One Soil Profile or Two Stations per Zone? How to Decide Before You Buy Sensors
The right sensor count is a representation question, not a hardware question. This guide separates the two actions that solve different errors—adding a station for lateral variation, adding a depth for vertical movement—and gives the pilot and evidence needed before one trigger controls a whole zone.
Disclosure: We may earn a commission if you purchase through links on this page, at no additional cost to you.
The right sensor count is not a hardware question first. It is a representation question: will the installed points stand in for the soils, rooting conditions, irrigation application and decisions that they are expected to control? A deep profile at one location can explain water movement through that profile. It cannot demonstrate that another part of the management zone behaves the same way. Conversely, several shallow single-depth stations may reveal lateral differences yet miss drainage below the active root zone. This guide provides a practical way to decide when to add a second monitoring station, when to add a depth, and when to divide a nominal irrigation zone before automating any decision.
The working conclusion is deliberately conditional. Start with management zones rather than acreage. Use representative sampling to test whether a zone is genuinely coherent. Add monitoring locations when soil, topography, crop development, irrigation uniformity or observed water response differs in a decision-relevant way. Add depths when the unresolved question is vertical storage, extraction or drainage at an otherwise representative location. The two actions solve different errors.
Start with a management zone, not a map label
An irrigation valve, field block or pivot sector is not automatically a management zone. For monitoring purposes, a usable zone is an area in which soil-water behaviour is expected to be similar enough that the same reading would support the same irrigation decision. Oregon State Extension frames monitoring locations around management zones that represent similar soil-water behaviour. Florida guidance likewise links sensor-number decisions to soil-water characterization and irrigation-zone distribution. These are design principles, not universal sensor-count prescriptions.
| Observed condition | Why one station can mislead | Design response |
|---|---|---|
| Different soil texture, depth or restrictive layer | The same irrigation event can store, drain or deplete water differently. | Split the management zone or install a comparison station before sharing one trigger. |
| Slope, low area, run-on or poor drainage | A station can capture a local wet or dry extreme rather than the decision condition. | Move away from abnormal flow paths and test an additional representative location. |
| Different crop age, canopy or root depth | Extraction pattern and useful sensing depths change even with the same soil. | Use a separate station or document why one profile remains representative. |
| Uneven emitter, sprinkler or pressure performance | The sensor may be reporting application variability rather than crop demand. | Audit irrigation uniformity; do not solve a delivery defect by averaging sensor data. |
Use two questions to choose stations and depths
Ask first, "Could another part of this zone lead to a different irrigation action today?" If yes, add a monitoring location or reconsider the zone boundary. Then ask, "At this representative location, could the water decision change because of what is happening deeper in the root zone?" If yes, add a depth or use a vertical profile. This order prevents a common procurement error: buying a multi-depth probe because it has more channels, then assuming those channels describe lateral field variability.
| If the uncertainty is mainly | Add first | Reason |
|---|---|---|
| Lateral soil or application variation | A second station | Independent locations test whether the zone is being represented. |
| Root-zone extraction versus deep drainage at one representative point | A second depth or multi-depth profile | Vertical readings show where water is being removed or moving below the targeted layer. |
| Both lateral variation and vertical movement | One additional station plus an appropriate depth plan | Neither more depths at one point nor more shallow points alone close both gaps. |
| No established reference location | A short pilot before a permanent design | The first installation should test representativeness, not become an unexamined control point. |
A practical sizing method
- Define the action that the sensor will change. Write the action in one sentence: delay a cycle, start an irrigation, stop a set, investigate a blockage, or document water status. A station intended only to prevent irrigation after rainfall can tolerate a different design from a station that drives a root-zone depletion rule. If the consequence of a wrong reading is high, require more evidence before allowing one location to control a larger area.
- Draft provisional zones from physical evidence. Overlay soil survey information, elevation, irrigation layout, crop age or variety, and known problem areas. Treat the result as a hypothesis. Do not declare zones uniform simply because they share a valve. The initial purpose is to decide where to sample, not to justify the fewest possible devices.
- Run a representative sampling pass. Before permanent installation, collect observations across the candidate zone using the relevant field method: hand samples, auger observations, pressure/flow checks, crop inspection and, where appropriate, temporary sensors.
- Install a reference station and a challenge station. Place the reference station in the most defensible representative condition. Place the challenge station where your desk review predicts a materially different response: a lighter soil, shallower profile, different irrigation-pressure reach or contrasting crop vigour. Operate both through wetting and dry-down events. If their trends and decision boundaries remain aligned, the evidence supports one monitoring design for that defined zone. If they diverge, retain both or redraw the zone.
- Decide depth after location is credible. At a credible station, select depths that answer irrigation questions rather than filling every possible port. Shallow observations can show dry-down; deeper observations can show whether water reached the intended lower root zone or moved beyond it. University of Minnesota guidance describes pairs near one-third and two-thirds of the root-zone depth and two or more field locations as a common installation pattern. The exact depths must follow crop rooting depth and local profile evidence.
What counts as enough evidence to share one trigger
A shared trigger needs more than two lines that look alike on a dashboard. Compare the timing of wetting, the depth and duration of response, dry-down rate, rain response, and the irrigation action each record would have produced. Keep field notes on irrigation events, rainfall, maintenance and crop stage. Interreplicate variability is a known limitation for electromagnetic sensors: identical instruments can differ because of microscale soil properties, installation or sensing-volume effects. That is why a single agreement check on one day is not enough.
| Pilot result | Interpretation | Decision |
|---|---|---|
| Stations agree through multiple wetting and dry-down cycles; no action would differ | The location hypothesis is supported for the observed period. | Keep one zone design, but document the condition and review after material field change. |
| Trend direction agrees but timing or threshold crossing differs | The difference may matter for automated control. | Keep both stations during another cycle; do not average away the difference. |
| One station repeatedly wets faster, drains faster or remains wetter | A physical or irrigation-delivery distinction is likely. | Separate the management zone or use independent schedules. |
| Output is erratic or conflicts with field observations | The evidence may reflect installation, cabling, configuration or sensor failure. | Troubleshoot and validate before treating the signal as spatial variability. |
Why a multi-depth probe is useful but not a field average
A multi-depth probe can reduce installation disturbance and give a vertical record at a single point. It can be a good choice when the buyer's unanswered question is whether water is reaching, remaining in or moving below successive depth bands at that location. It should not be used as evidence that a field is laterally uniform. The probe's depth channels share a site, its installation history and its surrounding soil. Verifying channel independence is a separate acceptance task that still matters before using a vertical profile for scheduling.
In a new zone, the strongest procurement plan is often staged. Buy or rent enough equipment to compare two locations first. Then choose whether the permanent design needs two discrete stations, a vertical profile at each location, or a simpler arrangement justified by the pilot. This approach avoids both under-instrumenting a variable zone and purchasing high-channel-count hardware that answers the wrong question.
Read the pilot as a decision test, not a dashboard comparison
Set the comparison up before looking at the traces. For each monitored location, specify the same reporting convention, time basis, wet and dry boundary, and candidate action. Then review the records event by event. A useful question is not whether the lines have the same shape; it is whether the irrigation action would have been the same. Two points may both trend downward while one reaches its dry boundary one irrigation earlier. That difference is operationally material even when a weekly graph looks broadly similar.
Separate a sensor-disagreement finding from a zone-design finding. If the challenge station disagrees, first inspect installation depth, soil contact, wiring, logger scaling, timestamp alignment and any configuration difference. Compare the suspected sensor with field observations and, where possible, a reference method appropriate to the measured quantity. Only after the measurement chain is credible should a persistent difference be used as evidence of lateral variability.
A practical comparison log has four columns beyond the readings: irrigation start and stop, rainfall, crop or field operation, and explanation status. The explanation status should be one of confirmed, plausible but unverified, or unknown. For example, a deeper wetting response may be confirmed by a long irrigation set, merely plausible because of a coarser soil lens, or unknown because the sensor was disturbed. This record makes it possible to revisit the zone decision instead of treating an early installation impression as permanent truth.
Avoid three costly shortcuts
- Averaging unlike stations. An average can conceal the very condition that needs a different action. If a shallow or sandy area reaches the dry boundary early while a heavier area remains wet, the mean is not necessarily representative of either. Use an average only after the buyer has shown that it maps to a defined decision objective and cannot mask a location that requires protection.
- Treating every depth as another location. A five-layer probe delivers five observations along one vertical installation. It does not deliver five independent samples across a management zone. When procurement comparisons quote channel count, ask whether those channels add the missing type of information.
- Scaling a pilot without documenting its assumptions. Conditions that make one location representative can change. Retain a short deployment note: zone boundary, reference-location rationale, installed depths, field observations, trigger logic and the date of the last review.
RFQ and acceptance questions
- Which supplied configuration records independent locations, and which options are multiple depths at one location?
- For each depth, what is the stated sensing position, output register, update interval and installation orientation?
- What information documents the delivered model, cable length, power requirement, interface and logger or gateway compatibility?
- Can the supplier provide a current manual, wiring diagram, register map, calibration statement and variant-specific data sheet?
- What test support is available for a two-location field pilot, including replacement, configuration and data export?
Sourcing candidates for a pilot
The two listings below are role candidates, not ranked product recommendations. Marketplace text is attributed to the listing and has not been independently verified as delivered configuration or field performance. The supplier age shown on the marketplace is a flexible stability signal only; it is not proof of manufacturing capability or suitability.
Candidate — Renke RS485 Modbus soil sensor (discrete station role)
Candidate — Renke multi-depth soil-moisture sensor (vertical profile role)
Decision rule
Do not ask "How many sensors per acre?" until you can state how many distinct irrigation decisions the field contains. Add a station when a second location could justifiably call for a different action. Add a depth when the missing information is vertical movement at a representative location. If neither decision is clear, conduct a short two-location pilot before committing to a permanent network.
For procurement, make the pilot design part of the purchase request rather than an afterthought. State the number of intended locations, the depth plan at each location, the data interval, the destination logger or gateway, and the records needed to evaluate agreement. A supplier can then quote the correct combination of probes, cables, power, enclosure and communications components. That does not prove the design is fit for the field, but it prevents a valid monitoring concept from failing because the delivered system cannot support the comparison.
Share this article