Combine space and in-situ evidence

Copernicus uses dedicated Sentinel missions, other contributing satellite missions and in-situ information collected on Earth or within its atmosphere. The in-situ component includes direct observations and geospatial reference data. A river gauge measures a local condition; a road map supplies a geographic reference; a satellite observes a wider area.

In an original flood-study exercise, the satellite image suggests water near a road, the gauge records river level, and an elevation model describes the terrain. These sources are complementary. Calling all three “satellite data” would hide their origins and make it harder to judge what each can establish.

The Kiruna station photograph illustrates ground infrastructure supporting Sentinel-5P. Its antennas receive spacecraft signals. A ground-based environmental sensor instead measures a condition such as air composition or temperature locally. Both are on Earth, but their roles in the information chain are different.

This distinction is useful when reading diagrams: a line from a satellite to an antenna represents communication, not a new independent measurement of the environment. A second copy of a received file can protect availability, but it does not provide the same validation evidence as a separately measured reference.

Copernicus uses well-characterized reference observations to assess satellite-derived products. Calibration and validation help connect instrument signals to interpretable measurements and test their quality. The comparison must consider what was measured, where, when and with what uncertainty; a reference is useful because its properties are understood.

Try an original example: an area-wide estimate is compared with one point sensor observed six hours later. A difference might reflect real spatial or temporal variation rather than only instrument error. Improve the comparison before choosing a culprit. Record the sampling mismatch instead of treating either number as automatically perfect.