The radiometric resolution of a sensor refers its sensitivity, which is the ability to detect small differences in signal strength as it records the radiant flux reflected, emitted, or back-scattered from the terrain.
The specification of the radiometric resolution is different in the optical domain of the electromagnetic spectrum than in the radar range.
In the optical domain, the radiometric resolution is given in bits. The maximum number of brightness levels available depends on the number of bits. The larger this number, the higher the radiometric resolution. As an example, the optical sensor Sentinel-2 has a radiometric resolution of 12 bits. This means that a pixel of an image acquired by Sentinel-2 can have 2^12 = 4096 grey levels.
In the radar domain, the radiometric resolution is usually specified as a backscatter level expressed as an logarithmic value. For instance, the radiometric resolution of Radar Scattermeters lies in the range of 0.1 to 0.3 dB, whereas the radiometric resolution of SAR sensors are in the range of 1.2 – 2.5 dB. This means that only differences in radar backscatter larger than these values can be interpreted as interpretable changes the of backscatter conditions at the Earth’s surface. Smaller measurement differences could have been caused by differences in backscatter conditions or just as well by instrument noise.
The capability of a sensor or EO product to resolve anything is a function of its (spatial, temporal, spectral and radiometric) resolution and of the detail at which a geographic phenomenon of interest manifests itself in time and space. A geographic phenomenon can be named or described, georeferenced and provided with a time interval at which it exists. The geographic phenomenon of interest is the one of which a user needs information to help him make a decision. Therefore, the geographic phenomenon needs to be resolved with a low enough uncertainty and a high enough quality that allows the user to make a decision with confidence.
For example, let’s consider a helicopter pilot that wants to know whether a specific site is suitable for an emergency landing. The decision to perform an emergency landing may be supported with an EO-derived digital map of emergency landing sites that are flat enough (as well as large enough for the pilot’s helicopter and free of any obstacles on the surface and in the approach area). If we only focus on the flatness of the terrain, we need a digital elevation model (DEM) of high enough spatial resolution and accuracy in the Z dimension to calculate slope within acceptable levels of uncertainty. The pilot probably can tell us what degrees of slope are okay for his helicopter and tell us sites (e.g. football fields) where such a landing would succeed. However, this is only the input to an analysis of different DEMs to identify the minimum spatial resolution and accuracy in the Z dimension to model slope products and associated uncertainty to derive an emergency landing site product that fulfils the requirements. Thereby the capability of different DEMs to resolve emergency landing sites can be analysed.
Spatial resolution is a measure of the smallest angular or linear separation between two objects that can be resolved by the remote sensing system. A useful heuristic rule of thumb is that in order to detect a feature, the nominal spatial resolution of the sensor should be less than one-half the size of the feature measured in its smallest dimension.
Other types of resolution of an EO dataset are available that determine for various geographic phenomena under investigation whether it is possible to resolve them in the data. These are radiometric resolution, spectral resolution and temporal resolution. Radiometric resolution is defined as the sensitivity of a remote sensing detector to differences in signal strength as it records the radiant flux reflected, emitted, or back-scattered from the terrain. Spectral resolution is the number and dimension (size) of specific wavelength intervals (referred to as bands or channels) in the electromagnetic spectrum to which a remote sensing instrument is sensitive. The temporal resolution of a remote sensing system generally refers to how often the sensor records imagery of a particular area. For time-series analysis, the temporal resolution determines the time granularity for resolving processes that underlie the change that is observable between subsequent images.