Image gallery: SCIAMACHY Carbon Dioxide

For latest images please see also:

GHG-CCI website (click on Image Gallery and/or CRDP)

Here we show some pictures of the greenhouse gas carbon dioxide (CO2) retrieved from the near-infrared nadir spectra of reflected and backscattered solar radiation measured by the SCIAMACHY instrument onboard the European environmental satellite ENVISAT using the retrieval algorithms WFM-DOAS and BESD developed at the University of Bremen. For more information please visite the WFM-DOAS main web page.

Image gallery (click to enlarge)

Greenhouse gas carbon dioxde

Carbon dioxide (CO2) and methane (CH4) are the two most important anthropogenic (man-made) greenhouse gases and contribute to global warming. The reliable prediction of future atmospheric greenhouse gas concentrations and associated climate change requires an adequate understanding of their (natural and anthropogenic) sources and sinks. SCIAMACHY on ENVISAT is the first satellite instrument whose measurements contain information on the vertical columns of both gases due to high measurement sensitivity down to the Earth surface where the major sources and sinks of these greenhouse gases are located. The vertical column of a gas is the number of molecules of this gas located in an air column which extends from the Earth's surface to the top of the atmosphere per surface area (unit: molecules/cm2). For the greenhouse gases (CH4 and CO2) we normalize the greenhouse gas columns with the corresponding (measured) number of air molecules (obtained, e.g., using oxygen (O2) measurements) to get dry air column-averaged mixing ratios or mole fractions of the greenhouse gases (unit: ppb for CH4 and ppm for CO2).

Here we show a collection figures generated using the SCIAMACHY derived data CO2 products generated at our institute:

The global maps show the distribution of CO2 obtained when averaging all observations obtained for several years (here: 2006-2011). The maps on the right show the seasonal variations. As can be seen, CO2 is low (blue) especially during the summer months at mid and high latitudes over the northern hemisphere as this is the growing season of plants which take up atmospheric CO2 thereby reducing the amount contained in the atmosphere. Therefore CO2 variies during the year with a maximum around April and a minimum around August. This is more clearly shown in the red curve at the bottom. This curve also shows that atmospheric CO2 is increasing by about 1.8 ppm per year mainly due to burning of fossil fuels (oil, coal, gas) despite all efforts undertaken so far to reduce CO2 emissions:

Here the same figure but in German:

The SCIAMACHY CO2 retrievals show a lot of details not visible in the Figures shown above as those have been smoothed. The regional figures below show more considerably more detail:

SCIAMACHY CO2 over China:

SCIAMACHY CO2, population density and EDGAR anthropogenic CO2 emissions for China:

SCIAMACHY CO2, population density and EDGAR anthropogenic CO2 emissions for Germany:

SCIAMACHY CO2, population density and EDGAR anthropogenic CO2 emissions for Japan:

SCIAMACHY CO2, population density and EDGAR anthropogenic CO2 emissions for western USA:

SCIAMACHY CO2, population density and EDGAR anthropogenic CO2 emissions for eastern USA:

Northern hemispheric carbon dioxide during March-June, where CO2 is relatively high mainly due to release of CO2 to the atmosphere by decaying vegetation, and July-October, where CO2 is relatively low mainly due to uptake of atmospheric CO2 by growing vegetation:

Three years of northern hemispheric CO2:

Hier das Bild mit deutscher Beschriftung

Our analysis of the SCIAMACHY CO2 starts 2003 (thick orange curve). The two thin red and blue curves starting earlier are highly precise and accurate CO2 measurements performed at the Earth's surface (obtained from the NOAA/ESRL/GMD website). The SCIAMACHY measurements confirm that atmospheric CO2 is increasing (by about 0.5 percent per year). The SCIAMACHY CO2 measurements also show nicely the (CO2) "breathing of our plant" due to quite regular uptake and release of CO2 mainly by the growing and decaying vegetation resulting in high atmospheric CO2 during late autumn to early spring and low CO2 during summer (note that during spring plants start to grow, taking up atmospheric CO2, but this is (over)compensated by decaying vegetation (this releases CO2 to the atmosphere) which could not be fully decomposed during the cold late autumn and winter months).

CO2 over Europe:

High values (shown in red) typically indicate a CO2 source region. Atmospheric CO2 is however very long lived and therefore atmospheric transport (wind) is important. High values may therefore also appear outside major source regions especially over areas where sampling (= measurement frequency) is low due to, for example, persistent cloud cover. For similar reasons low values may appear over source regions. Interpretation of the CO2 maps is therefore not straight forward and typically requires complex models. The large area extending from Amsterdam to Frankfurt is Europe's region where population density is highest and large amounts of CO2 are emitted. This is clearly visible in the SCIAMACHY measurements showing an extened region of elevated CO2 (shown in red). Anthropogenic (man made) CO2 emissions are primarily due to burning of fossil fuels (energy generation and consumption (e.g., domestic heating), traffic, industrial activites, etc.). SCIAMACHY is the first satellite instrument which can detected regionally elevated atmospheric CO2 from space which results from local anthropogenic CO2 emissions. The detection of regional antropogenic CO2 signals is quite challenging because the amount of CO2 in the atmosphere is high (at least compared to most of the other trace gases) and even a large source (such as a coal-based power plant or a large city) is expected to result locally only in a very small CO2 increase, if the CO2 is averaged over the entire air column - extending from the Earth's surface to the so-called top of the atmosphere - and when averaged over the large ground pixel size of SCIAMACHY which is 30 km x 60 km. Transport (wind) and large natural CO2 fluxes resulting from the growing and decaying vegetation result in further complications which can however be dealt with at least to a certain extent using models. Monitoring of CO2 emissions is important and required, for example by the Kyoto protocol, as CO2 is the most important anthropogenic so called greenhouse gas (a radiatively active gas which absorbs the Earth's infrared (long wave) radiation but is relatively transparent for the visible (shorter wavelength) solar radiation) and increasing CO2 concentrations result in an increase of the average temperature of our planet with adverse consequences such as rising sea levels and an increase of extreme weather conditions.

Here the same picture but with annotation in German: Hier das Bild mit deutscher Beschriftung: CO2 ueber Europa:

How the satellite CO2 retrieval works:

First of all we measure the CO2 total column, i.e., the number of CO2 molecules per area above the Earth's surface. This shows primarily where the mountains are simply because there are less (air) molecules above a mountain compared to sea level:

We do the same with molecular oxygen (O2) from which we compute the number of air molecules per area knowing that about 21 percent of all air molecules are O2 molecules:

As can be seen, the spatial variation of the CO2 columns and the O2 columns are very similar. This is because both gases are long-lived and therefore well mixed in the atmosphere. The CO2 signals we are interested in are the tiny differences between the CO2 column and the O2 or air column which tell us something about the CO2 surface sources and sinks. To make this signal visible, we compute the ratio of the CO2 column and the air column (obtained from the measured O2 columns), i.e., the CO2 mole fraction (= molecular mixing ratio). The unit is ppm which means parts per million. 380 ppm for example means that on average we have 380 CO2 molecules per 1 million air molecules (assuming a dry atmosphere without any water vapour molecules).

Here similar figures focussing on the Rhine-Main area:

Hier das Bild mit deutscher Beschriftung.

With black background.

Mit schwarzem Hintergrund.

Without annotation (except the cities).

Ohne Beschriftung (ausser den Staedten).

Schwarz-weiss mit Beschriftung.

Schwarz-weiss ohne Beschriftung.

Here the corresponding figures for the global data set:

CO2 columns:

The variability of the CO2 columns is mainly dominated by variations of the surface topography (the number of air molecules or CO2 molecules over a high mountain is lower compared to a region at sea level).

O2 columns:

As CO2 also the variability of the O2 columns is mainly dominated by variations of the surface topography (the number of O2 molecules over a high mountain is lower compared to a region at sea level).

CO2 column-averaged mole fraction (or mixing ratio) computed by dividing the CO2 columns with the corresponding O2 columns (times a constant factor):

$Global CO2 mole fractions$

Note that the figure does not show a "true" 3-years average due to the irregular sampling of the satellite data. Over the mid and higher latitudes of the northern hemisphere, for example, the average is strongly weighted towards summer, where CO2 is low due to uptake of the atmospheric CO2 by the growing vegetation. The sampling rate in winter is typically low due to persistent cloud cover, snow and ice covered surfaces (which have low reflectivity in the near-infrared spectral region), and low sun elevation (= high solar zenith angle). The interpretation of the map is therefore not straight forward and typically requires complex models. The high values over parts of the tropics are very likely partially due to the sampling but also due to high thin clouds which may result in disturbances (this potential issue is currently under investigation). The low CO2 especially in the southern part of the Himalaya region may be due to artifacts resulting from the complex terrain. The assessment of the quality of the map shown here is currently still under investigation. Further improvements of the retrieval algorithm may result in some modifications in the future.

In order to assess the quality of the retrieved CO2 we compare our data with ground based measurements (mostly FTIR) and global models such as NOAA's assimilation system CarbonTracker. This is illustrated in the following two figures (see Buchwitz et al., 2008 for details):

Below a comparison over North America for July 2003 (the white symbols in the right hand side panel show the locations where surface CO2 measurements have been assimilated in CarbonTracker). The main feature that is visible is low CO2 over the northern part and higher CO2 over the southern part separated by a wave-like boundary. During summer significant amounts of CO2 are taken up by the growing vegetation - such as the boreal forests - resulting in lower atmospheric CO2 over the northern part of North America as shown in the figure.

Comparison of SCIAMACHY and CarbonTracker CO2 over North America:

The CarbonTracker data set used here has a much lower spatial resolution compared to the SCIAMACHY observations. Therefore, perfect agreement is not to be expected. Recently NOAA has release CarbonTracker data at higher resolution which we will use for future comparisons.

Thanks to SCIAMACHY on the European ENVISAT satellite global CO2 observations with high sensitivity down to the Earth's surface are now possible (since 2003). In the near future (end of 2008 / beginning of 2009) CO2 observations will also be performed by other satellite instruments (which will also measure spectra of reflected and backscattered solar radiation in the near-infrared / shortwave-infrared spectral region in nadir (= downlooking) observation mode as currently only done by SCIAMACHY), most notably NASA/JPL's OCO and Japan's GOSAT missions which will continue the time series started by SCIAMACHY in 2003.

Here an other picture of the SCIAMACHY CO2 over North America, this time an average over the years 2003-2005. The data are shown as anomalies, i.e., the mean values have been subtracted to eliminate effects of small systematic biases between the data sets which better enables to compare the spatial pattern. Top row: vertical column of CO2 (left) and O2 (right). Bottom row: SCIAMACHY (left) and CarbonTracker (right) column-averaged CO2 mole fractions (= molecular mixing ratios).

Here a similar picture but for Japan:

Here the western part of central Africa:

Typically the observed CO2 spatial pattern is a complex mixture of CO2 resulting from anthropogenic emissions and (typically much larger) natural CO2 fluxes mostly aring from the uptake and release of CO2 by the terrestrial biosphere resulting in higher CO2 in the first half of the year and lower CO2 during the second half of the year (over the northern hemisphere). This is illustrated here for China where it is shown that not each and every city or industrialized area can be easily identified in the measured CO2 for the reasons mentioned above in combination with the irregular sampling of the satellite data (clouds etc.) and the long lifetime of the atmospheric CO2 which is tranported over long distances by the wind. The interpretation of the maps shown here is therefore not straight forward and typically requires complex models such as CarbonTracker.

CO2 over China: