3 The GSM campaign at Cerro Paranal Observatory

The GSM instrument has been transported to Paranal after the completion of the previous mission in October 1998 at Cerro Pachon (Chile).

The excellent quality of Paranal site has permitted to collect 1884 acquisitions during 19 nights, more than in any previous GSM campaign.

Micro-thermal data acquisitions have been operational only during 9 nights.

3.1 Results of the campaign

Throughout this report the universal time (UT) is given. A date always refers to the beginning of the night. The data summary for the mission is presented in Table 1. In the first columns we give date, median speed and direction of the wind at 10 m altitude. Next 3 columns contain the nightly median values of main atmospheric parameters measured by GSM, namely seeing $\varepsilon$, outer scale ${\cal {L}}_0$and isoplanatic angle $\theta _0$. In the last column the seeing values computed by the ESO DIMM are listed.

During our mission the overall median seeing was slightly worse than long-term average for Paranal (0.70''), but typical for the whole 1998 year, with its monthly average seeing of 0.9'' - 1''. We note a very good stability of atmospheric parameters. Notably, in the period of December 1-5 surprisingly little variation of all main parameters occurred. Similar values were found again on December 15, 16 and 17.

The graphic representation of the ensemble of GSM data is given in Fig.  2.

3.2 Outer scale statistics

GSM is the first instrument capable to measure the outer scale ${\cal {L}}_0$ continuously and to study the statistics of this important parameter. During its previous missions it was found that ${\cal {L}}_0$ follows a log-normal distribution with a dispersion of log( ${\cal {L}}_0$) of about 0.23, and, rather similar median values, ranging from 24 to 31 m, were obtained in all GSM campaigns (Ziad et al. [1999]). Large isolated values of ${\cal {L}}_0$, called "bursts'', were also noted at all sites.

At Paranal we, for the first time, registered during several periods relatively small and stable outer scales. Our overall median ${\cal {L}}_0$= 22.0 m is smaller than at other sites studied.

In the plot of ${\cal {L}}_0$given in Fig. 2a we note the occurrence of two nights with relatively large median values of ${\cal {L}}_0$(45 and 48 m). On both these nights the wind of 6 m/s was blowing from South, which is not typical for Paranal.

During any single night there is no obvious systematic correlation of ${\cal {L}}_0$with seeing, even if a poor correlation appears sometimes.

The bursts of ${\cal {L}}_0$ exist at Paranal, as well as at other sites.

Curiously, a burst occurred at about the same time, 03h30 UT, during several nights. We believe that this effect is not an artefact, but reflects some real changes in the turbulence regime at Paranal site. Persistent local effects occurring in the SL at the same local time during several nights were also found (see below).

 

 

Table 1: Data summary

Date	Wind at 10 m		$\varepsilon$	${\cal {L}}_0$	$\theta _0$	$\varepsilon_{{\rm DIMM}}$
	m/s		''	m	''	''
27.11	7.3	E $\rightarrow$N	0.81	23.8	1.95	0.79
28.11	0.6	S $\rightarrow$W	0.67	15.6	2.45	0.97
29.11	6.9	SE (stable)	0.99	45.3	1.68	1.05
30.11	3.6	S $\rightarrow$W	1.02	21.7	1.90	1.00
1.12	4.1	S $\rightarrow$NW $\rightarrow$NE	0.90	16.7	1.78	0.85
2.12	5.4	N $\rightarrow$E $\rightarrow$N	0.88	16.9	1.75	0.85
3.12	4.8	E $\rightarrow$N	0.79	16.8	1.97	0.76
4.12	4.5	E $\rightarrow$N	0.81	17.0	2.17	0.69
5.12	5.2	E $\rightarrow$N	0.90	16.5	2.09	0.87
6.12	8.3	N $\rightarrow$NE	0.86	24.5	1.74	0.84
7.12	6.5	N $\rightarrow$NE	1.03	19.4	1.48	0.95
8.12	3.2	S $\rightarrow$N	0.90	16.9	1.51	0.80
14.12	6.3	S (stable)	0.92	48.4	1.95	0.95
15.12	3.4	SE $\rightarrow$N	0.95	21.5	2.22	0.80
16.12	3.1	SE $\rightarrow$N $\rightarrow$NE	0.89	17.5	2.54	0.96
17.12	6.5	E $\rightarrow$N	0.89	23.6	1.41	0.99
18.12	8.8	N $\rightarrow$NE	1.31	37.0	3.17	1.46
19.12	10.8	N (stable)	0.73	33.9	2.09	0.83
20.12	6.5	E $\rightarrow$N	1.29	23.7	1.41	1.30
Average median values			0.88	22.0	1.90	0.87