What does a pixel represent?
A raster image is a grid of cells, or pixels. In satellite imagery, each cell represents a measurement associated with an area on the ground. Pixel spacing is often used to describe spatial detail, but recognizable detail also depends on the sensor and the scene.
A small feature may share a cell with other surfaces. Enlarging that cell on your screen changes its display size; it does not add an observation of what happened inside it.
Same landscape. Different measured detail.
Change the sample grid. Each larger cell combines the colors of smaller patches in this constructed landscape; the small building and narrow river become harder to separate.
A large cell mixes several surfaces. Enlarging this grid on a screen would make bigger cells, not recover the small feature.
Smaller cells separate more of the landscape, but a narrow feature can still share its measurement with neighbouring surfaces.
Finer sampling reveals more of this constructed landscape. Real sensor performance, noise, geometry, and validation still determine what can be identified.
Prepared teaching illustration. Landscapes, dates, and results are fictional; no live imagery search, analysis, or app generation is running.
Schematic only. For the distinction between pixel spacing and spatial resolution, see the USGS guidance on evaluating remote-sensing systems.
There is more than one kind of detail.
- Spatial: the detail that can be distinguished across the ground.
- Temporal: how frequently observations are made. Actual usable observations can still have gaps.
- Spectral: how finely a sensor separates wavelength ranges, usually described by the widths and positions of its bands.
- Radiometric: how finely recorded values distinguish differences in measured energy. More available values do not by themselves establish accuracy.
A source can be useful in one dimension and limited in another. “Higher resolution” is incomplete without saying which dimension matters to the question.
Definitions adapted in plain language from the USGS Landsat glossary and USGS introduction to remote-sensing resolution.
Match the detail to the work you need to do.
Begin with what you need to distinguish. Comparing vegetation across a region, reviewing a field and checking a narrow boundary feature call for different data. Ask how the method was evaluated at that scale as well as how the image looks.
Practice the decision
What kind of detail matters?
Choose a fictional task to see the first question you should ask about its observations.
Start at the scale of the decision.
Imagine reviewing a pattern across a large region. You may care more about consistent coverage and repeated observations than a close view of each small feature.
The feature needs enough evidence.
Imagine looking for a narrow change along a site boundary. If the feature occupies only part of a pixel, nearby surfaces can contribute to the same measurement. A sharp-looking basemap does not solve that limitation in the analysis layer.
Detail in time matters too.
Imagine an event happening between two available observations. Finer spatial detail in those images cannot recover the missing observation during the event.
Fictional examples for learning. No live analysis is running.
When several datasets are combined, inspect the source used for each result. A detailed background image can make an overlaid analysis seem more precise than the measurements behind it.
Use these questions to choose your data.
- What is the smallest feature or change I need to distinguish?
- Which measurements will the analysis use?
- What dates and parts of the site have usable observations?
- What evidence shows that the method works for this task and scale?
You can ask these questions in ordinary language when preparing your analysis. Keep the answers with the map or report so the next person understands the detail it can support.
See how the choice affects a practical site comparison and reporting workflow.