Correspondence of variations of <i>AE</i> and <i>Apo</i> indices in 23–24 solar cycles

The auroral electrojet index AE is often used in forecasting models as a characteristic of a source of the disturbance propagation in the geosphere from the pole to middle and low latitudes. However, these data are no longer available digitally since January 2020. Instead of the AE−index, we suggest using the recently introduced 1 h Apo−index, given the close proximity of magnetometer networks for these indices at high latitudes and the availability of the Apo−index in real time. To this end their correlation is analyzed during 276 intense storms for 1995–2017. Storm profiles are constructed by method of superposed epoch with zero epoch time t0 = 0 taken at the threshold value of AE ≥ 1000 nT. A comparison is made of the storm profiles of AE(t), Apo(t), the interplanetary electric field E(t) and the solar wind speed Vsw(t) within 72 hours: 24 hours before the storm peak t0, and 48 hours after it. A good agreement is obtained between the sets AE(t) and Apo(t) with a correlation coefficient of 0.70. Comparison with the interplanetary parameters testifies on the correlation of AE(t) and Apo(t) with the electric field E(t) but absence of their coupling with the solar wind speed Vsw(t). A two−parametric formula is derived for dependence of the auroral electrojet index AE(t) on the interplanetary electric field E(t) and the geomagnetic Apo(t) index for the geomagnetic storm forecasting. In the absence of E(t) data, formulae for the dependence of AE(t) on Apo(t) is introduced for implementation in real time and the inverse dependence of Apo(t) on AE(t) for reconstruction of the 1 h Apo−index before 1995. Validation of the proposed models with data for 5 intense storms in 2018 has shown a close resemblance of the model with observation data of the AE−index with a high coefficient of determination R2 ranging from 0.62 to 0.81.