GOME-2 Products Guide - Processing Algorithms

7.1 Algorithmic functions

This section briefly describes the major algorithmic functions of the GOME-2 Level 0 to 1b processor. Only algorithmic functions are described. Those related to data handling are excluded. Functions consisting of a Calculate Correction and an Apply Correction part (as shown in the previous functional decomposition diagrams) will be described together. Module numbers used in the GOME-2 Level 1 Product Generation Specification [RD7] (where exhaustive information, including variable lists, full algorithm descriptions and data handling modules can be found) are provided.

Functions will be presented in six groups: initialisation and preprocessing, geolocation, in-flight calibration parameters, scene-dependent corrections, absolute radiometric calibration, and quality flagging. For each function the type of data affected is indicated: radiometric (i.e., "signals"), spectral (i.e., "wavelengths"), housekeeping data, geolocation, quality flags, or other. Several functions affect more than one of these types.

7.1.1 Initialisation and pre-processing

Initialise orbit propagator			Geolocation
	Modes	Calibration key data / MME	PGS reference
Calculate from	N/A	N/A	AG.1
The orbit propagator function is initialised by calling it in its initialisation mode. A Cartesian orbit state vector in Earth-fixed coordinates close to Ascending Node Crossing (ANX) is required on input. UTC and mean Kepler elements at ANX are provided as output. This function is executed once at PPF start-up.

Preprocess Müller matrix elements			Radiometric
	Modes	Calibration key data / MME	PGS reference
Calculate from	N/A	Radiance response Irradiance response Polarisation sensitivity	A2.1 (AG.2)
Calibration key data on radiance, irradiance and polarisation response are converted to a representation as Müller matrix elements M. In this formalism, used throughout [RD7], elements M1 represent the radiometric response, M2 the s/p polarisation sensitivity, and M3 the +45°/-45° polarisation sensitivity. Conversion is performed to a wavelength and angular grid defined as initialisation parameters. This function is executed once at PPF start-up.

Receive and validate Level 0 data			Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	All	N/A	A1.1
Basic integrity checks are performed on the incoming Level 0 data. The GOME-2 Level 0 data flow is checked in three steps, of which the first two are related to the integrity of individual packets, and the last one to the integrity of the sequence of packets. GOME-2 data packets are identified in the data flow via their fixed fields. The length of the data packet is checked. For each packet, the checksum is recalculated and compared against the checksum contained in the packet. A basic check for duplicate packets and the time order of the packets is performed. In the event of problems, error messages are issued. The normal quality flagging mechanism via PCDs cannot be used here because corrupted packets have to be excluded from further processing. This function is executed for every scan.

Determine observation mode and viewing angles			Housekeeping Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	All	N/A	A2.3
Data from different observation modes will be sent to different branches of the processing. This function derives the observation mode for a scan from a combination of housekeeping data and the scanner viewing angles. The viewing angles themselves are calculated from the scan mirror readings in the data packet. Scans which do not match any of the available observation modes are classified as "invalid". Furthermore, both the PMD transfer and readout mode are determined, and the UTC times corresponding to the scan mirror positions are calculated. This function is executed for every scan.

Convert housekeeping data			Housekeeping Quality
	Modes	Calibration key data / MME	PGS reference
Apply to	All	N/A	AG.3
Selected housekeeping data, e.g., temperatures, lamp currents and voltages, are converted from raw instrument binary units into physical units. The conversion uses a polynomial expansion which is usually linear (two coefficients). Polynomial coefficients may vary between models and are defined as initialisation parameters. Converted temperatures, voltages and currents are checked against thresholds defined as initialisation parameters and quality flags are raised as needed. This function is executed for every scan.

Prepare PMD data			Radiometric
	Modes	Calibration key data / MME	PGS reference
Apply to	All	N/A	AG.4
For each PMD band, the GOME-2 on-board processor co-adds individual detector readouts (16-bit values) into a band sum, then divides the sum by a scale factor such that it fits again into a 16-bit value. Therefore, PMD signals in binary units have to be reconstructed from the scaled values and scale factors found in the Science Data Packet which is the task of this function. This function is executed for every scan containing PMD data in band transfer mode.

7.1.2 Geolocation

Calculate geolocation for fixed grid			Geolocation
	Modes	Calibration key data / MME	PGS reference
Calculate from	All	Instantaneous field of view	A2.6
A number of geolocation parameters are calculated from an orbit state vector, the UTC, and scanner viewing angles. Most geolocation parameters are calculated in granules of the shortest effective integration time for the main channels (187.5 ms, 32 times per scan), independent of the actual integration time in the main channels. The remaining geolocation parameters are calculated once per scan, either because their variation within a scan is small, or because they are not needed on a finer grid by higher-level processing. Basic geolocation parameters such as the sub-satellite point and solar angles at the satellite are calculated for all measurement modes. In addition, mode-specific geolocation parameters are calculated for Earth, sun and moon modes: Earth mode: solar and line-of-sight zenith and azimuth angles at a given height, corner and centre coordinates (latitude/longitude) of the ground pixel at ground level, satellite height and Earth radius. Sun mode: distance between satellite and sun, and relative speed of satellite and sun. Moon mode: Lunar elevation and azimuth angles, sun-moon distance, satellite-moon distance, lunar phase angle, illuminated fraction of lunar disc.

Calculate geolocation for actual integration times			Geolocation
	Modes	Calibration key data / MME	PGS reference
Calculate from	Earth	N/A	A3.2 (AG.19)
Geolocation parameters for the actual integration times in a scan are calculated from the geolocation parameters for the fixed 187.5 ms grid. This is done only for integration times resulting in "simple" ground pixel shapes (below 6 s: no change in scanning direction during integration; 6 s and higher: only full scans). The synchronisation between main channel readouts and 187.5 ms ground pixels is taken into account. Geolocations refer to the first detector pixel read in a main channel.

7.1.3 In-flight calibration parameters

A number of calibration parameters can be derived from the dedicated in-flight calibration measurements described in In-Flight Characterisation and Calibration. These are: dark signal, pixel-to-pixel gain (PPG) correction, spectral calibration parameters, Etalon correction, and the Sun Mean Reference spectrum. The frequency of in-flight calibrations varies between once per orbit (e.g. dark signal) and once per month (e.g. PPG correction). Corrections are generally sufficiently stable to be valid for more than one orbit.

In-flight calibration parameters are not immediately used or stored in a product after they have been calculated. Instead they are stored first in a database. When an Apply Correction step requires in-flight calibration parameters, they are selected from the database according to certain rules. In particular, parameters are only updated close to the northern hemisphere terminator. This ensures that in-flight calibration parameters are not changed on the dayside which could lead to artificial steps in higher level processing. Only those in-flight calibration parameters selected to be used are stored within the Level 1a product.

If a given in-flight calibration parameter depends on a parameter or parameter combination, this is indicated in the tables below. It is then calculated, stored, and applied separately for each parameter combination. For integer parameters such as integration time and PMD modes this is straightforward. For temperature values, bins are used whose width is defined in the initialisation parameters such that the change of the in-flight calibration parameter within a given bin can be tolerated.

Examples of in-flight calibration parameters (dark signal correction, PPG correction, spectral calibration and etalon correction) and their effect on the spectrum to which they are applied are shown in In-Flight Characterisation and Calibration.

Dark Signal Correction			Radiometric Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	Dark	N/A	A2.8
Apply to	All above Dark	N/A	AG.10
Depends on	FPA and PMD: Integration time, detector temperature PMD: PMD readout and transfer modes
The dark signal for each parameter combination is calculated as an average of measurements in dark mode during eclipse. Applying the dark signal correction means subtracting the dark signal from a spectrum. This is the most basic correction and the first of all the corrections to be applied in Level 0 to 1 processing. The dark signal has two components: the integration-time independent offset of typically 1500 BU, and the integration time and detector temperature dependent leakage current of typically 0.7 BU/s at a detector temperature of 235 K. When calculating and applying the dark signal correction, only the total dark signal is considered. There is no need to split the dark signal into its components. This is only done for long-term monitoring purposes, see Sensor Performance Assessment.

Pixel-to-pixel gain (PPG) correction			Radiometric Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	LED (WLS)	N/A	A2.11
Apply to	All above Dark	N/A	AG.12
Given a uniform illumination over the detector array, dark-signal corrected signals vary slightly between detector pixels, mainly because of small differences in pixel width. To compensate for this effect, the variation in pixel-to-pixel gain is determined from measurements in LED mode which provide a fairly uniform illumination. A PPG correction spectrum is obtained by applying a triangular smoothing to the LED (fall back: WLS) measurements and dividing the measured spectrum by the smoothed spectrum. Applying the PPG correction means dividing a spectrum by the PPG correction spectrum. The PPG correction is of the order of 10^-4 relative.

Spectral calibration			Spectral Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	SLS	FPA: SLS line positions PMD: PMD slit function, SLS Stokes fractions, FPA overlap wavelengths, MME	A2.13 A2.14
Apply to		N/A	AG.13
Depends on	Predisperser prism temperature
Spectral calibration is the assignment of a wavelength value to each detector pixel. For each GOME-2 channel, a low order polynomial approximation will be used to describe wavelength as a function of detector pixel. Polynomial coefficients are derived from preprocessed spectra of the on-board PtCrNeAr Spectral Light Source (SLS) which provides a number of spectral lines at known wavelengths across the GOME-2 wavelength range. Different algorithms are used for FPA and PMD channels because their spectral resolution is different. Individual spectral lines can only be resolved in the FPA channels. Positions of individual lines from a predefined set are determined using a Falk centre-of-gravity algorithm. For the PMD channels an iterative cross-correlation algorithm is used. The expected PMD signal is calculated from the measured FPA signal, taking into account the PMD slit function and ratio of radiometric response between PMD and FPA channels. The expected PMD signal is then spectrally matched with the measured PMD signal using cross-correlation. The detector pixel onto which light of a given wavelength is impinging depends on instrument temperature. The temperature at the predisperser prism is used as a reference. Typical shifts with temperature are of the order of 0.01 pixel / K for the FPA channels, but depend on detector pixel.

Etalon correction			Radiometric Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	WLS (Sun)	Reference WLS spectrum	A2.16
Apply to			AG.14
Interference in the thin detector coating layer causes a wave-like pattern on the radiance response (fixed etalon). When deposits settle on the detector coating, the interference pattern is changed (variable etalon). The etalon correction accounts for changes in the variable etalon between on-ground calibration (reference etalon) and the in-orbit situation. It is calculated from preprocessed spectra of the on-board quartz tungsten halogen white light source (WLS). The in-orbit WLS spectrum is ratioed to a reference WLS spectrum representative of the on-ground calibration of the radiance response. For each channel, a bandpass filter is applied to the ratio. Spectral components within the bandpass (typically 4^-10 oscillations per channel) are declared to be the Etalon correction. The remainder is called Etalon residual. Applying the Etalon correction means dividing a spectrum by the Etalon correction spectrum. The Etalon correction typically is of the order of 10^-2 relative.

Sun mean reference (SMR) spectrum			Radiometric Spectral Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	Sun	Irradiance response MME	A2.20 (AG.17, AG.18)
Apply to	Earth	N/A	A3.11
A fully calibrated solar spectrum is calculated from preprocessed spectra acquired in sun mode. Spectra within the central part of the sun field-of-view are absolutely radiometrically calibrated taking into account solar elevation and azimuth angles for each spectrum, and then averaged into a sun mean reference (SMR) spectrum. The radiometric calibration is performed for the actually measured spectra, i.e., SMR intensities are not normalised to an Earth-sun distance of 1 AU. The SMR spectrum is also spectrally calibrated, correcting the Doppler shift due to the movement of the satellite towards the sun. Applying the SMR spectrum means dividing an Earth-radiance spectrum by the SMR spectrum. The SMR irradiance is of the order of 5·1014 photons / (s cm² nm) at 550 nm.

7.1.4 Scene dependent corrections

Polarisation Correction			Radiometric Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	Earth	Polarisation MME	A2.21
Apply to	Earth	N/A	A3.10
GOME-2 is a polarisation-sensitive instrument: the measured signal depends on the polarisation state of the incoming light which in turn depends on the observed scene, in particular observation geometry and cloudiness. In the Müller matrix formalism, the polarisation state is characterised by Stokes fractions q (for the s/p component) and u (for the +45°/-45° component). For each ground pixel, these Stokes fractions are determined for a number of wavelengths from the observation geometry (single scattering value for the short wavelength end) and ratios of PMD s and PMD p band measurements. In contrast to GOME-1, FPA measurements are not used in this step. Stokes fractions are then interpolated to the wavelengths of the FPA detector pixels, and for each pixel a polarisation correction factor is calculated from Stokes fractions and polarisation MMEs. FPA signals are polarisation corrected by dividing them by the polarisation correction factor. The polarisation correction typically is of the order of 10^-2 to 10^-1 relative, but may be larger for strongly polarised scenes.

Stray light correction			Radiometric
	Modes	Calibration keydata / MME	PGS reference
Calculate from	Earth, Sun	Uniform stray light fraction Ghost stray light intensity and position polynomial coefficients	AG.15
Apply to	Earth, Sun	N/A	AG.16
Spectral stray light is (unwanted) light from other than the nominal wavelength measured by a given detector pixel. We distinguish between uniform stray light from diffuse scatter and ghost stray light from specular reflections. Both types have been characterised during on-ground calibration. It turned out that only intra-channel stray light has to be considered, the most important one being the uniform stray light in channel 1. For each ground pixel the uniform stray light contribution is calculated by weighting the sum of the measured signals in a channel with the corresponding stray light fraction from the calibration key data. The ghost stray light position and intensity is calculated from the "parent" position and intensity using polynomial coefficients from the key data. Applying the stray light correction means subtracting the total stray light (sum of uniform and ghost stray light) from the measured signals. The stray light correction typically is smaller than 10^-3 relative.

Spatial aliasing correction			Radiometric
	Modes	Calibration key data / MME	PGS reference
Calculate from	Earth	N/A	A3.14
Apply to
Spatial aliasing is caused by the finite readout time of FPA and PMD detector pixel arrays. The detector pixels are read consecutively while the scan mirror is moving and therefore each pixel observes a ground scene that is slightly shifted in space compared to the previous pixel. E.g., with an FPA integration time of 187.5 ms and a detector readout time of 46.875 ms, the shift between the first and last detector pixel amounts to one quarter of the observed scene. This will alias spatial patterns into spectral patterns whose magnitude will depend on scene variability (cloud/no cloud, land/water, etc.). At those main channel boundaries where the last pixel of a channel and the first pixel of the next channel are read at different times, spatial aliasing results in intensity jumps. Spatial aliasing correction makes use of the higher spatial resolution of the PMD channels as compared to the FPA channels. FPA signals are corrected to the time of the readout of the detector pixel at the short wavelength end of the detector. When calculating the correction factor, the relative timing between PMD and FPA readouts is considered. For each FPA channel, only one PMD band is selected, and the correction is interpolated in wavelength between the selected PMD bands to avoid spectral artifacts being introduced by the correction. The spatial aliasing correction typically is of the order of 10^-1 relative.

Calculate fractional cloud cover and cloud top pressure			Other Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	Earth	N/A	A3.15
An effective cloud fraction and cloud top pressure are retrieved for each GOME-2 ground pixel using the Fast Retrieval Scheme for Clouds from the Oxygen A band (FRESCO) developed by KNMI. The FRESCO retrieval method is based on a comparison of measured and simulated reflectivities in three approximately 1 nm wide spectral windows in and around the oxygen A band (758, 761, 765 nm). Information on cloud fraction is mainly coming from the continuum (high intensity for high cloud cover). Information on cloud top pressure is mainly coming from the oxygen A band itself (high absorption for low cloud). Cloud parameters derived in this function are reported in the products for use in higher level processing. They are not used any further in the Level 0 to 1 processor.

7.1.5 Absolute radiometric calibration

Normalise to integration time of 1 second			Radiometric
	Modes	Calibration key data / MME	PGS reference
Apply to	All	N/A	AG.11
Preprocessed signals are normalised to an integration time of one second by dividing them by the integration time. For the PMD channels the actual integration time for each block, as determined by the readout and transfer modes, is used.

Apply radiance response			Radiometric
	Modes	Calibration key data / MME	PGS reference
Apply to	Earth	Radiance response MME	A3.11
Absolutely calibrated radiances in physical units [photons / (s cm² nm sr)] are calculated from preprocessed signals in instrument units [BU/s]. This is done by dividing the preprocessed signals by the radiance response MME for the actual scanner viewing angle. For calculating absolute radiances from the PMD channels, signals from PMD p and PMD s channels are combined. Radiances are strongly dependent on observation geometry and ground scene. Even for a fixed geometry a variation of one order of magnitude between a cloudy and a cloud-free scene is common. For a solar zenith angle of 60° typical radiance values are between 1013 and 1014 photons / (s cm² nm sr).

Apply irradiance response			Radiometric
	Modes	Calibration key data / MME	PGS reference
Apply to	Sun	Irradiance response MME	AG.17
Absolutely calibrated irradiances in physical units [photons / (s cm² nm)] are calculated from preprocessed signals in instrument units [BU/s]. This is done by dividing the preprocessed signals by the irradiance response MME for the actual solar elevation and azimuth angle combination. As stated above for the SMR, the solar irradiance is of the order of 5·1014 photons / (s cm² nm) at 550 nm.

7.1.6 Quality flagging (product confidence data generation)

The task of product quality control is shared between the Level 0 to 1 processor and the Product Quality Evaluation (PQE) function. The Level 0 to 1 processor condenses information relevant for product quality monitoring into Product Confidence Data (PCD). These PCD records are then used by the PQE function to derive quality reports and statistics.

PCDs are provided at three levels:

PCDs related to in-flight calibration parameters are reported as numbers and flags in the corresponding database entries and VIADRs (see Product Formats).
PCDs related to individual scans are reported as numbers and flags in the MDRs.
PCD flags on scan level are counted and their total number is reported on product level in the SPHR. The number of scans per observation mode and the numbers of missing data packets and scans are also reported in the SPHR.

A number of functions create their own PCD records, as indicated in the function descriptions above. In addition, there are two dedicated functions performing specific quality checks and generating additional PCD records.

Determine PCDs from raw intensity			Quality
	Modes	Calibration key data / MME	PGS reference
Calculate from	All	N/A	A2.4 (AG.5, AG.6)
Raw signals are checked for the following conditions affecting their quality, and flags are raised accordingly: Saturated pixels, i.e., binary readouts above a predefined threshold, typically 50,000 BU. (The threshold should be defined in a conservative way such that it indicates the end of the linear response rather than the real saturation value.) Spectra with saturated pixels should not be used at all in higher level processing because of undefined side effects to non-saturated pixels in the same channel. Hot pixels, i.e., spikes caused by high-energy cosmic particles. This check is applied only to measurement modes having a small signal variation between adjacent pixels, i.e., dark, LED and WLS modes. Spectra with hot pixels could be used in further processing if the hot pixels are excluded. For this purpose a hot pixel mask is generated. Band averaged signal too low. If the band averaged signal is lower than the band averaged offset (around 1500 BU), this points to a problem. E.g., the corresponding FPA might be switched off. For all checks, threshold values are defined as initialisation parameters.

Determine PCDs from geolocation			Quality
	Modes	Calibration keydata / MME	PGS reference
Calculate from	All	N/A	A2.7 (AG.7, AG.8, AG.9)
Geolocation parameters are checked for a number of conditions affecting data quality, and flags are raised accordingly: Satellite within South Atlantic Anomaly (SAA) region, defined as a rectangular region in longitude and latitude. Ground pixel possibly affected by sun glint over sea, due to observation geometry. Ground pixel possibly affected by rainbow effect, due to observation geometry. Observation mode and geolocation not consistent (e.g., dark measurements performed on day side, solar measurements far away from northern hemisphere terminator, etc.). Threshold angles are defined as initialisation parameters. Sun glint and rainbow flags indicate a geometric possibility only. Actual occurrence and magnitude of the effect depends on cloud cover (sun glint, rainbow) and wind speed at ocean surface (sun glint).

7.1.6.1 Sensor performance assessment

The Sensor Performance Assessment (SPA) function is used for long-term monitoring of instrument performance. Performance-related monitoring parameters are: instrument housekeeping data, spectral data (in particular in-flight calibration parameters) and polarisation data. SPA functionality comprises extraction, preprocessing and analysis of the monitoring parameters. The SPA function consists of two components:

Monitoring data are extracted from Level 1a products, Level 1b products and in-flight calibration datasets. They are preprocessed and written to the SPA database (GOME-2 PGS A4.1).
Monitoring data for a given timeframe are retrieved from the SPA database and statistically analysed (GOME-2 PGS A4.2).

A typical use of the SPA function would be to extract and preprocess a selected monitoring parameter, e.g. leakage current, and to visually and statistically analyse its time series (trends, oscillations, jumps, etc.).

Extract and pre-process monitoring data			SPA
	Modes	Data source	PGS reference
Calculate from	All	1a, 1b, in-flight calibration	A4.1
Monitoring data from the Level 1a and 1b data products and the in-flight calibration database is extracted, preprocessed and written to the SPA database. Preprocessing reduces the volume of monitoring data, which is essential if time series over several years have to be analysed. The following preprocessing steps are applied to the extracted monitoring data: Housekeeping data are converted to physical units and mean/min/max/stddev values per scan are calculated. Spectral data are averaged per observation mode and interpolated to a fixed wavelength grid. Spectra from sun and Earth modes are normalised to a solar distance of 1 AU. Stokes fractions are determined for individual solar measurements (not the SMR). The spectral reflectivity of the on-board diffuser is determined in predefined wavelength windows from ratios of SLS over diffuser and direct SLS measurements. Dark signals are separated into offset and leakage current components.

Perform statistical analysis on time series			SPA
	Modes	Data source	PGS reference
Calculate from	All	SPA database	A4.2.1
This function provides basic statistical functions for the higher level functions described below.

Monitor housekeeping data			SPA
	Modes	Data source	PGS reference
Calculate from	All	SPA database	A4.2.3 to A4.2.6
Monitoring of housekeeping data covers thermal, electrical and mechanical instrument performance. For life-limited items such as on-board lamps, scanner and shutter, usage statistics are derived. The total time per instrument mode is calculated. Selected temperatures are statistically analysed, in particular with respect to orbital and annual periodicities and trends. SLS and WLS voltages are statistically analysed, per switch on-period and as a long-term time series. This includes monitoring of SLS ignition delay and ignition voltage. Scanner positions across the scan are compared to the nominal values, and the timeseries of deviations is monitored.

Monitor spectral data			SPA
	Modes	Data source	PGS reference
Calculate from	All	SPA database	A4.2.7 to A4.2.14
The variation of spectral data over time is analysed in order to monitor detector performance, overall instrument throughput and stability, and sun diffuser reflectivity. Time series of the following parameters can be statistically analysed: Dark signal offsets, leakage current and dark signal noise. On-board light sources: SLS, WLS, and LED signals. Spectral stability: Wavelength values for selected detector pixels. Etalon spectra and Etalon residuals. Pixel-to-pixel gain. Diffuser reflectivity. Instrument throughput: Ratios of solar spectra with a reference solar spectrum, Ratios of Earth spectra for selected scenes (deserts, cloud-free ocean pixels, etc.) with a reference Earth spectrum for the same scene.

Monitor polarisation data			SPA
	Modes	Data source	PGS reference
Calculate from	Sun, Earth	SPA database	A4.2.15 to A4.2.16
Selected Stokes fractions are monitored in order to detect changes in the instrument polarisation response and the performance of the PMD channels. Stokes fractions are selected such that they should be zero due to the measurement geometry. Time series of the following parameters can be statistically analysed (see [SCD20]): Stokes fractions for solar measurements. Stokes fractions for special Earth viewing geometries.

7.1.6.2 Product quality evaluation

The Product Quality Evaluation (PQE) function provides information about the quality of the generated Level 1 products. It uses PCD records from the Level 1a and 1b data products to derive quality reports, statistics and quick-look data. The PQE function consists of two components:

PCDs are extracted from the Level 1a and 1b data products and written to the PQE database (PGS A5.1).
Quality data for a given timeframe are retrieved from the PQE database and condensed to provide daily, weekly, monthly and yearly Product Quality Summaries and Quick Look information.

Extract quality data			PQE
	Modes	Data source	PGS reference
Calculate from	All	Level 1a and 1b products	A5.1
PCD records are extracted from Level 1a and Level 1b products and stored in the PQE database.

Generate daily quick-look data			PQE
	Modes	Data source	PGS reference
Calculate from	Earth	PQE database	A5.2.1
The following parameters are calculated (usually from 1 day of data, i.e., 14 orbits) and visualised on a global map: Ozone line ratio from two UV wavelengths (FPA channel 2) as a proxy for ozone column densities. Red-green-blue ("false colour") image from three PMD bands. Normalised Difference Vegetation Index (NDVI) from two wavelengths in FPA channel 4 as a proxy for vegetation. Visual inspection of these maps allows a number of instrument or processing problems to be spotted quickly (e.g., pointing problems, wrong commanding, saturation, errors in geolocation calculations).

Generate product quality summaries			PQE
	Modes	Data source	PGS reference
Calculate from	All	PQE database	A5.2.2
PCDs for a specified time interval (day, week, month, year) are condensed as follows: PCDs describing an occurrence number (counters) are summed. PCDs describing mean values are averaged. Flags on non-nominal situations are counted. A report with the condensed information is issued.