OASIS is a modular and open-access Python toolbox for processing multi-frequency GNSS data and computing ionospheric indices. It was developed to overcome limitations of proprietary or non-standardized tools and to promote scientific transparency and reproducibility in ionospheric research.

OASIS automates the detection and correction of cycle slips and outliers, performs arc-wise geometry-free (GF) leveling, and derives ionospheric indices directly from calibrated observations, without relying on external products such as Differential Code Biases (DCBs) or vertical TEC maps.

• Processes RINEX v2 and v3 files with 15- or 30-second sampling rates.
• Supports GPS and GLONASS constellations (Galileo and BeiDou coming soon).
• Fully autonomous detection of data gaps, cycle slips, and outliers.
• Arc-wise geometry-free leveling of carrier-phase combinations.
• Computation of ionospheric indices: ROTI, ΔTEC, and SIDX.
• No dependency on DCBs or global TEC maps.

To install the package, simply run:

pip install pyOASIS

• Python 3.8 or higher
• Required libraries (see requirements.txt for exact versions):
  • numpy
  • pandas
  • matplotlib
  • scipy
  • astropy
  • georinex
  • pyproj

To install all dependencies at once:

pip install -r requirements.txt

This repository includes example input files under the INPUT/ directory:

• boav0491.23o – RINEX observation file for station BOAV, day 049 of 2023.
• GFZ0MGXRAP_20230490000_01D_05M_ORB.SP3 – MGEX SP3 precise orbit file for the same day, provided by GFZ.

These files allow you to run a full test of the OASIS pipeline without additional downloads.

To use your own data, simply replace the files in the INPUT/ folder and update the parameters in main.py accordingly.

MGEX SP3 orbit files can be downloaded from the CDDIS archive by accessing the folder corresponding to the desired GPS week. A very easy registration is required to access the CDDIS archive. You can register here.

To convert a calendar date (dd/mm/yyyy) to GPS week and day-of-year (DOY), use gnsscalendar.com.

Each orbit file follows the naming pattern:

<prefix>_<YYYY><DOY>0000_01D_05M_ORB.SP3.gz

Where:

• YYYY is the year,
• DOY is the day of year (001–365/366),
• The prefix depends on the year:
  • For data before 2018, use: JAX0MGXFIN_
  • For data from 2018 onward, use: GFZ0MGXRAP_

Example: The orbit file for day 049 of 2023 is:

GFZ0MGXRAP_20230490000_01D_05M_ORB.SP3

After downloading, decompress the .gz file and place the resulting .SP3 inside your INPUT/ directory.

1. Input: RINEX observation files (e.g., boav0491.23o) and MGEX SP3 precise orbit files (e.g., GFZ0MGXRAP_20230490000_01D_05M_ORB.SP3).
2. Orbit Interpolation: SP3 orbits are parsed and interpolated to match GNSS epochs.
3. IPP Calculation: Coordinates of the ionospheric pierce point (IPP) are computed for each satellite-receiver pair.
4. MW Combination: The Melbourne–Wübbena combination is applied to detect and correct cycle slips.
5. Screening:
   • Initial outlier removal based on residuals from ∆MW polynomial fitting.
   • Fine-grained residual analysis defines mini-arcs based on sign changes.
6. Geometry-Free Leveling: Performed arc-wise on valid L1–L2, L1–L5, or L2–L3 combinations.
7. Index Derivation:
   • ROTI: Standard deviation of the Rate of TEC in 1-minute windows.
   • ΔTEC: Difference between 15-minute and 60-minute moving averages of geometry-free combinations.
   • SIDX: Mean absolute ROT over a 1-minute interval, sensitive to co-seismic and auroral disturbances.

1. Place the required RINEX observation files (.yyo) and MGEX SP3 orbit files (.SP3) inside the INPUT/ directory.
2. In main.py, define the station code, year, day of year (DOY), and paths to the input and output directories.
3. Run the pipeline:

python main.py

4. Results are saved in structured output folders, organized by station and satellite.

• SP3_INTERPOLATE.py – Interpolates MGEX SP3 precise orbit files and generates tabulated satellite positions in the format ORBITS_YYYY_DOY.SP3.
• RNX_CLEAN.py – Parses RINEX observation files and produces initial GNSS datasets organized by satellite and station. Assigns ionospheric pierce point (IPP) coordinates and flags preliminary gaps, outliers, and cycle slips. Output: STAT_SAT_DOY_YYYY.RNX1.
• RNX_SCREENING.py – Refines arc definitions and detects outliers and cycle slips using ∆MW residuals and polynomial fitting. Output: STAT_SAT_DOY_YYYY.RNX2.
• RNX_LEVELLING.py – Performs arc-wise geometry-free leveling on refined data arcs. Output: STAT_SAT_DOY_YYYY.RNX3.
• ROTI_CALC.py – Calculates the Rate of TEC Index (ROTI) from leveled geometry-free combinations in .RNX3.
• DTEC_CALC.py – Calculates the TEC anomaly index (ΔTEC) using 15-minute and 60-minute moving averages of leveled combinations.
• SIDX_CALC.py – Calculates the Sudden Ionospheric Disturbance Index (SIDX) as the mean absolute value of ROT in 1-minute windows.
• linear_combinations.py, gnss_freqs.py, settings.py, glonass_channels.dat, etc. – Supporting modules for GNSS frequency combinations, coordinate transformations, and IPP computation.

• Time series of arc-wise leveled geometry-free combinations for each GPS and GLONASS satellite.
• ROTI, ΔTEC, and SIDX indices computed per satellite and station.
• Visualizations of geometry-free combinations and derived ionospheric indices.

Example of arc-wise geometry-free leveling for GNSS signals at station BOAV (year 2023, day 049), using L1–L2 and L1–L5 combinations for GPS, and L1–L2 and L2–L3 for GLONASS. Each curve represents a leveled geometry-free combination over a continuous observation arc.

Example of Rate of TEC Index (ROTI) for GPS (L1–L2, L1–L5) and GLONASS (L1–L2, L2–L3) signals at station BOAV (year 2023, doy 049). This plot highlights ionospheric irregularities typically associated with nighttime equatorial plasma bubbles. Elevated ROTI values are observed between 00:00 and 07:00 UT, and again after 23:00 UT, consistent with post-sunset and pre-dawn plasma bubble activity.

Example of TEC anomaly index (ΔTEC) for GPS (L1–L2, L1–L5) and GLONASS (L1–L2) signals at station TONO (year 2024, doy 132). The ΔTEC index highlights positive and negative anomalies, particularly associated with wave-like ionospheric disturbances and density gradients observed during the May 2024 geomagnetic storm (the Mother’s Day Storm).

Example of the Sudden Ionospheric Disturbance Index (SIDX) for GPS at station PTBB (year 2017, day 249). The intense SIDX enhancements observed around 09:00 UT and 12:00 UT are associated with X-class solar flares during the September 2017 geomagnetic storm.

If you use this software in your research, please cite:

Developed by Giorgio Picanço (Ph.D. in Space Geophysics).

For questions or contributions, please visit:
https://github.com/giorgiopicanco/OASIS
or email: [email protected]

The OASIS toolbox is open-source and free to use under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license. You are free to use, adapt, and share this software for non-commercial purposes, as long as proper credit is given.