Chile Astroclimate, a Biannual Update

Not long ago (The ESO Messenger 97, September 99), climate change was identified as the main responsible for the degradation of observing conditions (seeing) at Paranal. It was pointed out in particular that the weakening of the traditional westerly wind pattern was more frequently allowing turbulent air from inland to blow over the coastal cordillera.
Six months later and in spite of much wishful thinking, the site quality has only marginally improved and remains way below the standards established during the extensive site survey (dashed lines, fig. 1). This means for the observatory that Period 64 should not be better than Period 63 which provided sub-half arcsecond seeing only 13% of the time (R. Gilmozzi, The ESO Messenger 98, December 99, to be compared 21% in the period 1989-1995). During that same period, La Silla, which is not undergoing any visible climate change but is rather on a favorable phase of its own cycles, had been producing 8% of such good quality observing time and promises even more in Period 64.

Figure 1: Seeing Statistics at Paranal since UT1 first light: monthly average (red), median (black) and 5th percentile (green). The dashed lines give the respective long term (89-95) site characteristics. Seeing is reconstructed from DIMM measurements taken at 6 m above ground, at 0.5 microm and at zenith. Because of the finite outer scale of the atmospheric turbulence, actual large telescope image quality can be better that predicted by DIMM (see eg: The seeing at the William Herschel Telescope , R. W. Wilson et Al., MNRAS 309, 379-387, 1999)

It was reported (The ESO Messenger 90, December 1997) that cloudiness at Paranal was obviously increasing with warmer see water, ie: El Niño events. The dependency of Paranal seeing to El Niño cycles had been indeed similarly tested over a decade in the past (1988-1997) but without unveiling any correlation (yellow squares in fig.2). It was thus concluded that the basic Paranal observing conditions were weather independent. The seeing increase of the past 20 months (green squares in fig.2 corresponding to the period shown in fig.1) is mainly due to a particular North-East wind pattern which lasts part of the night, a few times per month. As shown in fig.2, all these poor months belong to the current La Niña and the seeing trend even shows some correlation with the standardized Southern Oscillation Index (SOI) which is commonly used to define the state of the Pacific ocean surface temperature.

Figure 2: Correlation of the standardized monthly Southern Oscillation Index (SOI) with monthly average seeing at Paranal during 1988-1997 (yellow) and since April 1998 (green). The SOI represents the sea level pressure anomaly between Darwin and Tahiti (http://www.cpc.noaa.gov/data/indices/). A negative index corresponds to warmer waters (El Niño), a positive index to cooler ones (La Niña).

The El Niño and La Niña cycles are hardly predictable and many past attempts failed. Some success was apparently obtained by a model based on solar activity cycles which correctly predicted the 1997-1998 El Niño event ( http://www.microtech.com.au/daly/sun-enso/sun-enso.htm ). If one can believe such models, the next El Niño event should arrive in 2002, perhaps bringing to an end the current phase of poorer than average astroclimate on Paranal.
Moreover, recent analyses of sea surface elevation measured by the Topex-Poseidon satellite (NASA/JPL News release, Jan. 20, 2000) lead researchers to suspect the Niño-Niña oscillations to sit on, and therefore partially hide, a much wider (20-30 years period) so-called Pacific decadal oscillation. If this phenomenon was confirmed and quantified, it would provide new pespectives to astroclimatological surveys, let us thus wait and see.

See also Seeing and El Niño (ESO Wide Review, April 2000).