Printer Friendly

Bacterial community structure and denitrifier (nir-gene) abundance in soil water and groundwater beneath agricultural land in tropical North Queensland, Australia.


The Great Barrier Reef (GBR), off the north-east coast of Australia, is the largest coral-reef ecosystem in the world. It is of immense ecological and economic significance, and is listed as a World Heritage Area. However, the GBR and its lagoon are adversely affected by elevated levels of nutrients, sediments, and other pollutants moving through its catchments due to the agricultural, urban, and industrial development over the past 150 years (Greiner et al. 2005). Of the pollutants derived from agricultural land-use, nitrogen (N) is the main nutrient of concern due to its relatively high application rates, its mobility, and its effect on aquatic ecology. Nitrate, the most mobile and biologically available form of N, is transported through the landscape in groundwater and runoff, mostly as pulses during periods of high rainfall (Rasiah et al. 2003). Although there is some understanding of the magnitude of riverine discharge of nitrate to the GBR lagoon, data relating to the importance of groundwater are sparse. The role of microorganisms and associated ecosystem processes in the quality of groundwater discharging to the GBR is a significant knowledge gap (Post et al. 2007).

The microbial ecology of groundwater is poorly explored (Madsen and Ghiorse 1993; Chapelle 2001). However, interest has increased recently, due to (1) the recognition that groundwater ecosystem functions affect the quality of water discharging into aquifers, rivers, estuaries, and coastal waters, including microbially driven biogeochemical reactions relating to nutrient and mineral cycling (Chapelle 2001); (2) the requirement to bio-remediate those aquifers contaminated with organic and inorganic pollutants (Anderson et al. 2003; Farhadian et al. 2008); and (3) the need to understand the survival and fate of pathogenic protists, viruses, and bacteria from human and animal waste streams (Gordon and Toze 2003). By far the most important route for nitrate removal from groundwater is through microbiological activity. Microbial dissimilatory denitrification of groundwater represents an important ecosystem function that occurs before groundwater discharge (Slater and Capone 1987; Smith and Duff 1988). The fundamental physicochemical conditions conducive to heterotrophic denitrification, regardless of environment, are well known: hypoxic to anoxic conditions and presence of organic matter and N (nitrate, nitrite, etc). However, the factors that shape the microbial community structure and activity are still unclear, particularly as the diverse range of bacterial taxa capable of denitrification may not be actively involved in this process for much or all of their time. Dissimilatory denitrification is a step-wise process resulting in the conversion of dissolved nitrate to nitrite, and then to gaseous N compounds, removed to the atmosphere. The reduction of nitrite to nitric oxide is a key step, in that it involves formation of the first gaseous intermediate. The genes encoding nitrite reductases of the cytochrome [cd.sub.1]-type (nirS) and copper-containing type (nirK) have been used as functional markers for exploring the ecology of denitrifying communities in the environment (Philippot and Hallin 2005; Kandeler et al. 2006).

The aim of the work reported here was to examine microbial diversity in saturated soil zones above and below the water-table and under two contrasting but typical agricultural sites in the wet tropics of Australia. In particular, we sought to (1) compare community structure in the unsaturated, soil-water zone (lysimeter samples) and the saturated groundwater zone (bore samples) zone; (2) determine the environmental parameters that influence bacterial community structure in soil-water and groundwater; and (3) determine the abundance of ecologically significant, nitrite-reducing bacteria.

Materials and methods

Site description

Sugarcane is the dominant crop in the GBR catchments (188 000 ha; QLUMP 1999). Bananas and other horticultural crops occupy much smaller areas (10000ha of bananas; QLUMP 1999), but fertiliser application rates can be high. This study was carried out at one site under sugarcane and another site under bananas; both sites are in the Tully Murray catchment, between 17[degrees]30' and 18[degrees]30'S and between 145[degrees]30' and 146[degrees]05'E. The Tully and Murray catchments are regarded as a single system because their flows merge on the floodplain during the wet season. Mean annual rainfall (averaged monthly) varies from ~1500mm on the south-western margins of the catchment to >4500mm in the mountains of the northern margin, mostly falling between December and May, and followed by a dry winter. Some 72% of the catchment consists of mostly montane rainforest and sclerophyll forest, with sugarcane (13%), grazing (5%), forestry (3%), and bananas (3%) on the alluvial plains and fans. The Tully-Murray basins are considered of medium to medium--high 'potential ecological impact rating' in relation to diffuse-source pollution to the inner GBR (Greiner et al. 2005).

The study sites differed not only in crop type but also in other physical characteristics. The sugarcane site is flat, poorly drained, and with annual rainfall of 2575, 1307, and 2649 mm in the 2004-06 study period. The soil is a Hydrosol (Isbell 2002) with a sandy loam texture. The banana site, on the banks of the Tully River, is gently sloping with 2004-06 annual rainfall of 3585, 2390, and 3513 mm. The soil is a brown Dermosol with a loamy texture. The sugarcane site was cleared in 1964-65, sown to pasture for cattle grazing, and converted to rainfed sugarcane in 1996. During the study period, the sugarcane was in its first third ratoon phase. Annual fertiliser application (kg/ha) since 1996 in the plant phase of production has been 100-150 N as urea, 60 P as superphosphate, and 150 K as muriate of potash, with all of the P and K and about half of the N usually applied at planting in September and a second urea application side-dressed in October/November. In the ratoon crops, 100 kg N/ha as urea and 180 kg K/ha as muriate of potash have been applied after harvesting the previous crop. The cropping cycle has usually been one plant plus three or four ratoons per cycle. The banana site was cleared for beef cattle grazing in the 1940s and converted to bananas in the 1980s. Since then, bananas have been grown for periods of five years (one plant plus four ratoons) followed by one year of grass fallow. During the study period the banana crop was in its second--fourth ratoon phase, receiving irrigation and fertilisation via micro-sprinklers during July-December, depending on rainfall. Annual fertiliser inputs (kg/ha, 2004-06) were 300-450N, 30 P, 430K, 38 S, 3 Cu, and 5 Zn. The N was applied as ammonium sulfate and ammonium nitrate, at a ratio of applied ammonium-N to nitrate-N 1.6 : 1.

Around the sugarcane site, sugarcane fields extend for ~5 km in most directions. However, upslope of the site there is ~100 m of sugarcane, then 2km of grassland, then the forested mountains. Around the banana site, banana fields are the predominant land use; they extend for ~4 km upslope parallel to the Tully River and 1 km upslope perpendicular to the river.

Collection of water draining below the root-zone

At both sites, six lysimeters were installed (l.5m apart, 1 m deep) at the start of the study (March 2004) to collect leachate from below the root-zone. Each lysimeter consisted of an open-topped PVC drum (300 mm diameter, 280 mm tall) containing three ceramic suction cups set in silica flour (general use discussed in Fares et al. 2009). After installation, the soil was manually replaced according to the original profile. A vacuum line connected the ceramic cups in each lysimeter to a buried, PVC water trap (5-L capacity) with an above-ground access tube. To simulate natural drainage, leachate was extracted under low, constant vacuum (~15 kPa), generated by a falling column of water, regulated by float switches and a pump. Samples were collected only after the lysimeters had settled in, having received ~700 mm rainfall, and were taken weekly, or more frequently if rainfall was high. Between sample collections the water was held in below-ground reservoirs (at soil temperature). Samples of measured volume were placed in polypropylene bottles, chilled immediately (~4[degrees]C), and frozen the same day. Analysis was for pH, electrical conductivity (EC), ammonium-N (N[H.sub.4.sup.+]-N), nitrate-plus nitrite-N (NOx), dissolved reactive P (DRP), and dissolved organic C (DOC). As nitrite-N (N[O.sub.2]-N) concentrations are usually negligible, we hereafter refer to 'nitrate-plus nitrite-N' as 'nitrate-N'. At the banana site, samples were collected on 21 occasions (6 July 2004 to 21 September 2005), covering two dry seasons and one wet season. At the sugarcane site, the water table rose above the lysimeters during the wet seasons, so these samples can be considered as shallow groundwater and unsuitable to characterise the biogeochemistry below the root-zone. Consequently, fewer samples were taken than at the sugarcane site.

From the sugarcane site, five lysimeter samples were used for microbial analysis (lysimeters 1, 2, 3, 4, and 6), and these were collected on 25 January 2005. From the banana site, 16 lysimeter samples were used: a sample from each of the six lysimeters collected on 25 January 2005 and two samples collected from five lysimeters over the period 23 February to 4 March 2006. The microbial compositions of the lysimeter water samples would have been influenced by the process of drawing water through the silica flour and ceramic cups, and thus are not necessarily the same as microbial composition of the actual soil water. Unfortunately, it is not possible to sample water from the unsaturated zone in situ without drawing it through some sort of porous medium. This sampling bias needs to be taken into account when comparing microbial composition of the lysimeter and bore water.

Collection of groundwater

Both sites are underlain by an unconfined aquifer composed of Quaternary alluvium of variable texture. Bores were installed at three locations at the sugarcane site and four at the banana site. The bores were installed 50m apart in a spatial design that allowed for determination of groundwater flow direction. Bores were cased with a sealed-base PVC pipe (43 mm diameter). The bottom 3 m of the pipes (7-10 or 8-11 m depth) was slotted and wrapped with a 250-[micro]m pore-size polyester filter sheath. The hole surrounding the casing was back-filled with coarse sand to a depth of ~5-10 m, then bentonite (~4.85-5 m), soil (~4-4.85 m), and cement grout (0-~4 m). The depths to the water table were recorded and samples taken every 7-10 days during the wet seasons. Collection of samples followed the procedure of Alexander (2000). At each sampling, three bore volumes of water were removed using a hand bailer, and then the sample was taken using the same bailer. Sample treatment and analysis was the same as for lysimeter samples.

From the sugarcane site, six samples of groundwater were collected for microbial analysis, one sample from each of the three bores on both 16 and 22 February 2006. From the banana site, eight samples were used, one sample from each of the 4 bores on both 16 and 22 February 2006.

DNA extraction

DNA was extracted from samples taken during the wet season from bores and lysimeter samples, using the MoBio water DNA extraction kit (MoBio Inc., CA). Biological material in the water samples was collected onto sterile 0.22-[micro]m filters by vacuum filtration. Excess suspended material, which clogs filters, was removed by centrifugation (6000G for 10 min) from some samples and was not included in the extraction. After extraction, DNA was stored in TE buffer at -20[degrees]C. The concentration of DNA in each sample was fluorometrically quantified using PicoGreen dsDNA quantitation reagent (Invitrogen) against a standard curve of Lambda-phage DNA on a Stratagene MX3000P qPCR system. The standard curve was calibrated on a NanoDrop ND-1000 spectrophotometer.

Bacterial community structure

Bacterial community structures were compared using denaturing gradient gel electrophoresis (DGGE; Muyzer et al. 1993). Polymerase chain reaction (PCR) was used to amplify a variable region of the 16S rRNA gene using primers 27F-GC and 534R (Lane 1991; Muyzer et al. 1993), as described by Wakelin et al. (2008a). DGGE of PCR-amplified 16S rRNA genes was performed in the Ingeny PhorU system using urea--formamide denaturing gradients of 40-60% in a 8% linear acrylamide-bis-acrylamide (37.5:1) gel. Electrophoresis was conducted at 60[degrees]C at 110 V for 18 h. Post-electrophoresis, DGGE gels were stained in SYBR gold (1 x in TAE buffer; Molecular Probes) for 30min and visualised on a DarkReader (Clare Chemicals Inc., USA). Bands were digitally captured using an Olympus E-500 digital SLR camera and the position and intensity of bands determined using TotalLab[TM] software (Nonlinear Dynamics Ltd). Band intensity data were 4th-root transformed and a resemblance matrix generated using the Bray-Curtis algorithm. Ordination by non-metric multi-dimensional scaling (nMDS) was used to compare similarity of microbial community structures between sites and sample type. Treatment effects on the bacterial community structures were tested using two-way crossed analysis of similarities (ANOSIM; Clarke 1993).

Associations between water chemistry and biological community structure were tested using the BIOENV routine (Clarke 1993; Clarke and Ainsworth 1993). Correlations between the fixed matrix (bacterial community structure; Bray-Curtis) and environmental variables (pre-normalised; Euclidean distance) were compared using Spearman rank correlation. The entire biotic dataset (i.e. bore and lysimeter data across both sites) was compared against a subset of the total water chemistry data that excluded N[H.sub.4] data for groundwater samples and DOC for soil-water samples (not available due to logistic and operational constraints). All multivariate data analysis (MDS, ANOSIM, and B1OENV) was conducted in the Primer6 software package (PrimerE Ltd, UK) using routines described by Clarke and Warwick (2001).

Bacterial community composition

DNA from sample replicates was pooled into a single representative sample for each of the four treatments, i.e. lysimeter water from under sugarcane, lysimeter water from under banana, bore water from under sugarcane, and bore water from under banana. Universal bacterial primers U1 and U2 (Ash et al. 1993) were used to randomly generate long-length rDNA sequences from each of the four DNA samples. PCR mixes contained primers used at 0.5 [micro]M, 10 mM of each dNTP, 2.5 [micro]L of 10x Qiagen HotStar PCR buffer, 1 U Qiagen HotStar Taq DNA polymerase, and 2 [micro]L of undiluted template DNA. Thermocycling was conducted on an Eppendorf Mastercycler gradient PCR unit. After initial activation of the polymerase enzyme (5 min at 94[degrees]C), PCR conditions were 94[degrees]C for 30 s, 50[degrees]C for 30 s, and 72[degrees]C for I min (25 cycles), and a final extension at 72[degrees]C for 10 min. The PCR products were cloned into the TOPO vector system (Invitrogen). Bacterial colonies were picked onto a library plate and sent to the Australian Genome Research Facility (Adelaide) for capillary sequencing from the M13 region. Twenty-four colonies were submitted from each treatment, resulting in a total of 48 sequences from 'lysimeter-collected' treatments and so forth. Following removal of flanking vector regions, the taxonomic affiliation of the 16S rRNA sequences was determined using the 'Naive Bayesian rRNA Classifier' tool (Wang et al. 2007) of the Ribosomal Database Project V2 ( Assignment of sequences to known taxa was conducted at a 90% confidence threshold. Sequence libraries were compared between water source (lysimeter and bore) and site (sugarcane and banana).

nirK and nirS gene quantification

The abundance of Cu-containing (nirK) and cytochrome [cd.sub.1]-type (nirS) nitrite reductase genes in DNA extracted from the water samples was used as a marker for denitrification (Philippot and Hallin 2005). Quantification of nir-genes was conducted using real-time PCR (qPCR) with SYBR-based detection (Henry et al. 2004, 2005). For nirK, each 25-[micro]L PCR contained primers nirK878 and nirK1040 at 0.5 [micro]M, 12.5 [micro]L of QuantiTect SYBR Green PCR mix (Qiagen), 5 [micro]L of undiluted template DNA, 0.5 [micro]L of BSA (Promega), and sterile, DNA-free water (Sigma). Touchdown PCR conditions were as described by Henry et al. (2004). For nirS, qPCR was based on primers nirScd3aF and nirS3cd (Throback et al. 2004; Kandeler et al. 2006) with other chemistry as before. PCR thermocycling conditions were described in Kandeler et al. (2006).

The qPCR was performed on a Stratagene MX3000P system, with ROX as a passive reference dye. A standard curve was created by ligating a nirK or nirS gene PCR product, amplified from Alcaligenes faecalis and Pseudomonas stutzeri, respectively, into the pGEM-T plasmid (Promega). Plasmids were harvested from E. coli cells using a plasmid extraction kit (MoBio Inc.). The quantity of DNA was determined as previously and serial dilutions were made in sterile water. Following qPCR, products were separated electrophoretically to ensure amplification of a single fragment of the expected size. The number of copies of nirK genes per ng DNA was calculated and two-way ANOVA used to test the effect of site or water source on gene abundance (GraphPad Prism version 5).

Comparisons between nir-gene abundance and water chemistry were tested as described above for bacterial community structure. Resemblance matrices (Euclidean distance) of nirK gene copies were created for non-transformed, square-root, and log-transformed data (total sample set and for bore and lysimeter subsets). Environmental variables were treated as previously and compared with the nirK copies using the BIOENV routine (Primer6; PrimerE Ltd, UK).


Water chemistry

Over the study period (404 days), rainfall was 2972 mm at the banana site, of which 41% drained below the root-zone. Rainfall was 1760 mm at the sugarcane site, but drainage could not be calculated because the water table rose above the lysimeters for much of the study period. At both sites, the watertable showed rapid recharge and discharge in response to rainfall and dry periods. Depth to watertable ranged from 0.15 to 1.96 m at the sugarcane site and 1.07 to 4.77 m at the banana site (Table 1). At both sites, [C1.sup.-] concentrations were similar in the bores and nearby drainage channels, and their concordant fluctuations indicate they were connected. The range of pH values measured over the whole sampling period was very high (Table 1). However, the range of pH values in the samples analysed microbiologically was much smaller: for the lysimeter samples 5.57-6.70 for the sugarcane site and 5.78-6.71 for the banana site, and for the bore samples 6.80-7.16 for the sugarcane site and 5.17-5.60 for the banana site. For the lysimeter samples, the variation in pH was related to differences between individual lysimeters. For the bore samples, the variation in pH at the sugarcane site was related mostly to sampling date, being higher on the later sampling date. For the banana site, the variation was low, with no discernible effect of sampling date or individual bores.

Concentrations of nitrate were high in the soil-water and lower in the groundwater (Table 1). Chemistry of soil water at the sugarcane site can only be compared with the banana site for a limited number of samples collected during the short period when the water table was deeper than the lysimeters. High ammonium concentrations in sugarcane soil water suggested that nitrification rates were low at this site. In the groundwater, nitrate concentrations differed markedly between sites (Table 1). Mean nitrate concentrations in the soil water were 556 times higher than in groundwater at the sugarcane site but only 1.3 times higher at the banana site. When compared with the [CI.sup.-] values, it appears there were lateral inputs of nitrate to the groundwater at the banana site. At the sugarcane site, nitrate concentration decreased as water percolated through the soil, or more dilute water may have moved in laterally. However, the exact nature of processes in the soil water could not be determined at the sugarcane site due to the shallowness of the water table. At the banana site, the N collected in the lysimeters was equivalent to 114kg N/ha as nitrate and 2kg N/ha as ammonium, representing 38% of total N inputs during the sampling period.

Dissolved phosphate concentrations reflected fertiliser inputs, being higher in soil water and lower in groundwater, under bananas than sugarcane (Table 1). Groundwater DOC concentrations were higher under sugarcane than bananas.

DGGE community profiling

PCR-DGGE fingerprinting of the bacterial community showed high diversity of rRNA genotypes in all samples (Fig. 1). Community structure of bore- and soil-water was statistically significantly distinct (ANOSIM R = 0.75, P = 0.001; Figs 1 and 2). Within this primary effect (water source), there was a secondary site effect (banana v. sugarcane; ANOSIM R = 0.451, P = 0.001; Figs 1 and 2). It should be kept in mind that differences in sampling technique are likely to have influenced the nature of the communities identified in the two water sources. Site effect was driven mainly by differences within groundwater rather than soil water (Fig. 2).


165 rRNA gene library comparison

Sequences of all samples were dominated by Proteobacteria (>80% of total; Figs 3 and 4). In lysimeter samples, a few sequences were assigned to TM7, Spirochaetes, Verrucomicrobia, and Chlamydiae phyla. Non-proteobacterial sequences from groundwater included TM7, Actinobacteria, and Bacteroidetes (Fig. 3). There was no difference in distribution of phyla between bores and lysimeters. Within the proteobacteria, betaproteobacteria were the dominant class in groundwater (40% of total), with a mix of beta- and alphaproteobacteria in the soil-water (Fig. 3). Within the betaproteobacteria, groundwater had significantly more sequences assigned to the order Neisseriales (17% total bore sequences) than lysimeter samples (no Neisseriales sequences detected; Fig. 3; P< 0.01). Furthermore, 27% of total sequences in groundwater to belonged to a 'betaproteobacteria, Burkholderiales, Comamonadaceae' (Fig. 3; P<0.01), whereas only 2% of sequences in lysimeter samples were thus assigned. Although 16S rRNA sequence libraries were dominated by proteobacterial sequences, the banana site had slightly greater diversity of other phyla and unassigned sequences (Fig. 4). In particular, gammaproteobacterial sequences were present in the banana but not sugarcane sequence libraries (Fig. 4).


Significant differences existed between the banana and sugarcane sites in the distribution of sequences among families of Burkholderiales (Fig. 4). The banana site had a significantly higher proportion of sequences (15% of total) assigned to the 'Incertae sedis 5' family (Aquabacterium, Leptothrix, and unclassified genera), which were absent from the sugarcane samples. The samples from the sugarcane site were chiefly within the Comamonadaceae (Delftia, Acidovorax, and unclassified genera) and Burkholderiaceae (Cupriavidus, Ralstonia, and unclassified genera; Fig. 4).

Links between bacterial community structure and water chemistry

Significant links were found between physicochemical properties of water and bacterial community structure. Water pH accounted for 37-44% of variation in bacterial community structure across all samples (P<0.01; Table 2). Within this model, bore and lysimeter bacterial communities were then independently tested for association with environmental factors and for the groundwater community; addition of NOx and DRP to pH increased the correlation to 46% (P=0.03; Table 2). For the lysimeter community, addition of N[H.sub.4.sup.+] and EC to pH increased the correlation to 50% (P = 0.02; Table 2).

Nitrite reductase gene (nirK and nirS) abundance

The abundance of nirK ranged from hundreds of copies per ng DNA to tens of thousands of copies (Fig. 5) and was affected by an interaction between site and water source (P=0.01). The dominant factor affecting nirK gene abundance was water source; soil water had 2-3 orders of magnitude more nirK than corresponding groundwater (Fig. 5). The banana site had more nirK in the groundwater than the sugarcane site (P < 0.05), but the abundance of nirK in soil water was similar between sites (Fig. 5). No significant associations were found between the abundance of nirK genes and physicochemical properties of the associated water samples.

The abundance of nirS genes varied less with site or water source than did nirK genes (Fig. 5). Total copies were around [10.sup.5]/ng water DNA, which was similar to upper levels of nirK detected. There were no significant effects of site or water source on nirS copy numbers. No significant links were made between abundance of nirS gene copies and environmental properties.

Total amounts of nir-type genes in the samples were significantly greater in water sampled from the soil compared with the groundwater (Fig. 5; P = 0.05). These differences were due to the response of nirK type genotypes in the water from the soil and groundwater (Fig. 5).


The results show that soil water and groundwater in the GBR catchment region contain rich communities of bacteria, of which a fraction harbours the capacity to reduce nitrite. Thus, the microbial communities are likely to provide important ecosystem services with respect to improving groundwater quality and mitigating the undesirable effects of high NOx in down-stream ecosystems where groundwater discharges. Characterising these communities, establishing the links between physicochemical conditions and community structure and function are prerequisites of understanding their ecological significance.

Microbial communities in groundwater have been shown to have a taxonomically distinct suite of microbiota (reviewed by Madsen and Ghiorse 1993). They tend to have no, or low, abundance of endospore-forming bacteria (Firmicutes) and Actinobacteria, whereas these taxa are widely distributed in other aquatic environments or in soils. In the present study, which involved generation of nearly 100 random DNA sequences across four groundwater samples, these taxa were effectively absent (<1% abundance). Rather, the community was strongly dominated (>80%) by proteobacteria, most of which fell into the alpha- and beta-sections, and community composition between the two sites and water sources was driven primarily by differences in diversity of the proteobacteria.

The major factor affecting bacterial community composition and also nir gene abundance was water source, i.e. whether the water originated from the unsaturated soil or from the saturated water table beneath. A key difference was presence of Comamonadaceae (Burkholderiales) and Neisseriales in the groundwater (27% and 17% of sequences, respectively), whereas these taxa were either absent (Neisseriales) from soil--water or of very low abundance (2%, Comamonadaceae). This indicates that the bacterial community within the groundwater is Fig. 4. Taxonomic distribution of sequences from the sugarcane and banana groundwater samples within the Proteobacteria, Betaproteobacteria, and Burkholderiales. Numbers correspond to the total number of sequences assigned to that taxonomy, not a percentage value per pie chart. Taxa labels in bold indicate significantly significant difference between the two sample sets. distinctive, and not simply a derived version of that in the soil. However, the difference may have been at least partially due to differences in sampling techniques.



The family Comamonadaceae includes several bacteria that have been isolated from aerobic soil and fresh water, including the genera Comamonas, Acidovorax, Delftia, Hydrogenophaga, and Variovorax (Willems and Gillis 2005). The main Comamonadaceae genera found in this study were Acidovorax (10.4% of groundwater sequences), Delftia, as well as unclassified Comamonadaceae. All that is known of the ecology of Acidovorax spp. is that some are dominant in systems that involve the degradation of organic contaminants and that they are generally aerobic, although two species are capable of heterotrophic denitrification of nitrate (Willems and Gillis 2005). Delftia spp. are also aerobic and can reduce nitrate to nitrite, but not further reduce nitrite. Although these bacteria represent an important difference in the communities between soil water and groundwater, it is unlikely that they are the source of the denitrification capacity, as measured by nirK abundance in the groundwater samples.


The family Neisseriaceae contains several genera isolated from water samples, including Aquaspirillium (freshwater under aerobic to microaerophilic conditions), lodobacter (running water, lakes, and sediments), Prolinoborus (pond water), and Vogella (freshwater) (Tonjum 2005). Although further taxonomic resolution of sequences from the Neisseriaceae was not possible, these bacteria were shown to constitute a significant, characteristic component of the groundwater microbiota.

The similarity of bacterial community diversity in groundwater at the two sites suggests the existence of a distinctive microbiota reflecting the physicochemical characteristics of the environment, including saturated conditions within a matrix of varying porosity, providing surface habitat for microbial growth and movement, as well as provision of chemically reactive sites. Heterotrophy due to DOC inputs is dominant, and may represent an important driver of denitrification (Goodale et al. 2005), but chemolithoautotrophic processes also occur, reflecting conditions driven by underlying geology, affecting the solution chemistry of the system, and providing opportunities for a range of biogeochemical transformations of ecological importance.

In this study, the key factor associated with bacterial community composition was water pH. Whether pH effects are direct or indirect is unclear. Certainly pH directly influences bacterial enzyme kinetics, protein stability, and cellular metabolism as well as being related to chemical processes such as nutrient availability, pH is a powerful driver of microbial community composition, particularly in soils (Fierer and Jackson 2006; Lauber et al. 2008; Wakelin et al. 2008a, 2009) and sediments (Zeng et al. 2009). In aquatic systems, DOC and temperature appear to be primary drivers of community composition (Lindstrom and Bergstrom 2005; Crump et al. 2007), and their interaction with pH is likely to represent an additional level of complexity in determining bacterial communities in groundwater.

The large differences in nitrate concentrations between sites in both soil-water and groundwater can be attributed to the balance between (a) fertiliser inputs, (b) loss by denitrification, and (c) in the case of groundwater, inputs from lateral flow. Inputs of fertiliser N and groundwater N were higher at the banana site than the sugarcane site. However, neither of those factors fully explains differences in nitrate concentrations between the sites, and there appears to be a higher rate of denitrification at the sugarcane site than the banana site, as evidenced by the marked reduction in nitrate concentration of groundwater compared with soil-water. It is possible that groundwater with low nitrate concentration moved laterally into the sugarcane site, but total nir-gene abundance was higher in soil-water at the sugarcane site than the banana site, suggesting higher denitrification rates at the sugarcane site. Conditions were more favourable for denitrification at the sugarcane site than at the banana site (waterlogged conditions, high nitrate concentration, and high soil organic matter content). At the banana site, nirK abundance was lower in the soil-water than at the sugarcane site (due presumably to the more aerobic environment), leading to more nitrate leaching, evidenced by the higher nitrate concentrations in groundwater than at the sugarcane site, and consequently slightly better conditions for denitrification, and higher nir abundance in groundwater. Higher dissolved P and DOC concentrations and pH of the groundwater at the sugarcane site is also consistent with less aerobic conditions in the overlying soil at the sugarcane site than at the banana site. We acknowledge that a clearer picture of the effects of site and of crop on denitrification could be gained from a balanced, replicated study design, but logistical constraints on site selection and the installation of infrastructure limited the availability of sites in which crop type could be replicated. There is good evidence to suggest that crop type in the Tully-Murray Catchment may influence patterns and processes in the nitrogen cycle because sugarcane soils were found to contain the N-fixing m]H gene, associated with free--living and endophytic N-fixing bacteria (Colloff et al. 2008).

Waterlogged soils have been associated with low nitrate and high dissolved P and DOC concentrations in streams in grazed catchments (Nelson et al. 1996). These deductions about processes occurring in the soil-water and groundwater at the study sites are indirectly supported by the modelling of McKergow et al. (2005), who found that loss of N from denitrification in GBR fiver systems is generally low (<1% of mean annual loads). However, in wetter catchments, such as the Tully-Murray, total modelled N export values are higher than would be expected. This anomaly may be explained by denitrification losses in soil and groundwater.

Several links between nir abundance and environmental conditions are worth noting. The marked reduction in nir abundance with depth (from soil water to groundwater) was consistent with previous work showing the effect of reduced carbon availability with depth (Slater and Capone 1987; Smith and Duff 1988; Starr and Gillham 1993). Although we found no link between pH and nir gene abundance, pH strongly influences denitrifier community composition (Prieme et al. 2002; Enwall et al. 2005), and in turn, N transformations directly influence pH (van Breemen et al. 1983). Several other factors may also confound links between functional gene abundance and environmental data. These include (1) lack of direct association between gene abundance, expression, and enzyme activity; (2) nutrient limitation; and (3) less-than-complete recovery of nir genes from environmental samples provided by the current gene primers (Philippot and Hallin 2005; Deiglmayr et al. 2006; Kandeler et al. 2006). In this study, the links between nir and water chemistry are further affected by the episodic nature of groundwater recharge through high rates of soil infiltration. In cropping lands of the GBR catchments, 55-68% of rainfall moves below the root-zone as deep drainage (Prove et al. 1994), often as pulse events. This pattern of recharges results in rapid movement of nutrients into the groundwater, making the system highly dynamic (Rasiah et al. 2005). Thus, although groundwater that receives significant amounts of N-containing recharge may be expected to harbour larger quantities of nitrite-reducing bacteria, actual levels of N and other factors affecting microbial denitrification are temporally highly variable (Rasiah et al. 2005) and will collectively influence the microbial potential for N cycling at a particular place at a particular time

The presence of distinct, adapted, microbial communities in groundwater raises important considerations regarding the protection of subsurface-mediated ecosystem services these systems provide. Their microbiota are sensitive to anthropogenic disturbance, as is clearly demonstrated by bioremediation research (Anderson et al. 2003; Farhadian et al. 2008). As such, there is a need to understand, manage, and conserve groundwater microbial community diversity and function in order to protect water quality (Ghiorse and Wilson 1988; Chapelle 2001). Increasing recognition of the effects of land and water management practices on the nitrogen cycle, particularly on denitrification (Davidson and Seitzinger 2006), have led to increased awareness of the scope for identification of management practices that protect and enhance these important ecosystem functions and services (Wallenstein et al. 2006a). The application of molecular approaches for determining microbial community structure and functional gene occurrence and abundance provides important insights into the linkages between microbial biodiversity, functions, and the options available to manage for the delivery of those ecosystem services delivered by biodiversity (Wallenstein et al. 2006b; Colloff et al. 2008; Wakelin et al. 2008b). In this study, the groundwater ecosystems have hydrological connectivity to the GBR a world heritage conservation area. As such, the ecosystem service of denitrification provided by the microbiota ultimately affects the quality of water flowing to the reef, particularly in the shallow near-shore zone (Johannes 1980). Our findings echo previous calls (Post et al. 2007) for a greater level of research into the unique microbial biodiversity within these groundwater systems, and the ecosystem functions and services they provide to a unique marine ecosystem of international conservation significance.


Field sampling was carried out by Tracy Whiteing and Dale Heiner. Ms Adrienne Gregg provided technical support for DGGE community fingerprinting and clone-library generation. Installation of the lysimeters and bores and chemical analyses of the water were funded by the Cooperative Research Centre Catchment to Reef. The microbial analyses were funded by the Australian Research Council (LP0669439) and James Cook University (Career Development Grant). S. A. Wakelin was supported by a CSIRO OCE Julius Career Award. S. A. Wakelin and M. J. Colloff acknowledge funding and support from the CSIRO Water for a Healthy Country National research Flagship. Greg Rinder and Soussanith Nokham assisted with preparation of the figures. Drs Catherine Dandie and Lynne Macdonald (CSIRO Land and Water) kindly reviewed and commented on drafts of this work pre-submission.


Alexander DG (2000) Hydrographic procedure water quality sampling. Department of Natural Resources and Mines, Queensland, Australia.

Anderson RT, Vrionis HA, Ortiz-Bernad I, Resch CT, Long PE, Dayvault R, Karp K, Marutzky S, Metzler DR, Peacock A, White DC, Lowe M, Lovley DR (2003) Stimulating the in situ activity of Geobacter species to remove uranium from the groundwater of a uranium-contaminated aquifer. Applied and Environmental Microbiology 69, 5884-5891. doi:10.1128/AEM.69.10.5884-5891.2003

Ash C, Priest FG, Collins MD (1993) Molecular-identification of rRNA group 3 Bacilli (Ash, Farrow, Wallbanks and Collings) using a PCR probe test proposal for the creation of a new genus Paenibacillus. Antonie Van Leeuwenhoek Journal of General and Molecular Microbiology 64, 253-260. doi:10.1007/BF00873085

Chapelle FD (2001) 'Ground-water microbiology and geochemistry.' 2nd edn (John Wiley and Sons: New York)

Clarke KR (1993) Non-parametric multivariate analysis of changes in community structure. Australian Journal of Ecology 18, 117-143. doi:10.1111/j.1442-9993.1993.tb00438.x

Clarke KR, Ainsworth M (19931 A method of linking multivariate community structure to environmental variables. Marine Ecology Progress Series 92, 205-219. doi:10.3354/meps092205

Clarke KR, Warwick RM (2001) 'Change in marine communities: an approach to statistical analysis and interpretation.' 2nd edn (Primer-E Ltd: Plymouth, UK)

Colloff MJ. Wakelin SA, Gomez D, Rogers SL (2008) Detection of nitrogen cycle genes in soils for measuring the effects of changes in land use and management. Soil Biology & Biochemistry 40, 1637 1645. doi:10.1016/ j.soilbio.2008.01.019

Crump BC, Adams HE, Hobie JE, Kling GW (2007) Biogeography and bacterioplankton in lakes and streams of an arctic tundra catchment. Ecology 88, 1365-1378. doi:10.1890/06-0387

Davidson EA, Scitzinger S (2006) The enigma of progress in denitrification research. Ecological Applications 16, 2057 2063. doi:10.1890/1051-0761(2006)01612057:TEOPID]2.0.CO;2

Deiglmayr K, Philippot L, Tscherko D, Kandeler E (2006) Microbial succession of nitrate-reducing bacteria in the rhizosphere of Poa alpina across a glacier foreland in the Central Alps. Environmental Microbiology 8, 1600-1612. doi:10.1111/j.1462-2920.2006.01051.x

Enwall K, Philippot L, Hallin S (2005) Activity and composition of the denitrifying bacterial community respond differently to long-term fertilization. Applied and Environmental Microbiology 71, 8335 8343. doi:10.1128/AEM.71.12.8335-8343.2005

Fares A, Deb SK, Fares S (2009) Review of vadosc zone solution sampling techniques. Environmental Reviews 17, 215 234. doi:10.1139/A09-010

Farhadian M, Vachelard C, Duchez D, Larroche C (2008) In situ bioremediation of monoaromatic pollutants in groundwater: a review. Bioresource Technology 99, 5296-5308. doi:10.1016/j.biortech. 2007.10.025

Fierer N, Jackson RB (2006) The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Sciences of the United States of America 103, 626-631. doi:10.1073/pnas.0507535103

Ghiorse WC, Wilson JT (1988) Microbial ecology of the terrestrial subsurface. Advances in Applied Microbiology 33, 107-172. doi:10.1016/S0065-2164(08)70206-5

Goodale CL, Aber JD, Vitousek PM, McDowell WH (2005) Long-term decreases in stream nitrate: successional causes unlikely; possible links to DOC? Ecosystems 8, 334-337. doi: 10.1007/s10021-003-0162-8

Gordon C, Toze S (2003) Influence of groundwater characteristics on the survival of enteric viruses. Journal of Applied Microbiology 95, 536-544. doi:10.1046/j.1365-2672.2003.02010.x

Greiner R, Herr A, Brodie J, Haynes D (2005) A multi-criteria approach to Great Barrier Reef catchment (Queensland, Australia) diffuse-source pollution problem. Marine Pollution Bulletin 51, 128-137. doi:10.1016/ j.marpolbul.2004.11.033

Henry S, Baudoin E, Lopez-Gutierrez JC, Martin-Laurent F, Baumann A, Philippot L (2004) Quantification of denitrifying bacteria in soils by nirK gene targeted real-time PCR. Journal of Microbiological Methods 59, 327-335. doi:10.1016/j.mimet.2004.07.002

Henry S, Baudoin E, Lopez-Gutierrez JC, Martin-Laurent F, Baumann A, Philippot L (2005) Quantification of denitrifying bacteria in soils by nirK gene targeted real-time PCR (Vol 59, Pg 327, 2004). Journal of Microbiological Methods 61, 289-290. doi:10.1016/j.mimet.2004. 12.008

Isbell RF (2002) 'The Australian Soil Classification.' (CSIRO Publishing: Melbourne)

Johannes RE (1980) The ecological significance of the submarine discharge of groundwater. Marine Ecology Progress Series 3, 365-373. doi:10.3354/meps003365

Kandeler E, Deiglmayr K, Tscherko D, Bru D, Philippot L (2006) Abundance of narG, nirS, nirK, and nosZ genes of denitrifying bacteria during primary successions of a glacier foreland. Applied and Environmental Microbiology 72, 5957 5962. doi:10.1128/ AEM.00439-06

Lane DJ (1991) 16S/23S rRNA sequencing. In 'Nucleic acid techniques in bacterial systematics'. (Eds E Stackebrandt, M Goodfellow) pp. 115-175. (John Wiley and Sons: Chichester, UK)

Lauber CL, Strickland MS, Bradford MA, Fierer N (2008) The influence of soil properties on the structure of bacterial and fungal communities across land-use types. Soil Biology & Biochemistry 40, 2407-2415. doi:10.1016/j.soilbio.2008.05.021

Lindstrom ES, Bergstrom AK (2005) Community composition of bacterioplankton and cell transport in lakes in two different drainage areas. Aquatic Sciences 67, 210-219. doi: 10.1007/s00027-005-0769-2

Madsen EL, Ghiorse WC (1993) Groundwater microbiology: subsurface ecosystem processes. In 'Aquatic microbiology'. (Ed. TE Ford) pp. 167-214. (Blackwell Scientific Publications: Cambridge, MA)

McKergow LA, Prosser IP, Hughes AO, Brodie J (2005) Regional scale nutrient modelling: exports to the Great Barrier Reef World Heritage Area. Marine Pollution Bulletin 51, 186 199. doi:10.1016/j.marpolbul. 2004.11.030

Muyzer G, de Waal EC, Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental Microbiology 59, 695-700.

Nelson PN, Cotsaris E, Oades JM (1996) Nitrogen, phosphorus and organic carbon in streams draining two grazed catchments. Journal of Environmental Quality 25, 1221-1229. doi:10.2134/jeq1996.004724 25002500060008x

Philippot L, Hallin S (2005) Finding the missing link between diversity and activity using denitrifying bacteria as a model functional community. Current Opinion in Microbiology 8, 234-239. doi:10.1016/j.mib.2005. 04.003

Post DA, Waterhouse J, Grundy M, Cook F (2007) 'The past, present and future of sediment and nutrient modelling in GBR catchments.' (CSIRO Water for a Healthy Country National Research Flagship: Canberra, ACT)

Prieme A, Braker B, Tiedje JM (2002) Diversity of nitrite reductase (nirK and nirS) gene fragments in forested upland and wetland soils. Applied and Environmental Microbiology 68, 1893 1900. doi:10.1128/ AEM.68.4.1893-1900.2002

Prove BG, McShane TJ, Reghenzani JR, Armour JD, Sen S, Moody PW (1994) Nutrient loss via drainage from the major agricultural industries on the wet tropical coast of Queensland. In 'Water Down Under 94: Groundwater Papers'. pp. 439-443. (Institution of Engineers: Canberra, ACT)

QLUMP (1999) Queensland Land Use Mapping Program. Available at:

Rasiah V, Armour JD, Yamamoto T, Mahendrarajah S, Heiner DH (2003) Nitrate dynamics in shallow groundwater and the potential for transport to off-site water bodies. Water, Air, and Soil Pollution 147, 183-202. doi:10.1023/A:1024529017142

Rasiah V, Armour JD, Cogle AL (2005) Assessment of variables controlling nitrate dynamics in groundwater: Is it a threat to subsurface aquatic ecosystems? Marine Pollution Bulletin 51, 60-69. doi:10.1016/ j.marpolbul.2004.10.024

Slater JM, Capone DG (1987) Denitrification in aquifer soil and nearshore marine sediments influenced by groundwater nitrate. Applied and Environmental Microbiology 53, 1292-1297.

Smith RL, Duff JH (1988) Denitrification in a sand and gravel aquifer. Applied and Environmental Microbiology 54, 1071-1078.

Starr RA, Gillham RW (1993) Denitrification and organic carbon availability in two aquifers. Ground Water 31, 934 947. doi:10.1111/j.1745-6584.1993.tb00867.x

Throback IN, Enwall K, Jarvis A, Hallin S (2004) Reassessing PCR primers targeting nirS, nirK and nosZ genes for community surveys of denitrifying bacteria with DGGE. FEMS Microbiology Ecology 49, 401-417. doi:10.1016/j.femsec.2004.04.011

Tonjum T (2005) Family I. Neisseriaceae Prevot 1933, [119.sup.AL] emend. Denhirst, Paster and Bright 1980, 265. In 'Bergey's Manual of Systematic Bacteriology'. 2nd edn. Vol. 2, Part C. (Eds Brenner D J, NR Krieg, JT Staley, GM Garrity) pp. 775-863. (Springer: New York)

van Breemen N, Mulder J, Driscoll CT (1983) Acidification and alkalinization of soils. Plant and Soil 75, 283-308. doi:10.1007/ BF02369968

Wakelin SA, Macdonald LM, Rogers SL, Gregg AL, Bolger TP, Baldock JA (2008a) Habitat selective factors influencing the structural composition and functional capacity of microbial communities in agricultural soils. Soil Biology & Biochemistry 40, 803-813. doi: 10.1016/j.soilbio.2007. 10.015

Wakelin SA, Colloff MJ, Kookana RS (2008b) Effect of wastewater treatment plant effluent on microbial function and community structure in the sediment of a freshwater stream with variable seasonal flow. Applied and Environmental Microbiology 74, 2659-2668. doi:10.1128/AEM.02348-07

Wakelin SA, Gregg AL, Simpson RJ, Li GD, Riley IT, McKay A (2009) Pasture management clearly affects soil microbial community structure and N-cycling bacteria. Pedobiologia 52, 237-251. doi:10.1016/ j.pedobi.2008.10.001

Wallenstein MD, Peterjohn WT, Schlesinger WH (2006a) N fertilisation effects on denitrification and N cycling in an aggrading forest. Ecological Applications 16, 2168-2176. doi:10.1890/1051-0761(2006)01612168: NFEODA]2.0.CO;2

Wallenstein MD, Myrold DD, Firestone M, Voytek M (2006b) Environmental controls on denitrifying communities and denitrification rates: insights from molecular methods. Ecological Applications 16, 2143-2152. doi:10.1890/1051-0761 (2006)016[2143: ECODCA]2.0.CO;2

Wang Q, Garrity GM, Tiedje JM, Cole JR (2007) Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Applied and Environmental Microbiology 73, 5261-5267. doi:10.1128/AEM.00062-07

Willems A, Gillis M (2005) Family IV. Comamonadaceae Willems, De Ley, Gillis and Kersters 1991a, 447VP. In 'Bergey's Manual of Systematic Bacteriology, 2nd edn, Vol. 2, Part C.' (Eds Brenner DJ, NR Krieg, JT Staley, GM Garrity) pp. 686-763. (Springer: New York)

Zeng J, Yang L-Y, Li J-Y, Liang Y, Xiao L, Jiang L-J, Zhao D-Y (2009) Vertical distribution of bacterial community structure in the sediments of two eutrophic lakes revealed by denaturing gradient gel electrophoresis (DGGE) and multivariate analysis techniques. World Journal of Microbiology & Biotechnology 25, 225-233. doi:10.1007/s11274-008-9883-3

Manuscript received 5 March 2010, accepted 2 August 2010

Steven A. Wakelin (A,B,F), Paul N. Nelson (C,D), John D. Armour (D), Velupillai Rasiah (D), and Matthew J. Colloff (E)

(A) CSIRO Land and Water, Environmental Biogeochemistry Theme, PMB 2, Glen Osmond, SA 5064, Australia.

(B) AgResearch Ltd, Lincoln Research Centre, Private Bag 4749, Christchurch 8140, New Zealand.

(C) James Cook University, School of Earth and Environmental Sciences, PO Box 6811, Cairns, Qld 4870, Australia.

(D) Department of Environment and Resource Management, PO Box 156, Mareeba, Qld 4880, Australia.

(E) CSIRO Entomology, GPO Box 1700, Canberra, ACT 2601, Australia.

(F) Corresponding author. Email:

Table 1. Summary of water quality and watertable depth at the two
sites over the whole study period

DRP, Dissolved reactive phosphorus; DOC, dissolved organic carbon;
n.d., not determined

                                         Sugarcane site
Parameter          Water
                   source       Mean [+ or -] s.d.       Range (n)

NOx-N ([micro]g    Lysimeters   8899 [+ or -] 14913     5-54900 (39)
  L)               Bores          16 [+ or -] 16          1-128 (99)

NH4-N ([micro]g    Lysimeters   1373 [+ or -] 2778     28-12100 (39)
  /L)              Bores               n.d.                n.d.

DRP ([micro]g      Lysimeters    3.0 [+ or -] 2.0       1.0-9.0 (39)
  /L)              Bores        67.7 [+ or -] 47.3    2.0-138.3 (96)

DOC (mg/L)         Lysimeters          n.d.                n.d.
                   Bores        1.97 [+ or -] 0.72    0.78 6.80 (99)

Chloride (mg/L)    Lysimeters   42.2 [+ or -] 37.5    4.0-160.2 (39)
                   Bores         7.2 [+ or -] 4.1     2.0-20.0 (180)

pH                 Lysimeters   5.79 [+ or -] 0.52    4.62 6.83 (39)
                   Bores        6.96 [+ or -] 0.25    6.40 8.13 (180)

EC (dS/m)          Lysimeters   0.23 [+ or -] 0.17    0.09 0.71 (39)
                   Bores        0.15 [+ or -] 0.10    0.06 0.94 (180)

Depth to           Bores        0.82 [+ or -] 0.05    0.15 1.96 (180)
  watertable (m)

                               Banana site
                   Mean [+ or -] s.d.      Range (n)

NOx-N ([micro]g    5271 [+ or -] 4686   309-26 300 (165)
  L)               4134 [+ or -] 2600      50-9040 (131)

NH4-N ([micro]g      41 [+ or -]  167       1-1690 (136)
  /L)                     n.d.                n.d.

DRP ([micro]g        30 [+ or -] 59          1-390 (165)
  /L)              14.0 [+ or -] 16.4     0.7-99.3 (129)

DOC (mg/L)                n.d.                n.d.
                   0.77 [+ or -] 0.51    0.20-2.75 (131)

Chloride (mg/L)    28.1 [+ or -] 22.3    2.0-130.1 (175)
                   11.0 [+ or -] 16.4     1.8-36.0 (239)

pH                 6.32 [+ or -] 0.48    5.48-8.86 (175)
                   5.60 [+ or -] 0.24    5.00-6.32 (239)

EC (dS/m)          0.28 [+ or -] 0.14    0.02-0.72 (175)
                   0.09 [+ or -] 0.07    0.03-0.75 (239)

Depth to           3.04 [+ or -] 0.09    1.07-1.77 (239)
  watertable (m)

Table 2. Associations between bacterial community structures and
water chemistry

                                         Spearman rank
Sample set           Environmental       correlation
                     factor              (Rho)             P

All samples          pH                  0.367           0.01
Bore only            pH                  0.392           0.001
                     pH, NOx, FRP        0.459           0.03
Lysimeter only (A)   pH                  0.444           0.001
                     N[H.sub.4.sup.+],   0.500           0.02
                       pH, EC

Including both the banana and sugarcane sites.
COPYRIGHT 2011 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Wakelin, Steven A.; Nelson, Paul N.; Armour, John D.; Rasiah, Velupillai; Colloff, Matthew J.
Publication:Soil Research
Article Type:Report
Geographic Code:8AUST
Date:Feb 1, 2011
Previous Article:Evaluation of the effects of cation combinations on soil hydraulic conductivity.
Next Article:Temporal variability of soil water storage evaluated for a coffee field.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters