The Earth’s ocean absorbs about 8 billion tons of carbon dioxide (CO2) each year (Le Quéré et al., 2009), about 30% of all anthropogenic CO2. The ocean is a huge reservoir of CO2 and a better understanding on how the ocean absorbs CO2 is critical for predicting climate change. Carbon dioxide fluxes (F) between the ocean and atmosphere are normally calculated from difference between oceanic and atmospheric partial pressure of CO2 (pCO2) and gas transfer velocity (k):
F = kα ΔpCO2
Where α is the solubility depending on temperature and salinity (Weiss, 1974).
The gas transfer velocity k is typically calculated from parametrizations depending solely on wind speed as reviewed by Wanninkhof (2014) but these parametrizations miss in situ data from low wind speeds (Johnson, 2010). Therefore, they assume that for wind speed close to zero the gas transfer velocity is also zero.
The non-zero intercept is important for CO2 budgets because there are large oceanic regions with prevailing low wind conditions (Figure 3 in Archer and Jacobson (2005), and Figure 6 in Wurl et al. (2016)) (e.g., the Mediterranea Sea in summer and tropical Pacific). Low wind regimes with limited surface mixing leads to stratification of the upper layer and eventually to increased ocean productivity that would decrease the pCO2 in the ocean and would increase the ΔpCO2. With commonly used k parametrization, these regions will have a net fluxes close to zero during the low sea state periods, because k will be equal or close to zero, and their sink capacity will be clearly underestimated. To evaluate this bias with different approaches is important because underestimating CO2 carbon fluxes will propagate an error on the future scenarios that are used for policy makers.
Parametrization with non-zero intercept have been reported in (McGillis et al., 2004; Butterworth and Miller, 2016; Mustaffa et al., in revision; Ribas-Ribas et al., in revision) and range from 2 to 11 cm h-1. During calm weather, the wind speed is not the main factor driven gas transfer in the surface. Buoyancy fluxes, micro breaking waves and a more stable sea surface microlayer would play a major role (Wanninkhof, 2014).
Zhang and Cai (2007) reported potential biases based on an assumed non-zero intercept of 10 cm h-1 reported in published papers (Monahan and Spillane, 1984; Soloviev and Schluessel, 2002; Asher et al., 2004; McGillis et al., 2004) and pointed out that if such non-zero intercept is adapted the total sea-air global flux would increase 0.2 Pg C yr-1. More in situ data are required to compute and validate non-zero intercepts and to combine model and observational studies to draw more robust conclusion on relevant cases, i. e. CO2 fluxes. However, it is challenging to compare ocean model with in situ data due to different temporal and spatial scales.
The purpose of the study is to quantify the biases introduced by using a zero-intercept gas transfer velocity parametrization in the global carbon fluxes. In this study, we test the implications of a non-zero intercept gas transfer velocity in the framework of: (1) a global three-dimensional ocean carbon cycle model; (2) SOCAT global gridded data: For (1) a global three-dimensional ocean carbon cycle model; we discuss the short-term and long-term evolution of the ocean to atmosphere exchange flux of CO2, and the evolution of the carbon inventories in ocean and atmosphere after introducing a non-zero intercept to the gas transfer velocity. We present both the global evolution of these variables and spatial patterns of change. The model is run for 5000 years such that the system has time to approach new steady state conditions. For (2) in situ local data from various regions (open ocean, coastal ocean and fjords) and a range of meteorological forcing, we discuss the ocean-atmosphere fluxes differences depending on different wind speed parametrization and their relative importance of wind speed and the differences between ocean and atmosphere pCO2. For (3), SOCAT global maps from 1982 until 2017 are used to calculated difference in global and regional CO2 fluxes using five differences gas transfer velocities parametrizations (two commonly used with a zero-intercept and three with a non-zero intercept). The three approaches agree that the non-zero intercept matter, especially in areas with low wind speeds; (3) in situ local data.
Overall, ignoring a non-zero intercept might lead to an error on the global FCO2 of -0.73 Gt c yr-1, which is 28 % lower than the present estimation.

Area C - Board 255

The ongoing climate warming is due to continuously increasing accumulation of carbon dioxide (CO2) in the atmosphere. This process is regulated by many processes of both man-made and natural origin including the uptake of atmospheric CO2 by the oceans. The ocean CO2 sink efficiency significantly constrains the rise of global temperature. Therefore, any reduction of the atmospheric CO2 flux into the ocean is highly consequential. For a number of northern seas (the North, Labrador, North, Norwegian, Greenland, Barents, and Bering Seas) and the Black Sea we found on a quantitative basis that blooms of a marine microalga called Emiliania huxleyi (a coccolithophore) are capable of appreciably enhancing the partial pressure of carbon dioxide (pCO2) in ocean surface waters within the bloom area. Moreover, for the Back sea we directly prove that the content of CO2 in the atmospheric column above E. huxleyi blooms is indeed elevated (by about 1-2 uatm), and hence the downward CO2 flux into the ocean is at least sensibly weakened if not reversed there. As this phenomenon is driven by enhancement of pCO2 in surface waters that we revealed in all above seas we investigated, the impact of E. huxleyi must be important across the world’s oceans via turning the atmospheric column above E. huxleyi blooms into a recipient of oceanic CO2. Given that E. huxleyi is the most abundant and productive calcifying organism in the world’s oceans growing at subpolar, temperate, subtropical and tropical latitudes in both Hemispheres and in waters with the trophic status in eutrophic to oligotrophic range, and its blooms are very extensive covering marine areas in excess of many hundred thousand, sometimes up to one million, square kilometers, it implies a definite climatic dimension in the overall environmental impact of the phenomenon. with all ensuing consequences for the planet Earth. Importantly, the impact is not solely restricted to the pCO2 aspect but also encompasses the release by this alga of calcites, i.e. particulate inorganic carbon (PIC). We quantified the amounts of PIC within the bloom areas (which were also prequantified) and found for the period 1998-2017 that they varied in the above seas in the range 70 -700 kt. This imparts to the E. huxleyi phenomenon additional importance for the global ocean carbon cycle, and, through intricate interactions, for CO2 exchange between the ocean and the atmosphere.
We express our gratitude for the primary financial support of this study provided by the Russian Science Foundation (RSF) under the project # 17-17-01117.

Area C - Board 256

Operational monitoring of greenhouse gases is a fundamental aspect for future climate monitoring
missions. The current paper provides an insight into a monitoring system developed as an airborne demonstrator for a dedicated CO2 space mission. Here, CO2 distributions shall be observed, with a specific dense sampling and high spatial resolution. This will enable, by inverse modelling, a more accurate reporting of CO2 emissions from fossil fuel combustion.
In the context of the Copernicus programme and its evolution, one of the main objectives is to provide to the European and International Community appropriate means and knowledge for assessing the effectiveness of the Paris Agreement and COP21 decisions, with regard to their impacts on man-made CO2 emissions and overall CO2 budget at country and regional/megacity scales.
On the other hand, the definition studies of the CarbonSat mission, as a candidate mission of Earth Explorer 8, highlighted the difficulty to reach the demanding accuracy, at levels of a fraction of part-per-millions of dry air column averaged CO2 mole fraction (XCO2). Hence, the selection of Earth Explorer 8 ESAC identified the need to consolidate the error budget of XCO2 retrieval by airborne measurements.
Representative high-quality atmospheric spectral images, acquired in a representative imaging configuration, are needed to assist the development of retrieval algorithms, the assessment of mission performances, the understanding of instrument artefacts, the optimisation of the requirements for the space component and the validation of the mission once deployed. As a result, there is a strong need for a high-performance airborne imaging spectrometer, replicating the spectral channels and spectral resolution foreseen for the CO2 space mission.
Airborne Carbon Dioxide Imager for Atmosphere (ACADIA) precursor airborne demonstrator project has been defined by ESA to fulfill such demands, and OHB System AG was selected as a prime contractor of the airborne instrument development and its installation in an experimental aircraft.
The ACADIA instrument is a high-performance airborne imaging spectrometer that shall replicate the spectral channels and spectral resolution foreseen for future CO2 space missions (like CarbonSat), and demonstrate that the demanding accuracies can be met.
The ACADIA system comprises the ACADIA instrument (a fibre-coupled NIR/SWIR imaging spectrometer accommodated in an experimental aircraft), the Calibration Home Base (CHB) for regular instrument calibration before and after each flight cycle and the Processing and Archiving Facility (PAF) for processing and archiving the measurement data.
The ACADIA instrument is planned to be installed in a FA 20 FALCON aircraft.

The paper for the living planet symposium 2019 will give an overview of the ACADIA instrument concept, preliminary design, performance characteristics and operational concepts and will present as well synergies with CO2M mission, for the Phase-A/B study of which OHB is also a prime contractor.

Area C - Board 257

The impact of CO2 on the climate of the Earth's atmosphere is large and widely studied all over the world. One of the largest reservoirs of this greenhouse gas are peatlands and swamps. The Biebrza Wetlands are one of the biggest area covered with marshes, swamps, and wet meadows in Central Europe. The Biebrza Wetlands are flat with average altitude of 105 m above sea level. The main river of this region is Biebrza with mean flow of 35.3 m3s−1, length of 155 km, and water shed of 7051km2. The Biebrza Valley mainly consists of hydrogenic soils such as peat soils in various stages of mouldering. Spongy structure of peat turns the valley into a huge reservoir of fresh water which is restored every summer when Biebrza floods. The process of peat-forming is still active over a reascovered by sedges, reeds, rushes, moss, scrubs, and grasses. The area of the "Czerwone Bagno" nature reserve, part of Biebrza Middle Basin, is particularly interesting due to the area of the marshlands and the flora and fauna original composition. In this part of the Biebrza Wetlands Institute of Geodesy and Cartography has been conducting research since the 90s. Since 2015, accurate measurements of CO2 concentration, using the method of the Eddy Covariance, are also carried out. In addition, the research area is regularly tested using the satellite Sentinel -1/2/3. Using data from Sentinel observation we made a few simple preliminary models, connecting data from space with these obtained form in situ measurements. These models concern, for the time being, the components that have the final effect on CO2 emission and absorption.

Carbon emission and absorption depends on ecosystem and meteorogical conditions. From ecosystem parameters, the most important are soil moisture and biomass changes. From meteorogical ones - solar radiation, air temperature and humidity, precipitations and season. Of course, some of them e.g. season of the year and biomass changes, are naturally linked.

Soil moisture (SM) is important for estimates the vegetation growing conditions. Backscattering coefficient (σ˚) calculated from Sentinel-1 images gives information about humidity of the ground with resolution of 25 or 50 km (Dąbrowska-Zielińska et al., 2016).

A key role in carbon cycle play biomass changes. It was assumed that the vegetation index NDVI derived from Terra MODIS can play the role of vegetation descriptor. The vegetation biomass may be represented by the difference of sigma VH and sigma VV. This assumption has been used following the approach of Rao et al. (2014) - sigma VH - sigma HH can be used as the roughness of the vegetation (in this case NDVI).
Meteorogical parameters are measured by meteorogical station being a part of the Eddy Covariance measuring tower. Data are acquired every 15 minutes, and give information about air temperature, humidity and pressure, solar radiation, precipitations, wind direction and speed. Meteorological data allows to reject data that is unsuitable for analysis (e.g. when cloudiness prevents satellite observation). It gives also information how many energy has been accumulated in ecosystem (solar radiation), and in what condition vegetation may be (is there a drought?, does haymaking or harvesting take place?).

Eddy Covariance tower has been measured the Latent Heat Flux (LE), carbon and water flux, the Sensible Heat (H). The values have been obtained and averaged as hourly, daily values. In general the carbon sequestration was higher with larger values of SM, the highest when SM= 80-90%. During the period of measurements the LE increased obtaining the largest values when SM= 60-70%. We have analyzed the data collected for the years 2015-2018. In the years 2015 and 2018 occurred drought, while 2017 was very humid. In 2016, the humidity of the land was at an average level. In 2017 the emission of carbon due to respiration was almost the same as sequestration of carbon during day time. While in 2015 the emission of carbon due to respiration at night was much higher than during the day time.
Based on satellite data and in situ measurements, we obtained two types of models describing the soil moisture – one based on the NDVI parameter and second based only on satellite data (Dąbrowska-Zielińska et al., 2018).


Acknowledgements

The work is partially supported by the National project Narodowe Centrum Nauki 2016/23/B/ST10/03155.


References

Rao, S., Sahadevan, D., Wadodkar, M., Nagaraju, M., Chattaraj, S., Joseph, W., Rajankar, P., Sengupta, T., Venugopalan, M., Das, S., et al., 2014. Soil Moisture Model with Multi Angle and Multi Polarisation Risat-1 Data, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2,145.

Dabrowska - Zielinska K., Budzynska M., Tomaszewska M., Malinska A., Gatkowska M., Bartold M., Malek I., 2016. Assessment of Carbon Flux and Soil Moisture in Wetlands Applying Sentinel-1 Data, Remote Sensing 2016 8(9) 756; doi:10.3390/rs8090756

Dabrowska-Zielinska, K., Musial, J., Malinska, A., Budzynska, M., Gurdak, R., Kiryla, W., Bartold, M., Grzybowski, P. Soil Moisture in the Biebrza Wetlands Retrieved from Sentinel-1 Imagery. Preprints 2018, 2018100453 (doi: 10.20944/preprints201810.0453.v1).

Area C - Board 258

As a contribution to the Copernicus programme, the European Commission and the European Space Agency investigate the implementation of a future mission to monitor anthropogenic CO2 emissions from space at country, regional and city scales. To this end, measurements of the CO2 vertical column are required with high accuracy (<0.5 ppm) and precision (<0.7 ppm). The current mission design comprises a spectrometer covering the near and shortwave infrared spectral bands at 0.76, 1.6 and 2.0 μm with moderate spectral resolution. This distinguishes the spectral sizing of the instrument from the currently operational OCO-2 and GOSAT missions, which both have a significantly higher spectral resolution. To estimate the effect of this design choice on the quality of the CO2 data product, we apply a spectral smoothing to the OCO-2 measurements to reduce it to the envisaged resolution and compare the retrieved CO2 data product with independent ground-based measurements of TCCON. Although the spectral bands of the spectrometers are chosen to infer aerosol information from the measurement in addition to the atmospheric CO2 abundance, the aerosol induced error represents a significant error source, which is mitigated in current missions by strict data screening and empirical bias corrections. This results in a poor data coverage particular for areas with strong air pollution, like India and China, which are both areas of vital importance to be covered by the CO2 monitoring mission. Therefore, a multi-angle polarimeter is proposed as an auxiliary instrument of the mission to provide aerosol information for improved estimation of the atmospheric light path in the different spectral bands and so to improve coverage and quality of the CO2 data product. We demonstrate the mission performance including the gain in data quality and coverage by a synergistic use of both the CO2 spectrometer and the multi-angle polarimeter. For this purpose, we analyse generic cases as well as ensembles of simulated measurements for Central Europe and China. We conclude that this unique combination of the proposed state-of-the-art sensing concepts will be of great benefit to achieve the challenging mission objectives of the future European CO2 monitoring mission.

Area C - Board 259

The exchange of CO2 between the ocean and the atmosphere is a critical to both global climate and the biogeochemistry of the oceans. Gas exchange between ocean and atmosphere results in a net sink of atmospheric CO2. By taking up atmospheric carbon, the oceans mitigate the effect of increasing emissions on global climate, a process that leads to ocean acidification and its impact on the marine environment.
Understanding these processes relies on international monitoring programs such as the Surface Ocean CO₂ Atlas (SOCAT). In order to support these programs, common tools can provide consolidated, validated outputs to improve understanding of ocean-atmosphere gas exchange. These tools help to remove barriers to international collaboration and bridge across different disciplines, which accelerates advances in research and understanding.
The FluxEngine is a freely available, open-source toolbox for the calculation of ocean-atmosphere gas flux. It is designed to combine satellite Earth observation data with in-situ, model or large-scale monitoring. For global-scale estimates, the toolbox uses SOCAT data and satellite Earth observation to provide a regular, standardised and validated global data set for ocean-atmosphere CO2 flux. The toolbox also offers the user complete control over the input data, the geographical areas and the parameterisations used to calculate flux. As such, it can support in-situ data collection, or experiments at finer spatial and temporal resolution. For example, gas fluxes along cruise tracks can be calculated from data gathered on board (fig. 1), with other data sources easily incorporated to fill gaps and enable gas-flux calculations where they would not otherwise be possible. The ability to interchange data sets and methodologies allows easy interpretation of differences and simplifies ensemble analyses. As such, the system is particularly suited to investigating new laboratory or in-situ research, which can be immediately scaled to quantify the effects of new knowledge on basin or global-scale carbon sink estimates. Examples will be provided of how FluxEngine is progressing research, including demonstrations of how the community can use the FluxEngine and access standardised outputs.
The FluxEngine was originally developed within the ESA OceanFlux series of projects and it is now being developed further and exploited within two European Union Horizons 2020 Integrated Carbon Observing System readiness projects (Ringo and Integral) and it has been used within undergraduate teaching at the University of Exeter and Utrecht University.

Area C - Board 260

It is well know that Arctic Ocean is a sink for CO2 within all year with seasonal variability and there are a lot of speculations over the future trend of the carbon flux. The research hypothesis was to determine that the air-sea CO2 fluxes depend largely on the difference in partial pressure of CO2 (pCO2) in seawater and air than on gas transfer velocity. For this we used Artifical Neural Networks (ANN's), a powerful non-linear modeling tool for mapping performance, and tested those methods for so spare and difficult area-which is the Arctic Ocean. The ANN's were first used 20-30 years ago (Kohonen, 2001) and since then there has been a significant increase in their application to solving environmental problems. They are now commonly used in atmospheric, oceanography, and meteorology sciences.
While atmospheric pCO2 shows relative homogeneity, marine pCO2 varies strongly both temporarily and spatially, wherefore proving the hypothesis required achievement of the objective that was: estimate of marine pCO2 distribution dependent on the sea surface temperature, sea surface salinity, biological activity, gas transfer velocity, and/or wind speed. The analysis was based on data from 2013-2017 year collected from two sources: publicity available satellite observation (like SOCAT, ESA/GlobWave Altimeret, NCEP/NCAR reanalysis, GlobColour GSM), and archives of Air-Sea Interaction Laboratory IO PAS (in-situ observation using LI-COR 7550 obtained during 5 Arctic cruises operated by Air-Sea Interaction Laboratory IO PAS, from r/v Oceania).
High uncertainty about the size of the sink/source of CO2 in the Arctic Ocean reflects the storage of coordinated in situ measurements and difficulties in providing logistic support in a hostile environmental, wherefore the coverage of in sity pCO2 measurements in the North Atlantic, still remains unevenly distributed in time and space. A potential alternative solution is to provide Earth Observation data. The Arctic Ocean is one of the strongest sinks for CO2 in the world’s ocean throughout the year with seasonal variability, and so plays an important role in the carbon cycle. Despite of that, there are a lot of speculations over future trend of the carbon dioxide. Estimates of the current CO2 flux, for this region, still are burden with high uncertainty, given that there are competitive processes to decreasing of increasing the rate of exchange. The magnitude of the ocean sink can be determine using air-sea flux estimates based on satellite observation as well as in situ measurements.

Area C - Board 261

Cities and large industrial sites account for about 70% of the total CO2 emission from fossil fuel. A growing number of cities and regions being engaged in climate change mitigation have disclosed their plans to reduce greenhouse gas emissions. Atmospheric inversions can use CO2 measurements to provide an independent way of monitoring the fossil fuel CO2 emissions and to evaluate effectiveness of the emission reduction pledges. Measurements of vertically integrated columns of dry air mole fractions of CO2 (XCO2) from satellites offer the advantage of a dense set of observations with global spatial coverage to monitor the fossil fuel CO2 emissions at the scales of large cities, industrial sites, regions and countries that are relevant to the carbon policy. In this study, we first identified 11314 targets of emitting clumps (defined as clusters of adjacent emitting area) covering major cities and large point sources all over the globe and representing about 75% of total CO2 emissions. This study assesses the potential of the XCO2 imagery from one to multiple low earth orbiting satellites for monitoring the fossil fuel CO2 emissions from these clumps. Such imagery could be provided by future space borne passive spectral-imagers in the short wave infrared spectrum, at a high spatial resolution of 2 km x 2 km with a swath width of 300 kilometers and with individual sounding precision of less than 1 ppm. This assessment was based on Observing System Simulation Experiments (OSSEs) with an atmospheric inversion approach. The inversion system solved for the average emissions before a given satellite overpass, using a Gaussian plume model to link the average emissions and the XCO2. The inversion framework accounted for the existence of some prior knowledge about the emissions before the satellite overpasses with 30% uncertainty for annual emission of each clump. The results show that emissions from small clumps (annual emissions smaller than 0.5 Mton per year) can hardly be constrained by satellite observations. With one imager, the CO2 plumes generated by large emission clumps could be sampled by satellite observations for 20-50 days during one year. North America, South America, and Africa are more effectively sampled than Middle East and Asia. When aggregating the emissions at national and annual scales, the decrease of uncertainties largely relies on the spatial and temporal correlations in the prior uncertainty in the emissions. The posterior uncertainty in annual emission for a clump (whose annual emission is larger than 0.5 Mton C per year) can be reduced to 20%, given intermediate temporal correlation. The posterior uncertainty can be even reduced to 10% given the strong temporal correlation. The satellite observations have limited ability to improve the national annual emissions, unless one can assume strong spatial correlations in the prior uncertainties.

Area C - Board 262

Over the past two decades, a great number of space-borne missions, airborne and ground-based campaigns have been providing observations of the Earth’s atmosphere for the vertical profiling of atmospheric variables aiming at ensuring global and continuous measurements of atmospheric species. These measurements are also used as input to the physical and chemical models that are used to predict the evolution of the atmospheric state. When two or more instruments sound the same portion of atmosphere either in different spectral regions or with different observation geometries, their measurements can be combined in order to obtain a single vertical profile of improved quality with respect to that of the profiles retrieved from the single observations. In the framework of the increasing interest in developing innovative techniques to exploit all the available information from measurements of the same portion of the atmosphere to retrieve the best vertical profile estimate, a new method of data fusion was proposed: The Complete Data Fusion (CDF). This method uses standard retrieval products to combine independent measurements of the same profile into a single estimate for a comprehensive and concise description of the atmospheric state. The CDF is an a posteriori algorithm and requires very simple implementation.
In order to determine simultaneously atmospheric constituents reducing the systematic error caused by interfering species, multi-target retrievals are frequently applied to the analysis of remote sensing observations. It is thus crucial to adapt the CDF algorithm to fuse profiles obtained from multi-target retrievals, extending its use to a greater number of remote sensing data.
In this work, we present the results of the first application of the complete data fusion to multi-target retrieval products showing how the inputs of the CDF have to be modified to take into account that the state vectors of the fusing measurements may contain only the same atmospheric variables or include different variables as well. We applied the method to simulated measurements in the thermal infrared and in the far infrared spectral ranges, considering the instrumental specifications and performances of IASI-NG and FORUM instruments, respectively. The results obtained demonstrate that the CDF can deal with state vectors from multi-target retrievals both when they contain the same variables and when they have only a subset of variables in common, providing outputs of improved quality with respect to the input data.

Area C - Board 263

Following the 21st Conference Of the Parties (COP 21) and the adoption of the Paris Agreement, the MicroCarb mission was officially announced by the French government. The objective is to remotely measure, from a microsatellite Myriade platform, carbon dioxide atmospheric volume mixing ratios in order to characterize CO2 surface fluxes. This mission is one of the major projects in development phase led by the French Space Agency (CNES) in partnership with the United Kingdom Space Agency (UKSA); the launch is scheduled in 2021.
In the current context of climate change acceleration, understanding the carbon cycle is critical since it can help the science community to anticipate its evolution according to various possible climate change scenarios. Thus MicroCarb aims to be a precursor of a future operational system able to accurately monitor global fossil emissions.

This article first gives a global overview of the MicroCarb mission in the CO2 worldwide context, in particular the need for measurement precision and the presentation of the different mission contributors. The scientific objectives are then detailed to explain the hypotheses on which the MicroCarb project is built, before focussing on the MicroCarb mission centre definition.

CNES is indeed responsible for specifying and developing the MicroCarb mission ground segment, called PLGS (PayLoad Ground System). As its main activities are the control of the satellite and instrument, as well as the science data processing, it is a critical component for guaranteeing the success of MicroCarb project. This mission centre will process MicroCarb acquisitions from raw data to Level 3 products (monthly average maps of CO2) while the generation of Level 4 products (CO2 surface fluxes) will be computed outside of the PLGS by the scientific laboratories.
Definition of the data processing chains shows thereby that a large IT infrastructure is required to meet products availability need for the science community and to deal with the large volume of data planned to be produced during the lifetime mission. This article explains and describes MicroCarb major IT needs with a first estimation of its mission centre sizing.

Area C - Board 264

In collaboration with the research community and with close European partnerships, the French space agency, CNES is developing or co–developing two missions MICROCARB and MERLIN, that are dedicated respectively to the measurement of atmospheric concentrations of CO2 and CH4 in order to access surface carbon fluxes.

The MERLIN (MEthane Remote sensing LIdar missioN ) space segment consists of the new platform product line named MYRIADE Evolutions (range of 400 kg) developed under CNES control, and the first IPDA (Integrated Path Differential Absorption) LIDAR (Light Detecting And Ranging) instrument under DLR responsibility. This technique enables measurements in all seasons, at all latitudes and has the advantage of an extremely low level of systematic error on the dry-air column mixing ratio of CH4 thanks to the auto-calibration of the signal. The main science objective is to bring a significant improvement on the retrieval of CH4 fluxes which requires 1% accuracy on CH4 column averaged dry-mixing ratio, at a 50 km horizontal resolution.

The MICROCARB mission is based on a compact grating spectrometer (around 60kg) onboard a classical Myriade plateforme (170kg range). The instrument will measure the brilliance in four spectral ranges including, for the first time, an O2 band around 1.27 µm to get a better correction of aerosol signal. The objective is to measure column concentration of CO2 with a random accuracy better than 1 ppm and a biais better than 0,1ppm for each individual pixel (around 40 km2). Some experimental mode will be possible to characterize local emissions at megacity scale using smaller pixels (2km*2 km) by slowing down the instrumental scanning.

CNES is also engaged in the IASI and the IASI–NG programs and support research activities to deliver high quality CO2 and CH4 products in the troposphere.
These space missions are based on innovative instrumentation and need development of specific and innovative approaches for algorithm inversion, calibration and validation purposes as well as use of data for science, societal application and benefits.
The overview presents the French strategy including also balloon, aircraft and ground-based instrumentations development and organization of campaigns supported by CNES as well as adaptation of tools mainly based on different existing of new models. To build this national strategy, a specific care is made to keep of strong spirit of international coordination with space and scientific initiatives worldwide relative to carbon budget.

Area C - Board 265

Remote sensing measurements of methane and carbon dioxide have been performed at Sodankylä in northern Finland since early 2009, using a Fourier Transform Spectrometer (FTS). The instrument records high-resolution solar spectra in the near-infrared spectral region. From the measured spectra column-averaged abundances of methane, carbon dioxide and other gases are derived.The instrument at Sodankylä participates in the Total Carbon Column Observing Network (TCCON). The TCCON measurements have been widely used for carbon cycle studies and for the validation of satellite based observations. The relevant satellite missions include the TROPOspheric Monitoring Instrument (TROPOMI) on board of the Copernicus Sentinel-5 Precursor satellite; the SCanning Imaging Absorption spectroMeter for Atmospheric CHartographY (SCIAMACHY); the Orbiting Carbon Observatory-2 (OCO-2); the Greenhouse Gases Observing Satellite (GOSAT) and TanSat.

Here we report ground based FTS retrievals at the Sodankylä TCCON site during years 2009-2018 and comparisons with the satellite borne observations. In order to guarantee the optimal performance of the FTS, the optical alignment has been checked and adjusted on regular basis. In addition we have monitored the instrument line shape by taking HCl reference gas measurements. We find that the derived modulation efficiency at maximum optical path difference has remained relatively stable over time, indicating that the alignment has been maintained. Comparisons with the GOSAT measurements since 2009 show good agreement. The relative difference in XCH4 has been -0.07 ± 0.02 % and the relative difference in XCO2 has been 0.04 ± 0.02 %. Finally, in situ AirCore observations in the vicinity of the TCCON instrument at Sodankylä are compared to the TCCON retrieval results. The AirCore profiling in this study has focused on accurate measurements of methane, carbon dioxide and carbon monoxide. In Sodankylä we have performed AirCore measurements during all seasons. The novel AirCore method allows sampling both in the stratosphere and troposphere, which is a benefit, compared to the aircraft in situ measurements.

Area C - Board 266

Wetlands are the largest natural source of methane; approximately 30% of the total methane emissions are from wetlands (Saunois et al., 2016). Based on current knowledge, most of these emissions originate from tropical wetlands, but the wetland emissions from the rapidly changing Arctic and boreal regions involve open questions and considerable uncertainties.The methane emissions from a specific wetland depend on the soil properties and conditions, for example, soil moisture and the temperature of the ground. High seasonal temperature variations and snow cover over frozen ground are common features of the Arctic areas. Frost and snow have both direct and indirect effects on the wetlands as a methane source, and this has been previously studied and analyzed mainly with in-situ and ground-based measurements; for example, Mastepanov et al. (2013) showed with chamber measurements that there is a clear correlation between the timing of snow melt and rising methane concentration. In addition, Helbig et al. (2017) showed with flux and meteorological in-situ measurements that during the spring, after snowmelt, the warmer the soil temperature is, the more methane is emitted from the ground.

We study the dependencies of the seasonal cycles of frost, snow and column-averaged methane, all based on satellite observations for the time period from summer 2010 to summer 2018. The column-averaged methane (XCH4) observations from the Greenhouse Gases Observing Satellite (GOSAT) and the Tropospheric Monitoring Instrument (TROPOMI) onboard Copernicus Sentinel-5 Precursor Satellite will be used as methane data sets. To detect the soil freezing, we use the soil freeze-thaw (F/T) product which applies observations from the European Space Agency’s (ESA) Soil Moisture and Ocean Salinity (SMOS) satellite (Rautiainen et al., 2016). Snow extent and snow properties will be studied based on ESA GlobSnow snow extent (SE) and snow water equivalent (SWE) data sets (Globsnow, 9.11.2018). The results will show, for the first time, the connection of seasonal snow and frost to the seasonal cycle of methane in a larger scale over the remote boreal and Arctic regions which are not thoroughly covered with in-situ measurements.

Saunois et al., Earth Syst. Sci. Data, 8, 697–751, doi:10.5194/essd-8-697-2016, 2016.
Mastepanov et al., Biogeosciences, 10, 5139–5158, doi: 10.5194/bg-10-5139-2013, 2013.
Helbig et al., Environ. Res. Lett., 12, 115009, doi: 10.1088/1748-9326/aa8c85/, 2017.
Rautiainen et al.: SMOS prototype algorithm for detecting autumn soil freezing, Remote Sens. Environ, 180, 346-360, 2016.
Finnish Meteorological Institute and ESA, GlobSnow, http://www.globsnow.info/index.php?page=Home, 9.11.2018.

Area C - Board 267

The variability in atmospheric methane concentration is mostly described by the seasonal cycle that is profoundly connected to the sources and sinks of this potent greenhouse gas. In this study, we evaluate and study the patterns of the trend and the average seasonal cycle of the column-averaged dry air mole fraction of methane (XCH₄), retrieved by three different Greenhouse Gases Observing Satellite (GOSAT) XCH₄ retrieval algorithms (NIES, RemoTeC Full Physics, RemoTeC Proxy). The evaluation is made against the ground-based Total Carbon Column Observing Network (TCCON) retrievals at 15 sites across the world, from 2009 to 2015. We used, for the first time for methane observations, a dynamic time series method to fit for the seasonal cycle and a non-linear trend. In addition, the dynamic time series method allows us to study the uncertainties of the trend and the seasonal cycle. The method models the average seasonal cycle with harmonic functions.

We study the seasonal cycle by evaluating the peak-to-peak amplitude and the timing of the minimum and maximum day of the XCH₄ seasonal cycle. Our results show that if the temporal coverage of the observations throughout the year is sufficiently large, the seasonal cycle of methane can be reliably quantified using satellite observations. The peak-to-peak amplitude is captured at the majority of the sites and for most retrieval algorithms within the uncertainty ranges, and the maximum day of the seasonal cycle is captured better than the day of the minimum.

In addition to the evaluation and analysis at the TCCON sites, we study the latitudinal dependence of the seasonal cycle of GOSAT XCH₄ by dividing the latitudes from 53° N to 44° S into 31 latitude bands. In the latitude bands, we also compare the trend of methane to the trend from NOAA’s Marine Boundary Layer Reference data, and we find that the growth rates mostly agree in the Southern hemisphere but not at the lower latitudes in the Northern hemisphere, where the differences follow a systematic pattern likely affected by atmospheric dynamics

Area C - Board 268

The RAL Remote Sensing Group has developed an optimal estimation scheme to retrieve global height-resolved information on methane from IASI using the 7.9 µm band and produced a v1.0 mission data set for MetOp-A (R. Siddans et al., 2017, https://doi.org/10.5194/amt-10-4135-2017). This scheme has subsequently been improved through use of pre-retrieved temperature, water vapour and surface spectral emissivity from the RAL Infrared Microwave Sounder (IMS) retrieval scheme based on collocated data from IASI, MHS and AMSU. The improved methane retrieval scheme has been used to reprocess the IASI MetOp-A record, producing a 10-year interim dataset.

While providing information on two independent vertical layers in the troposphere, sensitivity in the 7.9 µm band decreases towards the ground, due to decreasing thermal contrast between the atmosphere and surface. Addition of the 3.7 µm band of IASI, which detects a surface-reflected solar component on the dayside, in addition to terrestrial thermal emission, is currently under investigation to help mitigate this. A combined retrieval scheme exploiting in addition the high signal-to-noise information from Sentinel-5P (SWIR/column) with that from IASI or CrIS (TIR/height-resolved) would enable lower tropospheric distributions of methane and CO to be resolved. Lower tropospheric concentrations are more closely related to emission sources than are column measurements and inverse modelling of surface fluxes should be less sensitive to errors in representation of transport at higher altitudes; a limiting factor for current schemes.

In this paper, we present a 10-year interim dataset, based on the improved 7.9 micron IASI methane scheme, and initial progress on the ESA Living Planet Fellowship project ‘Production of lower tropospheric methane and carbon monoxide distributions through combined use of ESA Sentinel-5 Precursor shortwave infrared and IASI/CrIS thermal infrared satellite data (PROMCOM)’. An update is also provided on the status of IASI methane processing via the RAL-RSG MetOp near real time chain.

Area C - Board 269

Copernicus Sentinel-5P (S5P) launched in October 2017 with its Tropospheric Monitoring Instrument (TROPOMI), and Sentinel 5 (S5) due for launch in 2022 with its Ultraviolet, Visible, Near infrared, Shortwave infrared (UVNS) instrument will enable daily global measurements of methane (and other atmospheric trace gases). Both instruments exploit the spectral range 2305-2385 nm in the Short Wave InfraRed (SWIR) to retrieve CH4 total column information. This spectral range has only been used to a limited degree in space-based instruments up until this point, with SCIAMACHY on ENVISAT and NASA’s MOPITT sensitive to this range (for which exploitation was limited due to issues on both satellite instruments). GOSAT-2 due for launch in late October 2018 will also be sensitive to this spectral range.
This new generation of instruments will therefore be relying on spectroscopy that has not been assessed to a significant degree by space based instruments (with Galli et al., (2012) one of the few examples), especially at the high spectral resolutions employed (0.45 cm-1 TROPOMI/UVNS and 0.2 cm-1 GOSAT-2/TANSO-FTS-2). This extends to the main validation source for these instruments, the Total Carbon Column Observing Network (TCCON).
In this study, we assess the systematic biases in TCCON retrievals of the two main methane isotopologues 12CH4 and 13CH4. Retrievals are made in the TROPOMI spectral band, and the traditional TCCON methane spectral bands using the TCCON GGG retrieval environment. Four separate spectroscopy databases are used for this study, the TCCON spectroscopy database; the HITRAN2016 database (Gordon et al., 2017); the GEISA2015 database (Jacquinet-Husson et al., 2016); and the ESA SEOM-IAS database for TROPOMI (Birk et al., 2017), which has been extended to allow for non-Voigt methane line shape parameters, while the remainder do not currently include non-Voigt parameters. Spectra are used from two TCCON sites, Ascension Island (for background variations) and Tsukuba, Japan (for variable seasons).
We find notable differences in the retrieved volume mixing ratios of methane isotopologues for both TCCON sites, with significant variations between the spectroscopic databases. These vary in magnitude depending on the spectral band under investigation (in some cases up to 5% for 12CH4), but are persistent throughout the day, and across different sites. In general, however, there is more bias between the spectral bands, rather than the spectroscopy databases (but not in all cases). 13CH4 is particularly challenging, with large variations in the retrieval quality both in spectral bands and databases (up to 20% retrieval volume mixing ratio bias), and non-uniformity between the 2300 nm and the 1650 nm spectral ranges.

Area C - Board 270

Greenhouse gas (GHG) measurements are of increasing importance to understand the evolving carbon cycle in the Anthropocene and for policy makers aiming to limit manmade climate change. Based on the research in the development of the Earth Explorer 8 CarbonSat Phase A/B1 activities and in preparation for a forthcoming CO2 monitoring mission within the European COPERNICUS program, an end-to-end (E2E) mission performance simulation system for GHG total column products has been successfully developed, implemented, tested and used. This mission E2E simulator is able to reproduce all significant processes, design and steps that impact the mission performance as well as output simulated data products in the aim to become a coherent test bed for L1PP and L2PP and to support the verification of space segment performance and associated sensitivity analysis.
The GHG total column performance simulator (GHG-TCPS) comprises geometry, scene, orbit, instrument, Level 1b processing and up to Level 2 simulation modules, all embedded in an OpenSF framework infrastructure. The GHG-TCPS aims at deepening the understanding of the expected performance of a GHG observation mission focusing on CO2 and CH4 based on detailed observation system simulations (OSS). It includes the simulation of realistic large-scale full swath 2D scenes, which requires sophisticated instrument and retrieval simulation modules to be implemented in a consistent way.
The GHG-TCPS system has been used to run a series of simulations to estimate the performance of a GHG imaging spectrometer system to measure CO2 and CH4 total column content for extended areas, city scale and point scale plumes.
In this study we describe briefly the GHG end-to-end performance simulation system and present results from simulations using realistic atmospheric scenarios. These include examples of the application for flux inversion for selected source regions. These results enable the performance of the future CO2 observing system to be assessed realistically up to Level 4 data products.

Area C - Board 271

Temperature regime is one of the main controlling factors of greenhouse gas (GHG) emissions from northern peatlands, which contain one-third of global soil carbon. Remotely sensed land surface temperature (LST) has a potential to become an efficient instrument in environmental monitoring of carbon dioxide and methane emissions from peat bogs, particularly from peat bogs, which possess a peat soil.
Here we examine the relationships between field-measured hydrometeorological variables and MODIS LST data in a hemiboreal raised bog within three different temporal frameworks: 1)annual (2008-2016); 2) monthly (from May till September) over the 2008-2016 years; 3) study-period average. Data for nine years for the period from May till September were available, representing a wide range of environmental conditions of the growing season.
The Tooma mire research station provided us with the hydrometeorological data for 2008 – 2016 years. Air temperature (TAir) was measured at a height of 2 meters every 3 hours. Precipitation (Prec) was measured once a day (Table 1). Water vapour partial pressure (WVPP) was measured every 3 hours. Soil temperature (TSoil) at depths of 5, 10, 15, 20 cm together with surface temperature (TSurf) were measured every 6 hours. We used water-level data from the WL 212 bog pool as well as in the soil at the location of WL 323. That was close to the woodland edge of the bog while WL 212 was located in the central part of the bog with sparse cover of dwarf pines.
The MODIS product MOD11A1 from the sixth collection at a 1 km spatial resolution provides daily long-term measurements of LST and emissivity. We used the LST and quality control (QC) scientific data sets of the LST product. The scientific data-set values were converted to real LST value in K using the coefficient of the calibration equal to 0.02. Subsequently the Kelvin values were converted to Celsius degrees. QC information was also provided by AρρEEARS and it was used to determine usefulness of the LST data.
The Pearson product-moment correlation was used to reveal the relationship between the field-measured parameters and LST over years and months. A multiple linear regression was chosen to model relationships between the hydrometeorological variables and LST by month.
The results revealed non-uniform relationships between LST and the field-measured hydrometeorological variables. The study found variation in correlation coefficients between LST and the field-measured hydrometeorological variables between years and months, while the correlation between LST and air temperature was the highest during different months and for the entire study period.
WVPP, cumulated precipitation and WL had a cooling influence on LST. That was more noticeable for the upper peat layers. In addition, deep WL in the peat increased the monthly correlation between LST and TAir. The linear models with hydrometeorological variables could explain from 67% (in June) to 81% (in August) of variability in LST. TAir was the main factor of monthly changes of LST and could explain from 63% to 78% of LST variation.
The spatial resolution used in this study is too coarse to consider the site-type relationships of LST and hydrometeorological variables. However, we do provide evidence that MODIS LST can be associated with the hydrometeorological conditions of a hemiboreal bog, which in turn determine ecosystem respiration. Further, the relationships might be implemented in parametrization of temperature conditions and in practice of assessment and modelling of GHG emissions from northern raised bogs.

Area C - Board 272

Satellite missions measuring atmospheric gases such as greenhouse gases (GHG) and other pollutants require ground-based measurements for validation. Ground measurements using the Sun as a target can reach much higher SNR on spectra than those obtained from orbit. This higher SNR translates to more precise gas concentration retrieval and allow peaking deeper into the list of trace gases. Ground networks have sprung to take advantage of those conditions (e.g., NDACC, TCCON among others). However instruments high cost and cumbersomeness have slowed the expansion of these networks over the globe.

ABB has developed an affordable field FT-IR instrument based on its rugged industrial line of process spectrometer referred to as the TALYS series. The spectral range covers the VIS-NIR-SWIR region from 4000 to 16500 cm-1 (0.6-2.5 um) with a 0.3 cm-1 spectral sampling (maximum optical path difference of +/–1.67 cm) offering more than 40 000 spectral channels. A variant covering the mid infrared region (MIR) is also under preparation. The instrument uses the Sun as a source and tracks it down to 1/600 of its diameter to provide a high stability of the instrument line shape.

This paper will present the instrument and initial results including SNR, tracking performance and instrument line shape (ILS).

Area C - Board 273

The Emissions Database for Global Atmospheric Research (EDGAR) estimates the emissions of air pollutants and green-house gases for all world countries, covering the period 1970 till nowadays. EDGAR combines activity data retrieved from international statistics (e.g. fuel consumption, production processes, agriculture-related activities, etc.) with technology based emission factors and spatial proxies to deliver spatio-temporally disaggregated emissions that are used to inform policy as well as to support scientific research worldwide.
Among the wealth of data used to build the emissions in EDGAR, remote sensing and satellite retrieval of Earth imaging certainly have and continue to play a key role in the continue development and outreach of EDGAR. A few examples, among the many, include:
• mapping the household emissions using the latest product of the Global Human Settlements Layers which is based on advanced population census data (CIESIN GPWv4) and satellite imagery (Landsat 8) (Figure 1);
• developing a global emission inventory for the Land Use, Land Use Change and Forestry (LULUCF) sector;
• emissions from flaring in EDGAR are based on gas flaring estimates per country in satellite observations of flaring lights (NOAA data);
• Inverse modelling activities, e.g. within the Monitoring Atmospheric Composition and Climate (MACC) consortium; and others.
In the perspective of widening its operating space, EDGAR is now expanding its involvement in the remote sensing arena, and satellite in particular. EDGAR is developing a new methodology to provide consistent emission and removal estimates for the LULUCF sector by combining statistical datasets with land cover maps produced within the European Space Agency (ESA) Climate Change Initiative. A particular attention will be put on China using remote sensing to monitor combustion and non-combustion emission sources.
The return to the satellite community of using EDGAR’s products lies, e.g., in the perspective of providing more insights on the composition of the total emissions and source decomposition at very fine spatial resolution. EDGAR can in fact provide emission sector shares at grid-cell level to be used, e.g., for the analysis of different sector emission shares for urban and rural locations, or for different countries, and to compare to the sector composition of total column measurements performed e.g. by satellites.
Moreover, EDGAR is now improving the temporal disaggregation of the emissions, moving from annual estimates down to monthly and hourly resolved data which will allow monitoring acute episodes of pollution and comparing bottom-up with top-down estimates, also in this case enabling a two-way exchange with the information collected through satellite imaging.
The presentation will provide an overview of the several advancements already incorporated in the EDGAR system as a direct use of satellite imaging as well as showing the potentiality of fruitful developments.


CAPTION
Figure 1 – Comparison of EDGAR CO emissions from combustion in the residential sector in 2012 using the old population proxy data and the newly developed proxies based on the Global Human Settlements Layers.

Area C - Board 274

The design and implementation of the CO2 monitoring (CO2M) satellite as one of the candidate missions of the Sentinel expansion series are currently being studied. Scattering due to aerosols and cirrus modifies the light path of the backscattered near-infrared sunlight and this has been known to introduce one of the dominant sources of error in the retrieved XCO2. To address this concern, a multi-angle polarimeter (MAP) is envisaged to be a part of the CO2M mission to support the XCO2 retrieval. Such an instrument represents a cutting-edge approach to aerosol remote sensing; it collects the necessary information on aerosol properties to correct the light path. In this way, MAP complements the main spectrometer observations and is expected to offer a substantial improvement on the XCO2 product accuracy.

We developed an algorithm to retrieve XCO2 and aerosol information using a spectrometer and MAP simultaneously. Here, the measurement vector includes both the intensity and degree of linear polarization measurements at multiple viewing angles and at multiple wavelengths from the MAP, as well as the radiance spectra in the near infrared (O2 A-band) and in two short-wave infrared bands (weak and strong CO2 bands) at a single viewing angle from the spectrometer. The state vector includes the height of aerosol layers, aerosol microphysical and optical properties (fraction of spheres, refractive index, size distribution parameters, optical thickness), and ultimately the total column concentration of CO2.

Using synthetic measurements, we demonstrate that MAP provides a considerable improvement in the XCO2 accuracy compared to retrievals using spectrometer alone, making XCO2 retrievals possible within the required accuracy (about 0.1% or roughly 0.4 ppm) for scenes with moderate to high aerosol loading. In addition, we also discuss the different aspects of MAP observations (e.g. spectral bands, number of viewing angles, polarimetric accuracy) that are important in the context of XCO2 retrievals.

Area C - Board 275