This repository contains the Python scripts developed for data acquisition, processing, analysis and visualization throughout my PhD research.
They provide a reproducible workflow for handling atmospheric and environmental datasets, including:
- Automated retrieval of TROPOMI satellite and MERRA-2 reanalysis products
- Transformation of raw data into analysis-ready formats
- Calculation of derived variables such as tropospheric ozone columns and boundary-layer mixing ratios
- Generation of figures and plots to interpret temporal and spatial patterns
By sharing these tools, the repository enables other researchers to replicate, validate and extend the analyses presented in the thesis.
Scripts in this group automate the retrieval of remote sensing and reanalysis datasets:
-
openeo_tropomi_download.py– Downloads TROPOMI satellite products through the OpenEO platform. -
earthdata_merra2_download.py– Downloads MERRA-2 atmospheric variables from NASA Earthdata.
These scripts prepare the raw datasets for subsequent analysis:
-
dataframes_selected_sites.py– Extracts data for custom selected sites and organizes it into structured dataframes. -
vcds_monthly_means.py– Calculates multi-year monthly mean VCDs from individual monthly NetCDF files, producing one NetCDF per month with averaged values and metadata for seasonal and long-term analysis. -
netcdf_to_csv.py– Converts NetCDF datasets to CSV format for general use or external analysis. -
csv_to_netcdf.py– Converts CSV-formatted data into NetCDF files suitable for plotting.
These scripts combine TROPOMI and MERRA-2 data to derive new atmospheric variables:
-
pbl_hcho_and_no2.py– Computes planetary boundary layer mixing ratios of HCHO and NO2. The script integrates TROPOMI vertical column densities with MERRA-2 PBL heights, interpolating PBLH fields to the TROPOMI grid. -
tropospheric_ozone_estimation.py– Estimates tropospheric ozone columns by scaling TROPOMI total ozone observations using the troposphere-to-total ozone ratio derived from MERRA-2 reanalysis. The script calculates mid-level pressures, identifies tropospheric layers and interpolates the MERRA-2 ratio onto the TROPOMI grid.
Scripts in this group generate figures and summary plots used for analysis and interpretation:
seasonal_meteorology_plot.py– Generates seasonal and daily time series plots of meteorological parameters (air temperature, relative humidity, air pressure and precipitation). Example output:
monthly_meteorology_plot.py– Produces monthly averaged bar plots of meteorological parameters (air temperature, relative humidity, air pressure and precipitation). Example output:
windrose_and_distribution.py– Creates wind rose diagrams and wind speed distribution histograms. Example output:
days_with_data.py– Calculates and plots the annual number of valid days with TROPOMI data for multiple trace gases at selected sites. Example output:
seasonal_average_grid.py– Generates a 3 x 3 grid of seasonal averages with error bars and smoothed splines, including annual trends for multiple selected sites. Example output:
hcho_x_no2_x_o3.py– Creates a scatter plot of HCHO vs NO2, colored by O3, with reference lines representing FNR thresholds. Example output:
fnr_x_prob_o3.py– Plots FNR against ozone exceedance probability with a third-order polynomial fit, 95% confidence interval and peak highlighting. Example output:
fnr_trends_plot.py– Produces multi-panel seasonal trends of HCHO and NO2 across urban, transition and forest areas with error bars and linear regression lines. Example output:
If you use these scripts, please cite my PhD thesis:
FREITAS, A. D., Satellite observations of trace gases vertical columns: exploring temporal and spatial complexities. Thesis (Ph.D. in Meteorology). Institute of Astronomy, Geophysics and Atmospheric Sciences, University of São Paulo, São Paulo, 2026.
This repository is released under the MIT License, allowing free use, modification and distribution with attribution.