In this paper, we propose a new simple method for obtaining fair-weather Atmospheric Electric Field (Potential Gradient, PG) data based on PG only. We design a filter called “two-dimension” (”2D”) median filter to reduce disturbed-weather PG noises rather than choosing fair-weather PG using conventional approaches. “2D” means that when filtering PG data, the filter considers PG from days that are close to the PG measurement time in addition to PG near the PG measuring time, as is the case with a standard one-dimension median filter. The open access GloCAEM dataset is used to assess this filter's efficacy. The results show that comparing to other simple filters, it has significant effect on reducing disturbed-weather PG noises, at the same time, has little effect on the fair-weather PG signals. Moreover, compared to conventional fair-weather PG selecting methods, PG after being filtered is very similar to fair-weather PG while do not have extra missing values in the PG. “2D” median filter can be an alternative method for obtaining mean diurnal variation of fair-weather PG. The PG after being filtered can also be used for researching other fair-weather related phenomena directly.
Read full abstract