Ott (CMS 2020): GEOS-Carb IV: Delivering low-latency carbon flux and concentration datasets in support of NASA's Carbon Monitoring System
Lesley Ott, NASA GSFC GMAO, lesley.e.ott@nasa.gov
Michael Long, NASA GSFC / SSAI, michael.long@nasa.gov (Presenter)
Since its 2010 inception, GSFC-based modeling teams have continuously provided complete and physically consistent set of global flux and atmospheric concentration data products to CMS. The proposed work will draw on the unique capabilities of NASA's Goddard Earth Observing System (GEOS) models and data assimilation system and consists of two main components: (i) production and refinement of observationally constrained 'bottom-up' atmosphere-ocean and atmosphere-land biosphere fluxes, and fossil fuel emissions; and (ii) production of global carbon analyses that incorporate multiple satellite and in situ datasets. A central theme has been the use of meteorological forcing provided by NASA's Modern Era Retrospective-analysis for Research and Applications, Version 2 (MERRA-2) to produce a consistent picture of the interactions between weather, climate, and carbon.
The activities proposed here have two overarching goals: 1) to reduce the latency of flux and concentration datasets and 2) to implement refinements in retrospective products that incorporate additional observations and improve quantification of uncertainties. COVID-19 provided a powerful example of the growing demand for low latency information about the carbon cycle. The GEOS-Carb system contributed to international and national efforts to track human emission reductions from space using components developed during earlier phases of the project: gap-filled analyses of carbon concentrations, an observationally-informed flux package, and a novel anomaly detection technique. Products were delivered to dashboards coordinated by NASA and international space agencies with ~2 month latency, providing the first quantification of recent changes in carbon dioxide concentrations. Despite this success, efforts to separate human emissions changes from natural variability were limited by the lack of year-specific information about land fluxes. We propose to address this limitation by making modifications to GEOS-Carb land fluxes that incorporate reflectance information available in near real time and integrate GMAO’s existing fire radiative power based fire emissions, which reduce latency over burned area approaches. This will feed into low latency carbon dioxide analyses, supporting preliminary attribution of atmospheric anomalies in support of carbon monitoring.
We also plan to extend and improve flux estimates to cover the 20-year period from 2003-2022. GEOS-Carb land flux estimates have been provided by a model that assumes an approximate balance between respiration and primary production. We plan to incorporate lessons learned from recently released estimates of gross primary production based on surface reflectances to improve the ability to simulate interannual and daily variability. Ocean flux estimates will be further constrained through incorporation of new data on the exports of carbon from surface to deep ocean and will incorporate updated uncertainty estimates. We plan to continue to provide updates to our nighttime lights based fossil fuel emissions inventory. In addition to improvements in bottom-up flux estimates, will also exploit the unique ability of GEOS to run with multiple subgrid physical parameterizations to refine transport uncertainty estimates. Ocean and land fluxes and fossil fuel emissions will be used together in the GEOS constituent data assimilation system (CoDAS) to provide the most complete, data-driven picture of atmospheric greenhouse gases over the past 20 years. GEOS-Carb products and system development will continue to support a wide range of stakeholders, including NASA science teams; state and federal government agencies; and international efforts aimed at constraining carbon budgets.
Associated Project(s):
Poster Location ID: 47
Presentation Type: Poster
Session: Poster Session 1
Session Date: Wednesday (9/27) 1:15 PM