to:Research Policy and International Division, Final Reports Unit
DEFRA, Area 301
Cromwell House, Dean Stanley Street, London, SW1P 3JH.
An electronic version should be e-mailed to
Project title / The Coates Farm Study II: Nitrogen Flows in a Changing Mixed Farming System
DEFRA project code / NT1853
Contractor organisation and location / Rothamsted Research
Harpenden, Herts.
AL5 2JQ
Total DEFRA project costs / £ £451,871
Project start date / 01/04/00 / Project end date / 31/03/03
Executive summary (maximum 2 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.
CSG 15 (Rev. 6/02) 3
Projecttitle / The Coates Farm Study II: Nitrogen Flows in a Changing Mixed Farming System
/ DEFRA
project code / NT1853
Coates Farm Study Phase I provided nitrogen (N) balances for a mixed farming system in the Cotswolds over a 5-year period and identified the major route of N loss from the system to be leaching. Phase II was set up to complete measurements for the 7-year crop rotation, and to investigate the alterations to N flows associated with the removal of the dairy herd from Coates Farm to Elkstone Dairy Farm and conversion of Coates to an arable/sheep system. In detail, the objectives were to:
01. Continue to measure the N flows in the farmlets established as part of Phase I of CFS to complete a full 7-year rotation, construct farm N budgets for another 2 years, and a full 7-year rotational budget, and confirm leaky practices and alternative management options.
02. Account for the ‘missing’ N in the N budget.
03. Investigate the short- and longer-term impact on the N balance of the transfer of the dairy herd from Coates.
04. Quantify the flush of nitrate leached when leys are ploughed after the dairy herd is transferred and explore management options to minimise the losses.
05. Study the impact of break crops on mineralisation and subsequent leaching.
06. Explore ways to improve the N-use efficiency of slurry and FYM.
07. Transfer this information to the MAFF, farmers and the public by all appropriate means (reports, scientific publications, open days and press.
In the conversion year, the sale of the cows increased apparent N outputs, while concentrate inputs were decreased, so the N efficiency (NE) of the farm was improved at 56%, cf 45% in Phase I. However, once the new system of arable and sheep was established, its efficiency of 47% (2-year average) was similar to that of the previous mixed system including dairy cows. The N surplus for the new system was 118 kg N/ha compared with 144 kg N/ha for Phase I.
Nitrogen balances for Elkstone Dairy, averaged over three quota years, show a NE of 20% and N surplus of 336 kg N/ha or 22.3 g N/l milk. For the combined farms of Coates and Elkstone the figures for this period were: NE = 32%, N surplus = 196 kg N/ha or 36.6 g N/l. These show that a specialist dairy farm uses N relatively inefficiently and generates a large surplus of N per ha. However, if the N surplus is expressed per litre of milk produced then the efficiency for the specialised dairy farm is better than that for the mixed farm.
The largest leaching losses were from grazed grass. Losses were particularly high in 2002-3 when sheep stocking rates were increased. Although there are differences in leaching losses related to crop type, the weather determines the loss, with greater leaching during periods of greater drainage volume. Amounts of nitrate leached following the conversion of long-term grass leys to arable depended on the cultivation techniques used. In the year of cultivation, the highest loss (82 kg N/ha) was from ‘overcultivation’ (consisting of ploughing and two passes with a power harrow), and the lowest (35 kg N/ha) from direct drilling. However, where less extensive cultivation was carried out, the losses in the second year following cultivation were increased, although total losses over the two years were still higher with more extensive cultivation.
A seven-year N balance shows total inputs of 1444 kg N/ha, total sales of 683 kg N/ha and total losses of 670 kg N/ha. This leaves 91 kg N/ha unaccounted for over the 7 years, i.e. 15 kg N/ha per year over the 7-year period. Measurements of N in the light fraction of organic matter, and of potentially mineralisable N, in soil samples taken at the beginning and end of the project indicated that the ‘missing N’ had not built up in these fractions. This agreed with analyses of total N in soils sampled at the beginning and end of the project: there was no significant increase in N content in any fields. There is some indication that N reserves in soil increase in dry years and decrease in wet years, but over time total N did not increase significantly in soil. In addition, two of the 10 fields sampled showed a significant decrease in total N under continuous arable cropping, as might be expected.
Measurements of soluble organic N (SON) leached showed that SON comprised c. 9% of the N leached from grass fields and 6% from arable fields. These figures were used to adjust the current and historical leaching losses, in proportion to the cropping pattern each year. The average increase in leaching by accounting for SON was 7% across all years. Adjusting the leaching figures in this way, and carefully reconciling monitoring periods for different processes, showed that there was no ‘missing N’: Theil’s regression analysis of total inputs and total outputs (sales + losses) of N showed no significant difference.
Small plot trials on the effects of break crop residues were unavoidably disrupted by restrictions caused by the outbreak of ‘Foot & Mouth Disease’ and could not be completed. However, field scale data collected through the rotation suggest that residues from oilseed rape do elevate leaching losses from the crops in the following one or two years.
A small plot trial investigating the implications for leaching of applying different rates of slurry to set-aside land in July, prior to establishing a winter wheat crop (a common local practice), showed no significant difference in N leached from plots receiving 0, 25, 50 and 75 m3/ha slurry. This suggests that the immediately available N from up to 75m3/ha slurry applied at this time was efficiently used by the early-established crop. However, the slurry was very dilute, supplying approximately 1 kg N/m3.
Formation of a farmers’ discussion group has facilitated the transfer of information generated by the project through three group meetings and individual farm visits. Farmers were genuinely interested in their own farm gate N balance figures, which were calculated using the FWAG Nutrient Budget spread sheet, developed in co-operation with Rothamsted and IGER. This acted as a good stimulus for discussion of practices that might be altered to improve the efficiency of N use. The topic of greatest interest was how to make more use of N in slurries and manures. We propose a small KT project to continue and develop the discussion group. A final open day for CFS will be held in June 2003.
CSG 15 (Rev. 6/02) 3
Projecttitle / The Coates Farm Study II: Nitrogen Flows in a Changing Mixed Farming System
/ DEFRA
project code / NT1853
Scientific report (maximum 20 sides A4)
To tab in this section press the tab key and the Control key together
Press the DOWN arrow once to move to the next question.
CSG 15 (Rev. 6/02) 3
Projecttitle / The Coates Farm Study II: Nitrogen Flows in a Changing Mixed Farming System
/ DEFRA
project code / NT1853
1. Introduction and Policy Rationale
The Coates Farm Study (CFS) examines nitrogen (N) cycling in a whole farm system, located at the Coates Manor Farm of the Royal Agricultural College (RAC), Cirencester, Gloucestershire. Phase I ran from 1995 to 2000. Flows of N in and out of the mixed farm system (dairy, sheep and arable) were recorded, providing a farm gate balance. Also, using detailed measurements made on ten representative ‘farmlets’, and with the assistance of modelling, the internal flows were measured or calculated. The N use efficiency (NE, defined as N output in product divided by N input) for Phase I varied from 39% to 49% over four years. On average c. 17% of N inputs was unaccounted for ('missing'), presumed to be either in unmeasured losses, stored in soil, or part of the error in the measurements.
Phase II began at the conclusion of Phase I and was established to allow measurements on a complete 7-year crop rotation, but especially to investigate the alterations to N flows associated with major changes to the farm system. In summer 2000 the relatively small dairy herd at Coates was moved to nearby Elkstone Dairy Farm and expanded to form a herd of 300 cows. This change, following the economic depression of the dairy sector in the late 1990s, was typical of the industry as a whole. The removal of a large animal component from Coates Farm, changing it to an arable/sheep system, had considerable implications for the N flows and N balance. These continued to be monitored in detail at Coates, and farm gate N balances and N efficiencies were calculated for Elkstone to demonstrate the effects of transferring milk production and expanding the dairy herd there. Following the removal of the dairy herd at Coates, 16 ha of long-standing grassland were ploughed and converted to arable. As this practice had serious implications for N leaching it was studied in detail in a small plot experiment. Explanations for the ‘missing N’ apparent in the Phase 1 budgets were sought. In particular, leaching data were re-analysed, the amount of soluble organic N (SON) in leachate measured, and changes in total and dynamic fractions of soil N examined.
This report covers the results from Phase 2 but also describes the whole farm balance for the entire eight years of the experiment, 1996-2003. The agronomic year 2000 to 2001 constituted a changeover year, during which the dairy herd was gradually reduced. It was therefore treated separately.
2. Objectives
01. Continue to measure the N flows in the farmlets established as part of Phase I of CFS to complete a full 7-year rotation, construct farm N balances for another 2 years, and a full 7-year rotational balance, and confirm leaky practices and alternative management options (Years 1-3).
02. Account for the ‘missing’ N in the N balance (Years 1-3).
08. Investigate the short- and longer-term impact on the N balance of the transfer of the dairy herd from Coates Farm (Years 1-3).
09. Quantify the flush of nitrate leached when leys are ploughed after the dairy herd is transferred and explore management options to minimise the losses (Years 1-2).
10. Study the impact of break crops on mineralisation and leaching (Year 1).
11. Explore ways to improve the N-use efficiency of slurry and FYM (Year 2).
12. Transfer this information to MAFF (now Defra), farmers and the public by all appropriate means (reports, scientific publications, open days and press) (Years 1-3).
2. Approaches and Methods
2.1 The system
This was described in Phase I (NT1803). The soils are mainly Sherborne and Moreton series, HOST classification 2, and were described in full in the final report for Phase I. The nature of the soils, especially the average depth of 30 cm over porous limestone, means that leaching dominates the loss processes, and droughting can occur in summer. Areas within ten fields dominated by the Sherborne Series soil, which covers 70% of the farm, and covering the range of cropping and different stages of the crop rotation, were selected as farmlets and used for detailed measurements throughout the study.
In Phase II, following the removal of the dairy herd, arable and sheep enterprises remained at Coates Farm. The general rotation continued to consist of two wheat crops, followed by barley and then a break crop of oats, oilseed rape, or a forage crop, depending on the relative product prices and forage requirements. Forage crops were either grass or stubble turnips for autumn/winter grazing. There was also a proportion of compulsory set-aside, which was fitted into the rotation as a break, following stubble turnips, and was used as an opportunity for early entry to cereals.
In year 6 the sheep enterprise consisted of weaned lambs or ewes ‘visiting’ from other college farms to graze grass or stubble turnips, or for winter housing. The equivalent stocking rate for this year was 0.23 LU/ha (Table 1). From February 2002 a dedicated flock of 550 ewes was established at Coates, which remained on the farm throughout the year. Other college sheep also grazed at Coates for part of the year, primarily on the stubble turnips in autumn (October - December). The stocking rate for this year was 0.57 LU/ha. A first cut of silage was taken from selected fields in late May/early June each year, to provide sufficient forage stocks for the sheep when housed. Thereafter the priority was for sheep grazing. Second cut silage was made in summer 2001 as there was excess grass. Some of this silage was exported, to be fed to cows at Elkstone. Fertiliser applications to grassland were 200-250 kg N/ha (Table 1), depending on weather conditions, stocking rate and silage stocks. The very dry weather in 1996-7 halted grass growth, so much less N was applied that summer.