Searching for imagery


Every purchase on Data Sphere starts at the search page. Search is built around where and when you need imagery, so the two things you always provide are an area of interest and, optionally, a capture-date window.

Draw an area of interest

Pan and zoom the map to your target, then draw a bounding box over it. The box defines the footprint we match imagery against, so results are ranked by how much they overlap the area you drew. You can redraw the box at any time to widen or narrow the search.

The tools sit in three groups down the right-hand side of the map. The first zooms in and out. The second draws the area: Box drags out a rectangle, Shape lets you click corner by corner, Reshape drags the corners of what you have already drawn, and Clear removes it. Reshape and Clear stay greyed out until there is an area to act on. Hover over any tool for a one-line description of what it does.

The third group is for placing an area without drawing it. Upload a zipped shapefile or a GeoJSON file and its outline becomes the search area. Locate asks your browser where you are, moves the map there and drops a marker, which is the quickest way to start from your own position. Your browser will ask permission the first time you use it, and nothing is sent to us if you refuse.

Set a date range

Add a start and end capture date to restrict results to imagery taken in that window. Leaving the dates blank searches across the whole archive. This is the fastest way to cut a large result set down to the period you actually care about.

Describe what you want in plain English

If you would rather not think in filters, use natural-language search and type a description such as "recent high-resolution optical imagery of a coastline". Press the search button and Data Sphere interprets the request, runs the search itself and takes you to the results, as long as the description named a place we could turn into an area on the map or you have already drawn one. If it could not, you get the equivalent filters pre-filled, with a line saying why, and you can adjust them and press the button again.

Run the search

Submitting the search opens a short progress view while we query providers and aggregate matching imagery. When it finishes you are taken to the results view, where imagery is grouped into product cards. The next step covers how to refine those results.

While providers are still answering, a small window over the results tells you how many of them have come back and which ones are still working. It shows four providers at a time and moves through the rest on its own, so it stays the same size however many providers your search reaches; you can scroll it yourself to look at any of them. Closing the window does not stop the search. Results already found stay on the page behind it, and anything still to come is added as it arrives.