Hay Testing with NIRS
Does It Make Sense for You This Year?
By Shelagh Niblock, PAS
Accurate laboratory analysis is the backbone of efficient ration planning for many horse owners today. Whether you have an elite competition horse or a senior who is essentially a pet, horse owners are becoming increasingly aware that informed forage buying decisions can make a significant difference in the health and performance of their horse.
Easily accessible forage analysis options are more common than they were even ten years ago, but maneuvering your way around them can be confusing. When choosing a lab to analyse your hay, always choose one that specializes in forage and provides the following nutrients as part of their analysis package: crude protein (CP), acid detergent fibre (ADF), neutral detergent fibre (NDF), water soluble carbohydrate (WSC), and ethanol soluble carbohydrate (ESC). Once you find the lab, they will generally have an easy-to-follow directory of the different forage analysis packages from which to choose.
What are you testing for?
One of the more important nutrients you will be looking for in the lab analysis is protein, and if you have an overweight horse, or a horse with the metabolic conditions - insulin resistance or Pars Pituitary Intermedia Dysfunction (PPID or Cushing’s Disease) - you may also be looking for the “sugar” or water soluble and ethanol soluble carbohydrate (WSC and ESC respectively) values. You might also need trace mineral analysis including copper, iron, manganese, or zinc values. Alternatively, if you have a horse afflicted with Hyperkalemic Periodic Paralysis (HYPP), you will most definitely be looking for a total potassium percentage in the hay you are considering buying. Your analysis objectives should dictate whether you choose a lab that utilizes wet chemistry methodology or, alternatively, a lab that utilizes Near Infrared Spectroscopy (NIRS) methodology.
NIRS or Wet Chemistry? What’s the difference?
Traditionally, laboratory analysis was completed in a lab using a combination of heat and reagents to determine the nutrient composition of a forage sample. This is called wet chemistry, and it is still the gold standard for accuracy in laboratory results. While wet chemistry methods are still widely practised today, another method of laboratory analysis has gained in popularity over the last 25 years due to its speed and efficiency. NIRS or Near Infrared Spectroscopy is a laboratory analysis methodology that utilizes near infrared light and the spectra reflected back when it is directed at an unknown test sample. The unique spectra “fingerprint” generated by the test material can be used to predict the nutritional components of the sample. The spectra of the unknown sample is compared to a data set of spectra from known samples, and similarities are compared using computer software. The “calibration set” used to predict the nutritional components of the unknown sample is constructed through the use of data derived through wet chemistry methods, mathematical calculations, statistics, and computer software designed for this purpose.
Good quality hay is the foundation of a balanced diet for horses of all ages. Horse owners are becoming increasingly aware of the importance of the nutritional components of the hay they feed. Photo: iStock/Alexa Khruscheva
NIRS is inexpensive to run because it requires no specialized reagents. It is also environmentally friendly and fast. Typically, a hay analysis completed using wet chemistry methodology will take at least a week, while the same forage sample could be analysed using NIRS in about one hour. NIRS is very accurate in predicting nutrients such as protein, but it is less accurate than wet chemistry at predicting nutrients such as fat content. If you want an analysis of your hay sample that includes trace minerals such as copper, zinc, manganese, or iron, NIRS is not an acceptable analytical method to use, and your forage sample must be analyzed using wet chemistry methods. Macro minerals such as calcium, phosphorous, magnesium, and potassium are often reported on a hay analysis done using NIRS methodology. The results can be considered reasonably accurate because of the association these minerals have in the plant with the organic components of cellulose and hemicellulose, which are accurately quantified through analysis using NIRS.
What is NIRS and how does it work?
NIRS analysis is based on the use of near infrared light, which is the fraction of light with a wavelength of 700 to about 2500 nanometres (nm). This is the fraction of light that the human eye cannot see, but that humans can feel as heat. When a beam of infrared light is shone onto a test sample of hay, the energy from the light is absorbed by the organic material in the sample. This organic material includes the nutritional components of protein, fibre, and sugar. Each of the nutrients present absorb the energy at different levels and reflect the energy back that is not absorbed. The energy reflected back is called the “spectra” and it will take on unique characteristics determined by the nutritional composition of the organic material. The spectra from a hay sample will be very closely correlated to its nutritional composition.
How do we use the spectra to determine the forage analysis? NIRS is based on developing a “calibration set,” which is a set of similar forage samples where the nutritional analysis of each is determined using wet chemistry methods and saved in a computer database. Each sample in the set is also “scanned” using near infrared light, and the spectra for each sample is also saved in the database. The data from the wet chemistry results for each sample is aligned with the spectra for each sample through the use of computer software. The resulting “calibration set” can be used as a point of reference when scanning an unknown sample. The computer software takes the spectra from the unknown sample and compares it to the calibration set of known analyses and “predicts” the nutritional analysis of the unknown sample. A calibration set could be created with as few as 10 samples of similar type hays, but in order to develop a “robust” calibration set that is very accurate, it should have no fewer than 100 samples in it. Often calibration sets will be comprised of data from thousands of samples. Labs such as Equi-Analytical/Dairy One or Cumberland Valley Analytical, which routinely analyse forages such as grass hay or alfalfa hay, will have very robust calibration sets with thousands of forage samples included. Any lab offering NIRS forage testing service will have individual calibration sets for a variety of forages, including grass hays, legume or alfalfa hays, grass silage or haylage and corn silage (predominantly used as feed for cattle).
Accuracy is important!
A good NIRS analytical process has a system of checks and balances built in to ensure that the results it generates for a test sample are accurate. Operators of NIRS analysis systems are trained to look for signs that the sample being analysed is an “outlier” or a sample that, for whatever reason, doesn’t fit the calibration set. This could be caused by the presence of foreign material in the sample, such as sticks or soil or a lot of weeds. Weeds will reflect the near infrared light differently from the grass in the hay and so can skew the nutrient prediction. The software will have parameters that must be met when generating the NIRS analysis. Samples that “don’t fit” the calibration set are generally sent out for wet chemistry to ensure the client gets an accurate analysis.
It is important that the hay you submit for NIRS analysis is being “scanned” using a calibration set of similar forages. Otherwise, it could be classed as an outlier when really, it just isn’t well represented by the data set. Again, this is where a good lab with trained laboratory personnel will be able to determine if the sample can be accurately analysed using NIRS. Sometimes, less common forages like cereal hays are difficult to analyse accurately using NIRS because they aren’t well represented in a grass hay calibration set. If in doubt about the accuracy of the NIRS result on a forage, it’s always wise to use wet chemistry methods as a back-up.
Which is better? Wet chemistry or NIRS?
Wet chemistry results are usually very well correlated with NIRS results on the same sample. A recent paper by the respected equine researcher Dr. Patricia Harris of the Waltham Institute in the UK, (PA Harris et al. Journal of Equine Veterinary science 71 13-20) has shown that statistically, wet chemistry results and NIRS results on the same forages are generally very similar for most nutrients. The only component that shows wider variation on both NIRS and wet chemistry analyses is sugar (WSC) and Harris speculates that it is probably due to the wide range of fructan content possible in cool season grass hays such as timothy, tall fescue, orchard grass, rye grass, and native grass species. Statistically, it is very difficult to predict WSC with great accuracy, and results on cool season grass hays, whether analysed using wet chemistry methods or NIRS methods, can vary plus or minus two percentage points between labs or samples on the same stack of hay. Those of us with the IR, EMS, or PPID horse know that finding low WSC hay with reliably consistent results can be frustrating.
If you are responsible for feeding an overweight horse, or one with a metabolic condition such as insulin resistance or PPID, have their hay analyzed for sugar content. Photo: Shutterstock/Katarzyna Mazurowska
Why doesn’t NIRS predict minerals well?
Minerals in a forage sample do not reflect the energy of infrared light and therefore do not generate a spectra. Macro minerals, such as calcium, phosphorous, magnesium, and potassium are usually associated with the fibre fraction of the plant and so can be approximated by NIRS, but if the levels of these minerals in your forage is particularly important it is advisable to get them determined using wet chemistry analysis.
Is NIRS accurate?
NIRS is well respected in the scientific, manufacturing, and processing fields as an accurate methodology for analysis of organic materials. It is recognized by the Association of Official Analytical Collaboration (AOAC) International under the Official Methods of AnalysisSM (OMA) program. Labs offering NIRS services are scrutinized carefully in order to ensure that high analytical standards are being met. It is important to remember though that the analytical results you get for your money will only be as good as the sample that you submit. Accurate sampling of forages such as hay is of great importance. Submitting a flake of hay in a garbage bag will get you an analysis that may not be representative of the stack of hay in your barn. It is also important to submit the hay in a form suitable for scanning. Hay samples are scanned using NIRS methodology in either a whole “as sampled” form or in a fine ground form, depending on the equipment in the lab. Either way, your sample must be submitted in a form with the pieces chopped up into segments approximately one inch or less in length. This is easily achieved if you are using a proper hay drill to collect your sample. If you aren’t, you will need to get out the scissors and chop up your sample before submitting. Long pieces of hay simply cannot be scanned accurately.
The only way to accurately determine the nutrient content of your hay is to have it tested. A representative forage sample is important for accurate analysis, and the best way to take a good test sample is with a hay probe. Photo: Pam MacKenzie Photography
Are there local NIRS Forage Labs?
The two best known and respected NIRS forage testing labs in North America, which generate reports suitable for horse owners, are Equi-Analytical/Dairy One in Ithaca, New York, and Cumberland Valley Analytical in Waynesboro, Pennsylvania. Both these labs have satellite labs in Canada. Satellite labs are permitted to use the very large calibration sets built and maintained by the parent company. This can be an advantage to the horse owner, as a large calibration set is usually a more robust one and therefore more accurate. Satellite labs will also have expertise available to help you to decide which analytical package is best to get the information necessary for your horse. If you need a wet chemistry package the satellite lab will look after getting your sample into the appropriate channels to get it done for you.
Does testing your hay with NIRS make sense for you?
Getting your hay analysed for nutrient content always makes good sense, and utilizing the quicker and less expensive NIRS methodology may as well. NIRS analysis on forage samples will meet the requirements for most equine rations. If you need trace mineral analysis on your hay, then you should be considering a wet chemistry analysis. However, remember that whether you are submitting a sample for wet chemistry analysis or NIRS analysis, the sample you take is very important in determining the usefulness of the results. A poor sample will generate poor results.
If you have a horse with PPID or insulin resistance, ensure that the analysis package you choose reports WSC and ESC and not just sugar. Keep in mind that WSC and ESC are notoriously variable. In other words, just because your analysis comes back with an NSC of slightly over 12 percent doesn’t necessarily mean that your hay is unsuitable for your metabolic horse.
You will always be further ahead in your goal of having a happy, healthy horse if you know the analysis of the forage he eats. Consider contacting the satellite lab in your area for information about analysis options and the costs to further your achievement of this goal.
This article was originally published in the Summer 2020 issue of Canadian Horse Journal.
Main Photo: iStock/Magbug