ESO in-kind Projekt

Overview

The purpose of SM-01 is to create an advanced sky model based on a data compilation, containing currently unconsidered and/or unknown physical properties. It is an extension and refinement of the existing DR05 sky model, including available updates of the incorporated external data and code packages. The sky spectrum model shall provide, over any requested wavelength range in the domain 300 nm to 28 μm, a sky spectrum with choosable resolution (R < 4x10⁷).

Status

Significant fractions of this project are already implemented (see Noll et al 2012) and build the framework for the current version of the ONLINE SKY MODEL. The Documentation can be found already in the draft for the ESO document.

Developments
The current sky model of DR05 contains several components: scattered moon light (Krisciunas et al. 1991), zodiacal light (Leinert et al. 1998), telescope emission (grey body) and, as the major component, emission by the Earth atmosphere. The model for the latter contains both thermal components, derived with radiative transfer codes and meteorological data (radiation/transmission, lower atmosphere), and a rudimentary model of chemi-luminescent airglow (upper atmosphere).
The project SM-01 will contain the following updates of the DR05 sky background model:

• Improved airglow model: The airglow lines and continuum are mainly observed in the optical and near-IR regime of the night sky spectrum. These phenomena emerge in upper atmospheric layers in a complicated chain of non-LTE photo-chemical reactions. The current airglow model of DR05 mainly consists of a list of airglow specific emission lines from molecules and atoms such as OH, O2, O, and Na, and their classifications in terms of variability. The new model will contain an improved and more consistent line list than the existing DR05 version, which is characterised by a mixing of observed (Hanuschik 2003) and theoretical line intensities (Rousselot et al. 2000). In addition, an enhanced variability and airglow continuum study on the basis of archival X-shooter data will be conducted. X-shooter is suitable for such a study since high-resolution, pure sky spectra covering the entire wavelength range relevant for airglow emission (U- to K-band) are frequently obtained for sky subtraction. The observed airglow spectrum depends on a large number of parameters like zenith distance, time after sun set, season, and solar activity. The effect of the latter on airglow emission during a solar minimum could not adequately be studied for DR05 (see Patat 2008 for the available data). Archival data taken in recent years will be a significant improvement. The investigation of airglow continuum, which is expected to have very low surface brightness (especially in the near-IR with reference to line emission), requires the highest possible quality of data reduction. A flux accuracy of 10-20% is desirable for the ETC application of the model. However, X-shooter shows significant artefacts emerging from internal stray light in the J- and H-band (Vernet et al. 2011). In the case that these data cannot easily be discarded, the assistance of the X-shooter instrument team will most likely be required to solve problems with the data reduction.
• Effective extinction curves for point-like and extended sources: Absorption and scattering in the lower atmosphere reduce and re-distribute the flux coming from external radiation sources such as moon, zodiacal light, and airglow. The effective, wavelength-dependent extinction depends on the line of sight and the intensity distribution of the source on the sky due to the scattering of light coming from other directions into line of sight. In the DR05 model such effects are only treated in a very simplistic way. Therefore, single-scattering radiative transfer calculations (see e.g. Wolstencroft & van Breda 1967 and Staude 1975) will be carried out for obtaining more reliable extinction estimates (see Figure and Noll et al. 2012). The findings of Patat et al. (2011) on the details of the Paranal extinction curve, and in particular, the results on Mie scattering by aerosols, will be applied. This will improve the DR05 model significantly.
  

  Geometry of scattering in the Earth's atmosphere. Point N is at top of atmosphere and, hence, not in the same plane as the other points. The azimuth angle of N as seen from X is A (not shown). On the other hand, X and T are at an azimuth angle A0 for an observer at O (see Noll et al. 2012 for more details).

  
  
• New meteorological data: Currently the incorporated meteorological input data of the DR05 and DR06 models are based on atmospheric standard profiles (MIPAS@ENVISAT satellite), profiles from the Global Data Assimilation System (GDAS), provided by the Air Resources Laboratory (ARL) of the National Oceanic and Atmospheric Administration (NOAA), and the ESO Meteo Monitor. The SM-01 project aims at a refinement of these data by incorporating data of the recently installed radiometer at Cerro Paranal and/or – if available – new satellite data from the European centre For Medium-Range Weather Forecast (ECMWF). This approach will help to better determine corresponding weather conditions of specific observations.
• New radiative transfer code versions: Radiance and transmission of Earth’s atmosphere are calculated with the help of the radiative transfer code packages Line-File/Line-byline- radiative-transfer-model (LNFL/LBLRTM, provided by Atmospheric and Environmental Research AER, US). The most recent code versions are LNFL V2.6/LBLRTM V12.1. Newer versions of the codes will be used if they provide significant improvements. Other than in DR05 the Reference Forward Model (RFM, Univ. of Oxford, UK) had been removed.
  
• New line database update: The basic line database for the radiative transfer codes is HITRAN V2008. Usually, this database undergoes a major update every four years. If a newer version is available before the acceptance release of SM-01 (31 January 2013), it will be included.
• New setup parameter to enable a user to choose a PWV value in the ETC. Currently, three possible ways for the implementation are discussed:
  • Scaling an average profile to the desired PWV value
  • Selecting the most matching profile from a database of profiles
  • Determining the PWV value from a set of spectra with DR06