PGN DataChecker Pandora193s1 Tsukuba

Author

PGN , LuftBlick

Published

November 4, 2025

Introduction

This HTML report provides an overview of the instrument’s time series using the official retrieval versions. The analysis is based on Level 2 (L2) data for direct sun total column measurements of O3, NO2, HCHO, SO2, and H2O, which represent the current operational data products. In addition, tropospheric columns of NO2, HCHO, and H2O, derived from sky measurements, are also included and visualized alongside the corresponding total column values from the direct sun retrievals.

Direct sun retrievals

  • NO2: rnvs3
  • O3 ExtRef: rout2
  • HCHO SyntRef: rfus5
  • SO2 SyntRef: rsus1
  • H2O ExtRef: rwvt1

Sky retrievals

  • NO2: rnvh3
  • HCHO: rfuh5
  • H2O: rnvh3

Aggregated overview

The following panels are based on daily aggregates using the highest available data quality for each day. The data quality is color-coded as follows:

  • Bright blue: high quality (codes 0, 10)
  • Light blue: medium quality (codes 1, 11)
  • Grey: low quality (codes 2, 12) or flagged periods due to quality issues (codes 20, 21, 22)

Each parameter is visualized using:

  • Thin vertical lines to represent the daily interquartile range (25%-75%)
  • A filled dot to indicate the daily median

From top to bottom, the panels show:

  • used calibration file (starting validity date, version number)
  • data quality summary in percent: high quality (bright blue), medium quality (faint blue), low quality (grey)
  • vertical column vc [mmol/m^2]
  • wrms (weighted with independent uncertainty) wrms_ind
  • wavelength shift wl_shift [pm]
  • effective wavelength temperature wl_temp [dg]
  • retrieved resolution change resol_change [%]
  • mean count inside the fitting window mean_value_fitting_window

Columns at minimum available AMF

This check monitors the time series evolution of a given parameter at the lowest available Air Mass Factor (AMF), using only high-quality data. Parameters such as total column amounts may exhibit intrinsic seasonal patterns, so any detected breakpoints should be interpreted with caution. Nevertheless, distinct jumps-for example, following a reinstallation-can indicate the need for recalibration.

no data left for dq=0,10

Found 4 breakpoints at: 
 2021-12-02, 2023-04-02, 2023-11-06, 2024-06-11

Found 1 breakpoints at: 
 2024-04-17

Found 3 breakpoints at: 
 2022-12-12, 2023-08-20, 2024-03-10

Found 2 breakpoints at: 
 2023-02-16, 2023-09-27

Resolution Change

Based on the quality codes (QC-codes), we use the literature-referenced configuration with a smoothing polynomial of 1 and a 70 nm wide fitting window in the 300-520 nm range for OPEN. While the absolute values cannot be used directly, the setup allows for the estimation of relative resolution changes between the laboratory characterization and the field conditions. These relative changes are shown in the following panels: black dots represent the daily median, and the 0.1 and 0.9 quantiles are indicated by grey bars.

In addition to the resolution change, a statistical breakpoint analysis is performed to assess whether the time series is better represented by multiple segments with constant values. The baseline (no-break) fit is shown in black. If a statistically significant split is detected for a minimum segment length of 0.15 of the available time series, the segmented time series is shown in blue, with the breakpoint date annotated and a vertical red line marking the point of change.

Found 2 breakpoints at: 
 2023-07-25, 2025-05-16

Found 3 breakpoints at: 
 2023-07-09, 2024-07-17, 2025-02-09

Found 4 breakpoints at: 
 2023-02-08, 2023-08-31, 2024-02-04, 2024-09-28

Found 3 breakpoints at: 
 2023-07-11, 2023-12-28, 2024-06-25