An Estimation of Human‐Error Contributions to Historical Ionospheric Data

AbstractGround‐based radar sounders are used to characterize the dynamics and chemistry of Earth's upper atmosphere by using measurements of ionospheric peak electron density (NmF2) and its associated altitude (hmF2). Continuous sounder observations of the E and F regions of the ionosphere have been carried out regularly at dozens of stations worldwide since the midtwentieth century. A deep understanding of short‐ and long‐term upper atmospheric variability depends on a fundamental understanding of these observational data. The manual analysis of historical analog (predigital age) ionograms to derive the plasma frequency profiles and the ionospheric parameters hmF2 and NmF2 is a tedious procedure and susceptible to human error. In order to better understand this human error, a study is conducted in which ionograms from vertical sounders are manually scaled by a team of ionospheric researchers. The results of the study are then used to estimate the variability of the hand‐scaled ionospheric parameters foF2 and foE. Those results are then used to estimate the downstream impact on ionospheric models that use foF2 and foE as input. The results demonstrate that there can be large variability in the manual scaling of foF2 and fmaxE from vertical incidence ionograms. However, the participants did typically better than 5% uncertainty for benign ionograms. A long‐term analysis of hmF2 modeling exhibits low sensitivity to statistical errors imposed on foF2 and foE, but a short‐term analysis showed that modeled hmF2, neutral winds, and electron densities can be very different when small adjustments are made to foF2 and fmaxE.