EGU2020-16308, updated on 12 Jun 2020
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Large-scale study on groundwater dissolved organic matter reveals strong heterogeneity and a complex microbial footprint

Astrid Harjung, Johannes Schweichhart, Grit Rasch, and Christian Griebler
Astrid Harjung et al.
  • University of Vienna, Department of Functional and Evolutionary Ecology, Unit of Limnology, Vienna, Austria (

Dissolved organic matter (DOM) in fresh groundwater is generally low in concentration compared to other fresh waters. However, the overall amount of groundwater DOM is huge, as there is 100 times more fresh groundwater than fresh surface water. To date, research on groundwater DOM has merely focused on specific threats to humans such as e.g. DOM and heavy metal complexations and DOM from hydrocarbon contamination. Only few studies targeted to understand DOM as energy source of groundwater food webs and the role of groundwater DOM in the global carbon cycle. While research on these two subjects in surface waters flourish, a comprehensive, large-scale study of groundwater is still missing. Since a major fraction of Earth’s microbial biomass is found in the subsurface, mainly in aquifers, this represents a major knowledge gap. Moreover, researchers found that groundwater DOM concentrations worldwide increase alarmingly. Here, for the first time, we examine DOM properties and heterogeneity in a large-scale approach with regards to aquifer characteristics and physical-chemical as well as microbial features. We hypothesize that groundwater DOM quality shows high diversity and plays an important, yet complex role in these ecosystems, where bioavailability is influenced by intrinsic molecular properties, as well as environmental conditions.

We analyzed 1000 water samples from 100 groundwater bodies all over Austria with regards to their DOM quantity, quality and bacterial abundance (BA). From fluorescence excitation-emission-matrices (EEMs) we derived indices and components to describe DOM quality. Next, we explored this data with principal component analysis, where we used convex-hull areas to estimate the heterogeneity of DOM composition within the groundwater bodies. In parallel, the similarity of DOM quality was evaluated with self-organizing maps on EEMs to test if we captured the heterogeneity of the data set sufficiently with the previous analysis. DOM quantity and quality was then related to BA and physical-chemical parameters.

Our results show that water from fractured and karstic aquifers exhibit significantly higher terrestrial DOM origin and less degraded DOM than porous aquifers. This result can be explained by abiotic factors such as adsorption of large, aromatic compounds, as well as biological factors, specifically, larger surface areas for biofilm development in porous aquifers. The latter is supported by our observation that porous aquifers showed higher BA values. Remarkably, we found that BA was related to different DOM quality in each aquifer type: In porous aquifers BA was related to large, aromatic DOM molecules indicating that these are important for bacterial growth, while in fractured and karstic aquifers BA was related to fulvics and highly degraded humic compounds. Bacterial growth and degradation of complex DOM might be limited by low retention times in some of these aquifers.  Also, we found that groundwater bodies located in river valleys display high heterogeneity in DOM quality spanning across the whole DOM compositional diversity found in this study. This finding could either be explained by surface water infiltration in some parts and younger groundwater or the fact that river valleys are main settlement areas.

How to cite: Harjung, A., Schweichhart, J., Rasch, G., and Griebler, C.: Large-scale study on groundwater dissolved organic matter reveals strong heterogeneity and a complex microbial footprint, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-16308,, 2020

Display materials

Display file

Comments on the display material

AC: Author Comment | CC: Community Comment | Report abuse

Display material version 1 – uploaded on 30 Apr 2020
  • CC1: Comment on EGU2020-16308, Bethany Fox, 06 May 2020

    Hi Astrid,

    It is nice to see some more of your work this year, and biological DOM is right up my street!

    The data set is impressively big and shows a tremendous amount of work. Your conclusions are also very interesting, especially regarding the impact of residence time on proteinaceous material etc. This is something we have not explored as we have been looking at fast-acting dynamics.

    Have you undertaken any proteomics or DNA sequencing to try to unpick the potential microbial pathways in actions and how this is impacted by notable anthropogenic activity, such as agriculture?

    Hope you are well, 


    • AC1: Reply to CC1, Astrid Harjung, 07 May 2020

      Hi Beth,

      Thanks a lot for your comment. Pity we couldn’t meet this year.

      I realized during the discussion that the terrestrial-like component C3 in the presentation is maybe actually a bacterial product as you and Eva Perrin found (e.g. peak C). I have this suspicion, because I only found it in the agricultural areas with high nutrient loading. Also a search in the fluorescence database confirmed this, as this component only matched with few compounds found in waste water impacted catchments. I will definitely look into this possibility.

      About the residence times -> good thing about going underground kind of forces to look at this really long time scales ;).

      Your two suggestions are definitely things we should do next!

      • ) Proteomics: We could find out if some of the protein-like components are really proteins. -> This would enable us to understand which of these fluorophores are energetically relevant (as groundwater usually has around 8%BDOC) and can be recycled and which are recalcitrant and persist without any further cycling, so it would help to understand how exactly DOC concentrations stabilize in depth.
      • ) DNA sequencing: It would be definitely interesting to do that! In terms of evaluating the impact of land use, as you suggested that would be great. For fluvial systems there is quite some work done. It would be interesting to see if pattern hold true for groundwater, as residence times are much longer and conditions are much different. Also, with regards to the questions that Magda Bieroza (tracing along flow paths for DOM and bacterial communities) and Andrea Butturini (heterotrophs and chemolithoautotrophs) had put in the chat doing more into this direction would be interesting.

      Hopefully, after this large-scale study it will be easier to define appropriate sample amounts and locations for a study including these analyses. If you are interested in discussing more ideas and maybe we find some way for collaboration, I would be happy!

      Greetings from (a to empty for this time of the year) Vienna,