Animal Food Parameters, Testing, Variation, Economic Implications, & NIR Spectroscopy

Introduction

Animal feed as a general term can be used to describe any food given to domestic animals that are raised for the purposes of either procuring their products or as household pets, although it is more specifically used to describe animal fodder.  Fodder is food grown for animals and specifically carried and fed to them, as opposed to forage which is consumed by animals in the pasture and field. 

Most animal feed is manufactured for swine, fish, poultry, and livestock.  Field grains make up the primary source of animal feed, especially maize and soybean.  Hay, straw, and silage are manufactured from forage material and are also considered a form of animal feed.  Other traditional sources of animal feed include household scraps and food byproducts from processes such as milling, brewing, and ethanol manufacturing.  Leftover material from peanuts, soy, and corn milling can be important sources of animal feed. 

Brewer’s Spent Grains (BSG) is the primary waste product from brewing and has been used as a livestock feed.  Distiller’s Dried Grains with Solubles (DDGS) are the byproducts of ethanol manufacturing for both gasoline and alcoholic beverages.  The manufacturing of DDGS has dramatically increased in recent years and it has become an efficient alternative to corn and soybean meal.  DDGS has a longer shelf-life than corn and soybean meals and comparable digestible energy and metabolizable energy to corn.  It is especially used in swine diets, with many producers using a 20% portion of DDGS. 

A very broad definition for types of animal feed is concentrates and roughages.  Concentrates consist of cereal grains, byproducts, and oil meals or cakes.  They are high in energy.  Grains used for concentrates include corn, oats, barley, sorghum, rye, and wheat.  Oil meals or cakes include soybean, canola, cottonseed, and peanuts.  Byproducts can be used from processing of sugar beets, sugarcane, animals, fish, and both BSG and DDGS.

Roughages are grass pasture or plant parts, such as grasses, hays, silage, root crops, straw, and cornstalks.  Poultry, swine, and fish are fed with concentrates.  The digestive requirements of livestock ruminants require a diet formulated mostly from roughages.

Parameters of Interest and Testing Methods

Parameters of Interest

The ultimate goal of any feeding program is to achieve a balance among available feed ingredients where the nutrient composition of the total ration meets the daily nutritional needs of the animal. There are a number of important parameters to measure through the process of manufacturing animal feed.

Dry matter (DM) is defined as the non-moisture portion of a feed ingredient or diet. The sum of DM and moisture content will always equal 100%. The dry matter portion contains all the essential nutrients. Pasture and liquid feeds have a DM content between 10% and 25% while dried feeds usually have less than 15% moisture. In the case of hay and most dried feeds, it is important to keep the moisture below 15% to prevent mold growth. The DM and moisture content are extremely important for comparing the nutrient content of different feeds on an equal basis.

The nutrient content of a feed can be determined on an “As Fed” (AF) basis which includes the moisture content or DM (moisture excluded) basis. A good comparative example is pasture and hay. Pasture will have a much lower nutrient content than dried hay on an AF basis but a DM correction will equalize the nutrient content. This type of analysis is essential in feed formulation where different raw materials are mixed to create the final animal food product.

Crude protein is considered a good indication of most animal food products. Protein is not directly measured but rather estimated from nitrogen content using traditional wet chemistry methods. All biological proteins contain about 16% nitrogen and nitrogen analysis is used to get crude protein content. However, this method is only adequate for determining apparent protein and cannot distinguish if some type of adulteration is present in the product.

An excellent example of this was shown in some high profile incidents with melamine adulteration in pet food. Melamine was intentionally added to wheat gluten to increase the protein content. The industry standard test was unable to detect the adulteration because it measures nitrogen alone to correlate to protein.

Generally, the relationship between proteins and fibers is a good indicator of quality. As forage plants mature, the crude protein is diluted as fiber content increases. Standard protein analysis does not distinguish between digestible and non-digestible proteins. Some feedstuffs, especially medium moisture silages, can undergo heating during storage which causes some proteins to bind to carbohydrates, rendering them indigestible. Analysis for acid detergent fiber (ADF) can be used to account for the indigestible proteins.

Fiber analysis is important in any animal feed or forage as measures the total cell wall content. Neutral detergent fiber (NDF) is defined as the portion of a sample that is insoluble in neutral detergent and contains the primary components of the plant cell wall which are mostly hemicellulose, cellulose, and lignin. As cell wall production increases with advancing plant maturity, NDF content will increase while decreasing dry matter intake.

For legume forages, NDF content below 40% is good quality while above 50% is poor. For grass, NDF below 50% is high quality while greater than 60% is poor. ADF is a subset of NDF and specifically measures poorly digestible cell wall components like lignin and cellulose. It is often used to predict the energy content of feeds.

Carbohydrates provide most of the energy for livestock. They make up 65% to 75% of the dry weight of most grains, forages, and roughages. They are broadly classified as structural which are found in the cell walls and non-structural which are found inside the cells of plants. The cell wall substances found in NDF make up the structural carbohydrates. Non-structural carbohydrates are composed of starches, sugars, fructans, and organic acids.

Determination of non-structural carbohydrates varies by the product. Starch is the primary type in corn, grains, and their by-products. It can be determined by testing for dextrose and multiplying it by a factor. Grass forages contain little if any starch but they do contain sucrose and fructans, which can appear in the starch fraction.

Fats are a good source of energy and aid in the absorption and transportation of fat soluble vitamins, such as Vitamins A, D, E, and K. In concentrates, they consist largely of glycerides of fatty acids that generally provide over two times the amount of energy per weight of carbohydrates. When animal fats are low in price, adding up to 5% of the total ration can not only provide a practical source of energy, it can also have a palatable effect on the softness and taste of processed feed rations. Fats and fatty acids can also improve energy intake and milk yield in dairy cows.

Ash is a measurement of total mineral content, although it does not separate out any individual minerals. The inorganic material that makes up ash mineral content can be divided into two types of minerals: endogenous and exogenous.

Endogenous minerals are minerals that plants normally contain like calcium, phosphorus, potassium, and magnesium. They have high nutritional value. Exogenous minerals are exterior to normal plant minerals and are primarily associated with soil.

Ash testing of final products will also contain added supplements of minerals, such as salt and buffers. Abnormally high ash content usually indicates soil contamination which is undesirable. While the ash testing procedure cannot distinguish between the three types of minerals, good estimates can be inferred from total ash and comparing it to known reference values of the product.

Energy content is used to compare feeds and evaluate quality. It is not directly measured but calculated from regression equations using nutrient values. ADF and CP are two nutrients traditionally used to calculate energy values. Some units used for feed energy values are total digestible nutrients (TDN), net energy (NE), digestible energy (DE), and metabolizable energy (ME).

Vitamins are needed for various body processes in very minute amounts. For most animals, there are around fourteen different vitamins that are recognized to have specific roles in metabolic functions. Vitamin A is very important for good cattle production. Good quality forage contains carotene, which is naturally converted to Vitamin A in the body. Prolonged storage of feeds or excessive bleaching during curing can destroy carotene.

Vitamin D supplementation is often needed in housed animals that get very little sunlight. It is important to keep feed free of mycotoxins, pesticide residues, and heavy metals as contamination can result in health problems for animals and even death. If there is a danger or concern then tests must be conducted to ensure raw materials are free of contamination.

Testing Methods

It is important to test for nutritional parameters in animal food, but traditional methods are often expensive, time-consuming, and often ill-suited for large scale testing. The variation in natural products and the raw materials used to make animal food can be very large but traditional methods have no way to account for it.

Some tests are physical tests and while these are often relatively simple, they are still time-consuming. Moisture and ash are two examples of such physical tests. The traditional method for determining moisture is the gravimetric method. Sample weight is measured before drying in an oven for two hours at 105°C. The weight difference divided by the original weight gives the moisture content. Some new equipment has evolved that can do the same test faster, such as halogen moisture meters. They still take a minimum of several minutes for analysis. Ash content is determined by burning a sample in a furnace above 500°C, which burns the organic compounds off and leaves the inorganic mineral materials remaining as ash. It is insufficient for determining individual mineral content, which must be done by expensive methods like Atomic Absorption Spectroscopy (AAS) and Inductively Coupled Plasma Analysis (ICP).

Nutrient tests use expensive equipment, toxic chemicals, and are very time-consuming. Protein is determined from the amount of nitrogen. Determination of nitrogen content is most commonly performed by the Kjeldahl method. The sample is digested for a period of time with sulfuric acid and the solution is transferred to a steam distillation unit that performs distillation in the presence of boric acid. The distillate is then titrated to determine nitrogen content. An alternate method for determining protein is the Dumas method, which requires total combustion. However, both these methods are unable to distinguish between protein types.

Fat content is measured by Soxhlet extraction. The sample is bound within a thimble and a boiling organic solvent such as hexane is heated in a round bottomed flask. It evaporates under the influence of heat and extracts soluble compounds when it crosses the sample and refluxes the solubles back to the flask along with the solvent. After the process recurs and the soluble extract (fats) is concentrated in the flask with the solvent, the solvent is separated from fat content by means of rotary evaporation. The dried fat substance is then weighed to determine the fat content.

Fiber analysis requires digesting the sample with sulfuric acid to get rid of sugars and adding sodium hydroxide to get rid of proteins. Carbohydrate content is traditionally determined through calculation from the total content after subtracting fats, proteins, and fibers. Direct measurement of carbohydrate requires liquid chromatography.

More detailed analyses for minerals, fatty acids, vitamins, amino acids, pesticides, toxins, and heavy metals require especially complicated and expensive methods like gas chromatography, mass spectroscopy, and high performance liquid chromatography. For even basic physical and nutrient analysis, the following equipment is required: analytical balance, drying oven, muffle furnace, heating device, extraction and reflux unit, digestion unit, distillation unit, automatic titration unit, filtration unit, temperature controlled water bath, and volumetric equipment.

The cost, labor, and time required to perform analysis of a large amount of samples is impractical and for this reason alone, animal food cannot be tested on the large scale necessary to account for sample variation. However, the consequences of insufficient testing methods can be great from both a nutritional and health standpoint for the animals as well as economically for buyers and sellers of both the raw materials and finished products of animal food.

Raw Material Variation

There have been studies conducted to determine the actual variation in samples of raw materials used for animal food manufacturing. One recent study examined the impact of nutrient variability on the profitability of chicken meat production. Feed costs account for over 65% of the live production costs of poultry. Chemical analysis of feed ingredient samples is impractical on a real-time basis due to the time and cost involved. Variability in feed ingredients originates from three sources: the raw ingredient itself, sampling techniques, and type of analysis (including the normal analytical variability and differences in analytical methodology).

Near-Infrared spectroscopy is often used within integrated operations to instantaneously estimate the nutrient content of feedstuffs. But NIR calibrations are only as accurate and representative of the feed as the sample that was taken and proper sampling methods are crucial to ensure as much accuracy as possible.

When NIR methods are not available, nutritionists often rely on historical book values to estimate nutrient values. But book values rarely include the standard deviation and data distribution of nutrients nor do they provide specific information on the region and season where the raw materials were grown. In order to compensate for discrepancies between book values and true values in feed, safety margins are applied to formulations to attempt to meet the minimum nutrient requirements. However, increased safety margins add to diet cost and are complicated by the difference between analysis of samples and the actual nutrient content of the ingredients being manufactured at a given time. Multiple batches of an ingredient can be delivered to the mill and stored within the same silo. If nutrients do not meet the minimum requirements shown on the label, they may be legally liable and insufficient nutrients affect the composition of the final product.

The study focused on protein variability, an expensive and crucial component of poultry diet. The variability in crude protein of wheat-based poultry diets was estimated from the nutrient content of feed ingredients. Starter, grower, finisher, and withdrawal diets were used in the calculations. Computer simulations were used to estimate the likelihood that a diet mixed to optimal specifications may in fact fall below the recommended specifications.

One specific example shown for withdrawal diets showed that diets formulated to 192 g/kg of crude protein from book values have a 10% probability of falling below 182 g/kg of crude protein. Subsequent analysis showed improved sampling and real-time analysis that decreased the variability of crude protein by 25% for each ingredient in the formulation reduced the probability of the diet falling below the minimum specification to less than 5%.

Based on these conclusions, the economic consequences of misestimating the nutrient content of feed ingredients was examined. Within this example, the results showed that it was possible to incur a 63% reduction in gross margin from one cycle of 30,000 broilers simply by overestimating the nutrient content of feed. This amount can be a difference up to $19,053 per cycle.

Assuming a large poultry company can produce one thousand broiler cycles per year, this equates to a loss of up to $19 million annually. Expanding this analysis to include the entire global chicken meat industry, which produces approximately 304 million metric tons of feed per year, and the total cost of more than $110 billion for feed shows staggering potential losses and economic consequences from variability within chicken feed ingredients.

While NIR spectroscopy cannot solve all the issues related to nutrient variation in poultry feed, it does offer a fast, non-invasive technique that can determine multiple nutritional parameters in a single measurement, offering big advantages over traditional wet chemical methods or estimating nutrient content from book values.

Another study examined variation in the dry matter content of corn silage and this study examined the change in feeding rations that would be practical if in fact the dry matter could be determined on a timely basis before feeding.

Feed cost is one of the largest expenses on dairy farms. Overfeeding, underfeeding, or an improperly balanced diet can impair cow health, decrease milk production, and result in negative environment impacts.

Dry matter content is extremely important in any type of animal feed and the variation in dry matter was examined. Over five days, samples of corn silage were analyzed for multiple parameters of interest. Samples were pulled hourly and tested for dry matter, ADF, ash, crude fat, crude protein, NDF, and starch. Variation was minimal for crude protein, fat, and ash. High variation was observed for the fiber parameters and for dry matter, and especially from day to day for the dry matter.

The dry matter value is assumed to be 33% and while many of the samples were in between 30% and 35%, most days showed a large fluctuation in values. Day 1 had a high value around 37% and a low value below 25%, as did Day 2. Day 3 showed less deviation but still had a sample around 32% and one just above 25%. Day 4 showed the highest deviation with a sample around 42% and the lowest sample around 24%. No samples above 35% were observed for days 4 and 5 but the low samples for both days were around 23%.

This type of analysis is extremely important in determining feed rations but is difficult to achieve in practice, even with a relatively simple test like moisture and dry matter. For example, if a ration calls for fifty pounds of corn silage on an as-fed basis and the dry matter is assumed to be 33%, 16.5 pounds of corn silage will be fed to the cow. For each one point deviation of the actual dry matter from the assumed dry matter, the farmer will overfeed or underfeed the cow by 0.5 pounds of corn silage per day. If the actual dry matter was 28%, then the cows would only be fed 14 pounds of corn silage on a dry matter basis.

Both the economics and nutritional implications of insufficient analysis and testing methods are strong. One testing method which has gained great traction for on-site analysis of raw materials used for animal food is NIR spectroscopy. It offers the advantages of being fast and non-invasive. It does not require sample destruction or the use of toxic chemicals and solvents. It has the ability to determine multiple chemical, physical, and nutritional parameters from a single measurement.

Most importantly of all, it offers a means to test large amounts of samples in a manner that cannot be done with traditional methods and the ability to account for variation in raw materials that are used for animal food.

Economic Implications of Using NIR Spectroscopy

Using NIR spectroscopy as a method for testing animal food can have a strong economic impact as well as better nutritional intake and health for animals. A study conducted in Italy examined the benefits of using an NIR spectrometer for ration analysis of dairy cows raised for the purpose of producing milk for a specific type of cheese. The study was conducted over a six month period. Eight breeding farms participated in the study and separate groups of cows were observed. One group was fed rations that were analyzed using NIR spectroscopy and the other group was fed rations tested with traditional testing methods. Two hundred total cows were used in the study.

The first analysis was for the cost of rations that were fed to the cows to meet the proper nutritional intake. The daily cost per cow that was fed rations not tested by NIR was €7.04 and the daily cost of those fed samples that were tested was €6.96. For a herd of two hundred cows over a monthly period, the total savings from using NIR tested rations was €487.

The benefits of increased milk production were even greater. Cows fed NIR tested samples produced an average of 32.5 kg of milk daily while those fed samples that were not tested produced just under 30 kg of milk per day. For a herd of two hundred cows, this resulted in over 500 kg of extra milk. At a milk price of €0.44/ kg, the increase in revenue from milk per month was €6795 and the total monthly margin increase including reduced feed costs was €7282. In American dollars, the feed costs were $0.09 less and milk production was 5.6 pounds more per cow per day. Estimated yearly increased revenue in milk sales is $73,584 and reduction in feed cost is $6,570 for a two hundred cow farm.

Increased production was attributed to a more consistent ration being fed to the herd, which has been proven to be beneficial to the health and milk production of cows. The lower feed cost was attributed to being able to feed more precisely to the needs of the herd, a decrease in feed waste, and a reduction in the risk of imbalances in the rations due to rainy events. Testing also showed improved health from the analysis of blood parameters and the index of mastitis in the milk. The study showed how using NIR spectroscopy to analyze feed rations of dairy cows can have strong economic and nutritional benefits.

Another study examined the potential economic benefits of using NIR spectroscopy for on-site feed and forage testing by performing Return of Investment (ROI) analysis to determine the feasibility of using an on-farm NIR testing system. The basis of the study was to establish the baseline economics of using an NIR system to test feed rations formulated on the farm.

A timeframe of two years was used for the ROI analysis. A discussion was presented about the variation in nutritional content of animal feed, impact of feed rations on the yield of animal products, and the shortcoming of traditional testing methods. One type of feed used in the analysis that shows a tremendous amount of variation and is increasing in its use due to low cost is Distillers Dried Grains with Solubles (DDGS), the leftover material of corn that is used for ethanol manufacturing. Variation in fat content in DDGS can be five times greater of that in corn and the crude protein content variation can be sixteen times higher of that in corn.

In order to assess the ability of an NIR system to analyze nutritional content in animal feed without conducting actual tests, a review of research literature was conducted. Twenty-three separate ingredients used for animal feed manufacturing were used. Data were obtained for both nutrient variation and market price for all ingredients. For costs of the NIR system, assumptions were made for the initial cost of the system and calibration work as well as labor costs on an hourly basis and per truck tested. An Excel model was used for analysis.

For all 23 ingredients, price values for protein, dry matter, and Total Digestible Nutrients (TDN) were calculated based on the mean and standard deviation values. An assumption was made that the number of samples required to test a 25 ton truckload of raw material was ten.

A “breakeven” analysis was conducted for each of the ingredients based on each of the three nutritional parameters. Statistical analysis was calculated for the savings based on a given parameter being one standard deviation increase from the mean and subsequent cost savings from this information. The number of truckloads required for the breakeven point was far lower for protein than the other two nutritional parameters, which makes sense because protein is one of the most expensive nutrients in feed rations. The ingredients with the lowest breakeven point in terms of trucks tested were fish meal (25), potato byproduct (30), and soy byproduct (50). Grains such as wheat, barley, corn, and oats were highest ranging from three hundred and thirty to four hundred and fifty-five. Fish meal is a very expensive product so proper protein analysis resulted in a fairly quick ROI.

The study showed the economic benefits of using an NIR system for animal feed raw material products. Results would show even more benefits if variation in multiple ingredients were considered simultaneously, as this study only considered variation in a single nutrient at a time.

The benefits of using NIR spectroscopy as an analytical tool for animal food analysis are vast and are advancing in practice as technology strides have enabled the use of it in all steps of the animal food manufacturing process.

Conclusion

The variability in raw materials used for animal food manufacturing, shortcomings in traditional testing methods, and subsequent economic impact on not only the manufacturing of animal food to proper specifications but also the output of animal products present massive challenges. Traditional means of measuring nutrient content during feed formulation and the subsequent manufacturing process have proven to be inadequate for properly accounting for variability.

NIR spectroscopy has emerged as a fast, non-invasive testing method for parameters of interest in animal food quality control. It offers the advantages of little to no sample preparation, the ability to be used for large-scale testing, and is able to determine multiple parameters with a single measurement. In some cases, using NIR spectroscopy can provide an even higher specificity and sensitivity than traditional methods.

A perfect example of this is the ability to distinguish melamine from crude protein content, which cannot be done using the industry standards tests for protein. This example and many others have shown NIR spectroscopy to be a powerful tool for adulteration detection in all segments of the food industry, including animal food.

NIR spectroscopy can be used for analysis from the beginning to the end of the animal food manufacturing process. It is used for raw material ID and component analysis as well as a tool for checking variation and shipment within batches. NIR spectroscopy is an aid for feed formulation by both determining variation in materials and giving manufacturers the ability to make mixing adjustments on the fly. It is used as a real-time process control tool in the feed mill. When products are finished, NIR spectroscopy can be used for the final quality control check.

While the principles of NIR spectroscopy have been well-known for many years, recent technological advances have enabled its advancement as a practical tool in industry. Handheld and portable instruments have enabled analysis in the field. NIR spectroscopy requires the creation of calibration models to correlate NIR spectra to parameters of interest. Companies have created pre-built calibration models for parameters of interest, reducing both the labor and costs required to implement NIR spectroscopy. Third-party programs have enabled the use of web-based technology that supports database management, quality control, and trend analysis to optimize processes and protocols for animal food analysis.

The practical use of NIR spectroscopy as a tool throughout the animal food manufacturing process and technological advances that have enabled its use are examined in the next section.