SWRainAlertMeasurement
Rain Gauge or Radar? How to Measure Rainfall for Compliance
A gauge measures one point exactly, if someone reads it. Radar measures an area approximately, always. Those failure modes are opposites, which is the argument for using both.
This is the first objection we hear, and it is a fair one: we already have a rain gauge on site. Good — in several states you are required to. The question is not gauge or radar. It is what each one is for, and what happens when the one you rely on fails.
What each one actually measures
A standard 8-inch rain gauge collects water through an opening of roughly 0.05 square metres. Whatever falls into that funnel is measured almost exactly. Everything else on the site — the far corner, the sediment basin, the stockpile — is inferred.
A radar grid cell is the opposite trade. It covers about a square kilometre and reports an area average, so it never misses a corner of the site. But it is an estimate derived from what a radar beam saw several thousand feet up, not a measurement of water in a bucket.
How a gauge fails
- Nobody reads it. The most common failure by a wide margin. A gauge is a manual instrument; a storm at 4pm on a Friday is measured on Monday, if at all.
- It overflows or was not emptied. A gauge that was full before the storm reports whatever fits.
- Siting. Placed near a building, a stockpile or under equipment, it reads low or high, consistently, and nobody notices.
- It disappears. Gauges get knocked over by machinery, buried, or walk off site.
- Freezing. A tipping-bucket gauge without a heater does not measure frozen precipitation as it falls.
The unifying problem is that all of these fail silently. A gauge that was not read and a gauge that read zero look identical in the record: a blank line, or a 0.00 entered on Monday for a Friday nobody checked.
How radar fails
- Beam blockage. Terrain can hide the lowest scan, so precipitation below the beam is not seen.
- Distance. The beam climbs with range. Far from the radar it can overshoot shallow rain entirely.
- Bright band. Melting snow reflects strongly and can be read as much heavier rain than reached the ground.
- No coverage at all. Some cells simply are not observed.
The difference that matters: these failures are knowable. NOAA’s grids encode “no coverage” and “missing” explicitly, which means a system reading them honestly can tell you it could not see your site, rather than reporting a confident zero.
This is the failure mode we are most careful about. Those sentinel values are negative numbers — read carelessly they become 0.0, and a site the radar could not observe becomes indistinguishable from a site where it did not rain. One of those needed an inspection. SWRainAlert renders a gap as a gap, never as 0.00″.
What your permit says about which one counts
Read the permit before deciding this on technical grounds. Several state permits explicitly contemplate an on-site gauge, and some name it as the record for determining whether a qualifying event occurred. Where a permit does that, the gauge is the system of record and a radar estimate is supporting information — useful, but not the thing being audited.
Other permits allow a nearby weather station or an equivalent data source. In those cases radar-derived data is generally acceptable, and often more defensible than a manual log with gaps in it.
The combination most mature programs land on
- Radar for the alert. It is continuous, it covers every site at once, and it does not depend on anyone being on site. It tells you to go look.
- The gauge for the record, where the permit requires it. The inspector reads it on arrival and logs it against the event.
- Both numbers preserved. When they disagree, that disagreement is information about the storm and about the gauge, and it belongs in the file rather than being reconciled away.
The failure mode this design eliminates is the one that actually causes penalties: a qualifying storm that nobody knew about until the deadline had already passed.
Common questions
Does a SWPPP require a rain gauge on site?
It depends on your permit. Several state construction general permits require or contemplate an on-site rain gauge, and some name it as the record for determining whether a qualifying rain event occurred. Others allow a nearby weather station or equivalent data source. Read your own permit text rather than assuming.
Can radar rainfall data be used for SWPPP compliance?
In many jurisdictions, yes, where the permit allows an equivalent data source rather than mandating an on-site gauge. Radar is generally accepted as supporting information everywhere, and it is the practical way to know a storm crossed a threshold at a site nobody was standing on.
What happens if the rain gauge and the radar disagree?
They frequently will, because they measure different things: one point versus an area average. Keep both readings. Where the permit names the gauge as the system of record, the gauge governs, but the radar reading is useful evidence that the event was monitored and the disagreement was known rather than hidden.
A note on what this is. General guidance, not legal advice. Your permit, your SWPPP and your state’s requirements govern, and they differ. SWRainAlert is a monitoring and recordkeeping aid, not a compliance guarantee — where a permit names an on-site rain gauge as the system of record, the gauge is the record.