Atmospheric Applications / Data Assimilation 2

Global atmospheric concentrations of carbon dioxide (CO2) and methane
(CH4)  are forecasted in real-time and analysed with a delay of 5 days
as part of the EU-funded Copernicus Atmosphere Monitoring Service
(CAMS). These systems (forecast and analysis) are using the
infrastructure of the European Centre for Medium-Range Weather
Forecasts (ECMWF) Integrated Forecasting System (IFS).

One of the strengths of the CO2 forecasting system is that the land
surface, including vegetation CO2 fluxes, is modelled online within
the IFS. Other CO2 fluxes and CH4 fluxes are prescribed from
inventories and from off-line statistical and physical models. The CO2
and CH4 forecasts also benefits from the transport modelling from a
state-of-the-art numerical weather prediction (NWP) system at high
resolution (currently 16 km globally).

The analysis is the result of the assimilation of all the operational
meteorological data and the assimilation of greenhouse gases products
retrieved from satellite data. The greenhouse gas analysis is
constraining the atmospheric concentration of CO2 and CH4  without
adjusting the surface fluxes.  The CO2 analysis used to assimilate the
AIRS and IASI  radiances and it is now assimilating a GOSAT level 2
product. The CH4 analysis  used to assimilate SCIAMACHY level 2 data
and it is now assimilating GOSAT and IASI level 2 products.  

The presentation will show the skill of the real-time  forecast
especially in modelling the variability of CO2 and CH4 on different
temporal and spatial scales. It will also show the added value and
limitation of the analysis when only the atmospheric concentration  of
CO2 and CH4 is constrained. We will also present the preliminary
results on the assimilation of the data from Orbiting Carbon
Observatory-2  (OCO-2), a relatively new mission.

We present a method for assimilating CH4:CO2 measurements from satellites for inverse modeling of CH4 and CO2 fluxes in TM5-4DVAR inverse modeling system. Unlike conventional proxy approach, in which retrieved CH4:CO2 ratios are multiplied by model derived total column CO2 and only the resulting CH4 is assimilated, our method assimilates the ratio of CH4 and CO2 directly and is therefore called the ratio method. It is a dual tracer inversion, in which surface fluxes of CH4 and CO2 are optimized simultaneously. The optimization of CO2 fluxes turns the hard constraint of prescribing model derived CO2 fields into a weak constraint on CO2, which allows us to account for uncertainties in CO2. We assimilate real GOSAT ratio and proxy observations to optimize surface CH4 fluxes. We do two proxy inversions with CO2 fields from separate models and quantify the impact of differences in the model on the posterior CH4 fluxes. Atmospheric fields from the posterior emissions of all inversions are validated with independent aircraft measurements. We find that the posterior fields of ratio inversion are in better agreement with independent aircraft measurements than proxy inversion, especially in the regions like Tropical South America. 

Smoke from fires plagued Indonesia during the second half of 2015. Slash and burn practices combined with reduced rain due to El Niño conditions intensified biomass burning and led to extremely poor air quality in large parts of Indonesia, Singapore, and Malaysia. Burning of tropical peat forests leads to large emission of carbon (CO₂, CO), and first estimates claim that the Indonesian fires in 2015 emitted more carbon than the entire United Kingdom. However, carbon emissions from peat fires are highly uncertain and depend, among others, on the thickness of the peat layer. Top down estimates using satellite measurements and modeling might help to constrain carbon emissions from the Indonesia fires. We will present first estimates of the carbon emissions using TM5-4DVAR, a data-assimilation system that derives CO emissions using satellite observations from IASI. This system has been previsouly applied to the 2010 Russian fires, to the Amazon, and to emissions from Tropical Asia. Using CO/CO₂ emissions factors for peat burning we will translate the derived CO emissions into total carbon emissions and validate these numbers with independent observations from e.g. the GOSAT instrument.