GRESO CO2 Growth Rate Analysis

Global Rate of change of CO2 from Space-based Observations

Figure: Annual CO2 growth rates from OCO-2 satellite observations (blue) compared to NOAA ground-based measurements (red).

A Python package for calculating annual CO2 growth rates from OCO-2 satellite observations using the MIPV11 format and comparing with NOAA reference data.

Overview

This project analyzes OCO-2 satellite data to compute global annual CO2 growth rates and validates results against NOAA ground-based measurements. The analysis uses 10-second resolution OCO-2 data to calculate spatially-weighted global averages and apply seasonal detrending to isolate long-term trends.

Quick Start

Prerequisites

• Python 3.7+
• Required packages: numpy, pandas, xarray, matplotlib, scipy, python-dateutil

Installation

# Clone or download the project files
cd /path/to/project/

# Install dependencies (if not already available)
pip install numpy pandas xarray matplotlib scipy python-dateutil

Running the Analysis

# Basic analysis with default settings (MIPV11 format)
python greso_analysis.py

# Analyze specific data type (OG, LNLG, or all)
python greso_analysis.py --data-type LNLG

File Format Support

The analysis supports two OCO-2 data formats:

1. MIPV11 Format (Default): Recommended format with improved calibration

   • Uses xco2_2019_scale variable
   • NetCDF4 format (.nc4 files)
   • Download from: https://gml.noaa.gov/ccgg/OCO2_v11mip/download.php
   • Direct file download: https://gml.noaa.gov/aftp/user/andy/OCO-2/OCO2_b11.2_10sec_GOOD_r2.nc4

2. Legacy Standard Format: Original format for backward compatibility

   • Uses xco2 variable
   • HDF5 format (.h5 files)

# Example: Using MIPV11 format (default)
from greso_analysis import AnalysisConfig, CO2GrowthRateAnalyzer

config = AnalysisConfig()
config.data_file = "/path/to/OCO2_b11.2_10sec_GOOD_r3.nc4"
# file_format is "mipv11" by default

analyzer = CO2GrowthRateAnalyzer(config)
results = analyzer.process_oco_data(["all"])

# Example: Using legacy standard format
from greso_analysis import create_standard_config

config = create_standard_config("/path/to/OCO2_b11.2-lite_10sec_GOOD_140906-250430.h5")
analyzer = CO2GrowthRateAnalyzer(config)
results = analyzer.process_oco_data(["all"])

Data Requirements

Input Data

OCO-2 Satellite Data (MIPV11 Format - Default):

• Recommended: MIPV11 format files (e.g., OCO2_b11.2_10sec_GOOD_r3.nc4)
• Download: https://gml.noaa.gov/ccgg/OCO2_v11mip/download.php
• Format: NetCDF4 with 10-second resolution OCO-2 observations
• Variable: Uses xco2_2019_scale for improved calibration

Legacy Standard Format (Optional):

• File: OCO2_b11.2-lite_10sec_GOOD_140906-250430.h5
• Format: HDF5 with 10-second resolution OCO-2 observations
• Variable: Uses xco2

NOAA Reference Data:

• Source: ftp://aftp.cmdl.noaa.gov/products/trends/co2/co2_gr_gl.txt
• Global CO2 growth rates from ground-based measurements
• Automatically downloaded during analysis

Data Structure

The OCO-2 data should contain:

• sounding_id: Unique identifier with embedded timestamp
• xco2: Column-averaged CO2 concentration (ppm)
• xco2_uncertainty: Measurement uncertainty
• latitude, longitude: Geographic coordinates
• data_type: Quality flags (6=OG, <3=LNLG)
• Additional variables for filtering and analysis

Usage Guide

Configuration

Modify analysis parameters in the AnalysisConfig dataclass:

config = AnalysisConfig(
    data_file="/path/to/your_data.nc4", # Path to your OCO-2 data file
    lat_range=[-50, 50],                   # Latitude bounds for analysis
    uncertainty_threshold=2.0,              # Maximum allowed uncertainty
    output_filename="annual_growth_rate.png" # Name for the output plot
)

Main Components

1. OCODataProcessor: Handles data loading and preprocessing
2. CO2GrowthRateAnalyzer: Performs growth rate calculations
3. VisualizationManager: Creates plots and visualizations

Output Files

• annual_growth_rate.png: Plot showing annual growth rates
• Console output: Tabular display of annual growth rates
• Log file: Detailed processing information

Methodology

Data Processing Pipeline

1. Quality Filtering: Apply uncertainty and data type filters
2. Spatial Filtering: Restrict to tropical/subtropical latitudes ([-50°, 50°])
3. Temporal Processing: Convert sounding IDs to datetime objects
4. Spatial Weighting: Area-weighted averaging by latitude
5. Seasonal Detrending: Remove seasonal cycles using harmonic analysis
6. Growth Rate Calculation: Compute annual growth rates from detrended data

Scientific Approach

• Spatial Coverage: Global analysis with tropical focus
• Temporal Resolution: 10-second observations aggregated to monthly means
• Detrending Method: Harmonic analysis to remove seasonal cycles
• Growth Rate Calculation: Annual differences in deseasonalized CO2
• Validation: Direct comparison with NOAA global mean growth rates

Results

Visualization

The analysis generates a plot comparing OCO-2 derived annual CO2 growth rates with NOAA reference data (shown at the top of this README).

Example Output

============================================================
ANNUAL GROWTH RATES - ALL DATA
============================================================
Year       | Growth Rate (ppm/yr) | NOAA Growth Rate (ppm/yr)
------------------------------------------------------------
2015       |             2.9885 | 2.9500
2016       |             2.7095 | 2.8400
2017       |             2.0508 | 2.1400
2018       |             2.4630 | 2.3900
2019       |             2.7487 | 2.5000
2020       |             2.3091 | 2.3300
2021       |             2.2961 | 2.3700
2022       |             1.9195 | 2.2900
2023       |             3.0562 | 2.7300
2024       |             3.2199 | 3.7300
------------------------------------------------------------
MEAN       |             2.5046 | 2.5044
============================================================

Validation

• OCO-2 Mean Growth Rate: 2.5050 ppm/yr
• NOAA Mean Growth Rate: 2.5044 ppm/yr
• Difference: 0.0006 ppm/yr (excellent agreement)

File Structure

greso_analysis.py    # Main analysis script
utils.py            # Utility functions and classes
README.md           # This documentation file
annual_growth_rate.png  # Generated plot

Development

Code Architecture

The codebase follows object-oriented design principles:

• AnalysisConfig: Configuration management
• OCODataProcessor: Data loading and preprocessing
• CO2GrowthRateAnalyzer: Core analysis logic
• VisualizationManager: Plotting and visualization

Key Functions

• get_10Sec_data_baker(): Load and filter OCO-2 MIPV11 data
• CO2GR: Main analysis class with binning and regional analysis
• deseasonalize(): Remove seasonal cycles
• monthlyGrowthRate(): Calculate monthly growth rates
• areaWeighted(): Spatial averaging with latitude weights

Troubleshooting

Common Issues

1. File Not Found: Ensure OCO-2 data file is in correct location
2. Memory Issues: Reduce dataset size or increase system memory
3. Network Issues: NOAA data download may fail - retry or use cached data

Error Messages

• "No valid observations": Check data file path and quality filters
• "Interpolation failed": Reduce spatial/temporal filtering
• "Memory error": Process data in smaller chunks

Testing

• Validate results against known benchmarks
• Test with different data types (OG, LNLG, all)
• Verify spatial and temporal filtering
• Check edge cases (empty datasets, missing data)

License

This project is provided as-is for scientific research purposes. Please cite appropriately when using this code or results.

References

Primary Publication: Pandey, S., et al. (2024). Toward low-latency estimation of atmospheric CO₂ growth rates using satellite observations: Evaluating sampling errors of satellite and in situ observing approaches. AGU Advances, 5, e2023AV001145. https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023AV001145

Acknowledgments

• OCO-2 Science Team for satellite data
• NOAA Global Monitoring Laboratory for reference data
• NASA/JPL for mission support and data access

Authors

Primary Developer: Sudhanshu Pandey
Affiliation: NASA Jet Propulsion Laboratory / California Institute of Technology
Contact: [email protected]

Project: GRESO (Global Rate Estimates from Satellite Observations)
Development Date: August 2025

Please cite the primary publication above when using this code or results in publications:

Citation Format:

Pandey, S., Miller, J.B., Basu, S., Liu, J., Weir, B., Byrne, B., Chevallier, F., Bowman, K.W., Liu, Z., Deng, F. and O’dell, C.W., 2024. Toward low‐latency estimation of atmospheric CO2 growth rates using satellite observations: Evaluating sampling errors of satellite and in situ observing approaches. *AGU Advances*, 5, e2023AV001145. https://doi.org/10.1029/2023AV001145

stand_alone_GRESO_v2