Uncertainty and Confidence of Ionospheric Specifications with the Digisonde 
ARTIST-5 Ionogram Autoscaler 
Ivan Galkin, Bodo Reinisch, Xueqin Huang, and Vadym Paznukhov 
 
The unsurpassed accuracy and detail of the ionospheric specification by means of 
HF radio sounding will keep ionosondes involved in space weather monitoring for 
years to come. Since the early 1960s many have tried to find solutions to one of 
the big ticket items in HF sounding, the natural intelligence needed to derive 
specification data from raw ionogram images. UMLCAR has made another step 
forward on this path with the new version 5 of the Automatic Real-Time Ionogram 
Scaler with True height (ARTIST). The new software is now released for the 
network of digisondes, which is currently used by such space weather agencies as 
AFWA, DIAS, JORN, and SWSA, for 24/7 real-time ionospheric monitoring at over 60 
locations worldwide. The ARTIST-5 software adds two new features to the ARTIST 
lineup whose heritage goes back to the 1980s, specification of the uncertainty 
bounds for the ionogram-derived data and the autoscaling confidence level (ACL). 
Risks involved in the carefree use of the ARTIST-derived data in operational 
ionospheric models have been recognized, and efforts to specify and mitigate 
the autoscaling errors commenced in the mid-1990s with the advent of the AFRL 
QUALSCAN automated quality control system. The new ARTIST-5 is able to self-
diagnose ionogram interpretation difficulties in addition to the commonly employed
sanity checks of the autoscaling outcome. Special effort has been made to describe 
uncertainty of the ionogram-derived data by a statistical study of automatic vs. 
manual processing of a representative set of ionograms. Combination of the ACL 
and uncertainty analyses allows better integration of the autoscaled data in 
operational models based on assimilation principles, e.g., GAIM. We discuss 
lessons learned from the ARTIST-5 evaluation.