THE SHORT ANSWER

A sensor's raw number, a relative scale and a calibrated water-content estimate are different things; compare readings in context before acting.

This is an interpretation guide, not a tested device build, automatic-watering recipe or laboratory calibration. The circuit-board photograph is illustrative.

Start with the unit

A garden dashboard might show 600, 50%, or a coloured bar. Before deciding whether to water, ask what the display represents. A raw sensor value can be an electrical reading or a number produced by an analogue-to-digital converter. A percentage might be a device’s relative scale, a calibrated estimate of water content, or something else defined by its manufacturer.

Those numbers are not interchangeable. Our introduction to soil sensors explains why an electrical signal is not a direct reading of how a plant feels. This guide takes the next step: how to keep a measurement useful without pretending it answers more than it does.

A relative scale is not a water-content measurement

Suppose a particular sensor gives a raw value of 800 in one dry reference sample and 400 in one wetter reference sample. A later reading of 600 is halfway between those two numbers on a straight-line scale. Calling that position 50% is convenient, but it only means halfway between the chosen reference readings. It does not mean half the pot is water, or that half the water available to the plant remains.

The example is invented to illustrate arithmetic; other sensors can increase rather than decrease as a medium gets wetter. If both reference readings are almost the same, the scale is unstable. A value outside the reference range is a reason to investigate, not something to hide automatically by rounding it to an endpoint.

What volumetric water content describes

University of Minnesota Extension defines volumetric water content as the volume of water divided by the volume of soil. Its 25% example represents 0.25 units of water per unit volume of soil. That definition differs from a relative scale built from two raw readings.

Some sensors estimate water content indirectly from electrical properties and a calibration relationship. Other instruments estimate soil water tension, which concerns how strongly water is held. Manufacturer specifications, the medium and installation matter. A percentage displayed by an inexpensive probe is not automatically a validated volumetric measurement.

Compare a trend with an observation

Keep the probe in the same position and record the date, sensor, medium, depth, raw value and any watering event. Note what a careful physical check of the mix suggests as well. A sequence can reveal a sudden jump after watering, a gradual change or a reading that remains suspiciously fixed. This is a practical observation method, not proof that the sensor is accurate.

Oklahoma State University’s guide to sensor trends discusses interpreting data for field irrigation, including why installation and calibration matter. A small pot is not an agricultural field: its size, mix and plant requirements differ. Use the idea of reading a trend, not field-specific irrigation thresholds, as a starting point for your own container.

Change one thing at a time

If you compare a drier and a wetter reference, use a separate sample of the same mix rather than stressing a living plant. Record how you prepared it, keep the probe placement consistent and repeat the readings. Follow the device’s instructions for the part that may be immersed or cleaned. Do not assume that a comparison in one mix will transfer unchanged to another.

An abrupt change after moving the probe, a loose connection or a changed growing medium can look like a moisture change. Oklahoma State’s sensor overview also discusses how soil properties can affect readings. If the signal and your observation disagree, check the setup before deciding which is wrong.

Leave automation for a later experiment

Start with observation, not a pump. A sensor can fail, drift or represent only a small part of a pot. A two-point relative scale does not establish a safe watering threshold. Use species-specific growing advice and the container watering guide alongside the record; a future automatic system would need separately tested hardware, limits and failure handling.

KEEP EXPLORING

Sources & further reading

This founding-edition article was drafted with AI assistance and checked against the sources above. It has not had independent expert review. Read our editorial approach.