21/11/2025
🌾 NIR vs Wet Chemistry — Why WA Hay MUST Be Tested Properly
(And Why Some Reports Can’t Be Taken at Face Value)
🎯 WA horse owners — if you’re relying on NIR hay reports, you may be making feeding decisions on numbers that aren’t real. WA hay is unique, and because it’s not in national calibration libraries, NIR often produces misleading results.
I’ve had a few of our WA hay producers ask why I send so many of our hay samples over to the USA for testing, and I completely understand the question. From the outside, it can look unusual, or like we’re being awkward or making things harder than they need to be.
The truth is much simpler. We use overseas labs because they give us the most accurate numbers for the unique chemistry of WA hay — especially for sugars, starch, and minerals. Our goal is never to complicate anything; it’s to protect horses, give producers honest data, and make sure the results we’re using are scientifically reliable. There are very real reasons why we choose these labs, and why it matters for equine health in WA.
☕ Settle in with a cuppa or a tipple of whatever takes your fancy. This Facebook post is for every WA horse owner, hay producer, and equine professional.
🔍 The Two Testing Pathways
🔬 Wet Chemistry (WC)
• Chemical digestion + combustion + enzymatic assays for sugars
• Proper mineral testing via ICP-OES or ICP-MS
• ⏳ Slower & pricier
• ✅ Globally the gold standard (Williams & Norris, 2001)
🌈 NIR (Near-Infrared Spectroscopy)
• ⚡ Quick, cheap, repeatable
• ❌ Does not measure nutrients — it predicts them using calibration libraries (Saha & Lumburg, 2016)
⭐ Why WA Breaks NIR
Most commercial NIR systems were built using east-coast forages such as ryegrass, lucerne, clover, vetch and east-coast oaten hay (Jeong et al., 2024).
WA forage grows under completely different conditions:
• 🟤 Iron-rich sands
• 🥉 Low copper & zinc soils
• ☀️ Hot, dry Mediterranean climate
• 🌾 Different cereal cultivars
• ⏱️ Rapid curing due to dry air + strong sun, which increases:
– bleaching (UV)
– leaf shatter (legumes & soft oaten cultivars)
– loss of soluble carbohydrates
– higher fibre from leaf loss
• 🌱 Variable ryegrass presence depending on paddock history
➡️ The spectral fingerprints don’t match.
NIR begins guessing outside its experience — and accuracy collapses.
📊 Calibration Reality
A valid NIR model requires:
• 800–1,000+ wet-chemistry samples per forage type (Saha & Lumburg, 2016)
• 200+ new wet-chem samples per year to stay accurate (AFGC, 2019)
❌ No Australian NIR system has this for WA hay.
➡️ NIR numbers drift — badly.
⚠️ Typical WA NIR Distortions
• 💪 Crude Protein → +15–20% too high
• 🍬 WSC + Starch → 20–30% too low
• 🌾 Fibre → underestimated
• 🧪 Minerals → not measurable
👉 This is why hay that “looks laminitis-safe” on NIR can still spike insulin.
❌ Why NIR Cannot Measure Minerals
NIR only detects vibrations of organic molecules — chemical bonds like
C–H, O–H, N–H.
What does “C–H, O–H, N–H” even mean?
These are the tiny chemical bonds inside plants that NIR can detect:
• C–H → found in carbohydrates, fats, fibre
• O–H → found in water, sugars, cellulose
• N–H → found in amino acids & proteins
When NIR light hits these bonds, they vibrate.
That vibration is what the machine “reads.”
But here’s the important part:
Minerals don’t have ANY of these bonds.
No C–H, O–H, or N–H bonds =
❌ no vibration
❌ no absorbance
❌ nothing for NIR to detect
Minerals like sodium, iron, zinc, copper iodine, selenium & cobalt are inorganic (Williams & Norris, 2001; Meyer & Coenen, 2014).
They cannot be measured by NIR under any circumstances.
👉 Only ICP-OES or ICP-MS can measure minerals accurately.
🧪 ICP Explained — Plain English
ICP-OES
The sample is vaporised in a plasma flame (~10,000°C).
Each mineral glows with its own colour.
The machine reads the colour spectrum → mineral levels.
ICP-MS
Same plasma, but the machine weighs each mineral ion individually.
Ultra-sensitive — parts per billion.
If your minerals were tested with ICP → they’re real.
If they came from NIR → they’re predictions, and for WA usually wrong.
🟡 The Elephant in the Room — Marketing Bias
Many WA hay producers avoid sending hay to USA wet-chemistry labs because those results often show:
• 📈 Higher NSC %
• 📉 Lower crude protein%
…which is the true chemistry of WA hay, but not ideal for marketing.
So some hay buyers are shown NIR results because they look “prettier.”
⚠️ The Cut-and-Paste Problem
• Over the years we’ve seen:
• 📑 Copied hay & ARGT reports
• ✏️ Numbers altered
• 🔤 Fonts altered in results
• 📄 Word docs pretending to be lab reports
• 🏷️ Samples rebranded
• ❌ Missing lab headers / sample codes
➡️ Always demand the ORIGINAL PDF, showing:
✔ Laboratory name
✔ Sample code
✔ Method (NIR vs WC vs ICP)
✔ Full carbohydrate panel (ESC, WSC, starch)
If any of that is missing — it’s not reliable.
This is why I now watermark all hay results being posted on social media or that are sent out to customers.
🐄 Ruminant vs Equine Reports
Many hay tests are designed for cattle/sheep, not horses.
Some Equine-unsafe reports may:
• ❌ Omit starch
• ❌ Omit WSC
• ❌ Omit ESC
• ❌ Use ME instead of DE
👉 Horse nutrition requires DE, starch, WSC and ESC — non-negotiable for EMS/IR horses.
🧪 Our Own Comparison
We sent the same bale:
• 📦 To an east-coast lab
• 🌍 To a USA wet-chemistry lab
Results? Wildly different.
👉 NIR smoothed out the sugars
👉 Wet Chemistry showed the truth
✅ WA Truth in One Line
🌈 NIR = screening only 🔬 Wet Chemistry = truth 🧪 ICP = the only way to get real mineral values 🐴 Horse reports must include starch, WSC, ESC, and DE — not ruminant figures.
📌 Summary
WA hay is chemically and environmentally unique. Because calibration libraries don’t include WA forage, NIR consistently produces inaccurate — and sometimes dangerously misleading — results.
For safe equine feeding decisions:
• Use Wet Chemistry for sugars and NSC (WSC, ESC, starch).
• Use ICP OES / ICP MS for minerals.
• Treat NIR as screening only, never decision making.
• Ensure reports are equine specific — not ruminant reports missing starch, WSC, ESC, or using ME.
WA hay is fantastic — but unique. When we use the right testing methods, we protect our horses, support honest hay producers, and keep the whole WA horse community better informed. Horses first, always.
📚 References (APA 7th Style )
American Forage and Grassland Council. (2019). Forage analysis by near infrared spectroscopy (NIRS) vs. wet chemistry: Proceedings of the AFGC annual meeting. AFGC Press.
Forage & Feed Testing Consortium. (2013). Accurate analysis: NIRS versus wet chemistry. Rock River Laboratory.
Harris, P. A., Ellis, A. D., Fradinho, M. J., Jansson, A., Julliand, V., Luthersson, N., Santos, A. S., & Vervuert, I. (2018). Review of the equine digestive system and associated nutritional implications. Animal, 12(8), 1727–1740.
Jeong, E. C., Lindquist, A., & Kallenbach, R. L. (2024). Application of near-infrared spectroscopy for hay evaluation at the farm level. Animals, 14(7), 122848.
Kellon, E. M. (2020). The importance of accurate forage testing for horses with insulin resistance and laminitis. ECIR Group Technical Bulletin.
Meyer, H., & Coenen, M. (2014). Forage analysis and calibration challenges in arid regions. Equine Veterinary Nutrition Review, 9(3), 44–51.
Saha, U. K., & Lumburg, R. K. (2016). Development and validation of NIRS calibration models for forage quality analysis. Journal of Near Infrared Spectroscopy, 24(5), 421–430.
Williams, P. C., & Norris, K. H. (2001). Near-infrared technology in the agricultural and food industries (2nd ed.). American Association of Cereal Chemists.