High resolution mesospheric sodium properties for adaptive optics applications

Abstract

Context. The performance of laser guide star adaptive optics (AO) systems for large optical and infrared telescopes is affected by variability of the sodium layer, located at altitudes between 80 and 120 km in the upper mesosphere and lower thermosphere. The abundance and density structure of the atomic sodium found in this region is subject to local and global weather effects, planetary and gravity waves and magnetic storms, and is variable on time scales down to tens of milliseconds, a range relevant to AO. Aims. It is therefore important to characterize the structure and dynamical evolution of the sodium region on small, as well as large spatial and temporal scales. Parameters of particular importance for AO are the mean sodium altitude, sodium layer width and the temporal power spectrum of the centroid altitude. Methods. We have conducted a three-year campaign employing a high-resolution lidar system installed on the 6-m Large Zenith Telescope (LZT) located near Vancouver, Canada. During this period, 112 nights of useful data were obtained. Results. The vertical density profile of atomic sodium shows remarkable structure and variability. Smooth Gaussian-shaped profiles rarely occur. Multiple internal layers are frequently found. These layers often have sharp lower edges, with scale heights of just a few hundred meters, and tend to drift downwards at a typical rate of one kilometer every two to three hours. Individual layers can persist for many hours, but their density and internal structure can be highly variable. Sporadic layers are seen reaching peak densities several times the average, often in just a few minutes. Coherent vertical oscillations are often found, typically extending over tens of kilometers in altitude. Regions of turbulence are evident and Kelvin-Helmholtz instability are sometimes seen. The mean value of the centroid altitude is found to be 90.8 ± 0.1 km. The sodium layer width was determined by computing the altitude range that contains a specified fraction of the returned sodium light. We find a mean value of 13.1 ± 0.3 km for the range containing 95% of the photons, with a maximum width of 21 km. The temporal power spectral density of fluctuations of the centroid altitude is well described by a power law having an index that ranges from −1.6 to −2.3 with a mean value of −1.87 ± 0.02. This is significantly steeper than the value of −5/3 that would be expected if the dynamics were dominated by Kolmogorov turbulence, indicating that other factors such as gravity waves play an important role. The amplitude of the power spectrum has a mean value of 34 +6 −5 m 2 Hz −1 at a frequency of 1 Hz, but ranges over two orders of magnitude. The annual means of the index and amplitude show a variation that is well beyond the calculated error range. Long-term global weather patterns may be responsible for this effect.

The Four-Laser Guide Star Facility: Design considerations and system implementation

“…The sodium dissipation rate is also strongly affected by the MLT temperatures, which are in the range of 180-200 K representing the coldest region in the atmosphere. Because the temperature of the MLT varies seasonally and with latitude, this lower boundary varies in extreme cases by 10 km, but usually lies between 80 km and 85 km [16]. Figure 1 shows an example of the sodium layer cross section as detected by a high-resolution lidar (light detection and ranging) system, located near Vancouver, Canada, at the University of British Columbia (UBC) Large Zenith Telescope [17].…”

Design of pre-optics for laser guide star wavefront sensor for the ELT

“…In a real system, they should be weighted according to the sodium intensity distribution model used (see Ref. [13] or [14], for example). The maintenance of the exit pupil position and size was one of the main boundary conditions during the optimization.…”

Monte Carlo modelling of multi-object adaptive optics performance on the European Extremely Large Telescope

“…It can be seen here that for larger sodium layer depths, it is favourable to have larger pixel scales. We have therefore selected 0.7 arcseconds per pixel as the default for this study, since performance remains good up to depths of 30 km which is a likely upper limit (Pfrommer & Hickson 2014). We note that a smaller pixel scale would lead to an improvement of a few percent in ensquared energy when sodium layer depth is small.…”