P. A. Greet

Seasonal variations and inter-year trends in 7 years of hydroxyl airglow rotational temperatures at Davis station (69°S,78°E), Antarctica

Abstract Seasonal variation of hydroxyl rotational-temperatures above Davis, Antarctica (68.6°S, 78.0°E) is determined from 1148 nightly means collected over 7 years (1990 and 1995–2000). Measurements are limited by sunlight at this latitude to between day-of-year (DOY) 49 and 296. Sharp temperature transitions, particularly in autumn ( 1.2 K d −1 ), bracket an extended winter maximum with a slow temperature decrease ( −0.03 K d −1 from DOY 85 to 260). Seasonal variability and absolute magnitudes ( 206+/−4 K for a 1 July average across the years measured) are very similar to measurements at a comparable northern latitude (Lubken and von Zhan, 1991). A solar cycle association of 0.06 K ( sfu ) −1 ( 10.7 cm solar flux unit =10 −22 W m −2 Hz −1 ), implying a 7 K variation over a 120 sfu range solar cycle, fits the measured winter temperatures.

Further observations of the thermospheric vertical wind at the auroral oval/polar cap boundary

Abstract We present further observations of thermospheric winds and temperatures, derived from observations of the λ630 nm oxygen auroral/airglow emission (from ∼240 km altitude), obtained with a Fabry-Perot spectrometer (FPS) at Mawson station, Antarctica (L = 9). We report further instances of large upward zenith winds, with velocities up to ∼80ms−1, often associated with increases in temperature of up to 200 K. These upward winds are mostly seen around 21 Z, which is when Mawson station passes under the poleward edge of the (discrete) auroral oval during intervals of low to moderate geomagnetic activity, and are yet further examples of the type of event presented in an earlier paper (Innis et al., 1996). We also find intervals, up to several hours in length, when the temperatures measured South and East of the station can be several hundred degrees higher than those measured in the North and West directions, which we ascribe to the expansion of the hot polar cap into the South and East viewing directions. We have observed examples of seemingly anomalous wind measurements in the South and East directions during these times, which appear to be related to the presence of an upward wind in the observing volume of the FPS. Our observations suggest, however, that the upward wind would cause only a small perturbation (relative to the horizontal component) of the high-latitude thermospheric neutral wind field. The size of the zenith wind events (up to ∼120 m s−1) seen in our observations at Mawson are comparable with the amplitude of oscillations seen in gravity waves propagating at ∼300 km altitude over the polar cap from the nightside to the dayside, detected by Johnson et al. (1995) from Dynamics Explorer 2 observations. This similarity and the fact that we see the upward winds at the night-time auroral oval/polar cap boundary suggest that the origin of the upward winds may be intrinsically linked to the processes that generate these gravity waves.

Are polar cap gravity waves a heat source for the high‐latitude thermosphere?

Ground based measurements of thermospheric temperatures from Mawson, Antarctica, often show a marked spatial gradient, with data taken poleward of the discrete auroral oval up to 200 K or more warmer than measurements taken in or equatorward of the oval. This region of increased temperature is identified as the polar cap. We suggest a possible contribution to polar cap heating may be from thermospheric gravity waves.

A Theoretical Model Study of F-Region Response to High Latitude Neutral Wind Upwelling Events

Abstract Neutral wind upwelling events are well documented in both the northern and southern high latitude ionosphere. The vertical winds in the F-region frequently exceed 100 m / s , and winds in excess of 200 m / s have been observed. These upwelling events occur in a latitudinal band of 4°–6° width that extends in local time in the midnight-morning sector; this band always lies poleward of the auroral precipitation. Using the time dependent ionospheric model (TDIM), a series of sensitivity simulations are carried out, based on observational constraints provided by the spectacular upwelling event seen at both the Mawson and Davis stations in the Antarctic on 08 June 1997 (Innis et al., 1999). The model simulations indicate that the F-layer density at any given point may either increase or decrease during an upwelling event, depending upon the past history of the plasma flux tube. Because this past history of the F-layer convection is unknown for the specific upwelling events a detailed case study cannot be undertaken. Instead a series of sensitivity simulations based upon a range of possible convection histories will be studied to determine the relative effect of the upwelling. The absolute density is not dependent upon solar EUV production because of winter conditions, but is sensitive to the auroral electron precipitation. The best F-layer indicator of the upwelling is the height of the layer, hmF2. For upwelling events with vertical drifts of 100 m / s h m F 2 can be increased by 100 km in 10 min . Upon leaving an upwelling region, the hmF2 almost as rapidly decreases to its normal height. Resulting from this lifting of the O+ layer is the reduction in O+ recombination and 630-nm emission; this latter consequence is observed as a standard feature of the upwelling events. In the topside ionosphere the electron density is responsive to the upwelling. The total electron content (TEC) is not, in general, sensitive to the uplifting events, however, low elevation slant path GPS TEC measurements might well detect the rapid uplifting of the F-layer. The upwelling event observations are insufficient to constrain our understanding of their impact upon the ionosphere. This model study does imply that upwelling events can modify the F-layer height severely. Such layer height modification can have measurable effects on radio frequency ray paths through the ionosphere. To quantify such effects a fuller description of the upwelling events as well as the past history of ionospheric plasma is needed. Experiments with higher time resolution of both the neutral parameters and F-region at multiple locations are necessary to unravel these complex events.

Mesospheric rotational temperatures determined from the OH(6-2) emission above Adelaide, Australia

Abstract Tide-like variations observed in data from an OH spectrometer and MF radar, co-located at the Buckland Park field station (34°38′S, 138°29′E) near Adelaide, South Australia, over six nights in September 1993 are presented. A prominent semi-diurnal tide-like variation in temperature with a magnitude of up to 40 K peak-to-peak is observed in the OH data. Temperature and wind amplitudes and phases are compared with a semi-diurnal model, as are the relationships between the quantities. In addition, the observed temperature variations are compared with those of previous researchers. With regard to the absolute values of the amplitudes and phases, the model and observations differ. However, the observed relationships between the quantities, in terms of phase, are found to agree well with the model for four of the six individual nights, as well as for data that are the average of the six nights.

A comparison of optical and radar measurements of mesospheric winds and tides

Optical measurements of mesospheric winds by Fabry-Perot spectrometers, FPSs, at Mawson, 67.6°S 62.9°E, and Davis, 68.6°S 78.0°E, Antarctica are compared with similar measurements obtained using a spaced-antenna MF radar at Davis. The FPSs observed the OH emission. Different analysis procedures, used to determine the mean wind, and amplitude and phase of the semidiurnal tide, have been compared. At these latitudes the diurnal tide is weak and the semi-diurnal tide, although highly variable in amplitude, is usually the dominant periodicity. When comparing the amplitude and phase of the semidiurnal tide good agreement is obtained between measurements by the two instruments.

Tidal periodicities in observations of the OH(6-2) emission from Mawson, Antarctica

High-resolution Fabry-Perot spectrometer observations of the OH (6-2) Q1(1) line, λ834.460 nm, have been made at Mawson, Antarctica on a limited campaign basis since 1993. In August 1995, some data were obtained on 14 days in a 15 day period. Periodicities have been identified using the Lomb-Scargle technique. The most significant periodicity in the zonal wind and the meridional wind and intensity is at twelve hours. These results are compared with other measurements and a model of high latitude semidiurnal tidal oscillations.

Thermal behaviour of the Antarctic thermosphere observed from Mawson

Abstract The behaviour of thermospheric temperature in the region of Mawson, Antarctica during 1992–1996 is presented. Temperatures have been derived from Fabry-Perot spectrometer observations of the 630-nm OI emission. A distinct nighttime temperature gradient is observed along the magnetic meridian with hotter temperatures towards the pole. Equatorward of Mawson the temperature data have been fitted by an equation based on diurnal and annual sinusoidal variations, sunspot number and a quadratic dependence on Kp. Temperature differences between observed temperatures and the mathematical description have a mean of 10K and a standard deviation of 102K.

Analysis of high-resolution mesospheric sodium twilight spectral emission profiles

Abstract High-resolution Fabry-Perot spectrometers have been used for observations of mesospheric sodium spectral emissions excited by resonant scattering of sunlight. These observations are used to estimate winds and temperatures in the emission region. Hollow-cathode lamps have been used to define the rest wavelength for calculation of wind estimates from the Doppler shift of sky emission. Careful analysis of both lamp and sky spectra show the presence of self-absorption. The nature of the absorption is different for lamp and sky profiles, in the former case absorption occurs after emission, and in the latter absorption occurs in the incident radiation prior to emission. There is a further complication in the latter case, namely, the Doppler shift of the absorption is determined by the wind component in the direction of the Sun while the Doppler shift of the emission is determined by the wind component in the direction of viewing. The effects of these absorption processes on wind and temperature estimates will be discussed.

Argon auroral emissions

Abstract Argon is the third most abundant constituent of the atmosphere, but has not been previously detected by any ground-based optical remote sensing method. We report observation of argon emissions at λ840.82 and λ842.46 nm during intense aurora. These are most likely excited by direct electron impact. The maximum argon intensity observed is 270 R . Argon is the fifth elemental emission conclusively observed in geo-aurora, joining N, O, H and He.