D.R. CHADWICK1, S.K.E. BROOKMAN1, J. WILLIAMS2, K.A. SMITH3, B.J. CHAMBERS4, I.M. SCOTFORD5 and T.R. CUMBY5
1Institute of Grassland and Environmental Research, North Wyke Research Station, Okehampton, Devon EX20 2SB UK. 2ADAS Boxworth, Cambridge CB3 8NN UK 3ADAS Wolverhampton, Woodthorne, Wolverhampton WV6 8TQ UK. 4ADAS Gleadthorpe, Nottinghamshire NG20 9PF UK. 5Silsoe Research Institute, Wrest Park, Silsoe, Bedfordshire MK45 4HS UK.
Animal manures contain significant quantities of nutrients that, if managed carefully, can be used to sustain crops and reduce the requirement for inorganic fertilizers. However, it is evident that farmers lack confidence in the nutrient content and availability from animal manures. In order to exploit this resource effectively, farmers and consultants need good quantitative information on nutrient supply from manures. Recent survey data has shown that UK farmers are more likely to trust information that they can gather themselves. Laboratory analyses of animal manures are relatively expensive and there is often a delay of up to 2 weeks in reporting back to the farmer. This time delay is seen as an inconvenience. For these reasons, the development of reliable, rapid and low-cost techniques for the assessment of manure nutrient supply should encourage the uptake of improved techniques for the management and utilization of manure nutrients.
Several on-farm and rapid techniques are available to quantify the nutrient content of manures, e.g. hydrometers, in-line nutrient sensors on slurry tankers, as well as near infra-red spectroscopy (NIRS) and farmer-operated test kits for analysis of the ammonium-N content of manures. In this article, we describe some recent studies assessing the accuracy and reliability of such techniques.
Quantofix meter for testing ammonium-N content in slurry manure.
Agros meter for testing ammonium-N content in slurry manure.
In a comparative study of on-farm ammonium-N (NH4-N) tests, we determined the accuracy of five different techniques, namely: Agros and Quantofix nitrogen meters, reflectometers, and conductivity and selective ammonium electrodes. The techniques were tested under laboratory conditions on 40 slurry, 25 farmyard manure (FYM) and 20 poultry manure samples from commercial units. The results obtained from each method were compared with those from standard laboratory analysis techniques for measuring NH4-N concentrations. There were strong relationships (P<0.001) between the slurry laboratory analyses and the five techniques. For slurries, regression analysis showed that the relationship between laboratory NH4-N concentrations and those measured by the Agros and Quantofix nitrogen meters had r2 values greater than 95%, with regression line slopes close to unity (Figs. 1 and 2). This shows that the relationship is highly reliable and that the methods give the same value.
Figure 1. Relationship between Agros estimates of slurry NH4-N content
and laboratory analyses.
Figure 2. Relationship between Quantofix estimates of slurry ammonium-N (NH4-N) content and laboratory analyses.
The conductivity electrode and reflectometer generally gave good agreement with the laboratory results for NH4-N, but the strength of the relationships varied according to animal slurry type (r2 >80% for pig slurry; >60% for cattle slurry). The relationship between the ion specific electrode and laboratory analysis of NH4-N was not linear but exponential, which resulted in problems when estimating slurry NH4-N at high concentrations. The readings also tended to drift with time, which made it difficult to decide on the ‘correct’ value.
The most successful techniques in the laboratory (N meters, conductivity meter and reflectometer) were tested by 16 farmers to assess their suitability for use on-farm. The on-farm slurry tests showed that the farmers could also obtain good agreement with the laboratory NH4-N analysis results with both nitrogen meters (the Agros and Quantofix) and the conductivity meter. However, the reflectometer was assessed as ‘impractical’ for on-farm use. The robust construction, simple operation and accuracy of the nitrogen meters meant that they were very suitable for on-farm use and were available for purchase in the market place. The simple mode of operation and ‘instant’ readout provided by the conductivity meter was also popular with the farmers, but a marketable product was not available for purchase by farmers at the time.
Slurry dry matter content is a useful indicator of nutrient content, particularly P content (Fig. 3). A commercially available, calibrated glass hydrometer was used to assess slurry dry matter content. In the laboratory assessments, there was a strong relationship between the hydrometer readings and laboratory measurements of slurry dry matter content (P<0.001, r2=78%). Also, there was reasonable agreement between the on-farm hydrometer measurements and laboratory dry matter measurements (P<0.001, r2=44%). However, the slope of the line was less than unity (0.65), which may have been due to the farmers not agitating the sample thoroughly before testing, or taking too long before reading the hydrometer.
Figure 3. The relationship between cattle slurry dry matter content
and slurry phosphate content. * kg/m3 = 0.1%
Automatic in-line nutrient sensing offers greater convenience to farmers and contractors compared with approaches based on manual sampling. For example, nutrient estimates are in realtime, and sampling errors are reduced since the method analyzes and records the nutrients in all of the slurry being spread load-by-load; thus complete mixing of stored slurry to obtain representative samples is unnecessary. Initially, this approach was developed in the UK as a small-scale prototype incorporating a combination of standard industrial sensors to measure a set of physical and chemical properties of the slurry. The measured properties of the slurry were electrical conductivity, ammonium ions, density, temperature, differential pressure, flow, pH and redox. This device was used to assess 160 different pig and cattle slurries in 4 European Countries. Each slurry was analyzed for NH4-N, total phosphorus (TP) and total potassium (TK) using conventional laboratory techniques. The output data from the individual sensors were compared with the laboratory values, and statistical correlations were identified from these data sets by linear regression analyses. Hence, algorithms were identified to relate the physical and chemical properties of the slurries to their NH4-N, total P and total K concentrations. NH4-N concentrations were related to the electrical conductivity of the slurry, P was related to slurry density and K related to ammonium ion concentrations.
Based on the promising results obtained from the smallscale prototype, a full-scale version of the in-line slurry nutrient sensing system was constructed with sensors fitted in the side of a 7 m3 (2000 gal) slurry tanker. Thus, as the slurry was mixed by recirculation in the tanker, the slurry properties could be measured and converted to nutrient concentrations in real-time by a computer using the previously derived algorithms. This system estimated the NH4-N, total P and total K to respective accuracies of +/-0.29, +/-0.29 and +/-0.79 kg/m3. The respective ranges over which these values were determined were 0.63 - 5.29 kg/m3; 0.12 - 0.71 kg/m3 and 0.81 - 6.49 kg/m3. Further developments to include global positioning systems (GPS), enhanced data management and in-line systems for solid manures are envisaged.
For solid manures, it would appear that analysis by near infra-red spectroscopy (NIRS) could be a promising approach to quantify several nutrients. This technique has been tested on a limited number of solid manure samples in the UK. NIR spectrophotometer calibration equations for organic matter, dry matter, total N, NH4-N and uric acid-N showed good correlations and low standard errors of calibration. Preliminary validation experiments were undertaken on further manure samples, which gave encouraging results for all the analytes, with the exception of pH, suggesting that it should be possible to develop NIRS into a routine technique. Sample introduction to the NIRS is known to be a critical factor influencing accuracy. Therefore, in a new study, a robust homogenization procedure is being developed.
A potential advantage of using NIRS over conventional laboratory techniques is the use of larger sample size. Also, the rapidity of analysis and potentially lower costs mean that a greater number of samples can be analyszed per unit cost. These advantages mean that analysis can be carried out on a more representative sample of the manure.
Robust and accurate on-farm and rapid tests are available, or are in the process of being developed. However, if the samples being analysed are not representative of the manure supply, then the farmer is not going to make most efficient use of the manure nutrients. A current study in the UK is assessing improvements to sampling protocols for slurry stores and solid manure heaps.
Hydrometers, Agros and Quantofix meters are available from:
