Nine months' statistics of seeing at the ORM, at two different sites, are presented. Measurements were taken on a 5-m tower to avoid spurious turbulence produced by the surface layer and with cross-calibrated seeing monitors. From the result of this monitoring, taken at a site near the location of the TNG with the DA/IAC DIMM, the excellence of the site, in terms of image quality is confirmed. Median seeing value of 0.64'' was achieved over the whole period, with seeing values below 0.3'' during 7% of the observing hours and minimum values down to 0.17''. The total percentage of observed time with the DIMM is 72%, well in agreement, within the limitations of operation of our experiment, with the 79% that is reported by Carlsberg Meridian Circle, over several years.
Our routine observations combined with long temporal series
meteorological data from the literature, allow us to identify
June to September as the best months
coinciding with the period of the year in which
the strength of the temperature inversion layer is larger, due to the
prevailing trade-wind regime. During this period, the height
of the inversion layer (
1200 m) is also lower. The existence of a
seasonal dependence of seeing is then confirmed, with median
values from 0.54'' in summer and 0.77'' during the rest of the year, most
probably associated with the general climatology of the Canary Islands. Within
this best period, it is worth mentioning the statistics corresponding to the
month of June, with 0.46'' median seeing value (0.17'' rms).
Although during the last years a very big effort has been made in site characterization programmes, little long-term climatology is available to compare with the data presented here. Very good quality sites as Hawaii and Mount Graham, for instance, have never been checked routinely using cross-calibrated seeing monitors. The largest database has been established by ESO at La Silla and Paranal, developing a very extensive and reliable seeing monitoring. From the comparison between ESO (Sarazin 1995) and our data, presented in this paper, the three places, La Silla, Paranal and ORM are excellent, ORM presenting better behaviour, with the best median values, and a slightly larger percentage of seeing values below 0.3''. The parameters corresponding to the best month are also better.
The comparison of two different areas at the ORM shows almost the same trend although the TNG slope is slightly better than the WHT slope during the summer period. This fact might be attributed to local disturbances of the WHT dome and terrace. Hence, no obvious change in the spatial localization of the surface-layer turbulence has been detected.
We have analysed the the temporal variation of the seeing in order to have a better insight into the processes which give rise to the optical turbulence. We found that during relatively stable nights, with about 0.5'' mean typical values, the seeing can deteriorate abruptly over short periods. This can be interpreted by saying that even under good seeing conditions, and hence quasi-laminar flow, the steady equilibrium can be broken by a perturbation and degenerates into turbulence within a few minutes. This process is not reversible, and therefore we never observed a rapid seeing improvement. Another pattern that may appear sporadically is an oscillatory behaviour in the seeing temporal evolution during a given night. This periodical variation has a typical time scale of 45 min, which falls within the buoyancy range of gravity waves. We computed an averaged autocorrelation function over all the available data during the nine months. No oscillations seem to remain in the correlation, and therefore a large spectrum of gravity waves is present in the atmosphere.
From the analysis of this autocorrelation function it is concluded a 1.2-hour exponential decay time of the optical turbulence.
Since October 1996, a site-testing programme led by the IAC to select the site for the future GTC has been under way. Data are taken simultaneously at two sites at the ORM. At each site we have installed a DIMM (5 m above the ground) and a 12-m meteorological mast. These will provide a large and updated database on both optical quality and meteorology at the ORM. We will then be able to begin a study the relationship between seeing and meteorology.
Acknowledgements
This work has been funded by the Instituto de Astrofısica de Canarias and the programme for collaborative research between France and Spain (Acciones Integradas 1990, 1991, 1992). We are grateful to Sergio Ortolani and Dario Mancini from the Galileo group, who provided the 5-m tower, as well as the design of the foldable dome to protect the DIMM. Also their support and comments are acknowledged. Our observers, Amadeo Jorge Martín and Carlos González Suárez, have performed a very good job and their dedication is truly appreciated. The Site characterization project was promoted by Francisco Sanchez, to whom we are grateful also for valuable comments on this paper. Communication with the Isaac Newton Group on this topic has always been very good, and we are grateful to Mike Breare and Steve Unger for making it possible. We would also like to thank the staff of the ORM for ther help and in particular, Mary Barreto, the then Site Administrator of the ORM,
for making things run smoothly. We thank Terry Mahoney for correcting the English of this paper. Many thanks are also due to our referee, Marc Sarazin, for his prompt revision of the paper.