Microbial Ecology

, Volume 76, Issue 1, pp 182–191 | Cite as

Microbial Reduction of Fe(III) and SO4 2− and Associated Microbial Communities in the Alluvial Aquifer Groundwater and Sediments

  • Ji-Hoon Lee
  • Bong-Joo Lee
Environmental Microbiology


Agricultural demands continuously increased use of groundwater, causing drawdown of water table and need of artificial recharge using adjacent stream waters. River water intrusion into groundwater can alter the geochemical and microbiological characteristics in the aquifer and subsurface. In an effort to investigate the subsurface biogeochemical activities before operation of artificial recharge at the test site, established at the bank of Nakdong River, Changwon, South Korea, organic carbon transported from river water to groundwater was mimicked and the effect on the indigenous microbial communities was investigated with the microcosm incubations of the groundwater and subsurface sediments. Laboratory incubations indicated microbial reduction of Fe(III) and sulfate. Next-generation Illumina MiSeq sequences of V4 region of 16S rRNA gene provided that the shifts of microbial taxa to Fe(III)-reducing and/or sulfate-reducing microorganisms such as Geobacter, Albidiferax, Desulfocapsa, Desulfuromonas, and Desulfovibrio were in good correlation with the sequential flourishment of microbial reduction of Fe(III) and sulfate as the incubations progressed. This suggests the potential role of dissolved organic carbons migrated with the river water into groundwater in the managed aquifer recharge system on the indigenous microbial community composition and following alterations of subsurface biogeochemistry and microbial metabolic activities.


Organic carbon Groundwater recharge Microbial community 


Vast diversity of microorganisms resides in aquifer, and certain metabolic activities of the microorganisms contribute to subsurface biogeochemical cycling and organic contaminant degradation [1, 2, 3]. However, usually low concentrations of organic matters are found in groundwater, and thus, the microbial population density in groundwater and subsurface sediments is usually very low compared to other microbial habitats such as soils and stream bed sediments [4]. In other words, supply of electron donors and acceptors could lead to metabolic activeness of indigenous microorganisms and subsequent influences on geochemical conditions in the groundwater and subsurface.

Respiratory electron donors representatively organic carbons (OCs), as well as electron acceptors such as oxygen, ferric iron, and nitrate, could be supplied into groundwater through a range of natural events such as annual recharge, unintended leaching of anthropogenic chemicals such as fertilizers and pesticides, biogeochemical redox cycling of some elements, and surface water mixing in hyporheic zone groundwater. In this study, as a collaborative effort, a river bank site was characterized to provide information on the site potentials as a managed aquifer recharge (MAR) test site, where river water would be injected into the aquifer by motorized system, and thus, organic matters in the surface water might be added to groundwater, potentially influencing microbial communities. We investigated potential microbial activities in subsurface sediments and groundwater of the hyporheic zone of the alluvial aquifer in the river bank by simulating migration of OCs transported with the flux of river water into groundwater.

Dissolved organic carbon (DOC) transported from river water into aquifer was suggested to act a strong influence on microbial community composition and diversity in the subsurface of a MAR site [5]. DOC in freshwater was also suggested to be significantly contributed by allochthonous organic matters and often from anthropogenic sources, such as agricultural and industrial wastewater [6]. In the process of infiltration of freshwater through subsurface aquifer systems, organic contaminants among the DOC could be attenuated by the indigenous microbial communities. Therefore, the role of OCs transported from freshwater recharging groundwater could be important in MAR systems, due to shift of the indigenous microbial community composition by utilization of the OCs.

Externally supplied OCs to the subsurface sediments from hyporheic zone were mimicked and suggested that microbial activities of reduction of ferric iron and sulfate increased in the microcosm incubations [7], which also showed microbial community shifts compared to the pristine communities [8]. Stegen et al. [9] showed the microbial community changes in the subsurface hyporheic zone groundwater due to alteration of OC composition by groundwater-surface water mixing, which stimulated heterotrophic respiration, leading to increased numbers of microbial taxa capable of degrading a broad suite of immigrated organic compounds. In this study, we aimed to examine whether there would be a microbial community shift in the subsurface reducing conditions of the sediment-groundwater from the river bank by the migrated DOCs and dissolved oxygen (DO) along with the pumping-derived influx of the river water. Along with physicochemical analyses, next-generation sequencing (NGS) was applied on the environmental DNAs extracted from the groundwater-sediment incubations, to elucidate the potentially elevated abundances and changes of heterotrophs, oxidizing the added DOCs coupled to the respiratory reduction of Fe(III) and sulfate as the terminal electron acceptors.

Materials and Methods

Sampling and Analyses of Subsurface Groundwater-Sediments

Groundwaters containing sediment suspensions were collected from a 14-borehole cluster installed for the feasibility test of the MAR in the river bank of the Nakdong River near Changwon city, South Korea [10]. One of the boreholes was selected for this study among the well cluster and samples of groundwater and sediments were collected from two depths of 10 and 33 m below ground surface (bgs) of the borehole MLW2 (Fig. S1 in Supplementary Information). There are two distinctively different aquifers vertically separated by an aquitard of silty clay layer at about 15 m bgs [10, 11]. The lower aquifer was characterized by high electrical conductivity (EC) of approximately 6000 μS cm−1 from about 30 m bgs, which is unusually high for terrestrial groundwater, compared to 200 to 300 μS cm−1 from the upper aquifer at about 10 m bgs. It was suggested that the groundwater in the lower aquifer was originated from the ancient trapped sea water [10, 11].

Groundwater was analyzed for temperature, pH, electrical conductivity (EC), dissolved oxygen (DO), and oxidation reduction potential (ORP) on site by portable probes. Concentrations of ferrous iron and aqueous sulfide were analyzed on site using the portable kits and portable spectrophotometer (HACH, Loveland, CO, USA). Cations and anions were analyzed using ion chromatography (IC) and inductively coupled plasma atomic emission spectroscopy (ICP-AES), respectively. The measurement results were summarized in Table 1. Groundwater samples were contained in 1-l plastic sampler bags without headspace as initially sampled, transported with ice packs, and stored at 4 °C until laboratory analyses and experimental preparations at the same time.
Table 1

Groundwater chemistry of the groundwater from the boreholes (MLW2, groundwater/sediments samples originated for this study; OBS1, reference for comparison)

Well, depth


OBS1 at 20 m bgs

10 m bgs

33 m bgs

On site

Water table (m)



Temp. (°C)








EC (μS/cm)




DO (mg/L)




ORP (mV)




Fe(II) (mg/L)

0.79 (after filtration)

28.93 (after filtration)

21.1 (after filtration)

S(−II) (μg/L)

17 (w/o filtration)

7 (after filtration)

1 (after filtration)

In laboratory

Ca2+ (mg/L)




Cl (mg/L)




NO2 (mg/L)




NO3 (mg/L)




PO4 3− (mg/L)




SO4 2− (mg/L)




HCO3 (mg/L)



Fe (μg/L)




Si (μg/L)




Mn (μg/L)




Sr (μg/L)




Incubation of the Groundwater-Sediments and Analyses

The sampled groundwater contained sediments from the silty sand aquifers. The sediment suspensions in groundwater were prepared for the two-depth samples to be similar in sediment-groundwater ratio visually. The sediments were in 3.3 ± 0.8 mg ml−1 and 5.1 ± 2.0 mg ml−1, respectively, for the 10- and 33-m samples. Three treatments were setup for each of the samples in 27-ml anaerobic pressure tubes with butyl rubber thick stopper and aluminum seal with the total volume of 20 ml. For the control, the suspensions were autoclaved at 121 °C for 15 min. Two live treatments were separated by adding external OC or not. For the OC setup, mixture of acetate and lactate was added to be the final concentration of 1 mM for each, while the as-prepared incubations were setup by adding 2 ml of sterile dH2O into 18 ml of the groundwater suspension to compensate the volumetric dilution factor in the OC setup. The three sets of suspensions of groundwater and sediments were purged with 100% N2 and the head space was purged with a mixture gas (4% H2 and 96% N2). All the experiments were in triplicate and the tubes were incubated at 25 °C. For the added concentrations of the organic carbons, we roughly calculated the DOC concentrations in the river water and in the alluvial aquifer, and combined those values for mimicking the mixing of groundwater and surface water. We used the frequently found low molecular organic matters of lactate and acetate at 1 mM each, in the incubations. Based on the chemical oxygen demand (COD) values in the Nakdong River water, approximation was made for the concentrations of lactate or acetate (details not included). DOC concentrations from alluvial aquifers in variety ranges with the rough average of 10 mg l−1 [12] were also converted to lactate or acetate.

For 0.5-N HCl-extractable Fe(II) analysis, the suspensions (0.5 ml) during incubation was removed by syringe needles and mixed with 0.5 ml of 1-N HCl. After 10 min of extraction, the solutions were filtered through 0.22-μm filters and the filtrates were analyzed by ferrozine assay [13]. Aqueous sulfide was analyzed by the methylene blue method [14] using HACH protocol and reagents (Method 8131, 181632/ 181732; Loveland, CO) with appropriate adjustment to the experimental volume. Aliquots of samples (1 ml) were removed from the incubation tubes and the reagents (each 40 μl) were added in the anaerobic glovebox (VS-5600A, Vision Scientific, Daejeon, Republic of Korea). After 5 min of reaction, the solutions were centrifuged (5000 × g and 22 °C for 2 min), and the supernatants were measured at 665 nm. Sulfide standard solutions were prepared freshly using Na2S·9H2O in the glovebox. To identify potential mineralogical alterations, X-ray photoelectron spectroscopy (XPS, Model AXIS Ultra DLD, KRATOS Inc.) and transmission electron microscopy (TEM, JEM-2200FS(HT), JEOL) were performed on the incubated sediment samples. The sediment samples were collected into the hole of a plastic disk with the bottom sealed with 2-mil Kapton tape and dried within the desiccator in the glovebox and transported for the XPS analysis with the top opening sealed with the Kapton tape after being dried, as similarly with the procedure for X-ray absorption spectroscopy [15]. For TEM preparation, an aliquot of the dark/black sediment suspension was dropped on the electron microscopic grid and dried in the anoxic glovebox as suggested in other document [16].

Genomic DNA Extraction and Microbial Community Analysis

During the incubation, the groundwater/sediments were sampled at selected time points and the genomic DNAs were extracted from the triplicate suspensions (approximately 300 mg) using the PowerSoil DNA extraction kit (MO BIO Inc., Carlsbad, CA, USA). For each time point, the three DNA extracts from the triplicate incubations were pooled into one. 16S rRNA gene libraries were constructed and the amplicons were sequenced by Macrogen (Seoul, Republic of Korea). The V4 region (length, ca. 250) of 16S rRNA gene was amplified using the barcoded primer sets of F515/R806 [17], and the amplicons were paired-end sequenced on the Illumina MiSeq platform (Illumina Inc., San Diego, CA, USA).

The sequence was processed using Mothur ver. 1.33.3 [18] by referring the suggested standard operating procedure for MiSeq sequences [19]. Sequence errors were reduced by a series of subroutine processes in the Mothur and screened for chimeras using UCHIME [20]. With removal of low-quality reads and non-bacterial sequences, the total number of sequences from all 8 samples was 78,691 ranging from 1038 to 24,644 reads of the samples. With an average length of 253 bases (246–256 bases), the sequences were aligned against the SILVA database ver. 102 [21], and classified against the Ribosomal database project (RDP) 16S rRNA gene training set ver. 9 to be assigned to operational taxonomic units (OTUs) at a 3% dissimilarity level. For the downstream analyses of community diversity comparisons, the sample sequences were normalized to the minimum sample number, 1038 by random subsampling, with the observed OTUs (species level) ranging from 342 to 742 (Table 2).
Table 2

A sequence summary and alpha diversity: richness and diversity indices of the samples at selected times


No. of sequences

Good’s coverage (%)

Observed OTUs (species)






Inverse Simpson

10 m, bgs

T = 0








T = 3 days








T = 20 days








T = 20 days, + OC








33 m, bgs

T = 0








T = 3 days








T = 20 days








T = 20 days, + OC








Results and Discussion

Microbial Reduction of Ferric Iron and Sulfate

Incubation of the sediment slurries in groundwater from both the depths, 10 and 33 m bgs, showed increases of the concentrations of acid-extractable Fe(II) in the live setups both with and without additional OC over time, while the heat-treated controls did not show increase of both acid-extractable and aqueous Fe(II) (Fig. 1, Fig. S2). This suggested that Fe(III) reduction in the incubations was biological. For the iron reduction in the absence of addition of OC, it was speculated that there might be microbe-accessible indigenous OCs in the groundwater/sediment suspensions. We also considered the potential role of autotrophic iron-reducing microorganisms on the iron reduction without the added OCs, as suggested for the participants of iron reduction in the batch incubations of the acidic mine pit sediments [22]. The microbial Fe(III) reduction was larger in the sediment from 33 m bgs (from the initial 663 to the final 2240-μM Fe(II) for no OC addition) than that from 10 m bgs (from the initial 326 to the final 1344-μM Fe(II) for no OC addition). Fe(III) reductions in the OC-added experiments were similar with those in the no OC-added incubations, but showed a bit increased concentrations than those w/o OC addition. While the incubations progressed, the sediments turned to dark/black. Acid-labile sulfide monitored over time indicated that sulfide production started mostly after Fe(II) reached almost the maximum concentrations, probably by microbial sulfate reduction (Fig. 1c, d). In the heat-treated control incubations, there was no obvious change of sulfide or color. There was indigenous sulfate (SO4 2−) in the groundwater at 95 and 223 mg/l, respectively, for 10 and 33 m bgs (Table 1). Therefore, it was likely that microbes in the groundwater and/or sediments reduced sulfate to produce sulfide and dark/black color sediments were formed presumably by iron (Fe2+)-sulfide surface coating.
Fig. 1

Cumulative concentrations of 0.5-N HCl-extractable Fe(II) (a, b) and acid-labile (methylene blue reactive) sulfide (c, d) from the incubations of the sediment-groundwater at depths of 10 m (a, c) and 33 m (b, d). The incubation tube photos at 17 days, respectively, for 10 and 33 m. The blue diamond symbols indicate the incubation without addition of OC, while the orange rectangle symbols indicate the incubation with added OC. The gray triangle symbols indicate heat-treated incubation. Error bars indicate standard deviations from the average values from triplicates

The XPS analysis on the pristine and incubated sediments that was performed to identify potential Fe(II)-sulfide materials could not resolve sulfur signal at the binding energy of near 162 eV (Fig. S3). By TEM analysis, lots of electron-dense nanoparticulates were observed (Fig. S4), which was speculated as Fe-sulfide materials precipitated and/or coated on the silicate minerals. Feldspars and micas were common in the sediments.

Shift of Microbial Community Structures

Observed OTU or species numbers from the normalized sample sizes showed a roughly declining trend in the shallow depth incubations (10 m bgs) and increasing numbers in the deep sediments (33 m bgs) with increasing incubation time (Table 2). These patterns seemed to be supported by the richness indices. Species richness, which is the number of different species, estimated by the Chao1 index, indicated that species increased in the shallow depth sediments and increased in the deep sediments (Table 2). The similar patterns were more obvious with the abundance-based estimator (ACE) index. This might be explained by the presumption that relatively diverse indigenous aerobic, facultative anaerobic and/or microaerophilic microorganisms in the shallow 10 m-bgs sediment decreased during the development of strictly anaerobic condition, while relatively less populated indigenous anaerobic microorganisms in the deep 33 m-bgs sediment were enriched in the incubations. Species diversity indices of Shannon and inverse Simpson estimators, reflecting richness and species evenness as well, showed increasing trends with time in both the depths. Index of inverse Simpson indicated more obvious increases in the bacterial diversities in both the samples with the incubation time. Thus, we could suggest that in general, bacterial species decreased with relative even distribution across the species in the shallow sediment incubations, and both bacterial species richness and evenness increased in the deep sediment incubations.

When the MiSeq sequences were clustered at 3% dissimilarity and classified for the taxonomy, a total of 19 phyla were identified from the total 8 samples (time, 0, 3, and 20 days from the no OC-added treatment and 20 days + OC for the OC-added treatment, for both 10 and 33 m bgs). The phylum Proteobacteria was the most abundant in the initial sediments at 87.7 and 86.9%, respectively, from 10 and 33 m bgs (Fig. 2), followed by the second largest groups of the unclassified Bacteria (6.7%) and Bacteroidetes (3.8%), respectively. Over time, the ratios of Proteobacteria decreased to 81.6 and 70.9%, respectively, at 3 and 20 days, in the sediments of 10 m-bgs w/o OC addition, and to 84.5 and 38.8%, respectively, at 3 and 20 days, in the sediments of 33 m-bgs w/o OC addition.
Fig. 2

Relative sequence abundances of bacterial phyla and classes in Proteobacteria of the groundwater/sediments from depths of 10 m (a) and 33 m (b) at the selected times, based on the OTUs clustered at 97% similarity (species-level). “+OC” indicates the incubations supplemented with organic carbons

In the 10 m-bgs sediment, phylum Acidobacteria increased for its relative abundance (0.3% at the beginning to 11.1% at 20 days) w/o OC addition, whereas Bacteroidetes (1.9 to 25.9%) and Firmicutes (0.9 to 16.6%) increased with added OC (Fig. 2a). The most abundant class, Betaproteobacteria (78.4%), in initial sediments from 10 m bgs gradually decreased to 30.0 and 16.7%, respectively, without and with OC addition after 20 days, while Deltaproteobacteria (1.5%) increased, respectively, to 24.4 and 12.1% at 20 days (Fig. 2a). In the 33 m-bgs sediment, Actinobacteria (2.0%) and Bacteroidetes (3.8%) increased to 7.6 and 25.2%, respectively, without OC addition at 20 days, while there were minor increases with OC addition at 20 days (Fig. 2b). Classes Beta- (26.6%), Delta- (4.1%), and Gamma-proteobacteria (47.9%) showed reduced relative abundances at 21.4, 0.5, and 7.3%, respectively, without OC addition, after 20 days (Fig. 2b). With addition of OC, however, Deltaproteobacteria increased to 30.5%. Looking at the horizontal dendrogram calculated by the Bray-Curtis dissimilarity, Beta-, Delta-, and Gamma-proteobacteria occurred more often together (Fig. S5).

At genus level, the sequences were clustered to 381 OTUs including 120 “unclassified.” The number of OTU whose abundance was larger than 3% in at least one of the samples was 23 (Fig. 3). Mostly, the genera Alibidiferax, Geobacter, Pseudomonas, and Geothrix were abundant across the samples, and interestingly, those microorganisms such as Alibidiferax ferrireducens, Geobacter sulfurreducens, and Geothrix fermentans are known for their ability to use Fe(III) as respiratory electron acceptor. In the shallow groundwater/sediment (10 m bgs), Alibidiferax was highly abundant at 30.8% in the beginning and decreased to 8.2 and 3.9%, respectively, without and with OC addition after 20 days of incubation (Fig. 3a). Abundances of Geobacter (0.6%), Geothrix (0.0%), and Pseudomonas (1.7%) increased respectively to 14.9, 9.2, and 4.4% without OC addition and to 6.0, 3.3, and 2.0% with OC addition (Fig. 3a). In the deep groundwater/sediment (33 m bgs), however, abundant Pseudomonas (41.2%) decreased to 0.2 and 4.6%, respectively, without and with OC addition (Fig. 3b). Unlikely, in the shallow sediment, only minor fractions of Geothrix were found, while increased abundances of Albidiferax and Geobacter were found only at 3 days without OC addition and at 20 days with OC addition. This seemed to suggest that the heterotrophic Fe(III)-respiring microorganisms, Albidiferax and Geobacter, could use indigenous or added, easily accessible OCs and the populations ratios decreased at 20 days, after depletion of the indigenous, residual OCs without OC addition. Microorganisms capable of sulfur and/or sulfate metabolisms, such as Desulfocapsa, Desulfuromonas, and Desulfovibrio, were mostly found from the OC-added sediment at 20 days, which suggests that heterotrophs respiring alternative electron acceptors to O2 were populated by aid of the added OCs, possibly after the previously accessible oxidant; Fe(III) was consumed by Fe(III)-reducing microorganisms. Those microorganisms were well clustered especially in the deep sediment with the added OC at 20 days, by the Bray-Curtis dissimilarity (Fig. 3c). In addition, although it was difficult to find a chronological good relationship between the samples by the vertical dendrogram, the community structures at the beginning from 10 and 33 m bgs (M10T0 and M33T0) were most independent rather than other samples. This could suggest that the anaerobic incubation has changed the microbial community structures that adapted to the formed environments, from the indigenous ones.
Fig. 3

Relative sequence abundances of bacterial genera with more than 3% from at least one sample, for the 10-m (a) and 33-m (b) incubations. c A heat map illustrating genus abundances with dendrograms on the vertical and horizontal axes based on the Bray-Curtis dissimilarity matrix. Names of the samples consist of depth (M10 or M33), time (T0, T03D, or T20D), and organic carbon addition or not (OC or non)

Dissimilarities between the community structures of the samples were measured by the Yue and Clayton estimator (theta-YC) [23] and presented as a non-metric multidimensional scaling (NMDS) analysis in Fig. 4a. Four communities from each depth (10 and 33 m bgs) were grouped as an ellipsoid (Fig. 4a). As similarly to the above-described result by the dendrogram of Bray-Curtis dissimilarity, both the communities (10 m and 33 m bgs) at the beginning were remote from other samples. At 10 m-bgs samples, the community shift was obvious probably by development of anaerobic condition, and the communities with and without OC addition at 20 days were rather close to each other than the community at 3 days. In order to identify major OTUs responsible for shifting the samples along the NMDS axes, correlations of the relative abundance of each OTU with the axes were calculated by Spearman’s correlation. OTU13 (Desulfocapsa, p = 0.045) and OTU19 (unclassified Moraxellaceae, p = 0.033) were responsible for moving the sample points in a positive direction along the axis 1 (NMDS1). OTU1 (Albidiferax, p = 0.047) was responsible for a negative shift, while OTU17 (unclassified Betaproteobacteria, p = 0.048) was for a positive shift along with the axis 2 (NMDS2). For the axis 3 (NMDS3), Pseudomonas (OTU4, p = 0.006), Desulfuromonas (OTU6, p = 0.007), Geobacter (OTU8, 9, and 16, p < 0.04), and unclassified Comamonadaceae (OTU15, p = 0.048) were responsible for a negative shift. This seems to support that the added OCs were responsible for the community separation from the community w/o OC addition along with the axis 3 by the above-described microorganisms. This was also supported by the community diversity index of inverse Simpson that those index differences between the samples with and without OC addition were very large in both the depths (Table 2).
Fig. 4

Visualized similarities of the bacterial communities by a NMDS (stress = 0.10, R 2 = 0.90, p < 0.013) of the theta-YC distances (green and yellow ellipsoids, respectively, for 10- and 33-m incubations) and b PCoA of the weighted UniFrac distances among the samples (loadings given in the axis parentheses)

In order to assess the similarity between the community structures, the weighted UniFrac [24] distance metrics were visualized by 2D principal coordinates (PCoA), resulting in approximately 42.7 and 20.0% of the variation (62.7% of the total), respectively, by the first and second axes (Fig. 4b). The separations of the sample points between with and w/o OC at both 10 and 33 m bgs were mainly contributed by the axis 1 (PCoA1), and the distance among the deep (33 m) samples was larger than that from the shallow (10 m) ones (Fig. 4b). We assumed that there might have been significant amount of indigenous OCs in the shallow (10 m bgs) groundwater/sediments and supported the microbial activities up to 20 days. In the 33 m-bgs samples, however, the added OCs might have influenced to separation of the communities with the long distance between “20 days” and “20 days + OC” (Fig. 4b). There might have been only minor amount of indigenous OCs in the deep sediments. We speculated that through the microbial uses of the OCs, the potential, alternative electron acceptors of Fe(III) and sulfate were used sequentially, respectively, at 3 and 20 days, and the responsible microorganisms were relatively populated.


Some microorganisms can survive only under selected substrates in given environmental conditions, and changes in the environment may affect the microbes to a greater extent than others [25, 26]. Compositional change of organic substrates in an environment could have a larger influence on taxa that directly metabolize the newly introduced organic carbons than those that utilize them as nutrients. It was suggested that groundwater-surface water mixing could stimulate and elevate abundances of the microbial taxa associated with the altered carbon composition [9]. Microcosms amended with organic substrates were also shown to indicate sweeping replacements of community members [27]. In a study that investigated microbial communities in the similar systems of managed aquifer recharge sites with this study system, dissolved organic carbon (DOC) was suggested as a strong driving factor influencing on microbial community composition, and positive correlations of Betaproteobacteria and Gammaproteobacteria with DOC concentration were suggested [5].

Based on the results in this study, we could suggest that the microbial community structures of the groundwater/sediment microcosms presumably shifted to microbial taxa that were capable of respiring less-favored oxidants, as the incubations progressed, and the community-wide changes were characteristic of the stimulated organic carbon. The shifts of microbial taxa to Fe(III)-reducing and/or sulfate-reducing microorganisms were in good correlation with the sequential flourishment of microbial reduction of Fe(III) and sulfate as the incubations progressed.

From this study, we may suggest that DOCs brought along with river water into aquifer could stimulate and change the microbial compositions, which would be developed further to anaerobic communities in the groundwater with time. Li et al. [5] and Stegen et al. [9] also proposed the similar suggestions that river water saturated the nearby sediment pores and transported soluble carbons that in turn stimulated microbial respirations and community shifts. By differentiating groundwater pumping for agricultural irrigation or municipal uses such as river bank filtration, artificial recharge of surface water could bring DOCs into groundwater, which could influence on the water quality and potentially public health as drinking water by the water chemistry change through the microbial community and metabolic change and accompanied potential changes of subsurface sediment geochemistry and groundwater chemistry.


Funding Information

This work was supported by the Korea Institute of Geoscience and Mineral Resources (Basic Research Project No. 14-3211) funded by the Ministry of Science, ICT and Future Planning and also in part by the Cooperative Research Program for Agricultural Science and Technology Development (Project No. PJ012263), Rural Development Administration, Republic of Korea.

Supplementary material

248_2017_1119_MOESM1_ESM.docx (5 mb)
ESM 1 (DOCX 5155 kb)


  1. 1.
    Griebler C, Lueders T (2009) Microbial biodiversity in groundwater ecosystems. Freshw Biol 54:649–677. CrossRefGoogle Scholar
  2. 2.
    Hohnstock-Ashe AM, Plummer SM, Yager RM, Baveye P, Madsen EL (2001) Further biogeochemical characterization of a trichloroethene-contaminated fractured dolomite aquifer: electron source and microbial communities involved in reductive dechlorination. Environ Sci Technol 35:4449–4456. CrossRefPubMedGoogle Scholar
  3. 3.
    McGuire JT, Long DT, Klug MJ, Haack SK, Hyndman DW (2002) Evaluating behavior of oxygen, nitrate, and sulfate during recharge and quantifying reduction rates in a contaminated aquifer. Environ Sci Technol 36:2693–2700. CrossRefPubMedGoogle Scholar
  4. 4.
    Eydal HSC, Pedersen K (2007) Use of an ATP assay to determine viable microbial biomass in Fennoscandian Shield groundwater from depths of 3–1000 m. J Microbiol Methods 70:363–373. CrossRefPubMedGoogle Scholar
  5. 5.
    Li D, Sharp JO, Saikaly PE, Ali S, Alidina M, Alarawi MS, Keller S, Hoppe-Jones C, Drewes JE (2012) Dissolved organic carbon influences microbial community composition and diversity in managed aquifer recharge systems. Appl Environ Microbiol. 78:6819–6828. CrossRefPubMedPubMedCentralGoogle Scholar
  6. 6.
    Jay TL, Liza EP (2005) Source and supply of terrestrial organic matter affects aquatic microbial metabolism. Aquat Microb Ecol. 39:107–119Google Scholar
  7. 7.
    Lee J-H, Fredrickson JK, Kukkadapu RK, Boyanov MI, Kemner KM, Lin X, Kennedy DW, Bjornstad BN, Konopka AE, Moore DA, Resch CT, Phillips JL (2012) Microbial reductive transformation of phyllosilicate Fe(III) and U(VI) in fluvial subsurface sediments. Environ Sci Technol 46:3721–3730. CrossRefPubMedGoogle Scholar
  8. 8.
    Lin X, Kennedy DW, Fredrickson JK, Bjornstad BN, Konopka AE (2012) Vertical stratification of subsurface microbial community composition across geological formations at the Hanford site. Environ Microbiol 14:414–425. CrossRefPubMedGoogle Scholar
  9. 9.
    Stegen JC, Fredrickson JK, Wilkins MJ, Konopka AE, Nelson WC, Arntzen EV, Chrisler WB, Chu RK, Danczak RE, Fansler SJ, Kennedy DW, Resch CT, Tfaily M (2016) Groundwater–surface water mixing shifts ecological assembly processes and stimulates organic carbon turnover. Nat Commun. 7:11237. CrossRefPubMedPubMedCentralGoogle Scholar
  10. 10.
    Lee J-H, Unno T, Ha K (2017) Vertical and horizontal distribution of bacterial communities in alluvial groundwater of the Nakdong River bank. Geomicrobiol J 30:1–7. CrossRefGoogle Scholar
  11. 11.
    Moon S-H, Ha K, Kim Y, Koh D-C, Yoon H (2014) Examination for efficiency of groundwater artificial recharge in alluvial aquifer near Nakdong River of Changweon area, Korea. Econ Environ Geol. 47:611–623CrossRefGoogle Scholar
  12. 12.
    Tye AM, Lapworth DJ (2016) Characterising changes in fluorescence properties of dissolved organic matter and links to N cycling in agricultural floodplains. Agric Ecosyst Environ. 221:245–257. CrossRefGoogle Scholar
  13. 13.
    Stookey LL (1970) Ferrozine—a new spectrophotometric reagent for iron. Anal Chem. 42:779–781CrossRefGoogle Scholar
  14. 14.
    Cline JD (1969) Spectrometric determination of hydrogen sulfide in natural waters. Limnol Oceanogr. 14:454–458CrossRefGoogle Scholar
  15. 15.
    Lee J-H, Zachara ZM, Fredrickson JK, Heald SM, McKinley JP, Plymale AE, Resch CT, Moore DA (2014) Fe(II)- and sulfide-facilitated reduction of 99Tc(VII)O4 in microbially reduced subsurface sediments. Geochim Cosmochim Acta 136:247–264CrossRefGoogle Scholar
  16. 16.
    Lee J-H, Kennedy DW, Dohnalkova A, Moore DA, Nachimuthu P, Reed SB, Fredrickson JK (2011) Manganese sulfide formation via concomitant microbial manganese oxide and thiosulfate reduction. Environ Microbiol. 13:3275–3288. CrossRefPubMedGoogle Scholar
  17. 17.
    Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Lozupone CA, Turnbaugh PJ, Fierer N, Knight R (2011) Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc Natl Acad Sci U. S. A. 108:4516–4522. CrossRefPubMedGoogle Scholar
  18. 18.
    Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ, Sahl JW, Stres B, Thallinger GG, Van Horn DJ, Weber CF (2009) Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol. 75:7537–7541. CrossRefPubMedPubMedCentralGoogle Scholar
  19. 19.
    Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD (2013) Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl Environ Microbiol. 79:5112–5120. CrossRefPubMedPubMedCentralGoogle Scholar
  20. 20.
    Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R (2011) UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27:2194–2200. CrossRefPubMedPubMedCentralGoogle Scholar
  21. 21.
    Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glockner FO (2013) The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41:D590–D596. CrossRefPubMedGoogle Scholar
  22. 22.
    Porsch K, Meier J, Kleinsteuber S, Wendt-Potthoff K (2009) Importance of different physiological groups of iron reducing microorganisms in an acidic mining lake remediation experiment. Microb Ecol 57:701–717. CrossRefPubMedGoogle Scholar
  23. 23.
    Yue JC, Clayton MK (2005) A similarity measure based on species proportions. Commun Stat Theory Methods 34:2123–2131. CrossRefGoogle Scholar
  24. 24.
    Lozupone C, Knight R (2005) UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol. 71:8228–8235. CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Knelman JE, Nemergut DR (2014) Changes in community assembly may shift the relationship between biodiversity and ecosystem function. Front Microbiol 5:424. CrossRefPubMedPubMedCentralGoogle Scholar
  26. 26.
    Krause S, Le Roux X, Niklaus PA, Van Bodegom PM, Lennon JT, Bertilsson S, Grossart H-P, Philippot L, Bodelier PLE (2014) Trait-based approaches for understanding microbial biodiversity and ecosystem functioning. Front Microbiol 5:251. CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Nakatsu CH, Carmosini N, Baldwin B, Beasley F, Kourtev P, Konopka A (2005) Soil microbial community responses to additions of organic carbon substrates and heavy metals (Pb and Cr). Appl Environ Microbiol. 71:7679–7689. CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2017

Authors and Affiliations

  1. 1.Department of Bioenvironmental ChemistryChonbuk National UniversityJeonjuRepublic of Korea
  2. 2.Groundwater DepartmentKorea Institute of Geoscience and Mineral ResourcesDaejeonRepublic of Korea

Personalised recommendations