The Second Global Ozone Monitoring Experiment (GOME-2) will perform operational global monitoring of ozone column densities and profiles, and column densities of other atmospheric trace gases such as NO2, BrO, OClO, HCHO, SO2 and H2O. GOME-2 is an improved version of the Global Ozone Monitoring Experiment (GOME-1) launched in 1995 onboard the second European Remote Sensing Satellite (ERS-2) It will be embarked on the MetOp series of three polar-orbiting operational meteorological satellites, to be launched in 2006, 2010, and 2014. Although GOME-2 is considered a "recurrent" instrument the basic design of GOME-2 being the same as that of GOME-1 a number of technical improvements were made based on the experience with GOME-1 operations and data analysis and in response to more stringent user requirements. Particular improvements concern spatial resolution, polarisation measurements, and calibration. Level 0 to 1 processing will take place in the Core Ground Segment (CGS) at EUMETSAT while level 1 to 2 processing will be performed by the partner institutes of the Ozone Monitoring Satellite Application Facility (O3MSAF). This paper presents the GOME-2 instrument characteristics and the main improvements as compared to GOME-1. Instrument calibration aspects are discussed, covering on-ground calibration results, in-orbit verification during the first weeks after launch, and the scenario for routine in-orbit calibrations. An overview of the on-ground processing of GOME-2 data from raw instrument packets (level 0) via calibrated (ir)radiances (level 1) to geophysical data (level 2) is presented. As GOME-2 data will provide long time series of a number of trace gases, GOME-2 instrument and processing performance monitoring is also discussed. Additionally the post-launch GOME-2 calibration and validation activities, planned to be carried out centrally at EUMETSAT will be described. This includes real-time verification and quality checking of the GOME-2 level 1 products, in addition to full validation activities taking into account feedback from the retrieval of geophysical parameters. 1 . INSTRUMENT OVERVIEW The GOME-2 instrument is a medium resolution nadir-scanning UV/visible spectrometer. It comprises four main optical channels which focus the spectrum onto linear silicon photodioide arrays of 1024 pixels each, and two Polarisation Measurements Devices (PMDs) containing the same type of arrays for measurement of linearly polarised intensity in two perpendicular directions. The four main channels provide continuous spectral coverage of the wavelengths between 240 and 790 nm with a spectral resolution (FWHM) between 0.26nm and 0.51nm. Compared to the main channels the PMD measurements are performed at higher spatial but lower spectral resolution, comprising 15 programmable spectral bands per PMD channel. The PMD channels are designed such that maximum similarity of their optical properties is ensured. The wavelength dependent dispersion of the prisms causes a much higher spectral resolution in the ultraviolet that in the red part of the spectrum. In order to calculate the transmission of the atmosphere which contains the relevant information on trace gas concentration, the solar radiation incident on the atmosphere must be known. For this measurement a solar viewing port is located on the flight direction side of the instrument. When this port is opened, sunlight is directed via a ~40° incidence mirror to a diffuser plate. Light scattered from this plate, or in general, light from other calibration sources such as the Spectral Light Source (SLS or HCl) for wavelength calibration, and the White Light Source (WLS) for etalon (and optionally pixel-to-pixel gain) calibration are directed to the scan mirror using auxiliary optics. Diffuser reflectivity can be monitored internally using light from the SLS. All internal calibration sources with their optics are assembled in a subsystem called the calibration unit. The only exceptions are the light emitting diodes (LEDs) which are located in front of the detectors to monitor the pixel-to-pixel gain. For more information on the GOME-2 instrument see [1]. The default swath width of the GOME-2 scan is 1920km which enables global coverage of the Earth's surface within 1.5 days (note that other swath widths are also commandable). The scan mirror speed can be adjusted such that, despite the projection effect, the ground is scanned at constant speed. The along-track dimension of the instantaneous field of view (IFOV) is ~40km which is matched with the spacecraft velocity such that each scan closely follows the ground coverage of the previous one. The IFOV across-track dimension is ~4km. The actual integration time used (and thus the ground-pixel size) will depend on the light intensity. The integration time can be separately set for each channel; in channel 1 and 2 it is possible to sub-divide the channel in two parts (called 'band 1a', 'band 1b' and 'band 2a, 'band 2b' respectively) having separate integration times. It is anticipated that a default integration time of 0.1875ms will be used in all channels with two exceptions where longer integration times are needed because of low light intensity: (i) Band 1a has a default integration time of 1.5 seconds (yielding three spectra per scan and one from the fly-back with the possibility of co-adding spectra to improve signal to noise characteristics). (ii) The integration time for all channels will be increased for low solar elevations (high solar elevation angles). For the 1920km swath, the maximum temporal resolution of 187.5ms for the main channels (23.4ms for the PMD channels) corresponds to a maximum ground pixel resolution (across-track x along-track) of 80km x 40km (10km x 40km for the PMDs) in the forward scan. A summary of the GOME-2 instrument characteristics and the main improvements as compared to GOME/ERS-2 is given in Table 1. Table 1. Summary of GOME-2 instrument characteristics and the main improvements as compared to GOME/ERS-2 (shown in red). Principle Nadir-scanning UV/VIS grating spectrometer Wavelength Range 240 – 790 nm in 4 channels 300 – 800 nm in 2 polarisation channels (s/p) Detectors 1024 element Reticon linear diode arrays Readout time 46.875 ms (complete array) (/2) Spectral sampling 0.12 – 0.21 nm (main channels) Spectral resolution FWHM 0.26 – 0.51 nm (main channels) Swath width Default 1920 km (*2) Swath type Earth-curvature compensating Min effective IT 187.5 ms (/8) Spatial resolution Default 80x40 km2 (/4) Internal calibration LED, Spectral lamp (PtCrNeAr), White lamp Sun diffuser Quartz quasi-volume Data rate 400 kbits/s or 300 MB/orbit (*10) 2 . ON-GROUND CALIBRATION AND CHARACTERISATION The GOME-2 instrument was built by an industrial team lead by Galileo Avionica (I) with support from Laben (I), TNOTPD (NL), Arcom Space (DK), Innoware (DK) and Finavitec (FIN) where TNO-TPD were responsible for the calibration and characterisation of the instrument. Calibration and characterisation measurements, needed to meet the accuracy requirements for measurements made in thermal vacuum and under ambient conditions, were taken during an extensive onground campaign. The detailed characterisation measurements are fully described in [2] and [3]. Characterisation measurements have been post-processed to provide Calibration Key Data files which are documented in terms of both content and format in [4], [5] and [6]. Verification of the Calibration Key Data was addressed during dedicated Calibration Results Reviews ([7], [8], and [9]) for each flight model. Verification of the on-ground instrument performance against instrument requirements was also carried out at this time. A sub-set of the Calibration Key Data are a required input to the GOME-2 level 0 to 1 processor e.g. the radiance, irradiance and polarisation response of the instrument. For a full list of those Key Data used by the GOME-2 level 0 to 1b processing chain see [10]. Other Key Data describe aspects of the on-ground behaviour of the instrument which will also be measured in-orbit using on-board calibration targets e.g. dark signal performance, pixel-to-pixel gain, spectral calibration, and etalon. For these aspects the Calibration Key data form the starting point for instrument monitoring activities. Additionally, the GOME-2 Error Assessment Study ([11] & [12]) showed for the first time that O3 profile retrieval is very sensitive to knowledge of the shape of the slit-function in the wavelength interval of the ozone Huggins bands (320-340nm), which is used for retrieval at low altitudes. It was therefore concluded that, for height-resolved O3 data products from GOME-2 to meet specified User Requirements ([13] & [14]) the slit-function shape must be characterised at sub-pixel resolution pre-flight, as this cannot be determined adequately from information available in-flight. As a result additional slit function characterisation data were acquired during the on-ground calibration and characterisation campaign [15] and further analysed by TNO-TPD and the Rutherford Appleton Laboratory to provide additional Calibration Key data that describe the slit-function shape at sub-pixel resolution, for use in level 1 to 2 processing [16]. Trace gas absorption spectra measurements have also been made with each flight model after completion of the other on-ground calibration and characterisation activities. This was a dedicated activity carried out by the University of Bremen with the support of TNO-TPD. In particular, absorption spectra of O3, NO2 and O2 were measured in the wavelength region 230-800nm for a range of temperatures. These data will also be made available for use in level 1 to 2 processing. 3 . GOME-2 TIMELINES The GOME-2 instrument may be operated using timelines (GTL). Timelines are used primarily to reduce the load on the satellite up-link and additionally to provide on-board autonomy. One GTL is pre-loaded as a series of up to thirty three indivi