92nd American Meteorological Society Annual Meeting (January 22-26, 2012)

Monday, 23 January 2012
Technological Advances for Testing for Systematic Errors On-Orbit
Hall E (New Orleans Convention Center )
John A. Dykema, Harvard Univ., Cambridge, MA; and J. G. anderson

This paper examines the technological basis and potential science returns of sensors that make measurements that are quantitatively tied on-orbit to international measurement standards. These measurement standards provide an objective basis for testing sensor calibration for systematic errors. In principle it is possible to build an instrument with the best technology available and characterize it exhaustively to quantify and adjust for systematic biases and potential long-term changes to instrument subsystems. In practice, the utilization of multiple calibration systems, with appropriate ties to recognized measurement standards, is potentially cheaper and facilitates a wider array of on-orbit empirical tests of instrument accuracy. To most efficiently realize maximum benefit from the redundant instrumentation approach requires a critical exposition of the underlying scientific logic. A suitable departure point is the examination of processes for assessing accuracy. Accuracy is assessed by testing against fundamental standards. The base units of the International System of Units (SI, from Système international d'unités) are an obvious choice because they are supported by an international research community and have a established quantitative relationship with fundamental physical constants such as the speed of light, Planck's constant, and the Boltzmann constant. For measurement of thermal infrared radiances, Planck's Law of blackbody radiation allows the reproduction of a very high accuracy calibration scale, limited only by the knowledge of the thermal and optical properties of the blackbody, and the viewing geometry. The uncertainty in the thermal properties may be dealt with via appeal to the International Temperature Scale of 1990 (ITS-90), which defines instrumentation and protocols for practical temperature measurements traceable with very low uncertainties to the Kelvin. The optical and geometric uncertainties must also be independently tested by some objective method. This requirement may be met by illuminating the blackbody calibrator in question with a known infrared light source, and measuring the radiant power that is reflected, according to the appropriate viewing geometry. A practical method for accomplishing this illumination and reflected light measurement is described here. This method utilizes a Quantum Cascade Laser, a compact, high-power mid-infrared source that produces narrow linewidth, continuous wave laser emission. A subsystem for realizing this on-orbit systematic error testing method for a space-borne infrared sensor has been developed and tested at a system level in a demonstration sensor. Results from this demonstration are presented. Together with a full complement of tests for other significant systematic errors, this method is a key component of a measurement strategy for obtaining demonstrably accurate on-orbit measurements with ties to international measurement standards. This measurement strategy provides several distinct benefits. First, because of the quantitative relationship between these international measurement standards and fundamental physical constants, measurements of this type accurately capture the true physical and chemical behavior of the climate system and are not subject to adjustment due to excluded measurement physics or instrumental artifacts. In addition, such measurements can be reproduced by scientists anywhere in the world, at any time, by appeal to the scientific literature and protocols supported by the international community of measurement scientists. This link to the international measurement community provides an established link to a rigorous body of knowledge for the assessment of measurement uncertainty, which is crucial to societal objectives of a quantitative nature. Finally, because enhanced quantitative weather and climate prediction directly serve decision support structures that embody critical societal objectives, the credibility of these predictions is paramount. Since the foundation of credibility of any scientific theory is the underlying observational evidence, the link between these measurements and internationally recognized measurement standards is a critical component in the portfolio of Earth observing systems. Clearly measurements that are unerringly compatible with fundamental physical and chemical relationships offer distinct advantages for supporting improved prediction.

Supplementary URL: