Linkage Between Dissolved Organic Matter Transformation, Bacterial Carbon Production, and Diversity in a Shallow Oligotrophic Aquifer: Results From Flow-Through Sediment Microcosm Experiments

Abstract

Aquifers are important reservoirs for organic carbon. A fundamental understanding of the role of groundwater ecosystems in carbon cycling, however, is still missing. Using sediment flow-through microcosms, long-term (171d) experiments were conducted to test two scenarios. First, aquifer sediment microbial communities received dissolved organic matter (DOM) at low concentration and typical to groundwater in terms of composition (DOM-1x). Second, sediments received an elevated concentration of DOM originating from soil (DOM-5x). Changes in DOM composition were analyzed via NMR and Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS). Carbon production, physiological adaptations and biodiversity of groundwater, and sediment prokaryotic communities were monitored by total cell counts, substrate use arrays, and deep amplicon sequencing. The experiments showed that groundwater microbial communities do not react very fast to the sudden availability of labile organic carbon from soil in terms of carbon degradation and biomass production. It took days to weeks for incoming DOM being efficiently degraded and pronounced cell production occurred. Once conditioned, the DOM-1x supplied sediments mineralized 294(±230) μgC L^-1 _sed d^-1, 10-times less than the DOM-5x fed sediment communities [2.9(±1.1) mgC L^-1 _sed d^-1]. However, the overall biomass carbon production was hardly different in the two treatments with 13.7(±4.8) μgC L^-1 _sed d^-1 and 14.3(±3.5) μgC L^-1 _sed d^-1, respectively, hinting at a significantly lower carbon use efficiency with higher DOM availability. However, the molecularly more diverse DOM from soil fostered a higher bacterial diversity. Taking the irregular inputs of labile DOM into account, shallow aquifers are assumed to have a low resilience. Lacking a highly active and responsive microbial community, oligotrophic aquifers are at high risk of contamination with organic chemicals.

Keywords: bacterial production; carbon cycling; carbon use efficiency; dissolved organic matter; groundwater; mass spectrometry; microbial activity; oligotrophy.

Figure 1

Experimental setup of the 2D microcosms. Two microcosms were run in parallel. Both microcosms were supplied with feeding solution DOM-1x, DOM-5x, and groundwater. In microcosm A, DOM-1x was infiltrated to the top zone, groundwater to the middle zone, and DOM-5x to the bottom zone. In microcosm B, the order was in reverse. DOM-1x and DOM-5x were supplied through four adjacent inlet ports at the bottom and top of the microcosm, separated by three ports receiving only groundwater. DOM-1x and DOM-5x were composed from soil extract (low and high conc.) and groundwater mixed 5:1 at the inlets to the microcosm. For water sampling, only the outflow of the three top and bottom ports (sampling ports; SP) was used. The two sediment zones receiving extra dissolved organic matter (DOM) and traversing each of the microcosms in a longitudinal direction are termed “treatment zones” (horizontal dashed lines). Vertical sediment sections are termed segments (segments I–VII, vertical black lines). Sediment was collected regularly in tanks A and B from the top (location is highlighted by the gray bars in microcosm A).

Figure 2

Depletion of dissolved organic carbon (DOC) during passage through the sediments of the two microcosms supplied with DOM-1x and DOM-5x. Data are mean values ± SD. Each of the two microcosm contained a DOM-1x and DOM-5x zone (biological replicates). For the individual sediment zones of each tank, additional technical replicate measurements were performed.

Figure 3

Hierarchical cluster analysis (A) and principal component analysis (B) derived from ¹H NMR spectra (800 MHz, CD₃OD) of porewater solid phase extracts (SPE)-DOM from DOM-5x (C) supplied sediments. HCA and PCA were computed with NMR section integrals of 0.01 ppm bucket width, with exclusion of HDO and HD₂COD.

Figure 4

(−)ESI Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS)-derived principal component analysis (A) and hierarchical cluster analysis (B) of groundwater and DOM-1x and DOM-5x feeding solution and eluates at different time points. (−)ESI FT-ICR mass spectra of groundwater and DOM-1x (C; left panel) and DOM-5x (D; right panel) feeding solutions and eluates; asterisk: doubly charged ions.

Figure 5

Dynamics of sediment bacterial biomass (A) and activity (B) over 171 days of incubation. The microcosms were infiltrated by the DOM-1x and DOM-5x feeding solution from left to right, entering the microcosms at sector I and leaving at sector VII. Samples were collected from the top sediment layer of each microcosm. Values are means of duplicate measurements of technical replicates.

Figure 6

Spatial distribution of sediment bacterial biomass (TCC) and activity (ATP) in the sediment zones that received DOM-1x (here upper segment) and DOM-5x (lower segment) feeding solution at the end of the experiment (t171). The contour plots combine data from sediment segments of the two microcosms that were subject to similar DOM supply.

Figure 7

Comparison of substrate utilization by the microbial communities from the DOM-1x and DOM-5x fed sediments.

Figure 8

Bacterial community composition in fresh aquifer sediment and in the sediments receiving DOM-1x and DOM-5x for 171 days.