Documentation - CTE2015
Biosphere Oceans Observations Fires Fossil Fuel TM5 Nested Model Assimilation
To learn more about a CarbonTracker component, click on one of the above images.
Or download the full PDF version for convenience.
TM5 Nested Transport [go to top]
1.   Introduction
The link between observations of CO2 in the atmosphere and the exchange of CO2 at the Earth's surface is transport in the atmosphere: storm systems, cloud complexes, and weather of all sorts cause winds that transport CO2 around the world. As a result, local events like fires, forest growth, and ocean upwelling can have impacts at remote locations. To simulate the winds and the weather, CarbonTracker uses sophisticated numerical models that are driven by the daily weather forecasts from the specialized meteorological centers of the world. Since CO2 does not decay or react in the lower atmosphere, the influence of emissions and uptake in locations such as North America and Europe are ultimately seen in our measurements even at the South Pole! Getting the transport of CO2 just right is an enormous challenge, and costs us almost 90% of the computer resources for CarbonTracker. To represent the atmospheric transport, we use the Transport Model 5 (TM5). This is a community-supported model whose development is shared among many scientific groups with different areas of expertise. TM5 is used for many applications other than CarbonTracker, including forecasting air-quality, studying the dispersion of aerosols in the tropics, tracking biomass burning plumes, and predicting pollution levels that future generations might have to deal with.

2.   Detailed Description
TM5 is a global model with two-way nested grids; regions for which high-resolution simulations are desired can be nested in a coarser grid spanning the global domain. The advantage to this approach is that transport simulations can be performed with a regional focus without the need for boundary conditions from other models. Further, this approach allows measurements outside the "zoom" domain to constrain regional fluxes in the data assimilation, and ensures that regional estimates are consistent with global constraints. TM5 is based on the predecessor model TM3, with improvements in the advection scheme, vertical diffusion parameterization, and meteorological preprocessing of the wind fields (Krol et al., 2005). The model is developed and maintained jointly by the Institute for Marine and Atmospheric Research Utrecht (IMAU, The Netherlands), the Joint Research Centre (JRC, Italy), the Royal Netherlands Meteorological Institute (KNMI, The Netherlands), and NOAA ESRL (USA). In CarbonTracker, TM5 separately simulates advection, convection (deep and shallow), and vertical diffusion in the planetary boundary layer and free troposphere.

The winds which drive TM5 come from the European Center for Medium range Weather Forecast (ECMWF) operational forecast model. This "parent" model currently runs with ~25 km horizontal resolution and 25 layers in the vertical. The carbon dioxide levels predicted by CarbonTracker do not feed back onto these predictions of winds. In contrast to earlier verions of CarbonTracker, we currently use the convection fields directly from ECMWF (whereas before they were calculated using the Tiedtke convection scheme).

For use in TM5, the ECMWF meteorological data are preprocessed into coarser grids. In CarbonTracker Europe, TM5 is run at a global 3x2 degrees resolution with nested regions over Europe (1x1 degrees) and North America (1x1 degree). The grid over Europe is shown in the figure. TM5 runs at an external time step of three hours, but due to the symmetrical operator splitting and the refined resolution in nested grids, processes at the finest scale are repeated every 10 minutes. The vertical resolution of TM5 in CarbonTracker Europe is 25 hybrid sigma-pressure levels, unevenly spaced with more levels near the surface. Approximate heights of the mid-levels (in meters, with a surface pressure of 1012 hPa) are:

LevelHeight (m)LevelHeight (m)
134.5149076.6
2111.91510533.3
3256.91612108.3
4490.41713874.2
5826.41815860.1
61274.11918093.2
71839.02020590.0
82524.02124247.3
93329.92229859.6
104255.62335695.0
115298.52442551.5
126453.82580000.0
137715.4

3.   Further Reading

Oceans Module [go to top]
1.   Introduction
The oceans play an important role in the Earth's carbon cycle. They are the largest long-term sink for carbon and have an enormous capacity to store and redistribute CO2 within the system. Oceanographers estimate that about 48% of the CO2 from fossil fuel burning has been absorbed by the ocean [Sabine et al., 2004]. The dissolution of CO2 in seawater shifts the balance of the ocean carbonate equilibrium towards a more acidic state (i.e., with a lower pH). This effect is already measurable [Caldeira and Wickett, 2003], and is expected to become an acute challenge to shell-forming organisms over the coming decades and centuries. Although the oceans as a whole have been a relatively steady net carbon sink, CO2 can also come out of the oceans depending on local temperatures, biological activity, wind speeds, and ocean circulation. These processes are all considered in CarbonTracker, since they can have significant effects on the ocean sink. Improved estimates of the air-sea exchange of carbon in turn help us to understand variability of both the atmospheric burden of CO2 and terrestrial carbon exchange.

2.   Detailed Description
Oceanic uptake of CO2 in CarbonTracker is computed using air-sea differences in partial pressure of CO2 inferred from ocean inversions, combined with a gas transfer velocity computed from wind speeds in the atmospheric transport model.

The long-term mean air-sea fluxes, and the uncertainties associated with them, derive from the ocean interior inversions reported in Jacobson et al. [2007]. These ocean inversion flux (OIF) estimates are composed of separate preindustrial (natural) and anthropogenic flux inversions based on the methods described in Gloor et al. [2003] and biogeochemical interpretations of Gruber, Sarmiento, and Stocker [1996]. The uptake of anthropogenic CO2 by the ocean is assumed to increase in proportion to atmospheric CO2 levels, consistent with estimates from ocean carbon models.

For CarbonTracker Europe, contemporary pCO2 fields were computed by summing the preindustrial and anthropogenic flux components from inversions using five different configurations of the Princeton/GFDL MOM3 ocean general circulation model [Pacanowski and Gnanadesikan, 1998], then dividing by a gas transfer velocity computed from the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA40 reanalysis. There are two small differences in first-guess fluxes in this computation from those reported in Jacobson et al. [2007]. First, the five OIF estimates all used Takahashi et al. [2002] pCO2 estimates to provide high-resolution patterning of flux within inversion regions (the alternative "forward" model patterns were not used). To good approximation, this choice only affects the spatial and temporal distribution of flux within each of the 30 ocean inversion regions, not the magnitude of the estimated flux. Second, wind speed differences between the ERA40 product used in the offline analysis and the ECMWF operational model used in the online CarbonTracker analysis result in small deviations from the OIF estimates.

Gas transfer velocity in CarbonTracker is parameterized as a quadratic function of wind speed following Wanninkhof [1992], using the formulation for instantaneous winds. Gas exchange is computed every 3 hours using wind speeds from the ECMWF operational model as represented by the TM5 atmospheric transport model. Other than the smooth trend in anthropogenic flux assumed by the OIF results, interannual variability (IAV) in the first guess ocean flux comes entirely from wind speed effects on the gas transfer velocity. This is because the ocean inversions retrieve only a long-term mean and smooth trend.

The initial release of CarbonTracker (2007A) used climatogical estimates of CO2 partial pressure in surface waters from Takahashi et al. [2002] to compute a first-guess air-sea flux. This air-sea pCO2 disequilibrium was modulated by a surface barometric pressure correction before being multiplied by a gas-transfer coefficient to yield a flux. Starting with CarbonTracker 2007B and in this CarbonTracker Europe release, the air-sea pCO2 disequilibrium is imposed from analysis of the OIF results, with short-term flux variability derived from the atmospheric model wind speeds via the gas transfer coefficient. The barometric pressure correction has been removed so that climatological high- and low-pressure cells do not bias the long-term means of the first guess fluxes. In either method, the first-guess fluxes have no interannual variability (IAV) due to pCO2 changes, such as those that occur in the tropical eastern Pacific during an El Niño. In CarbonTracker, this flux IAV must be inferred from atmospheric CO2 signals.

Air-sea transfer is inhibited by the presence of sea ice, and for this work fluxes are scaled by the daily sea ice fraction in each gridbox provided by the ECMWF forecast data.

The first-guess fluxes described here are subject to scaling during the CarbonTracker optimization process, in which atmospheric CO2 mole fraction observations are combined with transport simulated by the atmospheric model to infer flux signals. In this process, signals of terrestrial flux in atmospheric CO2 distribution can be erroneously interpreted as being caused by oceanic fluxes. This flux "aliasing" or "leakage" is evident in some regions as a change in the shape of the seasonal cycle of air-sea flux. Differences between CarbonTracker posterior air-sea fluxes and those of the OIF prior fluxes are minor, but do constitute an issue that we will be investigating in the future.

3.   Further Reading

  • NOAA Pacific Marine Environmental Laboratory (PMEL)
  • Ocean Acidification
  • Caldeira, K., and M. E. Wickett (2003), Anthropogenic carbon and ocean pH, Nature, 425365-365.
  • Gloor, M., N. Gruber, J. Sarmiento, C. L. Sabine, R. A. Feely, and C. Rödenbeck (2003), A first estimate of present and preindustrial air-sea CO2 flux patterns based on ocean interior carbon measurements and models, Geophysical Research Letters, 30, , 10.1029/2002GL015594.
  • Gruber, N., J. L. Sarmiento, and T. F. Stocker (1996), An improved method for detecting anthropogenic CO2 in the oceans, Global Biogeochemical Cycles, 10, , 809-837.
  • Jacobson, A. R., N. Gruber, J. L. Sarmiento, M. Gloor, and S. E. Mikaloff Fletcher (2007), A joint atmosphere-ocean inversion for surface fluxes of carbon dioxide: I. Methods and global-scale fluxes, Global Biogeochemical Cycles, 21, doi:10.1029/2005GB002556.
  • Pacanowski, R. C., and A. Gnanadesikan (1998), Transient response in a z-level ocean model that resolves topography with partial cells, Monthly Weather Review, 126, 3248--3270.
  • Sabine, C. L., R. A. Feely, N. Gruber, R. M. Key, K. Lee, J. L. Bullister, R. Wanninkhof, C. S. Wong, D. W. R. Wallace, B. Tilbrook, F. J. Millero, T. H. Peng, A. Kozyr, T. Ono, and A. F. Rios (2004), The oceanic sink for anthropogenic CO2, Science, 305, 367-371.
  • Takahashi, T., S. C. Sutherland, C. Sweeney, A. P. N. Metzl, B. Tilbrook, N. Bates, R. Wanninkhof, R. A. Feely, C. Sabine, J. Olafsson, and Y. Nojiri (2002), Global air-sea CO2 flux based on climatological surface ocean pCO2, and seasonal biological and temperature effects, Deep-Sea Research II, 49, , 1601--1622.
  • Wanninkhof, R. (1992), Relationship between wind speed and gas exchange over the ocean, Journal of Geophysical Research, 97, 7373--7382.
Fire Module [go to top]
1.   Introduction
Vegetation fires are an important part of the carbon cycle and have been so for many millennia. Even before human civilization began to use fires to clear land for agricultural purposes, most ecosystems were subject to natural wildfires that would rejuvenate old forests and bring important minerals to the soils. When fires consume part of the landscape in either controlled or natural burning, carbon dioxide (amongst many other gases and aerosols) is released in large quantities. Each year, vegetation fires emit more than 2 PgC as CO2 into the atmosphere, mostly in the tropics. Currently, a large fraction of these fires is started by humans, and mostly intentionally to clear land for agriculture, or to re-fertilize soils before a new growing season. This important component of the carbon cycle is monitored mostly from space, while sophisticated 'biomass burning' models are used to estimate the amount of CO2 emitted by each fire. Such estimates are then used in CarbonTracker to prescribe the emissions, without further refinement by our measurements.

2.   Detailed Description
The fire module currently used in CarbonTracker is based on the Global Fire Emissions Database version 4 (GFEDv4), which is used in the SiBCASA biosphere model as described here. The GFED4 dataset consists of 0.25x0.25 degree gridded monthly burned area for the time period spanning January 1997 - August 2012. The CO2 emissions are calculated in SiBCASA using the Burned Area and the vegetation types. The GFEDv4 burned area is based on MODIS satellite observations of fire counts. The full data set was produced by combining 500 m MODIS burned area maps with active fire data from the Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS) and the Along-Track Scanning Radiometer (ATSR) family of sensors.

Once burned area has been estimated globally, emissions of trace gases are calculated using the SiBCASA biosphere model. The seasonally changing vegetation and soil biomass stocks in the SiBCASA model are combusted based on the burned area estimate, and converted to atmospheric trace gases using estimates of fuel loads, combustion completeness, and burning efficiency. Between September 2012 and December 2013 we used climatological mean values.

GFED products were successfully used in recent studies of CH4, CO2, CO, and other trace gases.

3.   Further Reading

  • Giglio, L. et al. (2013), Analysis of daily, monthly, and annual burned area using the fourth-generation global fire emissions database (GFED4), J. Geophys. Res.: Biogeosciences, 118, 317-328
  • van der Werf, G. R. et al. (2006), Interannual variability in global biomass burning emissions from 1997 to 2004, Atm. Chem. Phys., 6(11), 3423-3441
  • van der Velde, I. R. et al. (2013), Biosphere model simulations of interannual variability in terrestrial 13C/12C exchange, Global Biogeochemical Cycles, 27(3), 637-649.
  • van der Velde, I. R. et al. (2014), Terrestrial cycling of 13CO2 by photosynthesis, respiration, and biomass burning in SiBCASA , Biogeosciences, 11, 6553-6571.
  • Giglio et al. (2006), Global estimation of burned area using MODIS active fire observations, Atmos. Chem. Phys., 6, 957-974
Biosphere Module [go to top]
1.   Introduction
The biospheric component of the carbon cycle consists of all the carbon stored in 'biomass' around us. This includes trees, shrubs, grasses, carbon within soils, dead wood, and leaf litter. Such reservoirs of carbon can exchange CO2 with the atmosphere. Exchange starts when plants take up CO2 during their growing season through the process called photosynthesis (uptake). Most of this carbon is released back to the atmosphere throughout the year through a process called respiration (release). This includes both the decay of dead wood and litter and the metabolic respiration of living plants. Of course, plants can also return carbon to the atmosphere when they burn, as described here. Even though the yearly sum of uptake and release of carbon amounts to a relatively small number (a few petagrams (one Pg=1015 g)) of carbon per year, the flow of carbon each way is as large as 120 Pg each year. This is why the net result of these flows needs to be monitored in a system such as ours. It is also the reason we need a good physical description (model) of these flows of carbon. After all, from the atmospheric measurements we can only see the small net sum of the large two-way streams (gross fluxes). Information on what the biospheric fluxes are doing in each season, and in every location on Earth is derived from a specialized biosphere model, and fed into our system as a first guess, to be refined by our assimilation procedure.

2.   Detailed Description
The biosphere model currently used in CarbonTracker is the Simple-Biosphere-Model-Carnegie-Ames Stanford Approach (SiBCASA) biogeochemical model. This model calculates global carbon fluxes using input from weather models to drive biophysical processes, as well as satellite observed Normalized Difference Vegetation Index (NDVI) to track plant phenology. The version of SiBCASA model output used so far was driven by year specific weather and satellite observations, and including the effects of fires on photosynthesis and respiration (see van der Velde et al., [2014], van der Werf et al., [2006] and Giglio et al., [2006]). This simulation gives 1x1 degree global fluxes on a 10-minute time resolution, which we average to monthly means for further processing.

3-Hourly Net Ecosystem Exchange (NEE) is derived directly from Gross Primary Production (GPP) and ecosystem respiration (RE) from SiBCASA.

3.   Further Reading

  • van der Velde, I. R. et al. (2013), Biosphere model simulations of interannual variability in terrestrial 13C/12C exchange, Global Biogeochemical Cycles, 27(3), 637-649.
  • van der Velde, I. R. et al. (2014), Terrestrial cycling of 13CO2 by photosynthesis, respiration, and biomass burning in SiBCASA , Biogeosciences, 11, 6553-6571.
  • Schaefer, K. et al. (2008), Combined simple biosphere/Carnegie-Ames-Stanford approach terrestrial carbon cycle model. Journal of Geophysical Research: Atmospheres , 113(G3)
  • Olsen and Randerson (2004), Differences between surface and column atmospheric CO2 and implications for carbon cycle research, Journal of Geophysical Research: Atmospheres, 109, D2, 27
  • van der Werf, G.R. et al. (2006), Interannual variability in global biomass burning emissions from 1997 to 2004, Atm. Chem. Phys., 6(11), 3423-3441
Fossil Fuel Module [go to top]
1.   Introduction
Human beings first influenced the carbon cycle through land-use change. Early humans used fire to control animals and later cleared forest for agriculture. Over the last two centuries, following the industrial and technical revolutions and the world population increase, fossil fuel combustion has become the largest anthropogenic source of CO2. Coal, oil and natural gas combustion indeed are the most common energy sources in both developed and developing countries. Various sectors of the economy rely on fossil fuel combustion: power generation, transportation, residential/commercial building heating, and industrial processes. In 2014, the world emissions of CO2 from fossil fuel burning, cement manufacturing, and flaring reached 9.8 PgC (one PgC=1015 grams of carbon) [CDIAC]. The largest share of CO2 emissions to the atmosphere from fossil fuel burning was in China: 27% in 2014, followed by the USA (15%), Europe/EU28 (10%) and India (7%). CDIAC has projected that the global total source will slightly decrease in 2015, to 9.7 PgC.

2.   Detailed Description
The fossil fuel emission inventory used in CarbonTracker Europe is the one constructed for the CARBONES project by USTUTT/IER. It uses emissions from the EDGAR 4.2 database together with country and sector specific time profiles derived by IER. A detailed description of the construction of the product is found here. The global total emissions for 2010-2014 were scaled to the global totals used in the Global Carbon Budget 2015.

3.   Further Reading

Observations [go to top]
1.   Introduction
The observations of atmospheric CO2 mole fraction by 25 different laboratories are at the heart of CarbonTracker. They inform us on changes in the carbon cycle, whether they are regular (such as the seasonal growth and decay of leaves and trees), or irregular (such as the release of tons of carbon by a wildfire). The results in CarbonTracker depend directly on the quality, amount and location of observations available, and the degree of detail at which we can monitor the carbon cycle reliably increases strongly with the density of our observing network.

2.   Detailed Description
This study uses CO2 observations from in-situ measurements or from air samples collected in flasks at 147 global sites by 25 institutions worldwide. All contributing laboratories are included under collaborators. These observations are included in ObsPack GLOBALVIEWplusv1.0. This ObsPack product contains 205 time series of surface flask samples, quasi-continuous in-situ observations also from towers and aircraft samples. Table 1 and the figure below summarize which time series have been used in our inversion. We assimilate a maximum of 1 time series per site (e.g. not 2 from the same location from different laboratories). Note that all of these observations are calibrated against the same world CO2 scale (WMO-2007).

For most of the quasi-continuous sampling sites, the time series consist of hourly averaged mole fractions. We assimilate only mole fractions from the afternoon hours, recognizing that our atmospheric transport model does not always capture the continental nighttime stability regime while daytime well-mixed conditions are better matched. At mountain-top sites (e.g. MLO, NWR, and SPL), we use the mole fractions from the nighttime hours as this tends to be the most stable time period and avoids periods of upslope flows that contain local vegetative and/or anthropogenic influence. The selection of hourly observations included in the assimilation is based on the flags as set in the ObsPack data sets. A set of coastal sites is moved by one degree into the ocean to force the model sample to be more representative of the actual site conditions (based on Transcom continuous simulations). Table 1 summarizes how data from the different measurement programs are included for this study.

The CO2 data from ObsPack used in CarbonTracker are freely available for download. Users are encouraged to review the literature and contact the measurement labs directly for details about and access to the actual observations.

We apply a further selection criterion during the assimilation to exclude non-marine boundary layer (MBL) and non-deep Southern Hemisphere observations that are very poorly forecasted in our framework. We use the so-called model-data mismatch (MDM) in this process, which is the random error ascribed to each observation to account for measurement errors as well as modeling errors of that observation. We scale the MDM with the amount of available observations per day, to represent both flask samples and quasi-continuous observations with equal weight. We interpret an observed-minus-forecasted (OmF) mole fraction that exceeds 3 times the prescribed model-data mismatch as an indicator that our modeling framework fails. This can happen for instance when an air sample is representative of local exchange not captured well by our 1x1 degree fluxes, when local meteorological conditions are not captured by our offline transport fields, but also when large-scale CO2 exchange is suddenly changed (e.g. fires, pests, droughts) to an extent that can not be accommodated by our flux modules. This last situation would imply an important change in the carbon cycle and has to be recognized by the researchers when analyzing the results. In accordance with the 3-sigma rejection criterion, less than 1% of the observations are discarded through this mechanism in our assimilations.

Table 1 gives a summary of the observing sites used in CarbonTracker and the assimilation performance. Model-data-mismatch ("R") is a value assigned to a given site that is meant to quantify our expected ability to simulate observations there. This value is principally determined from the limitations of the atmospheric transport model. It is part of the standard deviation used to interpret the difference between a simulation first guess ("Hx") of an observation and the actual measured value ("z"). The other component, HPHT is a measure of the ability of the ensemble Kalman filter to improve its simulated value for this observation by adjusting fluxes. These elements together form the innovation χ statistic for the site: χ = (z-Hx)/√(HPHT+r2). The innovation χ2 reported is the mean of all squared χ values for a given site. An average χ2 below 1.0 indicates that the √(HPHT+r2) values are too large. Conversely, values above 1.0 mean that this standard deviation is underestimated. The bias is a statistic of the posterior residuals (final modeled values - measured values). The bias is the mean of these residuals.

Table 1: Summary of observing sites used in CarbonTracker Europe and assimilation performance.
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
ABP surface-flask NOAA Brazil 12°46'S, 38°10'W, 1 masl 101 101 +1.50 +0.41 -0.77± 0.84 -0.30± 0.44 -1.26± 0.87 +0.59 ABP
ABP surface-flask IPEN Brazil 12°46'S, 38°10'W, 1 masl 104 0 +1000.00 +0.43 -1.08± 1.45 -0.60± 1.44 -1.56± 1.29 -99.00 ABP
ABT surface-insitu EC Canada 49° 2'N, 122°22'W, 100 masl 6765 802 +5.19 +13.11 -0.98± 4.19 -0.44± 4.07 -1.50± 4.37 +1.37 ABT
ACG aircraft-pfp NOAA United States Variable 1489 0 +1000.00 +1.17 +0.11± 1.78 +0.21± 2.05 +0.35± 1.23 -99.00 ACG
AIA aircraft-flask CSIRO Australia 40°32'S, 144°18'E, 0 masl 63 0 +1000.00 +0.14 +0.23± 0.53 +0.23± 0.53 +nan± nan -99.00 AIA
ALT surface-flask NOAA Canada 82°27'N, 62°30'W, 200 masl 823 0 +1000.00 +0.48 +0.20± 0.77 +0.02± 0.96 +0.37± 0.65 -99.00 ALT
ALT surface-flask CSIRO Canada 82°27'N, 62°30'W, 200 masl 509 0 +1000.00 +0.47 +0.25± 0.78 +0.10± 0.88 +0.45± 0.71 -99.00 ALT
ALT surface-flask SIO Canada 82°27'N, 62°30'W, 200 masl 348 0 +1000.00 +0.49 +0.41± 0.75 +0.23± 0.89 +0.55± 0.68 -99.00 ALT
ALT surface-insitu EC Canada 82°27'N, 62°30'W, 200 masl 111251 13786 +2.59 +0.47 +0.21± 0.76 +0.09± 0.90 +0.29± 0.71 +0.14 ALT
AMS surface-insitu LSCE France 37°48'S, 77°32'E, 55 masl 4900 4900 +2.85 +0.26 -0.10± 0.46 +0.10± 0.36 -0.55± 0.41 +0.04 AMS
AMT surface-pfp NOAA United States 45° 2'N, 68°41'W, 53 masl 914 0 +1000.00 +4.12 -0.12± 2.88 +0.29± 3.84 -0.34± 2.14 -99.00 AMT
AMT tower-insitu NOAA United States 45° 2'N, 68°41'W, 53 masl 78736 12922 +7.83 +7.79 +0.06± 2.88 +0.48± 3.79 -0.21± 2.14 +0.17 AMT
AOA aircraft-flask JMA Japan Variable 1123 0 +1000.00 +0.17 +0.51± 1.01 +0.38± 1.21 +0.69± 0.89 -99.00 AOA
ARA surface-flask CSIRO Australia 23°52'S, 148°28'E, 175 masl 22 0 +1000.00 +1.46 -1.10± 3.06 -0.06± 2.96 -0.29± 1.97 -99.00 ARA
ASC surface-flask NOAA United Kingdom 7°58'S, 14°24'W, 85 masl 1269 1269 +0.75 +0.18 -0.07± 0.74 +0.25± 0.65 -0.33± 0.73 +1.05 ASC
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
ASK surface-flask NOAA Algeria 23°16'N, 5°38'E, 2710 masl 655 655 +0.75 +0.13 +0.01± 0.59 -0.09± 0.57 +0.13± 0.60 +0.64 ASK
AZR surface-flask NOAA Portugal 38°46'N, 27°23'W, 19 masl 400 396 +1.50 +0.45 +0.36± 1.29 +0.47± 1.42 +0.41± 1.23 +0.83 AZR
BAL surface-flask NOAA Poland 55°21'N, 17°13'E, 3 masl 976 964 +5.02 +4.40 -0.61± 3.47 -0.91± 3.69 -0.34± 3.29 +0.47 BAL
BAO surface-pfp NOAA United States 40° 3'N, 105° 0'W, 1584 masl 2134 0 +1000.00 +1.34 -1.22± 3.32 -1.09± 3.75 -1.42± 2.99 -99.00 BAO
BAO tower-insitu NOAA United States 40° 3'N, 105° 0'W, 1584 masl 59794 9166 +5.89 +4.74 -1.30± 3.19 -1.01± 2.76 -1.63± 3.49 +0.39 BAO
BCK surface-insitu EC Canada 62°48'N, 116° 3'W, 179 masl 33017 4113 +5.17 +3.21 +0.20± 1.85 +0.15± 2.65 +0.28± 1.24 +0.19 BCK
BGI aircraft-pfp NOAA United States 42°49'N, 94°25'W, 355 masl 357 0 +1000.00 +2.77 +0.16± 2.50 +0.29± 3.32 +0.20± 1.45 -99.00 BGI
BHD surface-flask NOAA New Zealand 41°24'S, 174°52'E, 85 masl 197 197 +0.75 +0.23 +0.18± 0.65 +0.50± 0.70 -0.05± 0.57 +0.96 BHD
BHD surface-insitu NIWA New Zealand 41°24'S, 174°52'E, 85 masl 481 481 +0.79 +0.19 +0.28± 0.49 +0.45± 0.46 +0.12± 0.46 +0.58 BHD
BKT surface-flask NOAA Indonesia 0°12'S, 100°19'E, 845 masl 319 0 +1000.00 +0.89 +3.00± 4.09 +2.90± 4.61 +3.14± 3.81 -99.00 BKT
BME surface-flask NOAA United Kingdom 32°22'N, 64°39'W, 12 masl 236 230 +1.50 +0.54 +0.54± 1.26 +1.13± 1.21 +0.25± 1.28 +0.92 BME
BMW surface-flask NOAA United Kingdom 32°16'N, 64°53'W, 30 masl 502 498 +1.51 +0.60 +0.63± 1.10 +0.62± 1.02 +0.62± 1.09 +0.77 BMW
BNE aircraft-pfp NOAA United States 40°48'N, 97°11'W, 465 masl 1080 0 +1000.00 +2.27 +0.07± 3.40 +0.27± 3.81 +0.43± 1.66 -99.00 BNE
BRA surface-insitu EC Canada 51°12'N, 104°42'W, 595 masl 32795 4125 +5.18 +6.86 -0.14± 2.40 +0.11± 3.29 -0.19± 2.00 +0.35 BRA
BRW surface-flask NOAA United States 71°19'N, 156°37'W, 11 masl 864 0 +1000.00 +1.20 +0.13± 1.45 +0.12± 2.03 +0.18± 1.07 -99.00 BRW
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
BRW surface-insitu NOAA United States 71°19'N, 156°37'W, 11 masl 119410 11190 +2.67 +1.43 +0.27± 0.76 +0.28± 1.00 +0.26± 0.65 +0.22 BRW
BSC surface-flask NOAA Romania 44°11'N, 28°40'E, 0 masl 434 0 +1000.00 +4.01 -6.31± 9.28 -10.51±11.08 -3.98± 6.95 -99.00 BSC
CAR aircraft-pfp NOAA United States 40°22'N, 104°18'W, 1740 masl 5102 0 +1000.00 +0.45 +0.41± 1.04 +0.20± 1.31 +0.60± 0.75 -99.00 CAR
CBA surface-flask NOAA United States 55°13'N, 162°43'W, 21 masl 997 955 +1.51 +0.51 -0.33± 1.40 +0.57± 1.64 -0.77± 0.98 +1.02 CBA
CBA surface-flask SIO United States 55°13'N, 162°43'W, 21 masl 313 0 +1000.00 +0.53 +0.19± 1.88 +1.14± 2.57 -0.21± 1.03 -99.00 CBA
CBY surface-insitu EC Canada 69° 1'N, 105° 3'W, 35 masl 14337 1785 +5.16 +1.97 +0.42± 1.20 +0.33± 1.44 +0.47± 1.00 +0.10 CBY
CDL surface-insitu EC Canada 53°59'N, 105° 7'W, 600 masl 66546 8260 +5.13 +9.79 +0.08± 2.11 +0.45± 2.85 -0.10± 1.64 +0.26 CDL
CES tower-insitu ECN the Netherlands 51°58'N, 4°56'E, -1 masl 97340 0 +1000.00 +6.03 +1.75±10.01 +1.76± 9.89 +1.72±10.29 -99.00 CES
CES tower-insitu ECN the Netherlands 51°58'N, 4°56'E, -1 masl 97269 15765 +5.87 +3.79 -0.04± 4.16 +0.56± 3.52 -0.51± 4.54 +0.72 CES
CES tower-insitu ECN the Netherlands 51°58'N, 4°56'E, -1 masl 95578 0 +1000.00 +6.53 -3.44±13.66 -5.94±15.56 -1.96±12.48 -99.00 CES
CES tower-insitu ECN the Netherlands 51°58'N, 4°56'E, -1 masl 96668 0 +1000.00 +6.31 -0.00±10.44 -0.43±10.14 +0.26±10.77 -99.00 CES
CFA surface-flask CSIRO Australia 19°17'S, 147° 3'E, 2 masl 302 299 +1.64 +0.59 -0.54± 1.11 -0.12± 1.28 -0.85± 0.91 +0.56 CFA
CGO surface-flask NOAA Australia 40°41'S, 144°41'E, 94 masl 517 517 +0.50 +0.12 +0.12± 0.35 +0.36± 0.29 -0.08± 0.29 +0.57 CGO
CGO surface-flask CSIRO Australia 40°41'S, 144°41'E, 94 masl 762 0 +1000.00 +0.11 +0.10± 0.32 +0.31± 0.26 -0.08± 0.26 -99.00 CGO
CGO surface-flask SIO Australia 40°41'S, 144°41'E, 94 masl 317 0 +1000.00 +0.12 +0.31± 0.34 +0.54± 0.30 +0.11± 0.26 -99.00 CGO
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
CHL surface-insitu EC Canada 58°45'N, 94° 4'W, 29 masl 19945 2498 +5.16 +4.39 +0.18± 1.88 +0.20± 2.53 +0.27± 1.34 +0.17 CHL
CHM surface-insitu EC Canada 49°41'N, 74°18'W, 393 masl 22635 2802 +5.18 +4.05 +0.07± 2.43 +0.74± 3.10 -0.27± 2.11 +0.28 CHM
CHR surface-flask NOAA Republic of Kiribati 1°42'N, 157° 9'W, 0 masl 501 501 +0.75 +0.15 -0.42± 0.57 -0.27± 0.46 -0.50± 0.60 +0.98 CHR
CIB surface-flask NOAA Spain 41°49'N, 4°56'W, 845 masl 255 243 +2.51 +2.96 +0.52± 2.34 +0.45± 2.59 +0.55± 1.98 +0.72 CIB
CMA aircraft-pfp NOAA United States 38°50'N, 74°19'W, 0 masl 2039 0 +1000.00 +2.04 +0.33± 2.67 +0.08± 3.56 +0.46± 1.87 -99.00 CMA
CON aircraft-flask NIES Multiple Variable 3286 0 +1000.00 +0.14 -0.04± 0.76 +0.12± 0.67 -0.13± 0.78 -99.00 CON
CON aircraft-insitu NIES Multiple Variable 297018 0 +1000.00 +0.43 +0.06± 1.78 -0.45± 2.18 +0.18± 1.53 -99.00 CON
CPS surface-insitu EC Canada 49°49'N, 74°59'W, 381 masl 26002 3251 +5.19 +4.42 +0.09± 2.39 +0.36± 3.22 -0.08± 1.63 +0.26 CPS
CPT surface-flask NOAA South Africa 34°21'S, 18°29'E, 230 masl 190 0 +1000.00 +0.27 -0.17± 0.76 +0.00± 0.63 -0.36± 0.82 -99.00 CPT
CPT surface-insitu SAWS South Africa 34°21'S, 18°29'E, 230 masl 99307 99307 +3.41 +0.29 +0.07± 0.58 +0.33± 0.54 -0.11± 0.53 +0.05 CPT
CRI surface-flask CSIRO India 15° 5'N, 73°50'E, 60 masl 147 0 +1000.00 +6.21 -3.54± 6.78 -1.02± 4.21 -5.72± 7.75 -99.00 CRI
CRV aircraft-pfp NOAA United States 64°59'N, 147°36'W, 611 masl 1447 0 +1000.00 +2.49 -1.31± 5.09 -1.68± 5.66 +0.12± 2.25 -99.00 CRV
CRV surface-pfp NOAA United States 64°59'N, 147°36'W, 611 masl 584 573 +2.52 +2.01 +0.09± 1.98 +0.31± 2.62 -0.09± 1.37 +0.68 CRV
CRZ surface-flask NOAA France 46°26'S, 51°51'E, 197 masl 569 569 +0.50 +0.14 +0.13± 0.30 +0.22± 0.27 +0.05± 0.30 +0.43 CRZ
CYA surface-flask CSIRO Australia 66°17'S, 110°31'E, 47 masl 333 333 +0.58 +0.09 -0.05± 0.27 +0.11± 0.27 -0.15± 0.24 +0.27 CYA
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
DND aircraft-pfp NOAA United States 47°30'N, 99°14'W, 472 masl 1702 0 +1000.00 +1.59 +0.21± 1.97 +0.24± 2.90 +0.31± 1.11 -99.00 DND
DRP shipboard-flask NOAA N/A 59° 0'S, 64°41'W, 0 masl 182 182 +0.52 +0.20 +0.03± 0.36 +0.15± 0.38 -0.03± 0.33 +0.55 DRP
EGB surface-insitu EC Canada 44°14'N, 79°47'W, 251 masl 74853 9334 +5.19 +9.78 +0.03± 2.97 +0.36± 3.44 -0.22± 2.61 +0.48 EGB
EIC surface-flask NOAA Chile 27°10'S, 109°26'W, 47 masl 448 448 +1.50 +0.11 +0.35± 1.07 +0.82± 0.87 -0.05± 0.99 +0.57 EIC
ESP aircraft-pfp NOAA Canada 49°23'N, 126°33'W, 7 masl 3279 0 +1000.00 +2.82 +0.07± 2.79 -0.14± 4.08 +0.22± 1.38 -99.00 ESP
ESP surface-flask CSIRO Canada 49°23'N, 126°33'W, 7 masl 23 0 +1000.00 +0.88 +0.05± 1.29 +0.47± 0.67 -0.79± 1.35 -99.00 ESP
ESP surface-insitu EC Canada 49°23'N, 126°33'W, 7 masl 44437 5530 +5.19 +3.87 +0.07± 2.03 +0.36± 2.34 +0.10± 1.64 +0.20 ESP
EST surface-insitu EC Canada 51°40'N, 110°12'W, 707 masl 38993 4878 +5.18 +5.62 +0.03± 2.35 +0.49± 2.75 -0.18± 2.16 +0.31 EST
ETL aircraft-pfp NOAA Canada 54°21'N, 104°59'W, 492 masl 2416 0 +1000.00 +1.28 +0.21± 1.63 +0.45± 2.29 +0.19± 1.06 -99.00 ETL
ETL surface-insitu EC Canada 54°21'N, 104°59'W, 492 masl 75034 9356 +5.14 +7.27 +0.03± 1.95 +0.28± 2.58 -0.08± 1.50 +0.21 ETL
FNE surface-insitu EC Canada 58°50'N, 122°34'W, 361 masl 3972 417 +5.20 +9.13 -0.74± 4.20 +0.19± 4.05 -1.10± 4.10 +1.21 FNE
FSD surface-insitu EC Canada 49°53'N, 81°34'W, 210 masl 113321 14246 +5.14 +6.44 +0.19± 2.14 +0.57± 2.85 -0.01± 1.47 +0.25 FSD
FTL aircraft-pfp NOAA Brazil 3°31'S, 38°17'W, 3 masl 160 0 +1000.00 +0.25 -0.48± 1.31 +0.13± 1.33 -1.03± 0.90 -99.00 FTL
FWI aircraft-pfp NOAA United States 44°40'N, 90°58'W, 334 masl 378 0 +1000.00 +2.52 +0.03± 3.23 -0.27± 4.09 +0.73± 2.51 -99.00 FWI
GMI surface-flask NOAA Guam 13°23'N, 144°39'E, 0 masl 924 924 +0.75 +0.09 +0.24± 0.82 +0.23± 0.91 +0.34± 0.70 +1.33 GMI
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
GPA surface-flask CSIRO Australia 12°15'S, 131° 3'E, 25 masl 51 0 +1000.00 +0.99 +1.20± 3.07 +1.15± 3.25 +1.20± 2.63 -99.00 GPA
HAA aircraft-pfp NOAA United States 21°14'N, 158°57'W, 3 masl 1778 0 +1000.00 +0.11 +0.41± 0.81 +0.36± 0.85 +0.47± 0.73 -99.00 HAA
HBA surface-flask NOAA United Kingdom 75°36'S, 26°13'W, 30 masl 647 647 +0.50 +0.13 +0.10± 0.25 +0.27± 0.23 -0.01± 0.21 +0.32 HBA
HDP surface-insitu NCAR United States 40°34'N, 111°39'W, 3351 masl 53195 53194 +7.01 +0.36 -0.15± 1.23 -0.29± 1.50 -0.10± 1.03 +0.04 HDP
HEI surface-insitu UHEI-IUP Germany 49°25'N, 8°40'E, 116 masl 100021 0 +1000.00 +7.15 -7.47±14.93 -7.72±13.23 -7.30±16.09 -99.00 HEI
HFM aircraft-pfp NOAA United States 42°32'N, 72°10'W, 340 masl 1609 0 +1000.00 +2.10 +0.43± 2.66 +0.22± 3.77 +0.34± 1.57 -99.00 HFM
HIL aircraft-pfp NOAA United States 40° 4'N, 87°55'W, 201 masl 2065 0 +1000.00 +2.28 -0.27± 2.95 -0.68± 4.23 +0.08± 1.75 -99.00 HIL
HIP aircraft-insitu HU United States Variable 130016 0 +1000.00 +0.47 +0.04± 1.24 +0.18± 1.32 -0.10± 1.13 -99.00 HIP
HNP surface-insitu EC Canada 43°37'N, 79°23'W, 87 masl 4318 510 +5.19 +13.10 +0.43± 3.56 +0.08± 4.05 +0.57± 3.09 +0.72 HNP
HPB surface-flask NOAA Germany 47°48'N, 11° 1'E, 936 masl 379 372 +5.00 +4.33 +1.25± 4.10 +1.94± 4.06 +0.72± 4.06 +0.75 HPB
HUN surface-flask NOAA Hungary 46°57'N, 16°39'E, 248 masl 702 0 +1000.00 +7.08 -0.29± 5.26 +0.34± 4.38 -0.59± 6.02 -99.00 HUN
HUN tower-insitu HMS Hungary 46°57'N, 16°39'E, 248 masl 108610 0 +1000.00 +10.93 -7.59±16.51 -12.35±21.61 -3.73±10.42 -99.00 HUN
HUN tower-insitu HMS Hungary 46°57'N, 16°39'E, 248 masl 109990 17870 +5.98 +9.89 +0.03± 3.66 +0.48± 3.29 -0.23± 3.90 +0.63 HUN
HUN tower-insitu HMS Hungary 46°57'N, 16°39'E, 248 masl 112023 0 +1000.00 +10.57 -2.73± 9.93 -4.32±11.88 -1.25± 8.04 -99.00 HUN
HUN tower-insitu HMS Hungary 46°57'N, 16°39'E, 248 masl 108599 0 +1000.00 +10.18 -0.89± 8.42 -1.48± 9.82 -0.21± 7.20 -99.00 HUN
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
ICE surface-flask NOAA Iceland 63°24'N, 20°17'W, 118 masl 656 656 +0.75 +0.41 -0.44± 1.20 -0.15± 1.19 -0.58± 1.16 +3.05 ICE
INU surface-insitu EC Canada 68°19'N, 133°32'W, 113 masl 24741 3094 +5.19 +3.59 +0.08± 1.82 +0.24± 2.33 +0.03± 1.40 +0.18 INU
INX aircraft-pfp NOAA United States Variable 250 0 +1000.00 +5.14 -1.70± 5.44 -2.79± 7.64 -1.60± 4.43 -99.00 INX
INX surface-pfp NOAA United States Variable 1127 0 +1000.00 +8.08 -0.25± 9.33 +0.71±12.42 -1.06± 7.31 -99.00 INX
IZO surface-flask NOAA Spain 28°19'N, 16°30'W, 2372 masl 594 0 +1000.00 +0.15 +0.63± 0.99 +0.60± 0.95 +0.67± 1.02 -99.00 IZO
IZO surface-insitu AEMET Spain 28°19'N, 16°30'W, 2372 masl 107971 56648 +2.56 +0.15 +0.10± 0.76 +0.05± 0.82 +0.14± 0.76 +0.10 IZO
JFJ surface-insitu KUP Switzerland 46°33'N, 7°59'E, 3570 masl 72846 12076 +2.99 +0.77 +0.32± 1.64 +0.50± 1.45 +0.18± 1.78 +0.36 JFJ
JFJ surface-insitu EMPA Switzerland 46°33'N, 7°59'E, 3570 masl 31585 5236 +2.96 +0.79 +0.30± 1.42 +0.37± 1.26 +0.24± 1.52 +0.31 JFJ
KAS surface-insitu AGH Poland 49°14'N, 19°59'E, 1989 masl 77012 0 +1000.00 +1.54 +0.01± 4.94 +2.15± 4.96 -1.62± 4.30 -99.00 KAS
KEY surface-flask NOAA United States 25°40'N, 80° 9'W, 1 masl 498 494 +1.50 +0.79 +0.09± 0.92 +0.25± 0.83 -0.04± 1.01 +0.40 KEY
KUM surface-flask NOAA United States 19°31'N, 154°49'W, 3 masl 902 902 +0.89 +0.11 -0.01± 0.97 -0.07± 1.01 +0.09± 0.98 +1.34 KUM
KUM surface-flask SIO United States 19°31'N, 154°49'W, 3 masl 507 0 +1000.00 +0.11 +0.06± 1.14 +0.03± 1.17 +0.13± 1.21 -99.00 KUM
KZD surface-flask NOAA Kazakhstan 44° 5'N, 76°52'E, 595 masl 441 426 +2.50 +2.25 -0.32± 2.47 -0.86± 2.76 +0.01± 2.06 +1.03 KZD
KZM surface-flask NOAA Kazakhstan 43°15'N, 77°53'E, 2519 masl 393 392 +2.50 +0.89 +0.20± 2.22 +0.86± 2.11 -0.54± 1.83 +0.84 KZM
LEF aircraft-pfp NOAA United States 45°57'N, 90°16'W, 472 masl 2967 0 +1000.00 +2.73 -0.03± 2.39 -0.06± 3.31 +0.15± 1.52 -99.00 LEF
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
LEF surface-pfp NOAA United States 45°57'N, 90°16'W, 472 masl 2448 0 +1000.00 +4.53 -0.11± 3.66 +0.33± 5.30 -0.19± 2.26 -99.00 LEF
LEF tower-insitu NOAA United States 45°57'N, 90°16'W, 472 masl 115810 19169 +5.98 +4.40 +0.09± 2.39 +0.41± 3.07 -0.07± 1.80 +0.26 LEF
LJO surface-flask SIO United States 32°52'N, 117°15'W, 10 masl 307 303 +5.03 +0.84 +4.12± 2.61 +5.79± 2.80 +3.14± 1.99 +1.00 LJO
LLB surface-flask NOAA Canada 54°57'N, 112°27'W, 540 masl 159 0 +1000.00 +4.91 +0.12± 5.00 +0.95± 5.82 -0.14± 4.87 -99.00 LLB
LLB surface-insitu EC Canada 54°57'N, 112°27'W, 540 masl 59417 7351 +5.16 +6.95 -0.37± 2.97 +0.20± 3.38 -0.62± 2.72 +0.50 LLB
LMP surface-flask NOAA Italy 35°31'N, 12°37'E, 45 masl 337 331 +1.50 +1.35 +0.53± 1.36 +0.04± 1.45 +0.86± 1.18 +0.99 LMP
LUT surface-insitu RUG Netherlands 53°24'N, 6°21'E, 1 masl 51843 8651 +9.91 +6.64 -0.55± 4.76 -0.05± 4.22 -0.81± 5.04 +0.33 LUT
MAA surface-flask CSIRO Australia 67°37'S, 62°52'E, 32 masl 356 356 +0.58 +0.09 -0.03± 0.29 +0.16± 0.28 -0.16± 0.23 +0.30 MAA
MEX surface-flask NOAA Mexico 18°59'N, 97°19'W, 4464 masl 242 242 +2.50 +0.40 +0.83± 1.59 +1.49± 1.58 +0.22± 1.13 +0.50 MEX
MHD surface-flask NOAA Ireland 53°20'N, 9°54'W, 5 masl 585 580 +1.50 +0.69 +0.33± 0.97 +0.66± 1.08 +0.15± 0.92 +0.50 MHD
MHD surface-insitu LSCE Ireland 53°20'N, 9°54'W, 5 masl 41109 41050 +7.28 +1.53 -0.22± 2.75 -0.15± 3.51 -0.26± 2.11 +0.19 MHD
MID surface-flask NOAA United States 28°13'N, 177°23'W, 11 masl 686 686 +1.50 +0.21 +0.63± 0.92 +0.97± 0.97 +0.50± 0.89 +0.60 MID
MKN surface-flask NOAA Kenya 0° 4'S, 37°18'E, 3644 masl 138 138 +2.50 +0.22 +1.65± 1.95 +2.35± 2.23 +1.32± 1.50 +1.04 MKN
MLO surface-flask NOAA United States 19°32'N, 155°35'W, 3397 masl 1090 0 +1000.00 +0.11 +0.13± 0.59 +0.05± 0.65 +0.21± 0.56 -99.00 MLO
MLO surface-flask CSIRO United States 19°32'N, 155°35'W, 3397 masl 483 0 +1000.00 +0.10 +0.20± 0.67 +0.00± 0.55 +0.41± 0.73 -99.00 MLO
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
MLO surface-flask SIO United States 19°32'N, 155°35'W, 3397 masl 558 0 +1000.00 +0.10 +0.29± 0.64 +0.14± 0.59 +0.41± 0.70 -99.00 MLO
MLO surface-insitu NOAA United States 19°32'N, 155°35'W, 3397 masl 119104 14714 +1.42 +0.10 +0.19± 0.55 -0.06± 0.49 +0.36± 0.54 +0.21 MLO
MNM surface-insitu JMA Japan 24°17'N, 153°59'E, 8 masl 102234 0 +1000.00 +0.21 +0.30± 0.76 +0.22± 0.80 +0.47± 0.73 -99.00 MNM
MQA surface-flask CSIRO Australia 54°29'S, 158°58'E, 6 masl 440 440 +0.58 +0.14 +0.04± 0.38 +0.23± 0.39 -0.08± 0.33 +0.51 MQA
NAT surface-flask NOAA Brazil 5°31'S, 35°16'W, 15 masl 171 171 +2.50 +0.15 -0.64± 1.05 -0.56± 1.07 -0.71± 1.04 +0.25 NAT
NAT surface-flask IPEN Brazil 5°31'S, 35°16'W, 15 masl 89 0 +1000.00 +0.14 -0.52± 1.19 -0.45± 1.23 -0.48± 1.20 -99.00 NAT
NHA aircraft-pfp NOAA United States 42°57'N, 70°38'W, 0 masl 2910 0 +1000.00 +1.83 +0.38± 2.42 +0.48± 3.41 +0.36± 1.83 -99.00 NHA
NMB surface-flask NOAA Namibia 23°35'S, 15° 2'E, 456 masl 295 295 +1.50 +0.54 -0.24± 1.05 +0.21± 1.01 -0.73± 0.90 +0.60 NMB
NWR surface-flask NOAA United States 40° 3'N, 105°35'W, 3523 masl 669 0 +1000.00 +0.47 +0.54± 1.22 +1.29± 1.35 +0.16± 0.90 -99.00 NWR
NWR surface-insitu NCAR United States 40° 3'N, 105°35'W, 3523 masl 61718 61717 +11.67 +0.49 +0.12± 1.41 +0.19± 1.84 +0.17± 1.09 +0.02 NWR
NWR surface-pfp NOAA United States 40° 3'N, 105°35'W, 3523 masl 1841 0 +1000.00 +0.56 +0.76± 1.76 +1.37± 2.27 +0.41± 1.21 -99.00 NWR
OBN surface-flask NOAA Russia 55° 7'N, 36°36'E, 183 masl 133 133 +5.03 +3.63 +0.13± 3.70 -0.66± 4.09 +0.77± 3.56 +0.60 OBN
OIL aircraft-pfp NOAA United States 41°17'N, 88°56'W, 192 masl 424 0 +1000.00 +2.14 +0.64± 2.28 +0.71± 3.04 +0.44± 1.36 -99.00 OIL
OTA surface-flask CSIRO Australia 38°31'S, 142°49'E, 40 masl 139 0 +1000.00 +0.30 -1.43±19.80 -1.72±13.00 +2.49±17.59 -99.00 OTA
OXK surface-flask NOAA Germany 50° 2'N, 11°49'E, 1009 masl 319 319 +5.00 +1.74 -0.07± 3.58 +0.41± 3.83 -0.73± 3.41 +0.55 OXK
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
PAL surface-flask NOAA Finland 67°58'N, 24° 7'E, 560 masl 541 0 +1000.00 +2.54 -0.12± 2.43 +0.11± 3.10 -0.18± 2.01 -99.00 PAL
PAL surface-insitu FMI Finland 67°58'N, 24° 7'E, 560 masl 25459 0 +1000.00 +2.89 -0.16± 2.17 +0.07± 2.86 -0.16± 1.97 -99.00 PAL
PAL surface-insitu FMI Finland 67°58'N, 24° 7'E, 560 masl 22553 0 +1000.00 +1.46 +0.12± 1.22 +0.49± 1.66 +0.01± 0.98 -99.00 PAL
PAL surface-insitu FMI Finland 67°58'N, 24° 7'E, 560 masl 84326 84323 +10.73 +2.44 +0.01± 1.86 +0.36± 2.43 -0.09± 1.59 +0.05 PAL
PFA aircraft-pfp NOAA United States 65° 4'N, 147°17'W, 210 masl 3523 0 +1000.00 +1.01 +0.16± 1.68 +0.44± 2.40 +0.12± 1.19 -99.00 PFA
POC shipboard-flask NOAA N/A Variable 2169 2164 +0.88 +0.27 -0.07± 0.62 +0.08± 0.63 -0.15± 0.62 +0.72 POC
PSA surface-flask NOAA United States 64°55'S, 64° 0'W, 10 masl 722 722 +0.50 +0.27 -0.01± 0.31 +0.07± 0.30 -0.04± 0.28 +0.40 PSA
PSA surface-flask SIO United States 64°55'S, 64° 0'W, 10 masl 350 0 +1000.00 +0.27 +0.17± 0.34 +0.30± 0.28 +0.09± 0.33 -99.00 PSA
PTA surface-flask NOAA United States 38°57'N, 123°44'W, 17 masl 398 394 +5.01 +2.65 -2.35± 3.46 -1.81± 3.38 -2.44± 3.50 +0.70 PTA
PUY surface-insitu LSCE France 45°46'N, 2°58'E, 1465 masl 24990 4167 +4.99 +2.47 -0.75± 2.77 -0.74± 3.22 -0.74± 2.21 +0.42 PUY
RBA surface-insitu NCAR United States 36°28'N, 109° 6'W, 2982 masl 20181 20181 +11.77 +0.36 +0.17± 1.02 -0.09± 1.25 +0.33± 0.83 +0.01 RBA
RPB surface-flask NOAA Barbados 13°10'N, 59°26'W, 15 masl 690 690 +1.50 +0.29 +0.00± 0.69 +0.46± 0.63 -0.20± 0.58 +0.22 RPB
RTA aircraft-pfp NOAA Cook Islands 21°15'S, 159°50'W, 3 masl 2194 0 +1000.00 +0.11 -0.13± 0.68 +0.15± 0.49 -0.35± 0.74 -99.00 RTA
RYO surface-insitu JMA Japan 39° 2'N, 141°49'E, 260 masl 62540 0 +1000.00 +1.33 -0.35± 2.35 +0.28± 3.63 -0.27± 1.68 -99.00 RYO
SAN aircraft-pfp NOAA Brazil 2°51'S, 54°57'W, 78 masl 322 322 +8.00 +0.58 -0.06± 2.77 -1.09± 2.91 +0.79± 2.70 +0.13 SAN
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
SAN aircraft-pfp IPEN Brazil 2°51'S, 54°57'W, 78 masl 1641 1641 +8.13 +0.59 -0.21± 2.38 -0.33± 2.61 -0.06± 2.26 +0.09 SAN
SCA aircraft-pfp NOAA United States 32°46'N, 79°33'W, 0 masl 2288 0 +1000.00 +1.24 +0.13± 2.12 -0.03± 2.51 +0.15± 1.91 -99.00 SCA
SCT surface-pfp NOAA United States 33°24'N, 81°50'W, 115 masl 1453 0 +1000.00 +4.15 -0.32± 3.89 -0.20± 3.97 -0.58± 3.98 -99.00 SCT
SCT tower-insitu NOAA United States 33°24'N, 81°50'W, 115 masl 51664 8565 +5.98 +5.46 +0.17± 3.31 +0.20± 3.83 +0.00± 2.97 +0.44 SCT
SEY surface-flask NOAA Seychelles 4°41'S, 55°32'E, 2 masl 644 644 +0.75 +0.18 -0.15± 0.75 +0.04± 0.53 -0.36± 0.87 +1.10 SEY
SGP aircraft-pfp NOAA United States 36°36'N, 97°29'W, 314 masl 5157 0 +1000.00 +2.37 +0.21± 2.37 -0.17± 2.79 +0.57± 1.62 -99.00 SGP
SGP surface-flask NOAA United States 36°36'N, 97°29'W, 314 masl 584 561 +3.00 +4.64 -0.10± 2.33 -0.46± 2.74 +0.17± 2.04 +0.57 SGP
SGP surface-insitu LBNL-ARM United States 36°36'N, 97°29'W, 314 masl 81033 13360 +5.98 +9.96 +0.06± 2.68 -0.09± 3.03 +0.09± 2.41 +0.34 SGP
SHM surface-flask NOAA United States 52°43'N, 174° 8'E, 23 masl 492 490 +2.50 +0.51 +0.08± 1.96 +1.76± 2.22 -0.86± 1.04 +0.74 SHM
SIS surface-flask CSIRO Scotland 60° 5'N, 1°15'W, 30 masl 89 88 +1.51 +0.45 +0.67± 0.95 +1.40± 1.01 +0.21± 0.67 +0.77 SIS
SMO surface-flask NOAA American Samoa 14°15'S, 170°34'W, 42 masl 1161 0 +1000.00 +0.11 -0.18± 0.57 +0.17± 0.38 -0.51± 0.53 -99.00 SMO
SMO surface-flask SIO American Samoa 14°15'S, 170°34'W, 42 masl 408 0 +1000.00 +0.11 -0.13± 0.74 +0.19± 0.68 -0.41± 0.71 -99.00 SMO
SMO surface-insitu NOAA American Samoa 14°15'S, 170°34'W, 42 masl 114758 15361 +1.41 +0.10 -0.12± 0.52 +0.25± 0.35 -0.46± 0.45 +0.18 SMO
SNP tower-insitu NOAA United States 38°37'N, 78°21'W, 1008 masl 45495 7504 +7.97 +3.61 +0.13± 4.20 +1.99± 4.87 -1.00± 3.38 +0.41 SNP
SPL surface-insitu NCAR United States 40°27'N, 106°44'W, 3210 masl 61206 61200 +6.99 +0.51 -0.42± 1.62 -0.19± 2.04 -0.60± 1.32 +0.07 SPL
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
SPO surface-flask NOAA United States 89°59'S, 24°48'W, 2810 masl 774 0 +1000.00 +0.08 +0.11± 0.27 +0.35± 0.20 -0.05± 0.21 -99.00 SPO
SPO surface-flask CSIRO United States 89°59'S, 24°48'W, 2810 masl 147 0 +1000.00 +0.08 -0.04± 0.26 +0.14± 0.25 -0.18± 0.20 -99.00 SPO
SPO surface-flask SIO United States 89°59'S, 24°48'W, 2810 masl 346 0 +1000.00 +0.09 +0.13± 0.28 +0.36± 0.23 -0.03± 0.22 -99.00 SPO
SPO surface-insitu NOAA United States 89°59'S, 24°48'W, 2810 masl 122079 20070 +0.98 +0.09 +0.03± 0.26 +0.25± 0.20 -0.13± 0.19 +0.08 SPO
SSL surface-insitu UBA-SCHAU Germany 47°55'N, 7°55'E, 1205 masl 121836 20192 +4.99 +2.75 -0.56± 2.86 -0.57± 3.25 -0.43± 2.55 +0.51 SSL
STM surface-flask NOAA Norway 66° 0'N, 2° 0'E, 0 masl 853 850 +1.50 +0.86 +0.23± 1.03 +0.45± 1.07 +0.13± 0.98 +0.60 STM
STR surface-pfp NOAA United States 37°45'N, 122°27'W, 254 masl 3043 1365 +3.04 +1.94 +0.08± 2.26 +0.29± 2.39 -0.04± 2.23 +0.53 STR
SUM surface-flask NOAA Greenland 72°36'N, 38°25'W, 3209 masl 624 624 +0.75 +0.24 +0.25± 0.73 +0.45± 0.82 +0.14± 0.68 +1.17 SUM
SYO surface-flask NOAA Japan 69° 1'S, 39°35'E, 14 masl 347 347 +0.50 +0.10 -0.01± 0.26 +0.18± 0.23 -0.14± 0.21 +0.31 SYO
SYO surface-insitu TU Japan 69° 1'S, 39°35'E, 14 masl 5419 0 +1000.00 +0.10 -0.04± 0.23 +0.13± 0.22 -0.14± 0.19 -99.00 SYO
TAP surface-flask NOAA Republic of Korea 36°44'N, 126° 8'E, 16 masl 597 595 +5.50 +1.65 -0.02± 3.51 +0.82± 4.38 -0.37± 2.62 +0.46 TAP
TGC aircraft-pfp NOAA United States 27°44'N, 96°52'W, 0 masl 2086 0 +1000.00 +0.75 +0.23± 1.43 +0.18± 1.36 +0.37± 1.37 -99.00 TGC
THD aircraft-pfp NOAA United States 41° 3'N, 124° 9'W, 107 masl 1591 0 +1000.00 +2.03 +0.23± 2.55 -0.07± 2.24 +0.46± 2.73 -99.00 THD
THD surface-flask NOAA United States 41° 3'N, 124° 9'W, 107 masl 565 562 +5.09 +2.20 -1.90± 3.51 -2.34± 3.89 -1.46± 3.09 +0.64 THD
TPD surface-insitu EC Canada 42°37'N, 80°33'W, 231 masl 18132 2219 +5.19 +14.92 +0.16± 3.06 +0.34± 3.99 +0.07± 2.37 +0.61 TPD
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
ULB aircraft-pfp NOAA Mongolia 47°24'N, 106° 0'E, 1350 masl 517 517 +6.67 +0.71 +0.42± 1.59 +0.85± 2.20 +0.38± 1.29 +0.07 ULB
USH surface-flask NOAA Argentina 54°51'S, 68°19'W, 12 masl 351 351 +0.75 +0.17 -0.26± 0.55 -0.27± 0.46 -0.18± 0.57 +0.66 USH
UTA surface-flask NOAA United States 39°54'N, 113°43'W, 1327 masl 649 646 +2.50 +1.62 +0.29± 1.89 +0.81± 1.99 -0.19± 1.65 +0.63 UTA
UUM surface-flask NOAA Mongolia 44°27'N, 111° 6'E, 1007 masl 690 656 +2.50 +1.06 -0.16± 2.48 -0.64± 2.70 +0.31± 2.09 +1.09 UUM
WAO surface-insitu UEA United Kingdom 52°57'N, 1° 7'E, 20 masl 23679 3785 +9.60 +4.64 +0.93± 4.36 +0.80± 4.82 +1.14± 4.05 +0.29 WAO
WBI aircraft-pfp NOAA United States 41°43'N, 91°21'W, 241 masl 1644 0 +1000.00 +2.77 +0.15± 2.43 -0.33± 3.22 +0.50± 1.43 -99.00 WBI
WBI surface-pfp NOAA United States 41°43'N, 91°21'W, 241 masl 1785 0 +1000.00 +6.38 -0.47± 3.93 -0.71± 5.34 -0.35± 2.68 -99.00 WBI
WBI tower-insitu NOAA United States 41°43'N, 91°21'W, 241 masl 58271 9566 +5.98 +5.82 +0.02± 3.17 +0.33± 4.14 -0.21± 2.45 +0.51 WBI
WGC surface-pfp NOAA United States 38°16'N, 121°29'W, 0 masl 1683 0 +1000.00 +8.71 -2.08± 9.09 +1.13± 8.49 -4.22± 9.33 -99.00 WGC
WGC tower-insitu NOAA United States 38°16'N, 121°29'W, 0 masl 57897 9467 +5.99 +4.76 +0.17± 3.57 +1.09± 2.86 -0.21± 3.93 +0.51 WGC
WIS surface-flask NOAA Israel 29°58'N, 35° 3'E, 151 masl 738 735 +2.50 +0.48 -0.16± 1.94 +0.34± 1.65 -0.32± 1.96 +0.62 WIS
WKT surface-pfp NOAA United States 31°19'N, 97°20'W, 251 masl 1874 0 +1000.00 +3.29 -0.35± 2.91 -0.46± 3.12 -0.25± 2.65 -99.00 WKT
WKT tower-insitu NOAA United States 31°19'N, 97°20'W, 251 masl 89352 14774 +5.98 +2.97 +0.03± 2.48 -0.04± 2.41 +0.07± 2.51 +0.24 WKT
WLG surface-flask NOAA Peoples Republic of China 36°17'N, 100°54'E, 3810 masl 589 576 +1.53 +0.90 +0.04± 1.40 +0.19± 1.46 +0.21± 1.32 +0.89 WLG
WSA surface-insitu EC Canada 43°56'N, 60° 1'W, 5 masl 87000 10873 +5.19 +1.71 +0.26± 2.12 +0.84± 2.71 +0.02± 1.68 +0.22 WSA
Site code Sampling Type Lab. Country Lat, Lon, Elev. (m ASL) No. Obs. Available No. Obs. Assimilated √R (μmol mol-1) √HPH (μmol mol-1) H(x)-y (μmol mol-1) H(x)-y (JJAS) (μmol mol-1) H(x)-y (NDJFMA) (μmol mol-1) Inn. Χ2 Site code
YON surface-insitu JMA Japan 24°28'N, 123° 1'E, 30 masl 78139 0 +1000.00 +0.39 +0.03± 1.74 +0.22± 1.60 +0.14± 1.74 -99.00 YON
ZEP surface-flask NOAA Norway and Sweden 78°54'N, 11°53'E, 474 masl 757 754 +1.50 +0.48 +0.33± 0.88 +0.49± 0.92 +0.17± 0.88 +0.53 ZEP

3.   Further Reading

Ensemble Data Assimilation [go to top]
1.   Introduction
Data assimilation is the name of a process by which observations of the 'state' of a system help to constrain the behavior of the system in time. An example of one of the earliest applications of data assimilation is the system in which the trajectory of a flying rocket is constantly (and rapidly) adjusted based on information of its current position to guide it to its exact final destination. Another example of data assimilation is a weather model that gets updated every few hours with measurements of temperature and other variables, to improve the accuracy of its forecast for the next day, and the next, and the next. Data assimilation is usually a cyclical process, as estimates get refined over time as more observations about the "truth" become available. Mathematically, data assimilation can be done with any number of techniques. For large systems, so-called variational and ensemble techniques have gained most popularity. Because of the size and complexity of the systems studied in most fields, data assimilation projects inevitably include supercomputers that model the known physics of a system. Success in guiding these models in time often depends strongly on the number of observations available to inform on the true system state.

In CarbonTracker, the model that describes the system contains relatively simple descriptions of biospheric and oceanic CO2 exchange, as well as fossil fuel and fire emissions. In time, we alter the behavior of this model by adjusting a set of parameters as described in the next section.

2.   Detailed Description
The four surface flux modules drive instantaneous CO2 fluxes in CarbonTracker according to:

F(x, y, t) = λ(x,y,t) • Fbio(x, y, t) + λ(x,y,t) • Foce(x, y, t) + Fff(x, y, t) + Ffire(x, y, t)

Where λ represents a set of linear scaling factors applied to the fluxes, to be estimated in the assimilation. These scaling factors are the final product of our assimilation and together with the modules determine the fluxes we present in CarbonTracker. Note that no scaling factors are applied to the fossil fuel and fire modules.

2.1   Land-surface classification
The scaling factors λ are estimated for each week and assumed constant over this period. Each scaling factor is associated with a particular gridbox of the global domain. We chose an approach in which the ocean grid boxes are combined into 30 large basins encompassing large-scale ocean circulation features, as in the TransCom inversion study (e.g. Gurney et al., [2002]). The terrestrial biosphere grid boxes are combined up according to ecosystem type as well as geographical location. Thereto, each of the 11 TransCom land regions contains a maximum of 19 ecosystem types summarized in the table below for Europe.

Ecosystem types considered on 1x1 degree for the terrestrial flux inversions is based on Olson, [1992]. Note that we have adjusted the original 29 categories into only 19 regions. This was done mainly to fill the unused categories 16,17, and 18, and to group the similar (from our perspective) categories 23-26+29. The table below shows each vegetation category considered. Percentages indicate the area associated with each category for Europe rounded to one decimal.

Each 1x1 degree pixel of our domain was assigned one of the categories above bases on the Olson category that was most prevalent in the 0.5x0.5 degree underlying area.

2.2   Ensemble Size and Localization
The ensemble system used to solve for the scalar multiplication factors is similar to that in Peters et al. [2005] and based on the square root ensemble Kalman filter of Whitaker and Hamill, [2002]. We have restricted the length of the smoother window to only five weeks as we found the derived flux patterns within Europe and North America to be robustly resolved well within that time. We caution the CarbonTracker users that although the North American and European flux results were found to be robust after five weeks, regions of the world with less dense observational coverage (the tropics, Southern Hemisphere, and parts of Asia) are likely to be poorly observable even after more than a month of transport and therefore less robustly resolved. Although longer assimilation windows, or long prior covariance length-scales, could potentially help to constrain larger scale emission totals from such areas, we focus our analysis here on a region more directly constrained by real atmospheric observations.

Ensemble statistics are created from 150 ensemble members, each with its own background CO2 concentration field to represent the time history (and thus covariances) of the filter. In contrast to our earlier system design, we currently do not apply any localization to the statevector.

2.3   Dynamical Model
In CarbonTracker, the dynamical model is applied to the mean parameter values λ as:

λ tb = (λ  t-2a + λ  t-1 a + λ  p  )   ⁄   3.0

Where "a" refers to analyzed quantities from previous steps, "b" refers to the background values for the new step, and "p" refers to real a-priori determined values that are fixed in time and chosen as part of the inversion set-up. Physically, this model describes that parameter values λ for a new time step are chosen as a combination between optimized values from the two previous time steps, and a fixed prior value. This operation is similar to the simple persistence forecast used in Peters et al. [2005], but represents a smoothing over three time steps thus dampening variations in the forecast of λ b in time. The inclusion of the prior term λ p acts as a regularization [Baker et al., 2006] and ensures that the parameters in our system will eventually revert back to predetermined prior values when there is no information coming from the observations. Note that our dynamical model equation does not include an error term on the dynamical model, for the simple reason that we don't know the error of this model. This is reflected in the treatment of covariance, which is always set to a prior covariance structure and not forecast with our dynamical model.

2.4   Covariance Structure
Prior values for λ p are all 1.0 to yield fluxes that are unchanged from their values predicted in our modules. The prior covariance structure Pp describes the magnitude of the uncertainty on each parameter, plus their correlation in space.

In each of these regions on the northern hemisphere, individual λ(x,y) parameters are coupled through an isentropic covariance structure which makes two boxes i and j at a distance d of each other have a covariance C of

C = 0.64• exp(-d/L).

In this formula the covariance length scale L varies across the globe. Over Boral and Temperate North America where the observation network is relatively dense, L=300km, but in Boreal and Temperate Asia the number of observations constrains a much smaller number of parameters individually and we chose L=1000km. In Europe, with its strongly heterogeneous land-use and land management and large volume of observations available we took L=200km. In the rest of the world, the length scale is taken infiniely large, coupling fully all grid boxes and associated λ's in the tropics and southern hemisphere.

The figure shows ecoregions for Europe (click here for global land ecoregions). Note that there is currently no requirement for ecoregions to be contiguous, and a single scaling factor can be applied to the same vegetation type on both sides of a continent.

Theoretically, this approach leads to a total number of 9835 optimizable scaling factors λ each week, but the actual number is smaller since not every ecosystem type is represented in each TransCom region, and because we decided not to optimize parameters for ice-covered regions, inland water bodies, and desert. The total flux coming out of these last regions is negligibly small. It is important to note that even though many parameters are available to scale the fluxes, the imposed covariance structure reduces the number of degrees of freedom to about 1100 each week.

Furthermore, all ecosystems within tropical TransCom regions are coupled decreasing exponentially with distance since we do not believe the current observing network can constrain tropical fluxes on sub-continental scales, and want to prevent large dipoles to occur in the tropics.

In our standard assimilation, the chosen standard deviation is 80% on land parameters, and 40% on ocean parameters. This reflects more prior confidence in the ocean fluxes than in terrestrial fluxes, as is assumed often in inversion studies and partly reflects the lower variability and larger homogeneity of the ocean fluxes. All parameters have the same variance within the land or ocean domain. Because the parameters multiply the net-flux though, ecosystems with larger weekly mean net fluxes have a larger variance in absolute flux magnitude.

3.   Further Reading