Cost-E!ective Pollution Control by Shallow Injection of Pig Slurry into Growing ... Glasgow G4 0B4, UK. (Received 5 ... application can put a considerable burden on the pro- ducer as it ... may provide the farmer with a much enlarged &window'.
J. agric. Engng Res. (2001) 80 (4), 381}390 doi:10.1006/jaer.2001.0763, available online at http://www.idealibrary.com on SW*Soil and Water
Cost-E!ective Pollution Control by Shallow Injection of Pig Slurry into Growing Crops O. Pahl ; R. J. Godwin; M. J. Hann; T. W. Waine Institute of AgriTechnology, Cran"eld University, Silsoe, Bedford MK45 4DT, UK; e-mail of corresponding author: r.godwin@cran"eld.ac.uk Current address: School of the Built and Natural Environment, Glasgow Caledonian University, Cowcaddens Road, Glasgow G4 0B4, UK (Received 5 October 2000; accepted in revised form 1 October 2001; published online 23 October 2001)
In Europe, new legislation on environmental protection will probably require methods that reduce the emission of ammonia and odour from land spreading of animal slurry. Shallow injection (50}70 mm) of slurry is one of the methods recommended, albeit of higher cost than standard surface application techniques such as tanker broadcasting. The aims of this study were to investigate the agronomic e!ect of shallow injection of slurry into a growing crop, to examine the available information on the potential cost bene"ts of slurry}fertilizer utilization in the light of the new legislation and to determine the pollution impact of nitrate and ammonium leaching and ammonia and odour emissions. It was found that shallow injection of slurry into growing winter wheat during spring did not have a detrimental e!ect on the yield. Odour and ammonia emissions were reduced, nitrate and ammonium leachate levels were una!ected and it was of comparable cost to alternative application techniques that satisfy the requirements of the new legislation. 2001 Silsoe Research Institute
1. Introduction Agriculture is recognised as the major source of ammonia emissions in the UK (MAFF, 1998). New European legislation (Integrated Pollution Prevention and Control) aims to reduce ammonia emissions through implementation of abatement measures (EU, 1996). In intensive animal production, the abatement measures promoted in the UK are injection of slurry or rapid incorporation after surface application, reduction in emitting surfaces within animal buildings (e.g. smaller slurry channels under slats), and covering of waste stores. For these abatement techniques to be successful, it is of central importance that they are seen as a series of measures that aim to concentrate and contain ammonia in the slurry from its production to land application. The land application thus has to be of low-emission type and applied correctly. Bene"ts achieved from &upstream' abatement measures may otherwise be lost in this ultimate step, where ammonia losses of up to 65% have been reported (MAFF, 1999). However, low-emission land 0021-8634/01/120381#10 $35.00/0
application can put a considerable burden on the producer as it may necessitate short periods of intensive and somewhat specialized land application (e.g. injection and/or large machinery in order to handle the volume of slurry accumulated) in between long periods of storage. Shallow injection of slurry (50}70 mm) into growing crops can overcome this burden because it can be performed over a long period during the early stages of crop growth. This may result in a reduction in necessary slurry storage capacity (Warner et al., 1990). The utilization of slurry}fertilizer may be improved because slurry can be applied when crop nutrient requirements are at their maximum. A further reduction of ammonia and odour emissions may be achieved if the already small slurry surface after injection is reduced by use of special tines, for example tines that create a cavity into which the slurry is injected (Moseley et al., 1998; Godwin et al., 2000). In other words, shallow injection of slurry into growing crops can achieve the reduction in ammonia emissions required by legislation, and at the same time
381
2001 Silsoe Research Institute
382
O. PA H L E¹ A¸ .
may provide the farmer with a much enlarged &window' for application as well as savings in fertilizer cost. Whereas the emission abatement potential of shallow injection is well documented (Moseley et al., 1998; Pain et al., 1989; Lord & Mitchell, 1998), the adoption of the technique in wider farming practices is more likely to be linked with its agronomic implications and ultimately with its cost. However, little information is available on the agronomic e!ects on arable yield. Published cost comparisons between slurry injection and surface application are based on commercial farm-scale economic models. The aim of this study was to address these issues. The speci"c aims were as follows. (i) Assess the e!ect of shallow slurry injection on crop yield (winter wheat) and quality over two growing seasons. (ii) Review the data published on the cost of slurry injection and re-interpret it in the light of the new requirement for rapid incorporation of slurry after surface application. (iii) Determine the e!ect of shallow slurry injection on the leaching of nitrate and ammonium into ground water, an aspect of pollution abatement that has not yet been reported for shallow injection of slurry into a growing crop. (iv) Determine the e!ect of shallow slurry injection on ammonia and odour emissions and to review odour measurement techniques.
2. Materials and methods 2.1. Site preparation The 2-year experiment was conducted in the region of East Anglia on di!erent sites each year using
autumn-sown winter wheat. The slurry for the experiment was taken from the fully slatted weaner accommodation of a 270-sow pig farm (farrow-to-"nish operation). Slurry and mineral nitrogen fertilizer were applied to the growing crop in a block randomized plot design, as described in Table 1. Slurry was applied in a single dressing by shallow injection and low-level surface application (Fig. 1). The timing of the slurry injection was chosen to coincide with the application of mineral fertilizer on commercially operated farms (i.e. in line with the general agricultural practice in the area). The application was delayed in 2000 because of a wetter than average April (Fig. 2). In 1999, slurry was applied at the maximum permitted rate of 50 m ha\ speci"ed by current UK guidelines (MAFF, 1998). This rate in terms of total ammoniacal nitrogen (TAN"NH -N#NH -N) was equivalent to 150 kg [TAN] ha\ (Table 1), a rate reported to result in maximum yield at minimal risk of pollution (Lord & Mitchell, 1998; Silgram & Lord, 1999). In 2000, due to the di!erent composition of the slurry (from the same pig unit 1 year later), the equivalent rate was 77)4 kg [TAN] ha\. A single dressing of fertilizer was thus applied in order to make up the di!erence to 150 kg [TAN] ha\ (again in line with general agricultural practice, which would be to add mineral fertilizer rather than to inject twice). The slurry was applied with a modi"ed commercially available shallow injector (Moseley et al., 1998; Godwin et al., 2000). The injector was supplied with slurry by hose from a tanker. The same injector was used for low-level surface application to simulate band spreading, with the tines operating at around 0)1 m above the soil surface. Mineral nitrogen fertilizer (34)5% w/w nitrogen; 17)3% w/w as nitrate-N and 17)2% w/w as ammonium-N) was applied in a single dressing at a rate of 150 kg [N] ha\ in both years.
Table 1 Experimental design, slurry composition and application rates Component Slurry composition
Plots
Application date Application rates
Total Kjeldahl nitrogen (TKN), g kg\ Total ammoniacal nitrogen (TAN), g kg\ pH Total solids, g kg\ Design Replicates Size of each plot Soil texture Injection/surface Mineral fertilizer Injection/surface, kg [TAN] ha\ Mineral fertilizer, kg [N] ha\ Control
Year 1999 4)49 2)97 7)4 36)6 Randomized block 3 4 m by 15 m Sandy loam 4th March 1999 17th March 1999 150 150 0
Year 2000 2)258 1)547 7)0 21)5 Randomized block 5 4 m by 8 m Clay loam 3rd May 2000 12th May 2000 77)4 150 0
PO L LU TI O N CO N TR O L BY I NJ EC TI O N O F S LU R RY
383
Fig. 1. Slurry injection into winter wheat in 1999
2.2. Monitoring and analyses Leachate from the plots was collected with porous cups buried at 0)5 m depth in each plot. The leachate was analysed for nitrate and ammoniacal nitrogen by spectrophotometer (Phillips PU8730 UV}VIS, APHA, 1995). The water tension of the whole experimental "eld area was monitored at 0)5 m depth with vacuum-gauge tensiometers. Figure 2 shows the weather conditions for the duration of the experiment as well as the dates of leachate collection and the corresponding water tension measured in the "eld. The weather data were measured with an automated weather station (Casella, Bedford, UK). After ripening, two 0)5 m by 0)5 m areas were harvested manually from each plot. The grain and straw harvested from each plot were analysed for weight, moisture content and total Kjeldahl nitrogen (MAFF, 1986).
2.3. Ammonia emission measurement Ammonia emission from the "eld application of slurry was measured with hoods, which were installed on the plots immediately after slurry application (Fig. 3). These hoods, originally developed by Scotford and Williams (2001) for measurement of ammonia emissions from slurry lagoons, were made of lacquered wood, covered an area of 0)5 m and had an inner height of 0)18 m. Ambient air was drawn through a 100 mm diameter pipe of 0)5 m length with an internal fan that was resting on top
of the hood and was connected to the hood through a hole in its top. Air left the hood through a similar pipe (without fan) on the opposite site of the hood. The fan speed was set to result in a nominal air#ow above ground of 0)35 m s\. Ferm tubes (glass tubes coated with oxalic acid) (Sommer et al., 1996) were placed in the inlet and outlet pipes. Ammonia in the air passing through the pipes was absorbed by the oxalic acid, and the ammonia emission from the monitored soil was obtained from the di!erence between ammonia collected on the outlet and that collected on the inlet ferm tubes. The amount of ammonia absorbed in the ferm tubes was analysed photometrically.
2.4. Odour sample collection and analysis Odour samples were collected from the outlets of the same hoods as used for the ammonia measurements. However, cost of odour analysis prohibited di!erential measurement, so that only the odour in the air leaving the hoods was analysed. In order to avoid interference from neighbouring plots, the inlet pipes of the hoods were connected to 6 m long #exible hoses. Odour samples were taken in pairs (one plot each with surface application and injection), and the #exible hoses were placed so that both hoods received inlet air from the same point upwind of the plots. Outlet air was sucked through a polytetra#uoroethylene (PTFE) tube into proprietary polyethyleneterephthalate (PET, Nalophan2+) bags. Suction was applied by lung principle, i.e. the empty bags
384
O. PA H L E¹ A¸ .
Fig. 2. Weekly weather conditions and average xeld water tension for the duration of the xeld experiment, (a) 1999 and (b) 2000; , rainfall, mm; , xeld water tension, bar; average dry bulb temperature, 3C
were placed into an airtight drum, which was evacuated by an external pump. The advantage of this method is that the sampled gas is not in contact with the pump, but is drawn directly into the sample bags. The samples were analysed in accordance with the European draft standard for odour analysis (CEN, 1998). The laboratory repeatability and accuracy performance criteria during the experimental period were log repeatability r of 2)41 and accuracy A of 0)173, where the criteria (CEN, 1998) are r(3 and A(0)217.
2.5. Alternative odour assessment techniques Increased public concern has led to a greater number of complaints about agricultural odour. In extreme cases, the current technique employed by environmental health o$cers uses odour panel testing of samples. This technique is time consuming, labour intensive and expensive. There exists a need for a cheaper, quicker and more objective means to measure odour. Alternative methods of assessing odour using two electronic noses, both based
385
PO L LU TI O N CO N TR O L BY I NJ EC TI O N O F S LU R RY
Fig. 3. Ammonia emission equipment used to compare injection and surface treatments
on semi-conductor polymer arrays were considered by Misselbrook et al. (1997). The electronic noses were sensitive to odour from cattle slurry, but were not suitable for in-"eld use. During the odour sample collection for panel testing by Moseley et al. (1998), Waine (1999) used a portable gas detector (MGD-1, Environics OY, Finland) to simultaneously collect odour data. The portable gas detector works on the ion mobility principle described by Paakanen and Huttunen (1994), where the gas sample (pumped at a constant #ow) passes a radioactive source that ionizes the gas. Next, the ionized gas moves through an electromagnetic "eld which de#ects the charged particles so that they land on
di!erent parts of a six-segment sensor array according to their mass. Di!erent gas compositions lead to di!erent &time of #ight', changing the pattern of distribution on the sensor array. Higher concentrations give a larger cell unit output. Background air is used as a reference.
3. Results and discussion 3.1. Agronomy In both years the agronomic results (Table 2) indicate that physical disturbance of shallow injection of slurry did not have any detrimental e!ect on the yield. The plots
Table 2 Summary of agronomic results Agronomy component Grain yield*, t ha\
Total Kjeldahl nitrogen (TKN) in grain, g kg\ Harvest index, mass of grain/(grain#straw)
Treatment Injection Low-level surface Mineral fertilizer Control Injection Low-level surface Mineral fertilizer Control Injection Low-level surface Mineral fertilizer Control
* Expressed as dry matter. S Signi"cant di!erence at 75% con"dence level. R Signi"cant di!erence at 80% con"dence level. A Signi"cant di!erence at 95% con"dence level.
Year 1999 8)6 8)8 9)2 7)2R 20)3 20)3 21)1 15)8A 0)47 0)45 0)48 0)47
Year 2000 5)8 6)4 6)6 5)2S 19)8 18)8 20)0 14)8A 0)56 0)56 0)56 0)53
386
O. PA H L E¹ A¸ .
which had received mineral fertilizer showed a slightly higher yield than the plots which had received slurry by injection, but these results were not statistically signi"cant. Grain quality was assessed by measuring the total Kjeldahl nitrogen. Grain protein content can be estimated (Kent, 1983) by multiplying the nitrogen content by 5)7 (as 100 g of gluten contains on average 17)55 g nitrogen). There was no quality di!erence between the di!erent fertilizer and slurry applications and the control grain was of lower quality. Di!erences in harvest index, the proportion of grain to total above-ground biomass were not signi"cant. The Kjeldahl nitrogen results also suggested that there was no di!erence in the availability of nitrogen to the plants between the di!erent fertilization methods as the harvested crop from the fertilized plots contained similar amounts of nitrogen, whereas the harvested crop from the control plots contained less nitrogen. However, the analysis of leachate from the plots suggested that all plots contained residual nitrogen (see Section 3.3), which may have masked the di!erences between the fertilizer treatments and caused the apparent similarity in nitrogen availability. The leachate analysis also showed that very little ammoniacal nitrogen was lost through leaching. So it can be assumed that all TAN contained in the slurry was available to the plants.
3.2. Economy The additional cost of both low-level application and deep injection of slurry compared to tanker-broadcasting is well documented in the literature (Table 3). It was calculated that nitrogen savings through low-emission application of slurry do not compensate for the additional cost (McGechan & Wu, 1998). However, these calculations were based on the comparison between standard and low-emission application only. Legislation in
the European Union (EU) (MAFF, 1997) is likely to necessitate incorporation of slurry within 4 h after broadcast application adding the cost of ploughing to the surface application. Thus, slurry injection is cost e!ective if the additional cost of incorporation after broadcast application of slurry is considered (Table 3), and slurry injection is of similar cost to that of alternative emissionreducing application techniques. Fertilizer savings are not included in the costing in Table 3. In their study, McGechan and Wu (1998) calculated the fertilizer saving as 0)19 C m\ for deep injection compared with surface application, equating to an additional 0)37 C m\ for shallow injection over surface application, compared with the 0)62}0)82 C m\ additional cost of ploughing following surface application. Besides the direct cost considerations, shallow injection of slurry into growing crop allows application of slurry almost throughout the year and thus allows application in the spring to coincide with plant nutrient requirements. Surface broadcasting and deep injection is limited to the times between harvest and drilling. Another cost associated with injection of slurry, speci"cally into growing crop, is due to physical damage of the crop. Using a standard tractor most damage is caused by the tractor wheels. Low ground pressure &Terra-tyres' signi"cantly reduce the damage (Godwin et al., 1997). Di!erences between slurry injection and alternative emission-reducing application techniques are therefore likely to be small, since they all require the use of tractors. This was con"rmed by the lack of signi"cant yield or quality di!erences between shallow injection and low-level application (Table 2). The fundamental decision as to whether to apply slurry during the growing season must be made based on a farm-speci"c analysis taking into account cost, time requirements, machinery availability and weather conditions.
Table 3 Literature data on the additional cost in C m\3 of emission-reducing slurry application techniques compared to tanker broadcasting (normalized to an application of 50 m3 ha\1) Application (all from tanker)
Warner et al. (1990)
McGechan and Wu (1998)
Brewer (1999)
Nix (1999)
Deep injection (150}200 mm) Shallow injection (50}70 mm) Band spreader/Trailing shoe Incorporation (ploughing)
1)36 0)32* 0)18 *
1)13 0)52* * *
* 0)97 0)55}0)97 0)68
* * * 0)62}0)82
Application area [ha] on which costing is based
120
30
General
General
* Shallow tine cost estimated at 80% of deep injection cost, due to increased width of machine, reduced draft requirement (shallower tines) and increased #ow rate, hence increased forward speed, i.e. higher work rate. &&No data.''
387
PO L LU TI O N CO N TR O L BY I NJ EC TI O N O F S LU R RY
Table 4 Emission of nitrate, ammonia and odour during the experiments ¹reatment
Average nitrate in leachate after application, mg [NO3] l!1
1999 Treatment (no. of observations) Injection (7) Surface (7) Mineral (4) Control (6) Signi"cant di!erences between treatments* Signi"cant di!erences between treatments and control* 2000 Treatment (5 replications) Injection Surface Mineral Control Signi"cant di!erences between treatments* Signi"cant di!erences between treatments and control*
Ammonium in leachate, !1 mg [NH> 4 ]l
Ammonia emissions, mg [NH3] m!2 h!1
Odour emissions, OU m!3
63)9 57)2 62)2 62)0 No
(1 (1 (1 (1 n/a
No
n/a
57)2 45)5 56)7 49)0 No
(1 (1 (1 (1 n/a
203 1282 n/a n/a Yes
448 420 n/a n/a No
No
n/a
n/a
n/a
*At 95% con"dence level. OU, odour unit. n/a, Not applicable.
3.3. Pollution abatement 3.3.1. ¸eachate analysis The results of the leachate analysis in both years (Table 4) show that shallow slurry injection did not lead to higher leaching of nitrate or ammonium than the other treatments or the control. Ammonium could not be detected in the collected leachate. A slightly di!erent pattern emerged for the concentration of nitrate found in the leachate. Again, no di!erence was detected between nitrate leaching from the plots with shallow slurry injection and those with other treatments and the control plots (Table 4). However, in 1999 the overall level of nitrate detected in the leachate was above 50 mg l\ and thus above the concentration that in a water supply would be considered safe for human consumption. These elevated levels of nitrate were surprising, considering that only 150 kg [TAN] ha\ had been applied. The detection of similarly raised levels of nitrate in the leachate from control plots, which had received no nitrogen fertilizer, revealed that the observed leaching of nitrate was not due to the amount or type of fertilizer application, but due to residual nitrogen in the soil, likely to be the e!ect of overfertilization in previous years. This interpretation was supported by the pattern of nitrate release (Fig. 4), with around 100 mg l\ of nitrate found earlier in the year (February}March) but only 50 mg l\ by the end of the
leachate collection period (April}May). In 2000, the early nitrates were very high, indicating high residual nitrogen; this pattern may also indicate the release of nitri"ed nitrogen built-up over autumn/winter, during the wet season. In both years had the nitrogen originated from the applied slurry, the nitrate release could have been expected to increase*rather than decrease*in line with increasing activity of nitrifying bacteria due to rising temperatures and diminishing soil water-saturation (better aeration).
3.3.2. Ammonia and odour emissions Further to nitrate and ammonium leaching, the environmental considerations of pig slurry injection including the losses of ammonia through volatilization and the nuisance caused by odour emissions were evaluated in the year 2000 growing season. There was a signi"cant di!erence between ammonia emissions of injection and low-level surface application (Table 4) in the "rst hour. Of the total ammonia applied (77)4 kg ha\), 16)6% was lost in 1 h after low-level surface application compared with 2)6% lost after injection (84% reduction). These results are consistent with Moseley et al. (1998) where ammonia emissions, the greatest losses in the "rst 2 h, were also reduced using shallow injection methods rather than surface application.
388
O. PA H L E¹ A¸ .
odour emission (and ammonia emission) and the design of the wing, which had been optimized through soil bin testing was also altered. This demonstrated the need for the narrowest delivery tube commensurate with slurry #ow. The remaining ammonia emission and yield experiments were carried out using the prototype injector (as used in 1999).
3.3.3. Alternative Odour assessment techniques Comparing the results of the odour panel testing [Fig. 5(a)] and the gas detection [Fig. 5(b)], the odour concentrations of the improved injector were signi"cantly lower than above-ground treatments at the 85% level. The gas detector shows that all the treatment means are signi"cantly di!erent at 95% con"dence level, with no di!erence between the control and the injection. The "ndings suggest that the gas detector is more sensitive to odour concentration changes than the odour panel assessment. Further advantages of the gas detector are that the apparatus was portable and readings could be assessed immediately in the "eld.
Fig. 4. Nitrate concentration in leachate during the growing season in (a) 1999 and (b) 2000; , injectio; , surfac; , mineral fertilizer; , control; , average
In this experiment in the year 2000 (Table 4), there were no signi"cant di!erences between odour emissions from low-level application and injection. This was not expected as previous work of Moseley et al. (1998) [Fig. 5(a)] showed that odour emissions for pig slurry were lower with injection than with low-level surface application. Originally, it was planned to use the commercially available injector in the 2000 season as its design was based on the prototype. Having completed the injection for the odour sampling, and observing greater disturbance and wider than expected slots, close inspection of the injector tine slurry delivery tube dimensions revealed di!erences between the prototype injector used by Moseley et al. (1998), and the commercial injector (Fig. 6). On the commercial machine the delivery tube was wider which resulted in a wider slot allowing greater
Fig. 5. Pig slurry odour measurements of the same xeld plots using (a) odour panel testing (Moseley et al., 1998); (b) MGD-1 portable gas detector (Waine, 1999); OU, odour units; LSD, least signixcant diwerence
PO L LU TI O N CO N TR O L BY I NJ EC TI O N O F S LU R RY
389
Fig. 6. Injector width at working depth; (left) prototype 28 mm, and (right) commercial 39 mm
4. Conclusions
References
The "eld experiments, based on the operation of a commercial pig unit, considered the agronomic, cost bene"t and pollution implications of shallow slurry injection compared with low-level surface application of slurry and broadcast of mineral fertilizer.
APHA (1995). Standard methods for the examination of water and wastewater. American Public Health Association (APHA), Washington Brewer A J (1999). Developing Agricultural Controls*Manure and Slurry Application to Land. Farming and Rural Conservation Agency (FRCA), London SW1P 3JR (Presentation given to the National Farm Waste Management Register on 13 September 1999, Royal Agricultural College, Cirencester, UK) CEN (1998). CommitteH EuropeH en de Normalisation working group CEN/TC264/WG2 &Odours'. Draft European Standard for Dynamic Olfactometry. European Union, Brussels EU (1996). Council Directive 96/61 (concerning integrated pollution prevention and control). European Union, Brussels Godwin R J; Grundon P M; Hann M J; Pullen D W M; Wheeler P N (1997). Incorporation of biosolids into arable crops. Final Report for Anglian Water Services, Cran"eld University at Silsoe, Bedford, UK Godwin R J; Pullen D W; O:Dogherty M J O; Hann M J (2000). The design of tines for injection of biosolids into soil at shallow depths through growing arable crops. Proceedings of the 4th International Soil Dynamics Conference, Adelaide Kent N L (1983). Technology of Cereals (3rd Edn.), pp 118. Pergamon Press, Oxford, UK. Lord E I; Mitchell R D J (1998). E!ect of nitrogen inputs to cereals on nitrate leaching from sandy soils. Soil Use and Management, 14, 78}83 MAFF (1986). The Analysis of Agricultural Materials. MAFF Reference Book 427, Ministry of Agriculture, Fisheries and Food Publications, Admail 6000, London SW1A 2XX, UK MAFF (1997). EC directive 96/61 concerning IPPC*implementation in the pig and poultry sector*consultation paper. Ministry of Agriculture, Fisheries and Food, Environmental Protection Division, Branch A, London SW1P 3JR, UK MAFF (1998). The air code (Revised 1998). Ministry of Agriculture, Fisheries and Food Publications, Admail 6000, London SW1A 2XX, UK MAFF (1999). Making better use of livestock manures on arable land. Booklet 1 Managing Livestock Manures. Ministry of Agriculture, Fisheries and Food, Rural and Marine Environment Division, London SW1P 3JR, UK
(1) Shallow injection of slurry into growing crops: (a) had no detrimental e!ect on the yield compared to the use of mineral fertilizer or low-level application of slurry; (b) did not lead to increased levels of ammonium or nitrate leaching compared to application of mineral fertilizer and low-level slurry application; (c) is cost e!ective when compared to alternative low-level application or the combined cost of surface application and subsequent rapid incorporation, which is a requirement of UK environmental protection regulations; and (d) allows application of slurry during a much larger &window' than conventional techniques. (2) Small changes in injector tine design can have a signi"cant e!ect on the e!ectiveness of odour prevention and ammonia emissions. Hence, the soil engaging delivery spout needs to be no more than 28 mm wide. (3) Portable ion mobility gas detectors o!er a good possibility for a rapid and cost-e!ective method for in"eld measurement of agricultural odour. Acknowledgements This work was funded by a Teaching Company Directorate (TCD) contract between the Meat & Livestock Commission and Cran"eld University at Silsoe. Injection equipment was supplied by Greentrac Ltd.
390
O. PA H L E¹ A¸ .
McGechan M B; Wu L (1998). Environmental and economic implications of some slurry management options. Journal of Agricultural Engineering Research, 71(3), 273}283 Misselbrook T H; Hobbs P J; Persaud K C (1997). Use of an electronic nose to measure odour concentration following application of cattle slurry to grassland. Journal of Agricultural Engineering Research, 66, 213}220 Moseley P J; Misselbrook T H; Pain B F; Earl R; Godwin R J (1998). The e!ect of injector tine design on odour and ammonia emissions following injection of bio-solids into arable cropping. Journal of Agricultural Engineering Research, 71, 385}394 Nix J (1999). Farm Management Pocketbook. Wye College Press, Wye, Kent, UK, No. 30 Paakkanen H; Huttunen J (1994). Gas Detector based on Ion Mobility, Finnish Air Pollution Control Technology. Environics Industries OY, PO Box 1750, 70211 Kuopio, Finland Pain B F; Phillips V R; Clarkson C R; Klarenbeek J V (1989). Loss of nitrogen through ammonia volatilization during and following the application of pig or cattle slurry to grassland. Journal of the Science of Food and Agriculture, 47(1), 1}12
Scotford I M; Williams A G (2001). Practicalities, cost and e!ectiveness of a #oating plastic cover to reduce ammonia emissions from a pig slurry lagoon. Journal of Agricultural Engineering Research, doi: This is jaer 20192 accepted for publication on 27 July 2001 Silgram M; Lord E (1999). Large-scale control of nutrient loss: the UK nitrate sensitive areas scheme & HARPNUT initiative. Proceedings of Agriculture and the Environment (Challenges and con#icts for the new millennium), 14}16 April 1999, University of Warwick, UK, pp. 101}107 Sommer S G; Sibbesen E; Nielson T; Schj0rring J K; Olesen J E (1996). A passive #ux sampler for measuring ammonia volatilization from manure storage facilities. Journal of Environmental Quality, 25(2), pp. 241}247 Waine T W (1999). Non-invasive soil property measurement for precision farming. EngD Thesis, Institute of AgriTechnology, Cran"eld University, Silsoe, Bedford MK45 4DT, (unpublished) Warner N L; Godwin R J; Hann M J (1990). An economic analysis of slurry treatment and spreading systems for odour control. The Agricultural Engineer, 45, 100}105
