The crew is on site at 6 am. The forecast says fine and clear. Two hours later, the operator calls a no-go and reschedules. From the buyer's side this looks like risk aversion or scheduling drama. From the operator's side it's the result of a nine-variable checklist, half of which the buyer never sees and any one of which can degrade the deliverable in ways that don't show up until QA. This article walks through what's actually being evaluated, why marginal-weather captures cost more than they save, and how schedule pressure corrupts the call.
If you've ever booked a drone LiDAR capture, watched the crew arrive on a "fine" day, and then received a phone call by mid-morning announcing a postponement, you've met the go/no-go problem. The forecast may have looked acceptable; the BOM radar may have been clear; the field crew is fully mobilised and clearly motivated to fly. And the operator still says no.
The buyer's instinct is sometimes to push back — "the forecast says clear, can't we just try?" — and the operator's instinct is to absorb the schedule pressure rather than explain. The conversation usually ends with a defensive operator and an unhappy buyer.
Both are reading off incomplete information. The operator is evaluating nine variables most of which aren't in any consumer forecast. The buyer is evaluating one (general weather appearance) plus schedule cost. Bridging the gap requires laying out what the operator is actually checking.
This article is the checklist — the nine variables that determine go/no-go, why marginal captures degrade quality more than they save schedule, and the small behavioural pattern that keeps the call honest under schedule pressure.
The single most disqualifying weather variable. Drone LiDAR pulses can't penetrate cloud (the laser scatters off water particles and the captured point cloud includes spurious returns at cloud altitude). Operating below cloud is the workaround, but it requires sufficient ceiling height above the planned flight altitude.
For a typical 80 m AGL drone LiDAR capture, the operator wants at least 200 m of clear air above flight altitude — comfortable separation, allows for cloud descent, retains visibility of the aircraft from the ground. Ceiling at 150 m means flying lower than planned (reducing coverage rate, possibly compromising swath geometry); ceiling at 100 m is a no-go.
Cloud ceiling rarely appears in consumer forecasts but is the first thing operators check via aviation weather products (METAR, TAF) and visual confirmation on arrival.
Multirotor drones tolerate up to about 8-10 m/s sustained wind in their certification spec, with most operators self-imposing a tighter operational limit of about 7 m/s. Fixed-wing drones tolerate more (up to 12-15 m/s) but become harder to land safely as wind increases.
The variable most operators actually watch is gust rather than sustained — sustained 6 m/s with 12 m/s gusts is harder to fly than steady 8 m/s. Gusts introduce attitude perturbations the IMU has to compensate for; large gusts can put the trajectory solution outside the bounds of post-processing recovery.
Wind also affects safety margins for launch and landing. A landing in 10 m/s crosswind on a confined launch pad is materially harder than the same landing in 5 m/s.
Drone LiDAR sensors are not waterproof. Light rain on electronics during flight is a hardware-damage risk; sustained rain is a guaranteed no-go.
Even forecast precipitation 2-3 hours ahead is a concern — the operator has to commit to a flight plan, fly it, return safely, and demobilise before weather arrives. A 3-hour capture with a 2-hour rain forecast window leaves no margin.
What the operator is reading: BOM radar for active precipitation, AusEPCN bureau forecasts for short- horizon outlook, and local visual conditions for storm cell development.
Aviation visibility requirements apply: drone operations require visual line-of-sight (VLOS) with the aircraft unless the operator holds BVLOS approvals. In practice, this means usable visibility of at least 1.5-2 km on small drones and further on fixed-wing.
Mist, light fog, smoke or dust that drops visibility below the operational threshold is a no-go regardless of other conditions. The aircraft can be physically controllable; the regulatory requirement still applies.
In southern Australia mornings, fog and low cloud are the dominant winter no-go driver. The variable the operator is watching is the dewpoint depression — the gap between current air temperature and dewpoint. Small gap (under 2 degrees) means saturated air, fog formation or persistence likely. Larger gap (over 5 degrees) means dry air, fog won't form or will burn off quickly.
A typical winter morning: arrival at 6 am, air temp 5 degrees, dewpoint 4 degrees, dense fog throughout the project area. Forecast says clearing by 9 am. Operator watches the gap widen as the sun warms the air; depending on rate of widening, calls go at 9 am, 10 am or postpones.
This is the most common source of "looked fine to me but the operator said no" disputes. The forecast may have called clear by 9 am; the actual fog burn-off ran 90 minutes late; the operational window closed before capture could complete.
Particularly relevant during fire season but also for agricultural burns and dust events. Smoke and dust particulates scatter the LiDAR laser the same way cloud does — captured returns include spurious points at smoke altitude, the trajectory solution may degrade if GNSS satellite reception is attenuated by haze.
Air quality readings (PM2.5, PM10) above project-area thresholds prompt a no-go. Some operators have specific numerical limits; others use visual assessment combined with sensor performance data from recent captures.
(See seasonal pricing article for how smoke risk fits into the seasonal cost picture.)
Less obvious to non-specialists but operationally important. GNSS positioning accuracy depends on the geometric distribution of visible satellites — when satellites cluster in one part of the sky, the positioning solution is geometrically weaker (high PDOP). When satellites are spread evenly across the sky, positioning is geometrically strong (low PDOP).
Operators check PDOP for the capture window via satellite ephemeris prediction. Most of the day, PDOP is fine. Specific windows of 1-2 hours can have elevated PDOP that degrades trajectory accuracy even with PPK correction. A capture scheduled into a high-PDOP window will produce technically valid data with measurably worse positioning than the same capture scheduled an hour earlier or later.
(See RTK vs PPK article for the GNSS technical background.)
Pure LiDAR is largely indifferent to solar angle — the laser provides its own illumination. Hybrid captures that include orthophoto, oblique imagery or multispectral are sensitive to solar position.
Low-angle sun produces long shadows and side-lighting that complicates photo processing; vertical sun (midday in summer at low latitudes) produces flat lighting that obscures terrain shape. The optimal capture window for camera work is mid-morning to early afternoon for most Australian latitudes.
For pure LiDAR jobs, solar angle is irrelevant. For hybrid jobs, it can constrain the operational window to 4-5 hours within the daylight envelope.
(See hybrid workflows article for when this matters.)
Late spring and summer afternoons in inland Australia produce convective thermals that affect drone stability. Multirotor drones cope; fixed-wing platforms experience attitude oscillations that degrade trajectory solution quality even when the aircraft remains controllable.
Thermal activity is correlated with surface temperature and time of day — strongest in the 2-4 pm window in summer. Operators flying fixed-wing on a hot summer afternoon may call no-go even with perfect clear-sky conditions because the captured data will be post-processable but operationally degraded.
The schedule-pressure argument — "we're here, let's fly anyway" — feels economically rational from the buyer's side. The schedule cost of postponement is visible; the deliverable-quality cost of marginal capture isn't.
In practice, marginal captures produce one of three outcomes:
1. The QA pack passes, accuracy is at the bottom of the spec. No re-fly needed but the deliverable is near the threshold. Subsequent design decisions inherit less margin than expected.
2. The QA pack fails or is borderline. Re-fly required anyway. The marginal capture day cost has been incurred and the standard capture day still needs to happen. Net cost of pushing through: 1.5-2x the cost of postponement.
3. The QA pack passes but problems surface later in downstream use. Volumes are slightly off, contours have artefacts, classification is patchy in specific areas. These get attributed to "lidar limitation" rather than "marginal capture" and aren't fed back into the operator's decision-making.
The expected-value calculation for the operator strongly favours postponement under any meaningful weather marginality. The buyer's calculation often doesn't — but the operator is the one with deliverable accountability, so they're the one whose call rules.
There is a failure mode worth naming. When schedule pressure is high (deadline-driven project, mobilised crew, weather window won't reopen for two weeks), the honest no-go call gets harder to make even when the operator knows it's right.
The pattern looks like:
Three behavioural mitigations that good operators build into their process:
Pre-mobilisation forecast review at T-24 hours. If the forecast is marginal a day out, the conversation happens then rather than at the launch point. Crew movement can be deferred without on-site sunk cost.
Stated no-go criteria in writing. Operator's formal "we will not fly if X, Y or Z conditions are observed". This converts go/no-go from a judgement call to a checklist application. Buyers who try to push past stated criteria have to argue against documented policy rather than against the field crew's mood.
Reschedule cost transparency. Buyer knows the cost of postponing a day in advance; can make the trade-off decision with full information. The behavioural problem is worst when the buyer doesn't know what reschedule actually costs.
Four questions that prepare the buyer for a potential no-go call without creating defensive posture:
1. "What's your no-go criteria checklist?" Honest operators have one and will share it. The conversation pre-establishes that no-go is a checklist outcome, not a judgement under fire.
2. "What's the reschedule cost if conditions are marginal?" Both directions — what the buyer pays for postponement and what the operator absorbs. Knowing the numbers up front prevents the schedule- pressure dynamic.
3. "Who makes the final call and at what time?" Pre-mobilisation review at T-24 hours, on-site review at T-2 hours, final call at start of operational window. Establishing the cadence prevents last-minute confusion.
4. "What's the secondary capture window if today is no-go?" Operator's next available slot — keeps the project moving and constrains the impact of any single weather disruption.
A small but important reframe: a no-go is not a failure of planning or a sign of operator incompetence. It's the system working as designed. The operator's job is to deliver an accurate deliverable; weather marginality is one of the major risk inputs to that job; declining capture in marginal conditions is the operator doing their job properly.
The signal worth weighting isn't whether an operator calls no-go sometimes (all reputable operators do). It's whether they call no-go consistently against documented criteria with transparent reschedule mechanics. Operators who never call no-go in marginal conditions are either flying perfect weather every time (improbable) or pushing through marginal captures and absorbing the quality cost silently.
The go/no-go decision on capture morning is a nine-variable checklist most of which buyers don't see: cloud ceiling, wind speed and gust, precipitation current and forecast, visibility, dewpoint depression and fog burn-off, smoke and air quality, GNSS satellite geometry, solar angle for hybrid captures, thermal turbulence.
Marginal-weather captures cost more than they save — the schedule benefit is visible, the deliverable- quality cost isn't, and the expected-value calculation favours postponement under any meaningful marginality.
Schedule pressure corrupts go/no-go calls in predictable ways. Three behavioural mitigations help: pre-mobilisation forecast review at T-24 hours, written no-go criteria, reschedule cost transparency.
Four buyer-side questions prepare the relationship for a potential no-go call: criteria checklist, reschedule cost both ways, who calls and when, secondary capture window.
No-go is the system working. Operators who never call no-go are silently absorbing the quality cost of marginal captures and pushing it into the deliverable.
If you're scoping a capture and want to understand exactly what conditions would prompt a no-go call before mobilisation day, we'll send the criteria document. Honest operators publish theirs; cagey ones don't. Either way, it's the right thing to ask for before any operator quotes you.
The seasonality companion to this day-level piece — covers the structural reasons winter pricing reflects more frequent no-go calls, and why scheduling around the seasonal sweet spot reduces the on-site go/no-go drama.
Sites without operational history are harder to call go/no-go on because the operator doesn't have local-conditions context. Useful pairing for understanding why first captures cost more.