CEOS Development Environment

CEOS Greenhouse Gas Roadmap

Article 14 of the Paris Climate Agreement sets out the concept of the Global Stocktake (GST) as a means to evaluate global progress towards the goals of the Agreement. Earth observations naturally play a key role in this process, and CEOS & CEOS Agencies have been working hard in the years leading up to the first GST, to be held in November 2023, at the 28th Conference of Parties of the United Nation Framework Convention on Climate Change (UNFCCC). This series of articles details the work of CEOS & CEOS Agencies to support the GST process, and includes:

Recognising that high-quality observations of atmospheric carbon dioxide (CO2) and methane (CH4) should be an essential component of an integrated global carbon observing system, CEOS undertook a series of studies, which ultimately culminated in the publication of the CEOS Greenhouse Gas (GHG) Roadmap in 2020. 

System functional overview an atmospheric CO2 and CH4 monitoring system, showing the key
inputs, outputs, and interactions. Figure 1 of the CEOS GHG Roadmap.

The 2017 CEOS Chair, the United States Geological Survey (USGS), commissioned the CEOS Atmospheric Composition Virtual Constellation (AC-VC) to write a White Paper defining the key characteristics of a global architecture for monitoring atmospheric CO2 and CH4 concentrations and their natural and anthropogenic fluxes from instruments on space-based platforms. This paper aimed at reducing the uncertainty of national emission inventory reporting, while identifying additional emission reduction opportunities, focusing on the ability to provide nations with timely and quantified guidance on progress towards their emission reduction strategies and pledges, including tracking changes in the carbon balance of managed  ecosystems. The final chapter of the White Paper proposed an architecture of a future greenhouse gas constellation designed to address the objectives listed above and recommends a three-step plan to implement this architecture. The GHG Roadmap takes this final chapter as a starting point, and aims to deliver a prototype end-to-end system that yields estimates of CO2 and CH4 fluxes, alongside an initial operational system for producing future atmospheric CO2 and CH4 flux products for use in future Global Stocktakes and informing national inventory development.

In such systems, the space-based column-averaged dry air mole fractions of CO2 and CH4 estimates (sometimes referred to as XCO2 and XCH4, respectively) complement the spatial resolution and coverage of the ground-based and airborne in situ measurements. If the ground-based, airborne, and space-based datasets can be harmonised, they can be assimilated into atmospheric inverse systems to yield topdown global inventories of CO2 and CH4 fluxes with the accuracy, precision, resolution and coverage needed to serve as a complementary system for estimating national emission inventories, which can then be used for improving nationally determined contributions for the Global Stocktake, as proposed in the 2016 New Delhi Declaration. In addition, if these atmospheric data products were distributed freely and openly, in compliance with the CEOS data policy, they could support the foreseen Enhanced Transparency Framework of the Paris Agreement.

The implementation approach in the GHG Roadmap  follows the White Paper objectives, and outlines usage scenarios, expected outputs, functional scopes and schedules. These are supported by the required implementation actions, which are detailed in Annex C of the Roadmap and regularly updated. 

The delivery of pilot datasets to enhance the uptake of Earth observation datasets aims to provide support to inventory builders in their development of GHG inventories through the provision of timely global high-resolution atmospheric CO2 and CH4 flux products addressing the corresponding emissions. Meanwhile, the delivery of an initial operational system plans to integrate observations from the evolving pre-operational ground-based, airborne, and space-based atmospheric GHG observing network into flux inversion models. This will yield global, gridded estimates of atmospheric CO2 and CH4 fluxes with the precision and accuracy needed to support estimating the corresponding emissions at national scales. The system aims to be operational for supporting the second Global Stocktake in 2028. The refinement of user requirements requires a formulation of the user expectations and needs on the deliveries from the operational system. 

The CEOS-CGMS Joint Working Group on Climate (WGClimate) formed the dedicated GHG Task Team in March 2019, with the primary goal to initially develop and then implement the GHG Roadmap. Therefore, the GHG Task Team led the authoring of the GHG Roadmap, and now reports on progress towards the actions outlined. The group is now reviewing, revising and updating the actions outlined in Annex C of the GHG Roadmap, alongside a more broad update of the Roadmap. 

Initial pilot products have been produced in support of the first Global Stocktake and can be found on the CEOS GST Portal. For CO2, the products are reported in the paper by Byrne et al. (2023), while for CH4, the products are reported in the paper by Worden et al. (2022). These datasets have received substantial interest from both the public and policymakers in the product, showing a clear need for this information, with over 50 countries not reporting emissions in the last decade. 

Estimate of the fractional uncertainty reduction (FUR) on the v10 OCO-2 MIP estimates for each experiment based on Bayesian uncertainty estimates from the TM5-4DVar inversion. Figure 10 of Byrne et. al (2023)

The development of these datasets have also identified a number of challenges and lessons learned. Foremost is the challenges in isolating the quantities desired by inventory developers and policy makers. Top-down constraints are sensitive to the net surface-atmosphere GHG flux, but source attribution is needed for comparison with bottom-up estimates. This was found to be particularly challenging for CO2, where changes in carbon stocks on managed lands are desired. This requires accounting for lateral movement of carbon (due to the water cycle and crop/wood trade) and differentiating carbon fluxes between managed and unmanaged lands, which can differ between countries and are often not well defined.

Another key challenge has been quantifying uncertainties in the top-down estimates. The total error in these estimates is the combined result of systematic errors in the inversion approach, systematic errors in the greenhouse gas measurements, and Bayesian posterior uncertainty. Systematic errors from the inversion approach can be due to model transport biases, biases in prior constraints, or limitations of the inversion approach. These errors can be quantified through use of an ensemble of inversion systems that differ in each aspect. This was possible for the CO2 dataset but not for the CH4 dataset. Systematic errors in the observations can be identified through comparisons of posterior simulated GHG fields with independent observations. Finally, Bayesian posterior uncertainty estimates can only be calculated by certain inversion methods, which was possible for the CH4 dataset, but not for the CO2 dataset, where many ensemble members do not estimate this quantity.

The lessons learned provide the context and motivation to advance and improve on these products, and further implement the GHG Roadmap.