How to Set Soil Moisture Sensor Thresholds for Automatic Irrigation

A threshold is a control decision, not a universal percentage. A soil moisture threshold tells an irrigation system when a reading is dry enough to permit or start watering. It is not a portable number that can be copied from one field, crop, sensor output or substrate to another. Virginia Cooperative Extension describes the practical rule as irrigating when soil moisture falls below a minimum threshold, then applying enough water to refill the root zone without runoff or drainage below it [E01]. The threshold therefore belongs to a measurement basis, a soil profile and a control strategy.

This guide covers the commissioning task that follows sensor selection and installation: choose the control mode, establish a site reference, calculate a defensible trigger, define the wet-side stop or bypass condition, and test the complete sensor-to-valve path.

Choose the control mode before entering a setpoint

There are two common architectures. In bypass mode, a timer still defines the irrigation window and the sensor permits or blocks the scheduled event. In on-demand mode, the controller starts irrigation at a low threshold and stops at a high threshold. UF/IFAS describes both modes and notes that on-demand control is used when the buyer needs more control over timing and multiple zones [E02][E03]. The mode changes what the threshold means, so it should be selected before a number is typed into the controller.

Control mode	What the sensor does	Commissioning question
Bypass	Allows or blocks a scheduled irrigation cycle when the reading is on the dry or wet side of one setpoint.	If the sensor is offline or stale, does the timer fail open, fail closed or raise an alarm?
On-demand	Starts at a low threshold and ends at a high threshold or a configured maximum run time.	Can the controller enforce a deadband, maximum runtime and communication-failure state?
Supervisory	Provides a moisture signal to a PLC, logger or platform while another program decides the valve action.	Which unit, scaling, timestamp and alarm state are passed to the control logic?

A bypass system can be appropriate when a conventional irrigation timer already controls the zones. An on-demand system needs more explicit safeguards because a bad value can influence both the start and stop of an event. The controller documentation must define whether a higher number means wetter soil and whether the signal is VWC, water potential, a normalized percentage or an arbitrary sensor count.

Define what the reading actually means

Volumetric water content (VWC) is the volume of liquid water per volume of soil and is commonly expressed as a percentage [E04]. A VWC threshold can be related to field capacity, permanent wilting point and available water holding capacity. A kPa or other water-potential threshold is a different measurement basis. A raw ADC value, a vendor "soil humidity" percentage and a calibrated VWC value are not interchangeable merely because they all appear as numbers on a dashboard.

Before setting a threshold, record the sensor model, measurement principle, output unit, scaling equation, depth, soil or substrate, and the controller input configuration. If the listing or manual does not identify the unit or the conversion, treat the value as unknown. Do not calculate a crop threshold from an unverified display percentage.

Establish field capacity before calculating the dry trigger

For VWC-based scheduling, first establish what "full" means at the installed station. UMN Extension states that a sensor reading taken after 12 to 24 hours following heavy irrigation or rain can be used as a field-capacity observation, provided the site and sensor are suitable [E04]. UF/IFAS similarly describes saturating the soil at the probe and allowing approximately 24 hours before using the controller reading as a site-specific reference [E02]. These are commissioning procedures, not a claim that every soil drains on the same schedule.

Then determine the dry-side trigger from the crop, root-zone depth, soil water holding capacity and irrigation capacity. UMN gives management allowable depletion (MAD) as a common concept and notes that many crops begin to experience stress when depletion reaches roughly 30 to 50 percent of available water holding capacity; the correct value varies by crop, growth stage and system capacity [E04]. Use that range as a starting hypothesis, not a universal preset.

Step	Field action	Record
1. Wet reference	Irrigate or wait for a representative wetting event, then observe the sensor after drainage has largely settled.	Timestamp, irrigation amount or rainfall, reading, weather and soil condition.
2. Dry reference	Observe the same station as the root zone dries toward the point at which the crop or operating plan requires water.	Reading, crop stage, visible stress evidence and reference measurement.
3. Trigger calculation	Convert the selected MAD or water-potential rule into the sensor unit used by the controller.	Formula, constants, unit conversion and approver.
4. Control test	Force or simulate a dry-side value and verify the controller permits or starts irrigation.	Input value, valve state, timestamp and alarm behavior.
5. Wet-side test	Confirm the upper threshold or bypass condition prevents unnecessary watering and stops an event when specified.	Input value, valve state, maximum runtime and event log.

Convert site observations into a low and high threshold

For a VWC example, assume field capacity is 32% VWC and the selected dry trigger is 40% depletion of an available range from 16% to 32%. The available range is 16 percentage points. The dry trigger is therefore 32 − (0.40 × 16) = 25.6% VWC. This is an illustration of the calculation method, not a recommendation for a particular crop or soil.

The high threshold should be set from the observed wetting response and the controller behavior. If the sensor reaches the high value while water is still moving through the profile, the event may stop too early; if the system runs past field capacity, it may waste water or leach nutrients. For bypass control, the wet-side setpoint usually blocks the scheduled event. For on-demand control, keep a deliberate gap between the low start point and high stop point so normal measurement noise does not chatter the valve.

Threshold	Decision	Evidence required
Low or dry threshold	Permit or start irrigation when the root-zone signal reaches the selected depletion limit.	Crop stage, field-capacity reference, depletion rule, sensor unit and reference check.
High or wet threshold	Stop an on-demand event or block a bypassed cycle once the root-zone condition is adequately refilled.	Post-irrigation response, drainage risk, controller stop behavior and maximum runtime.
Deadband	Hold the current valve state between the low and high thresholds.	Repeatability of the signal and observed recovery after an irrigation pulse.
Fail-safe value	Define what happens when data are missing, stale, out of range or contradictory.	Controller manual, alarm test and written operating rule.

Test the control loop, not only the probe

A correct threshold on a sensor does not guarantee a correct irrigation event. Test the complete loop: probe contact, power, cable, address, input scaling, timestamp, controller decision, valve output and event record. UF/IFAS shows the sensor as part of a switching path between a timer and a solenoid valve; the physical wiring and switch state matter as much as the displayed reading [E02][E03].

• Inject or simulate a dry-side input and confirm the intended zone is permitted or started.
• Return the input to the wet side and confirm a bypass or stop action without waiting for an unrelated schedule.
• Disconnect communications or provide a stale timestamp and verify the documented fail-safe state.
• Run a short irrigation pulse, observe the sensor response at the correct depth, and check that the event is recorded with start, stop and reason.
• Compare neighboring stations or a reference method before enabling unattended control. A plausible number from one sensor is not proof that the control decision is safe.

EPA WaterSense treats consistent enable or disable behavior at a preset moisture threshold as a performance attribute for labeled landscape controllers [E05]. The practical lesson for a buyer is to write the threshold repeatability and response test into acceptance criteria rather than accepting a dashboard screenshot as proof.

Account for placement and zone differences

A threshold only represents the soil volume around the probe. UF/IFAS recommends representative soil, root-zone placement and, for a single sensor serving multiple zones, locating it in the zone that is normally driest or most in need of irrigation [E02][E03]. That placement rule affects the control consequence: a sensor in a wet corner can block irrigation for a dry zone, while a sensor beside an emitter can start an event too late or too early.

Do not solve a placement problem by continuously moving the threshold. If zones have different soils, emitters, crops or root depths, use zone-specific sensors or a documented control strategy. Keep threshold changes versioned with the station, crop stage and irrigation program so the history remains interpretable.

Two sourcing candidates for threshold-based control

The offers below are sourcing candidates for different control architectures, not ranked recommendations. The fields are marketplace or supplier statements unless noted.

Candidate — BGT Soil 8-in-1 online monitoring sensor

Candidate — JXCT JXBS-3001-WSD NB-IoT soil sensor

Write the operating rule before enabling automatic watering

A commissioning record should state the low threshold, high threshold, units, sensor and station identity, control mode, minimum and maximum run time, lockouts, stale-data timeout, rain or weather interlock, and the person authorized to change the setpoint. It should also say what happens when the threshold is crossed during a maintenance visit or when the sensor is replaced.

A practical first release is usually conservative: run the system in monitor or bypass mode, compare sensor decisions with field observations, then allow automatic starts for a limited period with a maximum runtime and a manual override. Promote the system only after the event log shows that dry triggers, wet stops, missed data and alarms behave as written.

Bottom line

Set a soil moisture threshold from a measured site reference and a defined irrigation objective. Confirm the unit, calculate the dry trigger, use a separate wet-side stop or bypass rule, add a deadband and fail-safe state, and test the sensor-to-valve loop before unattended control. A number that looks reasonable on a dashboard is not a commissioning result until the field response and the control action agree. BGT is a wired multi-parameter control-input candidate; JXCT is a remote cellular input candidate. Both require a register map or API path, a dry/wet input test and an exact-variant RFQ before their readings control a valve.

Evidence and source notes

• E01 — Virginia Tech and University of Georgia Extension, Scheduling Agricultural Irrigation Based on Soil Moisture Content: https://www.pubs.ext.vt.edu/BSE/BSE-339/BSE-339.html
• E02 — University of Florida IFAS, Smart Irrigation Controllers: How Do Soil Moisture Sensor Systems Work?: https://ask.ifas.ufl.edu/publication/AE437
• E03 — University of Florida IFAS, Smart Irrigation Controllers: What Makes an Irrigation Controller Smart?: https://ask.ifas.ufl.edu/publication/ae442
• E04 — University of Minnesota Extension, Soil Moisture Sensors for Irrigation Scheduling: https://extension.umn.edu/natural-resources/conservation/agricultural-soil-and-water/irrigation/soil-moisture-sensors-for-irrigation-scheduling
• E05 — US EPA WaterSense, Soil Moisture-Based Irrigation Controllers: https://www.epa.gov/watersense/soil-moisture-based-irrigation-controllers
• E06 — Alibaba marketplace listing, BGT Soil 8-in-1 online monitoring sensor, Product ID 1600845882411: https://www.alibaba.com/product-detail/BGT-Soil-8-in-1-Online-Monitoring_1600845882411.html
• E07 — Alibaba marketplace listing, JXCT NB-IoT soil humidity and temperature sensor, Product ID 1600088138408: https://www.alibaba.com/product-detail/JXBS-3001-WSD-NB-IoT-Soil-Humidity_1600088138408.html