A web-enabled expert system for quality control of data from an observing system

A primary concern of any observing system is its ability to produce accurate data products. Quality control (QC) procedures must be well defined and reliably executed during each operational phase including data acquisition, processing, and delivery. Issues that must be considered during the design and development of a data QC system include computational efficiency, the number of and type of QC measures performed, and remedial actions in response to problems. An expert system, which uses a set of defined rules and an inference engine to identify when an action must be taken, is one alternative for QC in observing systems.

Computer languages such as Jess (the Java Expert System Shell), developed at Sandia National Laboratories, should be useful for developing robust QC programs. Since Java-based applets are supported by Jess, a QC system built on this framework can be monitored and maintained via the internet. This paper describes the initial results of an investigation of the feasibility of implementing such a system for the NOAA/Forecast Systems Laboratory Ground-Based GPW Water Vapor Demonstration Network.