Science data validation
The AIRS instrument suite measures infrared radiation from the atmosphere and turns these measurements into the quantities of temperature, humidity and cloudiness. But how do we know if the AIRS measurements are accurate? Validation attempts to answer this question.
The basic approach to validation is as simple as comparing the temperature measured by a thermometer with the temperature inferred by AIRS as it looks down on that thermometer. In practice the procedure is complicated by a need for special measuring equipment (often deployed in the dead of night), and careful analysis of the measurements. For example, AIRS can infer atmospheric conditions several miles above the earth's surface, so the thermometer must be on a balloon or aircraft. Moreover, that balloon must be in the part of the atmosphere being observed by the AIRS. Other complications arise from the different sampling characteristics of AIRS and the instruments within the atmosphere: AIRS infers conditions over a region about 50 km (30 mi), in diameter while a thermometer measures a point.
Because these ideal measurements are necessarily fewer in number, more standard measurements must be exploited. Standard measurements include operational weather balloons and surface weather stations. Standard measurements, while numerous, are often close to–but not exactly located with–what AIRS observes. In such cases, distinguishing actual differences from the natural variability of the atmosphere can be challenging. These are some of the reasons the validation activities are an important and challenging part of the AIRS science effort.
The validation plan devised before launch is available here: AIRS Science Data Validation Plan
Current validation states and uncertainty estimates for the latest AIRS Data Products are summarized in the About the Data section on this website.