Agricultural industry has been obsessed with "ideal" soil health numbers. But, the question is, how to measure these numbers and how to know, that soil health management are really improving, what is the reference we are comparing to.
In 2020, when I was in Nebraska, the question hits, like everyone is trying to do good for their soil, but how to tell or define, am I doing good, what is my lands "soil health" compared to my neighbor? Am I doing everything good or not good?
So, the real problem was measurement.
Lot of literature and cultural practices end up in a discussion like: “Do this agronomic management, and this practice will improve soil health.”
But almost nobody could answer the question that actually matters:
"Improves soil health… compared to what? by how much?"
A farmer would show me a soil test and ask, “Is this better?” But, it was so hard to answer or practically say, this good enough or this not enough, you need to change practice to reach next level.
We saw, even two soils can have the same management and still behave differently because their potential is different. Texture, structure, mineralogy, precipitation, potential evapotranspiration, and landscape position quietly set the ceiling before management even enters the room.
That’s when we stopped thinking about soil health as a “score” and started thinking about it as a “distance.” How far we are from ideal soil in terms of management and improvement. That idea became the Soil Health Gap.
Part 1: Soil Health Gap (SHG) — turning “improvement” into a measurable concept
We framed soil health improvement like this:
Soil Health Gap (SHG) = soil health under reference condition − soil health under managed condition (cropland)
In equation form:
SHGₓ = (SH)ₙ − (SH)ₘ
Where:
- (SH)ₙ = reference/native soil health
- (SH)ₘ = managed soil health
- x = the indicator you care about (it can be SOC, aggregate stability, BD, infiltration, etc.)
The logic is straightforward: if the gap shrinks over time, management is working.
So we started sampling across Nebraska with that idea in mind: pair managed cropland sites with reference sites and quantify the gap.
Then the next reality hit fast. Each soil is different based on its texture and the environment influencing its potential.
Part 2: Why “statewide comparisons” break soil health science
Within a few sampling rounds, we had a problem and challenge to solve.
Comparing a soil in the Sandhills to a soil in eastern Nebraska is not an honest comparison. Even when both are “healthy.” Even when both are “improving.”
Because each soil has its own potential. A sandy soil under semi-arid conditions may never reach the same SOC or aggregate stability as a humid-region fine-textured soil. Not because the farmer is failing. Because the system constraints are different.
So even if Soil Health Gap is the right idea, we still needed one more thing:
A way to compare soils only against other soils that share the same ecological constraints. That became the Cropland Reference Ecological Unit (CREU).
Part 3: CREU — making soil health comparisons fair by controlling for potential
We defined CREU as a way to classify land into units with similar soil-forming and climatic context, so that reference benchmarks are meaningful.
Conceptually:
CREU = f(MLRA, ES, Precipitation, Soil)
Where:
- MLRA = Major Land Resource Area
- ES = Ecological Site
- Precipitation = rainfall range (often needs careful grouping, especially in semi-arid systems)
- Soil = soil grouping, often texture/series-level constraints
The point of CREU is not to make a fancy label.
The point is this: When you compare soils within the same CREU, differences are more likely to reflect management, not geography.
At that stage we had two pieces:
a measurable definition of “improvement” (gap shrinking), and
a fairness unit to control ecological potential.
But we still had the hardest operational question. Where to find the reference sites?
Part 4: Reference sites — the fence line problem (and why it matters)
Everyone uses fence lines. I understand why. Fence lines look undisturbed. They’re easy. They’re right there. But fence lines are often poor reference sites.
They can be contaminated by runoff, dust deposition, nutrient drift, compaction patterns, invasion dynamics, and long disturbance history. In windy regions they may even be systematically altered by aeolian processes.
And more importantly: a fence line is not necessarily a representation of productive ecological potential.
So we asked a different question:
How do we identify reference conditions in a way that is ecological, defensible, and repeatable?
This is where State-and-Transition Models (STMs) became essential.
Instead of guessing, we used STMs and Ecological Site Descriptions to identify the Reference Plant Community and the expected functional state for that ecological context.
Then the reference site selection becomes a method:
match soil + ecological site + vegetation reference state, then sample it systematically with a paired managed site from the same CREU.
Now we had the third equation of the puzzle:
reference isn’t a location convenience. It’s a ecological state.
So far the system looked like this:
Define “improvement” as closing a gap (SHG).
Define comparison units by potential (CREU).
Define reference sites ecologically (STM/ESD reference state).
Then we ran into the fourth problem that farmers and researchers both feel immediately.
Part 5: What do we measure, realistically?
Soil health has too many indicators to measure everything, everywhere. If we ask farmers (or programs) to measure 25–40 indicators repeatedly, adoption collapses. Budgets collapse. Time collapses. So we needed a minimal set of indicators; small enough to be operational, strong enough to explain most variation in soil functioning.
That’s where the Two-Tier framework comes in.
Tier 1: A universal Minimum Data Set (MDS) that is affordable and scalable.
Tier 2: CREU-based interpretation (thresholds derived from reference distributions).
In the framework development, the Tier 1 MDS converged on four indicators:
Organic matter, pH, nitrate-N, bulk density
Those four captured >85% of the variance present in a much larger indicator suite.
And Tier 2 sets the interpretation ranges using reference distributions, often using the interquartile range (25th–75th percentile) for each MLRA × texture x precipitation class.
This matters because “good OM” is not universal.
In reference conditions, OM medians can be ~2–3% in sandy semi-arid systems but >6% in humid clay loams. Both can be “healthy” in context and potential of the soil
Same word. Different number. Same ecological truth.
A quick data example: SHG for carbon becomes a clear, decision-ready metric
In one dataset from the semi-arid High Plains:
SOC was 4.4% in native grassland, 2.2% in no-till, 1.8% in conventional tillage, and 0.7% in exposed subsoil.
The Soil Health Gap for carbon becomes:
- SHG_C (no-till) ≈ 44 − 22 = 22 g C kg⁻¹
- SHG_C (conventional till) ≈ 44 − 18 = 26 g C kg⁻¹
- SHG_C (exposed subsoil) ≈ 44 − 7 = 37 g C kg⁻¹
Now the farmer doesn’t just hear “your SOC is low.” They hear:
“This is the size of the gap relative to your soil’s potential.”
And that gap can be tracked over time with respect to different managements. That’s a measurable story of degradation and recovery.
Practical implications for farmers
If you’re a farmer reading this, here’s what changes when you think in SHG + CREU + Two-Tier terms.
You stop chasing generic targets.
Instead, you can ask four practical questions:
- What CREU am I in?
Meaning: what are my soil + climate constraints? - What is my reference-informed potential range?
Not a national “ideal.” Your local ecological baseline. - How big is my gap right now?
SHG makes “improvement” quantitative. - Is my management shrinking the gap over time?
That’s the real definition of soil health improvement.
It also changes how we talk about success.
Success becomes “closing the gap,” not “hitting a universal number.” That’s fairer. And it’s more scientifically honest.
Practical implications for researchers and programs
If you’re a researcher, consultant, or program manager:
This framework gives you defensible interpretation.
- SHG gives you a metric that is explicitly baseline-referenced.
- CREU reduces confounding by controlling for agroecological variability.
- STM/ESD-based reference selection avoids the fence line trap by grounding reference in ecological state.
- Two-tier reduces indicator overload while maintaining explanatory power (>85% variance).
And this is big for incentives and carbon markets: If thresholds are not ecological, we reward geography instead of management. The Tier 2 concept is built to avoid that trap by using CREU-calibrated reference distributions.
The future: where this needs to go next
This is where Soil Health Exchange enters as more than a discussion platform. Right now, CREU is still logical and conceptual in many places.
The next step is operational:
- Digitize CREU
Build a map layer that can automatically identify CREU for a parcel using geospatial inputs (MLRA + precipitation zone (based on potential evapotranspiration) + soil texture/series). - Build reference distributions for each CREU
The system needs reference-based thresholds for Tier 2 interpretation. - Validate Tier 1 indicators in real-world test cases
We need paired, longitudinal datasets to show that the MDS tracks meaningful soil health changes across management transitions. - Make it farmer-usable
A user should be able to input:
location + soil type/texture + a few lab values (OM, pH, NO3-N, BD), and get back:
- “This is your CREU.”
- “These are your reference-informed thresholds.”
- “This is your soil health gap.”
- “Here’s what it implies, and what management pathways typically close that gap in this context.”
That’s the dream: simple measurement, smart interpretation, and ecological fairness. Because soil health will never be a universal scorecard. But it can absolutely be a universal method.
Further readings: