Collection Planning

When Can This Place Next Be Imaged?

Before you pay to task a satellite, it is worth knowing whether a free one is already on its way. Pass prediction answers that — across the systematic collectors that image on their own schedule and the agile platforms that only image when someone orders it.

Ask Delta Agent See the API

Quick answer — what Off-Nadir Delta is

Satellite pass prediction works out when a given location falls within a satellite’s imaging geometry, by propagating public orbital elements with SGP4 and testing the target against each platform’s swath or off-nadir pointing cone. The practical value is the decision it unlocks: systematic collectors — Sentinel-1, Sentinel-2, NISAR, Landsat — will image the area for free on their own repeat cycle, so if one lands inside your deadline you can simply wait. Agile commercial platforms — WorldView, ICEYE, Capella, SkySat — offer far more windows, but each is an access opportunity that needs a paid tasking order, not a collect that will happen by itself.

The question is always “wait, or pay?”

Tasking a commercial satellite is expensive and slow to justify. Waiting for an open-data pass is free but only useful if it arrives in time and the sensor can actually see what you need. Predicting the windows turns that from a guess into an arithmetic problem.

Free and scheduled

Systematic collectors fly a fixed repeat cycle and image along a wide swath. You cannot redirect them — but when the swath covers you, the data is open.

Paid and steerable

Agile platforms slew off-nadir on command, so opportunities are frequent — but an opportunity is not a collect until someone places the order.

Weather decides too

An optical pass over cloud yields nothing usable and a night pass yields nothing at all. SAR passes are usable regardless — often the earlier real answer.

Which satellites are considered, and what each one means for you

Facts last reviewed 2026-08-23.

Satellite family	Sensor	Collection	Nominal swath	Access	What a window means
Sentinel-2 (Copernicus)	Optical, ~10 m	Systematic	290 km	Free and open	It will image, if it is daylight and clear
Sentinel-1 (Copernicus)	C-band SAR, IW ~5×20 m	Systematic	250 km	Free and open	It will image, cloud or dark, looking right
NISAR (NASA/ISRO)	L-band SAR, 24 cm	Systematic	240 km	Free and open via ASF	It will image, cloud or dark, looking left
Landsat-8/9 (USGS/NASA)	Optical, ~30 m	Systematic	185 km	Free and open	It will image, if it is daylight and clear
Maxar WorldView	VHR optical, ~0.3–0.5 m	Agile	16 km	Commercial tasking	It could image — only if you order it
ICEYE	X-band SAR, ~0.25–1 m	Agile	30 km	Commercial tasking	It could image — only if you order it
Capella Space	X-band SAR, sub-metre spotlight	Agile	10 km	Commercial tasking	It could image — only if you order it
Planet SkySat	Agile VHR optical, ~0.5–0.7 m	Agile	8 km	Commercial tasking	It could image — only if you order it

Positions are propagated with SGP4 from public CelesTrak element sets, so no account or key is involved. Sensor specifications are the published mission figures. Access geometry is not a guarantee that a scene will exist: for the systematic missions the repeat cycle is the better guide to when an archive will actually gain a scene, and for the agile platforms a window is an opportunity requiring a paid order. Off-Nadir Delta is an independent project and is not affiliated with ESA, NASA, ISRO, USGS, Maxar, Vantor, ICEYE, Capella Space, or Planet.

A pass is not the whole answer — geometry is

Two passes over the same place can differ completely in what they measure. Whether the orbit is ascending or descending changes the illumination and the look direction. For radar, the side the antenna looks and the incidence angle decide which slopes are foreshortened, which are laid over on top of each other, and which fall into shadow — Sentinel-1 looks right, NISAR looks left, and in real terrain that is not a detail.

So the useful output is not just “a satellite goes over on Thursday.” It is which satellite, on what geometry, and whether that geometry can actually see the thing you care about. For an area in relief you can check the layover and shadow fractions for a given incidence and look azimuth before deciding a pass is worth waiting for.

How It Works

Give it a place

A point or a bounding box — the strike site, the port, the burn scar, the area you have been asked about.

Choose which satellites to consider

Free systematic collectors only, if the question is "can I just wait?" — or include the agile commercial platforms if you are deciding whether to pay for a tasking order.

Read the access windows

Each window says which satellite, when, and on what geometry — ascending or descending, and for SAR which side the antenna looks.

Decide: wait or task

If an open-data pass lands inside your deadline, waiting is free. If it does not, you know exactly how long the gap is and what tasking would buy you.

Frequently Asked Questions

How is a satellite pass predicted?▼

From orbital elements. Each satellite’s position is propagated with SGP4 — the standard model applied to two-line element sets — using public element sets published by CelesTrak, and the result is tested for whether the target falls within that platform’s access geometry. For a systematic collector that test is whether the target lies inside the imaging swath; for an agile commercial platform it is whether the target falls inside the off-nadir pointing cone the spacecraft can slew to.

What is the difference between systematic and agile collection?▼

Systematic collectors — Sentinel-1, Sentinel-2, NISAR, Landsat — fly a fixed repeating orbit and image near-nadir along a set swath. You cannot ask them to look at your area; you can only work out when their swath will cover it, and the data is free and open when it does. Agile commercial platforms — WorldView, ICEYE, Capella, SkySat — can slew off-nadir on command, so they have far more opportunities over any given place, but each one is an opportunity that requires a paid tasking order rather than something that will happen on its own.

Does an access window mean an image will exist?▼

No, and this is the distinction that matters most. Geometric access says a place can be imaged at that time. Whether a scene actually reaches a catalogue depends on the mission’s acquisition plan, on downlink and processing, and — for agile platforms — on whether anyone paid to task the collect at all. Predicted windows for commercial platforms are access opportunities, not guaranteed collects. For the systematic missions, the repeat cycle is a much better guide to when the archive will actually gain a scene.

Why does cloud matter for some satellites and not others?▼

Optical sensors need sunlight and a clear line of sight, so a Sentinel-2 or Landsat pass over a cloudy area yields no usable scene, and a night-time pass yields nothing at all. Synthetic-aperture radar provides its own illumination and passes through cloud, so a Sentinel-1, NISAR, ICEYE, or Capella pass is usable regardless of weather or daylight. That is why the honest answer to "when can I next see this place?" is usually a SAR pass sooner and an optical pass later but easier to interpret.

Which side does the radar look, and does it matter?▼

It matters a great deal in terrain. Sentinel-1 illuminates the ground to the right of its track, and NISAR is left-looking. Combined with the incidence angle, the look direction determines which slopes are foreshortened, which are laid over, and which fall into radar shadow — so two passes over the same valley can differ enormously in how much of it is actually measurable. If the target sits in relief, checking the terrain geometry for a given incidence and look azimuth before committing is worth the minute it takes.

Is pass prediction available over the API?▼

Yes. Pass prediction is exposed through both the REST API and the MCP server, so an AI agent or your own pipeline can ask when a place can next be imaged and by which platform before deciding what to do. It is metered by tokens; the pricing page has current rates.