Water has very dark tones in day thermal IR images and moderately light tones in night images compared with the land. This is due in part to a rather high thermal inertia relative to typical land surfaces, as controlled largely by water's high specific heat. Thus, it heats less during the day and holds that heat more at night (an obvious condition you may have experienced when swimming in a pool), giving rise to intrinsic cooler daytime temperatures and often warmer nighttime temperatures than the bulk of materials on the land. Also, being nonsolid, water in natural conditions (rivers, lakes, oceans) is likely to experience disruption of its thermal gradient by convection (e.g., upwelling) and turbulence (e.g., wave action) so that its near-surface temperatures vary by only a few degrees at most (temperature "smoothing").

In general, it is difficult to compensate for, correct, or otherwise remove effects of many of the factors mentioned above. Consequently, temperatures and derivative functions such as apparent thermal inertia (ATI; "apparent" is a qualifier indicating true values are not obtained unless the influence of atmospheric processes and other factors are taken into account) are approximations subject to (sometimes serious) errors. Field measurements of the more critical variables help to alleviate the uncertainties. These, and other sources of ancillary data, can be incorporated into mathematical phenomenological models that attempt to duplicate the roles played by the physical factors.

Some comments about thermal sensors: For scanners designed to sense in the 8 to 14 µm interval, the detector is usually an alloy of mercury-cadmium-tellurium (HgCdTe) that acts as a photoconductor in response to incoming photons in this thermal energy range (Mercury-doped germanium [Ge-Hg] is also used for this interval, although it is effective over a broader range to about 6 µm; over the 3-5 µm interval indium-antimony [In-Sb] is the alloy used in detectors operating in that range). Efficient operation requires onboard cooling of this detector to temperatures between 30 and 77°K, depending on detector type. This is done either with cooling agents such as liquid nitrogen or helium (in a container called a Dewar, that encloses the detector) or, for some spacecraft designs, with radiant cooling systems that take advantage of the cold vacuum of outer space. This cooling is needed to improve the signal-to-noise (S/N) ratio of the detector to a level at which it has a stable signal response. This signal is, of course, an electrical current related to changes in detector resistance that are proportional to the radiant energy.

To obtain a quantitative expression of radiant temperatures, the detector response must be calibrated. Calibration sources (e.g., thermistors) at different temperatures near the extremes expected from the ground are used to provide a correction function. The scanner normally has a glow tube or other device in which a wire passes a current that causes it to glow [giving off radiant energy] at some temperature. Two such thermistors, one glowing at a temperature near the low value anticipated from most targets, the other near the high value, are commonly used. These temperature/radiance relations are usually determined in advance in the laboratory prior to the scanner becoming operational and, for aircraft scanners, are periodically recalibrated. In operation, the signal is either sent to a separate recording unit or is sampled from the main beam through a chopper. The radiant temperatures are not normally converted to kinetic temperatures because emissivities of the diverse surface materials are usually not sufficiently well know to permit this.

Now that you have warmed up to this subject, lets look at some notably hot images - unless you decide to cool it and go on to something else!

---

[an error occurred while processing this directive] [an error occurred while processing this directive]