The Use of Modeling for Site Surveys

1
The Use of Modeling for Site Surveys

• Marc Sarazin
• European Southern Observatory

2
The Ideal Sites of the Past

• In the 40s
• On the national territory
• Close to a major educational center
• Cloud free area
• In the 60s
• Cloud free area
• Good seeing
• In the 80s
• Cloud free area (higher operation cost)
• Good seeing (better optics)
• Low precipitable water vapor (IR)

3
The Ideal Sites of the Present

• Common Trends
• Telescopes and instruments are seeing limited
• Observatories operate in flexible scheduling
  modes
• Observatories are equipped with seeing monitoring
  facilities
• Observatories are equipped with cloud monitoring
  facilities
• Observatories receive meteorological forecasts
• Basic Site Requirements
• Cloud free area (spectroscopic sky gt 80 of the
  time)
• Good seeing (median lt1 at 0.5 micron)

4
The Ideal Sites of the Present

• Requirements by Observing Technique
• Photometry and Spectroscopy
• Photometric sky (stable and isotropic extinction
  coefficients)
• Low light pollution (evolution controlled by
  state laws)
• Good seeing
• Classical Imaging
• Good seeing and slowly varying
• Imaging assisted by classical NGS adaptive optics
• Good seeing (nr. actuators) and low wavefront
  velocity
• Low high altitude turbulence (large isoplanatic
  angle)

5
The Ideal Sites of the (near) Future

• Requirements by Observing Technique
• Observations assisted by classical LGS adaptive
  optics
• Same as above
• Photometric sky (photon return of sodium laser
  guide star)
• Observations assisted by multi-conjugate LGS
  adaptive optics
• Same as above
• The increase in the corrected field of view
  scales as the isoplanatic angle

6
The Ideal Sites of the (near) Future

Have all good sites been discovered? Can a
continental site provide excellent seeing
conditions? Comparison of the seeing
distribution at Maidanak (Uzbekistan) with
conditions at ESO Observatories (Ehgamberdiev et
al., AA suppl, Aug. 2000)
7
The new parameters for site surveys

The wind velocity at the tropopause is a good
indicator of the global isoplanatic
angle ESO-Paranal DIMM isoplanatic angle
compared to ECWMF 200mB wind during 9
months. (correction for exposure time averaging
of scintillation after A. Tokovinin, ESO internal
memo, July 2000)
8
The new parameters for Site Surveys

The performance of AO systems is ultimately
determined by the real time vertical distribution
of the turbulence Gemini Web site Scidar
profiles at Mauna Kea (F. Roddier et al. SPIE Vol
1236, 485, 1990)
9
The new parameters for site surveys

Climate Change is a reality also for the
astronomical seeing
Oscillations with periods from a few months to a
few decades
10
The new parameters for site surveys

Pacific decadal oscillation, El Nino, global
warming?
Surveying a site for a few years long is not
safe climate change must be taken into account
11
The new parameters for site surveys

The apparent magnitude of Sodium Laser Guide
Stars is naturally variable in time. It also
depends on the atmospheric transmission (A.
Quirrenbach, session VII) Variation of the
sodium profile through the night of August 31,
2000 at La Palma. Source Dainty et al., Imperial
College http//op.ph.ic.ac.uk/research/
12
The Ideal Sites of the Future

• Additional Requirements
• Low Seismic Activity
• An ELT cannot be made stiff enough to survive
  earthquakes
• Low Wind at ground level
• An ELT is more sensitive to wind shake during
  tracking
• Crystal Clear Skies
• Laser guide star efficiency depends on
  atmospheric extinction

13
The new tools for site surveys

Coarse Cloudiness Maps are readily available. A
few km resolution is accessible (A. Erasmus,
session V) Source Surface Meteorology and Solar
Energy Data Set (SSE) of NASA's Earth Science
Enterprise Program (1x1 degree grid).
14
The new tools for site surveys

Peak Ground Acceleration up to 5m/s2 10
probability of exceedance in 50 years
Source http//www.seismo.ethz.ch/GSHAP/
15
The new tools for site surveys

Global Models The wind at 700mb is not
accurately represented over mountains
ECMWF analysis over northern Chile, 60km (0.5
degree) grid.
16
The new tools for site surveys

Global Models The wind at 500mb is accurately
represented
ECMWF analysis over northern Chile, 60km (0.5
degree) grid.
17
The new tools for site surveys

Global Models The relative humidity is hard to
model accurately
ECMWF analysis over northern Chile
18
The new tools for site surveys

Global Models Seasonal Variations are accurately
represented
ECMWF analysis over the VLT Observatory Paranal
in 1993
19
The new tools for site surveys

Mesoscale models reveal local orographic effects
on the flow The terrain model, covers 120x120 km
with 500 m resolution. The initialization is
done at one corner with the grid point values of
a global model
First attempts to model the seeing at ESO Paranal
Observatory using Meteo-France Meso-Nh model
(Masciadri et al., 1997)
20
The new tools for site surveys

Non-hydrostatic models can account for local
orographic effects on the potential temperature
because they allow vertical motion
First attempts to model the seeing at ESO Paranal
Observatory using Meteo-France Meso-Nh
model (Masciadri et al., 1997)
21
The new tools for site surveys

Vertical cut of the turbulence Cn2 along the E-W
direction centered on Paranal
First attempts to model the seeing at ESO Paranal
Observatory using Meteo-France Meso-Nh
model (Masciadri et al., 1997)
22
The new tools for site surveys

First attempts to model the seeing at ESO Paranal
Observatory using Meteo-France Meso-Nh
model (Masciadri et al., 1997) Map of the
integrated seeing in an area centered on Paranal
23
The new tools for site surveys

MM5, a mesoscale model is available as freeware.
It is used at the Mauna Kea Weather Center
(http//hokukea.soest.hawaii.edu/forecast/mko/) to
produce vertical profiles of the
turbulence. Comparison of MM5 profiles above
Mauna Kea Observatory with in situ SCIDAR
observations