Skip to main content
European Commission logo print header



Reporting period: 2017-02-01 to 2019-01-31

Rivers are the primary conduits through which carbon is transported from the land to the ocean, and therefore play a significant role in the global carbon cycle. However, aside from their function as a means of carbon transport, carbon can also be transformed and processed within rivers. For example, dissolved inorganic carbon (DIC) can be either taken up by aquatic plants or released by biological respiration and decay. As well, in addition to carbon fluxes in river waters, carbon can also be lost from water surfaces through CO2 emissions from oversaturated waters. This gives rivers a potentially important role as sources of atmospheric CO2, with implications for global climate given the role of CO2 as a strong greenhouse gas.

This project aimed to reduce current uncertainties in the global carbon budget by investigating aquatic carbon cycling within European karstic watersheds. The primary tools used to achieve this objective are the stable isotope and flux determinations of dissolved and particulate carbon, and dissolved CO2. Rivers situated in karst bedrock are thought to emit significant quantities of CO2, and due to their large combined surface areas and the abundance of karst terrains worldwide, these features play a crucial but as yet unclear role in the global carbon cycle. EURO-KARST addressed this lack of information through the innovative use of stable isotopes, together with the modeling of CO2 evasion and simultaneous flux determinations of dissolved and particulate riverine carbon.
In the first year of work, data from previous projects were incorporated into a new journal article draft, which was published in Applied Geochemistry that same year. As well, initial results were presented at the International Association of Hydrogeologists (IAH) 2017 Annual Meeting in Dubrovnik, Croatia. Another publication draft, with the Marie Curie recipient as contributing author, was completed for publication the following year in Hydrological Processes (van Geldern et al. 2018). In addition, preparations were made for fieldwork intended for the following year. This entailed preliminary analyses of existing datasets to help plan sampling schemes, and the establishment of selected rivers and springs for water collection.

The following year, a fieldwork campaign was carried out based on these plans, to obtain new data pertaining to carbon cycling in springs and karstic headwaters. This was carried out in 6 separate excursions between the months of February and November in order to discern seasonal trends in dissolved/particulate carbon fluxes and CO2 outgassing. For this work, water samples were analyzed for 13C and concentrations of dissolved organic and inorganic carbon (DOC/DIC) and particulate organic carbon (POC), and CO2 effluxes were measured with sensors installed in static chambers. Initial findings from the headwater sampling campaign was presented at another conference in September, the IAH 2018 Annual Meeting in Daejeon, South Korea (September 9-14, 2018). As well, data from an ancillary project, concerned with the quantification of nitrate fluxes in the Pegnitz River and determination of wastewater contributions, was presented as a poster at the European Geosciences Union (EGU) 2018 Meeting in Vienna (April 8-13, 2018).

An associated MSc thesis completed by Maxmilian Schmidt was also successfully completed, which incorporated some data from the 2018 fieldwork campaign. The findings from this thesis will be used to plan projected future work that will extend our research on CO2 outgassing from karstic springs, with focus on CO2 emissions within the first 200 metres of source springs. This new research will determine the gradient of CO2 outgassing within river headwaters, which may constitute the bulk of CO2 loss in a given basin. A related proposal has already been incorporated into a DFG grant for review. If successful, this grant will extend our work for another 3 years.
Our study showed that karst basins are particularly significant sources of atmospheric CO2, with CO2 outgassing rates that match or surpass even those observed in more productive tropical regions. This aids in the clarification of the global carbon balance, in which the importance of karstic river basins have been previously under-estimated. In addition to the results already published and presented, we also laid the groundwork for future research on karstic rivers and their catchments and springs (see section 1.2). Previous studies have shown upstream river areas to be important in fluvial carbon cycling, with springs representing the most significant endmembers within these sections. Specifically, we began planning for an anticipated study focusing on source springs and headwaters in karstic basins, as most of the CO2 efflux occurs in these upstream areas. These preparations included the requisition of new CO2 sensors, the selection of new sampling sites, and a reconnaissance sampling campaign incorporating the new locations (section 1.2).

A number of samples from this initial fieldwork have already been analyzed and results plotted. Preliminary findings show that the bulk of CO2 is outgassed within the first 50 meters of source springs. Based on these results, we hypothesize that over 90 % of the initial groundwater-sourced CO2 is lost within the first few hundred metres after their emergence to the surface in springs. This concurs with previous studies that suggest that CO2 emissions are greatest in stream and river headwaters, with significantly lower efflux rates further downstream. We also anticipated further collaboration with Dr. Jens Hartmann of Hamburg University for use of his GLORICH database for comparison of our target watersheds to those worldwide (Hartmann et al. 2014) . To this end, we plan to incorporate our own data into GLORICH and help fill in the information gaps regarding the importance of carbonate basins and springs in the global carbon cycling.

In addition, we have kept in contact with Drs. Abril and Polsanaere of the University of Bordeaux for further development of their “streamCO2-DEGAS” model, which is a MATLAB-based computer model that uses 13C and DIC concentration data to estimate carbon fluxes from river water surfaces. In the future, should funding be available, we plan to strengthen our collaboration with these researchers to develop a CO2 efflux model that can be universally applicable to watersheds globally, regardless of geological setting or climate.


Hartmann, J., R. Lauerwald, and N. Moosdorf. 2014. A brief overview of the GLObal RIver CHemistry Database, GLORICH. Procedia Earth and Planetary Science 10: 23-27.

van Geldern, R. and others 2018. Insights into agricultural influences and weathering processes from major ion patterns. Hydrological Processes 32: 891-903.