5 Comparison Oukaimeden/ORM

Table 6 presents both meteorological and photometric characteristics of the two sites. Both sites are in a subtropical zone of high pressure. Therefore, despite the difference in orographical surroundings, they provide similiar observational conditions. Also, at both sites, the temperature inversion layer is usually quite low. When this condition occurs, we occasionally notice the presence of a cloud layer below the site, there being a direct correlation between image quality and cloud height (Muñoz-Tuñón et al. 1997).

The mean temperature difference between the two sites where the altitude difference is approximately 374 m, is $1.9~^{\circ}$C, which is very close to the value of $1.8~^{\circ}$C predicted by Triplet & Roche (1986).

Figure 3 shows that the temperature at both sites decreases very slowly during the night, (by less than one degree), due to thermal cooling. The low surface layer contributes to excellent observing conditions at the ORM, already measured by Vernin & Muñoz-Tuñón (1994), leading us to predict the same excellent conditions at Oukaimeden.

The variation in temperature between two successive nights is approximately the same at both sites, i.e. $\sim 1.6~^{\circ}$C.

In Fig. 7, we show that relative humidity is higher during the winter months at Oukaimeden, and that the monthly mean values can differ by more than a few percentage points between the two sites. Thus the relative humidity variation at night is much higher at Oukaimeden.

We can observe a correlation between temperature and relative humidity: the warmest summer months (typically June and August) are marked by extremely low relative humidity, whereas the coldest winter months (from October to April) have the highest relative humidity and lowest temperature.

Wind statistics show the prevalence of low wind speeds at both sites. The nocturnal data show that mean wind velocity is less than 3 m s^-1, so that the OUK site may be considered to be suitable for astronomy. The night-time and daytime wind speeds at both sites are lower than 15 m s^-1 for 99% of the time.

There is a change in the wind roses for day- and night-time at each site and no obvious comparison can be made.

Due to the presence of a semi-permanent anticyclone, which follows the change in declination of the Sun (Font Tullot 1956), the trade winds are predominant in the Canary Islands. In the summer, when the anticyclone extends to Morocco, the weather becomes dry and hot. However, low pressure zones in the southern part of the country often cause local storms. In the winter the Azores anticyclone moves towards south. At the same time, depressions are formed in the Northern Atlantic and moves eastwards. The weather in these conditions becomes humid and cold (Benkhaldoun 1994).

Because no sky quality measurements at Oukaimeden were available, we tried to compare the extinction trend at both sites by using daytime data at Oukaimeden and night-time data at the ORM.

Figure 15: Extinction measurements from 01/10/1988 to 30/04/1991 at Oukaimeden (left) and ORM (right). The increase in atmospheric extinction appears clearly in summer 1989 and 1990 at both sites

We selected a range of CAMC data (1988 October to 1991 April) that coincided with measurements made at Oukaimeden. We compared 620 extinction values at Oukaimeden and 511 at the ORM (Fig. 15), during a period which was not affected by volcanic eruptions (Guerrero et al. 1998). The mode and median values of the atmospheric extinction coefficient at ORM are equal to 0.12 mag/airmass. These results confirm the excellent sky transparency above the ORM and converge toward those found by Guerrero et al. (cf. Sect. 4.5). At Oukaimeden the mode value is 0.10 mag/airmass and the median value is 0.12 mag/air mass, which equal Hill's findings (1994a,b).

We can see a clear seasonal effect at both sites with an increase in the extinction coefficient in the summer. The occasional irruption of spectrally grey Saharan dust (see Stickland et al. 1987; Whittet et al. 1987) leads to a temporary increase in the extinction coefficient. This effect is moderate at Oukaimeden due to the presence of the intervening Atlas Mountains, whose altitude exceeds 4000 m, which shield the site from dust and wind.

Figure 16 shows a subset of days for which simultaneous measurements were made. The data are reduced to 360 extinction values and there is no correlation between the two sites. The median is about 0.12 mag/airmass at both sites, the mode is 0.12 mag/airmass at ORM and 0.10 mag/airmass at Oukaimeden. Even if only 360 extinctions are analysed, daytime condition at Oukaimeden are comparable to night-time conditions at ORM.

These values compare favourably to previous observations made at other sites, e.g. 0.15 mag/air mass for Mt. Wilson (Schmidt-Kaler 1982) and 0.124 mag/air mass for La Silla (Rufener 1986).

Figure 16: Simultaneous extinction measurements at both sites from 01/10/1988 to 30/04/1991. The Oukaimeden data is displayed in X axis and ORM data in Y axis