What your TEROS-12 measures and how to steer irrigation on it
HomePrecisionWhat your TEROS-12 measures and how to steer irrigation on it
Precision · Root zone

What your TEROS-12 measures and how to steer irrigation on it

A TEROS-12 probe buried in your growing media reports three numbers: how much water is stored, how salty that water is, and the temperature. This guide explains what each number means, why the raw reading is not yet the truth, and how to convert it into irrigation decisions you can stake a crop on.

Precision12 diagramsEvidence-linked · 8 sources~18 min read
In this guide
01 Purpose and scope02 Definitions03 Evidence and limitations04 How the probe measures water without touching it05 Why you must calibrate to your exact substrate06 What the EC reading tells you and where it breaks down07 Steering irrigation from TEROS-12 readings08 Diagnosing a bad reading before blaming the sensor09 Expected results and limitations10 References
Start here

Purpose and scope

A TEROS-12 is a small probe you push into growing media — coco, rockwool, or soil. It sends three numbers down a single digital wire: how much water the media holds, how salty that water is, and the temperature. This guide starts from zero, explains what each number actually represents, how the probe arrives at it, and why the raw reading alone should never open a valve.

The probe samples a pocket of roughly 1010 mL (34.2 fl oz) of media around its prongs, not the whole root zone.[8] Everything that follows turns that one local, uncertain reading into a number you can actually steer irrigation on.

  • The TEROS-12 reports volumetric water content (VWC), bulk electrical conductivity (EC) and substrate temperature over a digital protocol called SDI-12.
  • Its sensing volume is only about 1010 mL (34.2 fl oz) of media around the prongs — one local spot, not the average of a tray or a zone.
  • The goal is to convert that noisy local reading into a trustworthy estimate of stored water, with the uncertainty stated openly rather than hidden.
  • No prior knowledge of soil sensors is assumed. Every term is defined the first time it appears.
hardware
The probe
Three steel prongs create a high-frequency electric field and read it back.
~1010 mL
Volume of influence
Only ~1010 mL (34.2 fl oz) of media around the prongs is sensed.
unseen
The rest of the pot
Everything outside that pocket is invisible to this probe.
Figure 1. What the probe sees: a small ellipsoid of media around the prongs, not the whole root zone.[8]
Who this is for

This is for anyone putting a moisture probe in a pot who wants to steer on it honestly. It pairs with the smart watering (VWC/EC) guide and the coco crop-steering paper.

Vocabulary

Definitions

Six words underpin everything in this field. Each is defined here in plain English first, then used precisely from that point on. They come back in context as you read.

Volumetric water content (VWC)Think of your pot as a fixed container. Some of that container is solid media, some is air in the gaps, and some is liquid water. VWC is the liquid water's share of the whole, expressed in cubic metres of water per cubic metre of media (m³/m³). So 0.34 means 34% of the pot's total volume is liquid water. This is the primary number you steer irrigation on.
Permittivity (dielectric constant)Water responds to an electric field roughly twenty times more strongly than dry growing media does — like how a wet sponge conducts electricity far better than a dry one, because the water is doing almost all the electrical work. Permittivity is the name for that responsiveness. Water’s is ~80; dry media is ~3–5; air is ~1. That contrast is the entire trick the probe uses to detect water.[1]
Capacitance sensorA sensor that measures permittivity by applying a high-frequency electric field and reading how strongly the surrounding material stores it. The TEROS-12 is one of these. It infers VWC from that permittivity reading, without touching the water directly.
Bulk EC vs pore-water ECThe probe measures the conductivity of the whole wet-media mixture together — solids, water, and air — which is called bulk EC (0–20000 µS/cm on this probe). What the roots actually experience is pore-water EC: the salt concentration in the liquid sitting in the gaps between media particles. You cannot read pore-water EC directly; you estimate it from bulk EC using the Hilhorst (2000) model.
DUL / container capacityDrained upper limit: how much water this specific pot holds after gravity has pulled out all it can. It is your steering ceiling — and it is a property of your pot and media, not a textbook constant, so you measure it from your own runoff events.
Resolution vs accuracyResolution is the smallest change the probe can report: 0.001 m³/m³ VWC. Accuracy is how close that number is to reality: only ±0.03 m³/m³ with the generic calibration. The probe prints a precise-looking number, but precision is not the same as accuracy — a very precise number can still be consistently wrong.[8]
Relative permittivity: why water dominates the readingWater's permittivity towers over everything else, so it controls the sensor's response.0224568901Air4Dry substrate80Water
Figure 2. Liquid water’s permittivity (~80) is roughly twenty times that of dry media (~3–5) and eighty times that of air (~1). Even a small amount of water swings the probe’s reading hard.[1]
fraction
Solids
The coir, rockwool fibre or soil grains themselves.
fraction
Air
Air-filled pore space between the solids.
VWC
Water
Liquid in the pores. VWC = this volume / total volume.
Figure 3. A unit volume of media splits into solids, air and water. VWC is the water slice divided by the whole.
Saturation, field capacity and drybackthe same block at three points in a daySaturatedjust irrigated, almostno air→Field capacityfree water drained,daily peak→Dryback lowplant drank, air +oxygen in
Diagram. What the sensor reads: saturation, field capacity (container capacity / DUL) and the dryback low.
A daily water-content cycleSaturate to field capacity, hold it, allow a controlled dryback, repeat. The size of the dryback is the steering lever.working bandtoo dry0255075100offP3 minonP1FCP2offwater content %
Diagram. A day of water content as the probe sees it.
Evidence assessment

Evidence and limitations

How sure is this paper?

We've gone to great lengths to keep these guides honest. One of the main ways we do that is self-review: we actively look for claims that are subjective, only lightly backed by literature, or based on grower practice rather than a controlled study — and we call those out instead of dressing them up as settled science.

Often there simply is no paper for the decision you're making. In those cases we're drawing on what other growers report and what has worked in our own rooms. That can still be useful — but it is not a lab proof. Do what works for your plants, your room, and your meters. If a table disagrees with your crop, believe the crop and log the difference.

Solid
Well supported by plant science, standards, or broad multi-source consensus
  • Capacitance probes estimate VWC via permittivity; media calibration matters
  • Pore-water EC estimation has real limits (Hilhorst-class caveats)
Operational
What many growers and rooms actually run — start here, then tune
  • Install depth, volume of influence, and multi-pot placement habits
Grain of salt
Subjective, thin literature, single studies, or “this works for us” practice
  • Manufacturer accuracy specs as guaranteed on every uncalibrated pack

See something glaringly wrong? Tell us and we'll fix it. Please open a GitHub issue with the paper name and what looks off (include a source if you have one): Report an accuracy issue. Local law, labels, and licences always override any recipe here. Inline notes labelled grain of salt flag the highest-risk over-trust points in the text.

Core content

How the probe measures water without touching it

The TEROS-12 is a capacitance probe. Its prongs push a high-frequency electric field into the surrounding media and read how strongly the media stores that field — a property called permittivity. Because water’s permittivity (~80) is roughly twenty times that of dry media (~3–5) and eighty times that of air (~1), the bulk permittivity of the media rises steeply and predictably as water content rises. That makes permittivity a reliable stand-in for VWC.[1]

The probe then applies a calibration equation — a generic mineral-soil curve by default — to map measured permittivity to a VWC number, reporting it to 0.001 m³/m³ resolution. The catch is built in from the start: the mapping is media-specific, and the generic curve is only accurate to ±0.03 m³/m³.[8]

  • Permittivity is the physical quantity the probe measures. VWC is a derived, calibrated estimate — one layer of math on top of that measurement.
  • The permittivity-to-VWC curve is nonlinear, especially near saturation, where the response flattens. Near-full media can report a ‘full’ reading even when it is not.
  • Substrate temperature shifts the dielectric response — a known physical effect that can look like a change in water content if you do not account for it.[4]
  • The probe outputs data over SDI-12. A stale, NaN, or railed value (pinned at 0 or full-scale) is a hardware or cable fault, not a data reading.
Permittivity to VWC: a nonlinear calibration curveBottom axis is permittivity. The curve flattens at the top, so response near saturation can mimic 'full'.00001381524354862VWC (m³/m³)
Figure 4. VWC rises with permittivity but the curve bends and flattens near saturation. The same VWC step covers very different permittivity steps depending on where you are on the curve.[1]
The measurement chain1Electric fieldprongs energisethe media→2Permittivitymedia stores thefield→3Calibrationcurve maps towater→4VWC + EC + tempreported overSDI-12Bulk EC and temperature branch off the same measurement; all three leave on the SDI-12 wire.
Figure 5. From electric field to a number: every output is downstream of the same physical measurement. Calibration is where media-specific error enters.
Core content

Why you must calibrate to your exact substrate

Out of the box the TEROS-12 uses a generic mineral-soil calibration, accurate to only ±0.03 m³/m³. A substrate-specific calibration for your exact coco or rockwool tightens that to ±0.01–0.02 m³/m³.[3] That difference matters in practice. Crop-steering dryback windows are often narrower than the ±0.03 generic error band. Steering on uncalibrated VWC means steering inside the noise.

A worked headroom example makes the consequence concrete. A naive 256 mL (8.7 fl oz) of ‘room to water’ shrinks to a safe ~109 mL (3.7 fl oz) once you account for ±0.02 accuracy, and to just ~54 mL (1.8 fl oz) under the generic ±0.03. Same pot, same probe — the only thing that changed is how honestly you treat the error band.[7]

How accuracy erodes usable headroom (same pot)Wider error band = less water you can safely add before risking overshoot.075150225300256 mLNaive estimate109 mLSafe @ ±0.02 cal54 mLSafe @ ±0.03 generic
Figure 6. Usable safe headroom collapses as calibration error grows: from 256 mL (8.7 fl oz) naive to ~54 mL (1.8 fl oz) on the generic curve.[7]
Resolution is the same either way. Accuracy is what changes. Calibrate to your media before you steer on it.
Calibration typeVWC accuracyResolutionTight steering?
Generic mineral (default)±0.03 m³/m³0.001 m³/m³No — error band wider than a typical dryback window
Substrate-specific±0.01–0.02 m³/m³0.001 m³/m³Yes — required for tight steering
Calibration does not fix everything

Calibration corrects an additive offset in the reading, but gain error and nonlinearity near saturation remain and do not cancel in later math. Treat substrate-specific calibration as mandatory for tight steering, not optional — and still respect the residual error that remains after you calibrate.

Core content

What the EC reading tells you and where it breaks down

The probe measures bulk EC — the conductivity of the whole wet-media mixture, 0–20000 µS/cm. What growers care about is pore-water EC: the salt concentration in the solution actually in contact with the roots. To get pore-water EC from what the probe reports, you combine bulk EC, VWC, and temperature using the Hilhorst (2000) model.[2] Think of it like measuring the saltiness of a wet sponge by pushing current through the whole thing — sponge fibre and water together. The Hilhorst model then estimates the saltiness of the water alone.

That model is useful but parameter-sensitive — roughly ±20% — and unreliable below VWC 0.10 m³/m³, where you should not use it at all. The deeper limit is representativeness: the probe integrates one ~1010 mL (34.2 fl oz) spot. Channeling, dry pockets, or poor probe-to-media contact can make a perfectly functioning probe report a number that does not represent the zone.[3]

  • Bulk EC (0–20000 µS/cm) is what the probe measures directly. Pore-water EC is estimated from it — they are not the same thing.
  • The Hilhorst (2000) conversion is ~±20% sensitive and invalid below VWC 0.10 m³/m³.[2]
  • A representativeness fault: the probe’s 1010 mL (34.2 fl oz) pocket can be unrepresentative of the zone due to channeling, an air gap, or a pulled probe — while the probe itself is working perfectly.
  • The sensor cannot see per-pot runoff volume, effective substrate volume (which shrinks as roots fill the pot), or whether a commanded irrigation shot was actually delivered.
measured
Bulk EC
Conductivity of solids, water and air together. This is what the probe measures directly.
~±20%
Hilhorst model
Combines bulk EC, VWC and temp. ~±20% sensitive; invalid below VWC 0.10.
derived
Pore-water EC
Salt in the liquid the roots actually contact. This is what you want to know.
Figure 7. Bulk EC is measured directly. Pore-water EC is derived through Hilhorst (2000) and degrades fast as the media dries.[2]
What one TEROS-12 can and cannot tell you, and the second witness that fills each gap.
The probe CAN seeThe probe CANNOT seeWorkaround
VWC (local spot)True zone average across potsMultiple probes, a cohort model
Bulk ECPer-pot runoff volumeRunoff trays / drain sensors
Substrate temperatureEffective substrate volume (shrinks with roots)Periodic re-learning of DUL
Derived pore-water ECWhether an emitter actually firedFlow meter or load-cell weight jump
How-to

Steering irrigation from TEROS-12 readings

The practical method is one demotion and one promotion. Demote the raw VWC reading from ‘truth’ to ‘one noisy witness with a confidence score’. Promote a small running water-balance model that holds the best estimate of stored water and is only nudged by trusted readings.

You track dryback — the VWC fall between shots — watch specific yield (how much VWC rises per mL delivered) to sense when the pot is approaching capacity, and anchor your ceiling on the observed DUL rather than a guessed number. Steer on trends — the shape and slope of the dryback — more than the absolute level, because trends are insensitive to additive calibration offset.[6] Never act on a trend alone without a second witness such as runoff timing or pot weight.

  1. 1
    Calibrate to your substrate first
    Substrate-specific calibration is the first requirement. Without it you are steering inside the error band — the reading cannot be trusted tightly enough.
  2. 2
    Verify probe contact and position
    The probe must be seated firmly in the media at a representative, fixed spot. A loose probe or an air gap reports its surroundings, not your root zone.
  3. 3
    Learn this pot's DUL from corroborated events
    Anchor the capacity ceiling on approximately five runoff or weight events — not one — and express it as water-volume space rather than a raw VWC number.
  4. 4
    Steer on the dryback slope, bounded by safe headroom
    Act on dryback slope and specific yield, bounded by a headroom estimate that accounts for calibration error, not the naive point estimate.
  5. 5
    Confirm with a second witness before moving water
    Runoff onset or load-cell mass must confirm before any signal reaches a valve. The probe reading never drives a valve on its own.
A day of dryback: shots, decay and the DUL ceilingSharp rises are shots; the decay between them is the dryback. The top is the learned DUL ceiling.00000P0 endshot+2h+4hshot+2hpre-darkVWC (m³/m³)
Figure 8. Irrigation shots push VWC up toward the DUL ceiling; the slope of the fall between shots is the steering signal, not the absolute level at any one moment.
The steering loop: the reading never touches the valve alone1Probe readingraw VWC / EC /temp→2Confidence scorehow much totrust it→3Belief modelwater-balanceestimate→42nd-witness gaterunoff or weightconfirmsOnly after the confidence gate AND a second witness does a valve action happen.
Figure 9. Every reading passes through a confidence score, a belief model, and a second-witness gate before it is allowed to move water.[6]
When it goes wrong

Diagnosing a bad reading before blaming the sensor

Most TEROS-12 problems are not the sensor failing. They are the sensor being believed when it should not be. Draw the first distinction clearly: a wrong reading — where the probe is working but its 1010 mL (34.2 fl oz) does not represent the zone — is a representativeness fault. It should lower your trust in the absolute VWC number. A railed, flatline, NaN, or stale value is a hardware or cable fault. It should stop all automated action.

Watch for VWC that tracks the daily substrate-temperature cycle. That is a contact or calibration artifact, not a real change in stored water.[5] Watch for wetting and drying paths that diverge abnormally — a sign of channeling or hydrophobic media. Watch for one pot drifting away from identically-treated neighbours — likely a blocked emitter or a dud probe, not a plant problem.

Temperature and EC should move your trust, not the water number

Temperature and EC should move your trust in the reading, not the stored-water estimate directly. A diurnal temperature cycle can produce a real dielectric shift in dry media that looks like a change in water content.[4] Let that shift write VWC and you will be irrigating in response to physics, not plant need.

Fault triage: hardware first, then artifacts, then local faults1Railed / NaN / stale?hardware fault →fallback→2Tracks temperature?contact artifact→ down-trust→3One pot unlike siblings?local fault →inspect→4Otherwisetreat as a noisybut usablewitnessHardware faults stop you acting; representativeness faults only lower your trust.
Figure 10. Work down the list in order: rule out hardware faults first, then artifacts, then local faults, before accepting the number as usable.
Symptom to cause to response. The ‘what it is NOT’ column is the most useful — misdiagnosis is the real cost.
SymptomLikely causeWhat it is NOTResponse
VWC railed / flatline / NaN / staleHardware or cable faultReal water readingStop steering. Run a bounded safe routine. Alert a human
VWC swings with diurnal tempPoor contact / calibration artifactA real water changeDown-trust absolute VWC; check probe seating[5]
Wetting vs drying diverge oddlyChanneling / hydrophobic mediaSensor failureInspect media; re-wet; check probe seating
One pot unlike its siblingsBlocked emitter or dud probeA plant problem (yet)Inspect emitter and probe before blaming the plant
Anchors must be earned from multiple events

A single manual reading or one human observation should never hard-write a capacity anchor or override a safety interlock. Anchors earn their place from corroborated events across multiple fills, not from one good look.

Reality check

Expected results and limitations

The honest account

A single TEROS-12 will not give you a per-zone, ground-truth picture of your root zone. Treating it as one is the most common and most expensive mistake. With substrate-specific calibration you can realistically resolve dryback trends and approximate stored water to about ±0.01–0.02 m³/m³ in the spot the probe occupies. That is enough to steer on — if you account for the uncertainty and cross-check it.[3]

Trust earned: single probe vs probe + witness + calibrationAdding a calibration and an independent witness is what moves you from caveats to control authority.single probe alone, wide band, many caveatsprobe + 2nd witness + calibration, steering-ready0% trust50% trust100% trust
Figure 11. A lone probe sits in a wide, caveat-heavy uncertainty band. Add substrate-specific calibration and one independent witness and the band tightens to a range you can steer from.[7]
  • Best case with substrate-specific calibration: trustworthy dryback shape and ~±0.01–0.02 m³/m³ stored-water accuracy for the probe’s local spot.
  • Not achievable with one probe alone: per-zone runoff, delivery verification, or a true zone average across pots.
  • A load cell (pot weight) is the highest-value add-on because it measures water stored directly, without routing through dielectric physics.
  • Design for explicit ‘I cannot tell’ outputs rather than false precision. Control authority should be earned from the data, not assumed.

The mature approach is to require the system to say when it cannot tell, rather than emit a confident number it has not earned. Calibrate first, add one independent witness, then read the smart watering (VWC/EC) guide for how those signals drive shots, and the signal-and-noise paper for separating a real trend from sensor noise.

Related papers
Smart watering · VRWE
The watering brain, plain English
Coco & crop steering
Steer the plant with water
Signal & noise
Real signal vs sensor noise

References

  1. Topp, G. C., Davis, J. L., & Annan, A. P. (1980). Electromagnetic determination of soil water content: Measurements in coaxial transmission lines. Water Resources Research, 16(3), 574-582. https://doi.org/10.1029/WR016i003p00574
  2. Hilhorst, M. A. (2000). A Pore Water Conductivity Sensor. Soil Science Society of America Journal, 64(6), 1922-1925. https://doi.org/10.2136/sssaj2000.6461922x
  3. Fragkos, A., Loukatos, D., Kargas, G., & Arvanitis, K. G. (2024). Response of the TEROS 12 Soil Moisture Sensor under Different Soils and Variable Electrical Conductivity. Sensors, 24(7), 2206. https://doi.org/10.3390/s24072206
  4. Nasta, P., Coccia, F., Lazzaro, U., Bogena, H. R., Huisman, J. A., Sica, B., Mazzitelli, C., Vereecken, H., & Romano, N. (2024). Temperature-Corrected Calibration of GS3 and TEROS-12 Soil Water Content Sensors. Sensors, 24(3), 952. https://doi.org/10.3390/s24030952
  5. Kapilaratne, R. G. C. J. & Lu, M. (2012). Correcting the Temperature Influence on Soil Capacitance Sensors Using Diurnal Temperature and Water Content Cycles. Sensors, 12(7), 9773-9790. https://doi.org/10.3390/s120709773
  6. Tavan, M., Wee, B., Brodie, G., Fuentes, S., Pang, A., & Gupta, D. (2021). Optimizing Sensor-Based Irrigation Management in a Soilless Vertical Farm for Growing Microgreens. Frontiers in Sustainable Food Systems, 4, 622720. https://doi.org/10.3389/fsufs.2020.622720
  7. Nemali, K. S. & van Iersel, M. W. (2006). An automated system for controlling drought stress and irrigation in potted plants. Scientia Horticulturae, 110(3), 292-297. https://doi.org/10.1016/j.scienta.2006.07.009
  8. METER Group, Inc. (2023). TEROS 11/12 User Manual & Specifications. METER Group, Pullman, WA. (industry/manufacturer or non-journal source) https://metergroup.com/products/teros-12/

Citations marked in-text as [n] map to this list. Primary literature and official guidance except where noted. Cannabis tissue culture is strongly genotype-dependent, verify dilutions, hormone doses and local regulations against the primary sources before relying on them.