PMSE-observations with the EISCAT VHF and UHF-radars: Statistical properties

Abstract

Observations of polar mesosphere summer echoes (PMSE) have been carried out with the EISCAT VHF radar during five summer seasons between 2003 and 2007 as well as with the EISCAT UHF radar in July 2004 and 2005. From a total of 523 h of observations with the VHF radar, PMSE have been detected for more than 300 h overall. In July 2004 and 2005 when PMSE were detected by both radars, the occurrence rate of the VHF PMSE is as large as 85.0% whereas the UHF PMSE are much rarer with an occurrence rate of 5.3%. Comparison of the year-to-year variation between the occurrence rate of PMSE at 224 MHz and noctilucent clouds (NLC) observed with a nearby lidar shows similar properties. The year-to-year variation of the occurrence rate of PMSE at 53.5 MHz, however, is quite different. This might imply that PMSE at 224 MHz require the presence of larger ice particles than at 53.5 MHz, and hence show similar behavior as NLC. In contrast, the correlations of the occurrence rate of PMSE at 224 MHz with the solar and geomagnetic activities both show positive correlations which are, however less significant compared with the PMSE at 53.5 MHz. In addition, the occurrence rate and volume reflectivities of PMSE derived from the observations at 224 MHz both show remarkable characteristics of seasonal variation and diurnal variation. As expected from previous PMSE observations at 53.5 MHz, the seasonal variation is determined by the condition of supersaturation with respect to ice. The diurnal variations, however, are much more complicated, as they can be significantly influenced by the temperature changes due to tidal activity as well as the ionization level due to the solar radiation and high-energy particle precipitation. Taking into account the ‘turbulence with large Schmidt number’-theory of PMSE, a comparison between the volume reflectivities of simultaneous VHF PMSE and UHF PMSE is carried out and shows 99% of the observations with reflectivity ratios larger than a minimum value predicted by theory. Last but not least, we considered electron densities derived from UHF incoherent scatter observations at altitudes above the PMSE-layers (i.e., 88 and 90 km) during cases with VHF PMSE with and without UHF PMSE as well as a comparison of spectral widths from observations with both radars. This reveals that the occurrence of UHF-PMSE most likely requires larger electron densities and larger spectral widths (corresponding to enhanced neutral turbulence) at altitudes above 84 km.