A horizontal permeable reactive barrier stimulates nitrate removal 1

and shifts microbial ecology during rapid infiltration for managed 2

recharge 3

Sarah Beganskasa*, Galen Gorskia, Tess Weathersa, Andrew T. Fishera, Calla Schmidtc, Chad 4

Saltikovb, Kaitlyn Redfordb, Brendon Stoneburnerb, Ryan Harmona1, Walker Weira2 5

6 aEarth and Planetary Sciences, University of California Santa Cruz, Santa Cruz, California, 95064 7 bMicrobiology and Environmental Toxicology, University of California Santa Cruz, Santa Cruz, 8

California, 95064 9 cEnvironmental Science, University of San Francisco, San Francisco, California, 94117 10 1Present address: Hydrologic Science and Engineering, Colorado School of Mines, Golden, 11

Colorado, 80401 12 2Present address: Hydrologic Sciences, University of Nevada Reno, Reno, NV, 89557 13

*Corresponding author, sbeganskas@ucsc.edu 14

15

SUPPORTING INFORMATION 16

25 pages; 5 figures; 9 tables 17

18

DETAILED MATERIALS AND METHODS 19

Infiltration testing. 20

The infiltration capacity of soils has been measured using a variety of tools and methods in the 21

last century (e.g., Alagna et al., 2016; Burgy and Luthin, 1956; Youngs, 1991). The de-facto 22

standard for such measurements remains the single- or double-ring infiltrometer (Ahmed et al., 23

2014; ASTM Standard D3385, 2009; Fatehnia et al., 2016; Ruggenthaler et al., 2016), despite the 24

well-known issue of lateral flow in soils below the instruments (Bouwer et al., 1999; Marshall 25

and Stirk, 1950). For studies that seek to resolve fundamental soils properties such as saturated 26

hydraulic conductivity or saturation parameters, this requires application of analytical corrections 27

to account for lateral flow, and/or specialized instrumentation that establishes boundary 28

conditions needed to focus on parameters of interest (Alagna et al., 2016; Bagarello et al., 2016; 29

Bouwer et al., 1999). 30

31

The issue of lateral flow is related to infiltration measurement scale: for a larger area of testing, 32

there is a smaller ratio of test site circumference to test area. In addition, all else being equal, 33

larger test areas are more likely to encounter the fastest infiltration pathways in heterogeneous 34

soils; this helps explain why tests of larger areas tend to yield higher infiltration rates compared 35

to tests of the same sites using smaller areas, and why tests of smaller areas tend to indicate high 36

heterogeneity (Burgy and Luthin, 1956; Fatehnia et al., 2016; Heeren et al., 2015, 2014; 37

Khodaverdiloo et al., 2017; Lucke et al., 2014). Larger test areas also present practical 38

challenges, generally requiring greater construction effort, higher cost for supplies, and a larger 39

water supply. The latter is a particular limitation for tests run for more than a few hours. 40

41

In this study, we were mainly interested in determining the total and vertical infiltration rate over 42

a 10–15-day experimental period, rather than transient response during early time. And we were 43

limited as to the rate at which water could be supplied, conveyed, and applied to the test plots. 44

Given expected rapid infiltration rates in coarse and vegetated soils, we elected to construct 1 m2 45

plots to measure total infiltration by mass balance and vertical infiltration using heat as a tracer 46

(Hatch et al., 2006; Racz et al., 2011). Past experience with the test plot configuration suggested 47

that we could expect a ratio of total to vertical infiltration of 8:1 to 12:1 (as was subsequently 48

observed). This plot size is larger than that commonly applied to field infiltration testing using 49

rings or disk infiltrometers and is consistent with plot sizes used in other field studies (ASTM 50

Standard D3385, 2009; Heeren et al., 2015, 2014; Khodaverdiloo et al., 2017; Reynolds et al., 51

2000; Youngs, 1991). In addition, temperature probes placed near the center of these plots to 52

assess vertical infiltration should be far enough from the plot margins to avoid being influenced 53

by them: the characteristic length scale of measurement for these probes can be estimated as L = 54

(! t)1/2, where ! = thermal diffusivity (~10-7 m2/s), and t = time (86,400 s for diel temperature 55

variations). For this experimental configuration, L ~10 cm. 56

57

Plot construction, operation, and sampling. 58

We hand-excavated four 1 × 1 m plots for the tests. The native soil plots were ~60 cm deep, 59

whereas the PRB plots were ~90 cm deep to allow for installation of a 30-cm-thick layer of 60

redwood chips at the bottom. The redwood chips were sourced from a local landscape supply 61

store. The plots were lined with fiberglass siding to prevent water from flowing sideways above 62

the base of the plots. We assembled the four plot walls as a mechanical unit and lowered them 63

into position using rope handles. We filled in gaps around the outsides of the walls with 64

bentonite chips at the base, activated with water to form a seal, and topped with native soil. 65

66

We installed two thermal probes in each plot, one attached to a real-time data telemetry system, 67

and one autonomous (recording internally). For each probe, a pair of thermal sensors (resolution 68

± 0.015°C) were installed in polyvinyl chloride tubes, placed at 5 and 20 cm below the base of 69

the plot. The probes were placed in hand-augered holes (7 cm diameter) and surrounded by silica 70

slurry to ensure good thermal contact with the soil and prevent fluid from flowing vertically 71

along the annulus around the probes. Data with a 15-minute sampling rate from these probes was 72

processed to calculate vertical infiltration rates. Filtered thermal records from these probes were 73

compared to assess the magnitude of amplitude reduction with depth over time, using a band-74

pass filter to isolate the frequency of variation with the greatest power.(Hatch et al., 2006). 75

Greater downward infiltration is associated with a smaller amplitude reduction in a monotonic 76

but non-linear relationship. This method is based on a transient equation for conductive-77

advective-dispersive heat transport during steady-state fluid flow, allowing assessment across a 78

wide range of infiltration rates (cm/d to m/d). We used physical and thermal sediment properties 79

from a previous analysis at this field site (Racz et al., 2011). Autonomous probes were recovered 80

after each experiment; real-time probes telemetered data using a dedicated logger and cellular 81

modem, with hourly scheduled uploads. 82

83

We installed two piezometer nests in each plot to allow sampling at depths of 30, 55, and 80 cm 84

below the base of the plot. Each fluid sampling piezometer was constructed from 1-cm diameter 85

rigid polycarbonate tubing (6 mm inner diameter, ID), perforated at the base and wrapped with a 86

fine mesh nylon screen. A piezometer was also installed within the redwood chip layer for the 87

PRB plots. Nylon tubing with a 6 mm ID extended from the piezometers to the side of the plot 88

for fluid sampling. Each piezometer was assembled and acid-washed prior to deployment. The 89

piezometers were installed at targeted depth into a 7-cm outer diameter (OD) hand-augered hole 90

in the base of the plot. The walls of these holes were gently scoured to break up a “skin” that 91

may have developed during augering. A coarse, well-rounded quartz sand/gravel filter (10–20 92

mesh) was installed around each screen, followed by a 15 cm-thick bentonite seal topped with 93

native soil. Piezometers were developed after installation (and after allowing time for the 94

bentonite seals to set) to ensure a good connection with the formation. We used a peristaltic 95

pump to flush de-ionized (MilliQ) water back and forth (gently) through the piezometer screens 96

and the surrounding sand/gravel pack. 97

98

The same peristaltic pump was used with a manifold to collect fluid samples daily from the 99

piezometers during plot operation. We pumped sampled fluid through a 0.45 µm cellulose 100

acetate filter into acid-washed polyethylene sample bottles. The bottles were pre-rinsed with 101

sample water prior to filling. The pump, manifold, and filter system were flushed with de-ionized 102

(>12 MOhm) water between samples. We collected individual replicate fluid samples to be 103

analyzed for nitrogen species, dissolved organic carbon (DOC), and nitrogen and oxygen 104

isotopes. The samples were put on ice in the field immediately and frozen upon return until 105

analysis. Samples could only be collected if the soil surrounding the piezometer was saturated. 106

To verify the accuracy of our internal results, twice during each test we collected additional fluid 107

samples to be run at Monterey Bay Analytical Services (MBAS, ELAP #2385). These samples 108

were collected, placed on ice in the field, delivered directly to MBAS, and analyzed within 48 109

hours. 110

111

Before and after the two-week infiltration test, we collected 1-m soil cores using a hand auger. 112

We photographed and sampled the cores at 10–20 cm intervals. We collected separate samples 113

for grain size analysis, total C and N, and microbiological analysis. Soil samples were put on ice 114

immediately and frozen until analysis. 115

116

For the PRB plots, after the subsurface instruments were installed, we added a 30-cm thick layer 117

of redwood chips. An additional sample tube terminated like a mini piezometer was installed in 118

each PRB. We put a nylon mesh screen (18x18 thread per cm2) on top of the woodchips and 119

weighted it down with washed river rocks from a local garden supply. For consistency, we 120

placed identical nylon mesh and river rocks above the native soil plots. 121

122

A hose and pump delivered water to the plots from a shallow groundwater well, recovering water 123

from the Harkins Slough MAR system. For most of the tests, the water drawn from the well was 124

entirely shallow groundwater, but it occasionally included a mixture of local groundwater, 125

groundwater from farther back in the basin, and recycled wastewater ("project water" delivered 126

to agency customers). The hose was anchored at the plot corner, and an electronic flow meter 127

recorded flow with a pulse logger. A float valve connected to a solenoid valve at the end of the 128

hose controlled water delivery, stopping inflow when the water level reached a desired height, 129

and restarting flow when the water level dropped below a second height. Each fill cycle lasted 130

10–20 minutes. This system was powered using batteries, trickle charged with a solar panel. We 131

operated each test for 13–15 days. 132

133

We installed an autonomous pressure gauge (2-minute sampling rate) in a PVC stilling well at 134

the base of each plot. We converted pressure to water depth, correcting for barometric pressure, 135

measured simultaneously on site with an atmospheric gauge. We converted pressure gauge 136

readings to water level using a staff plate positioned on the side of the plot and monitored 137

throughout the tests. For each fill-empty cycle, we divided the change in water level by the time 138

elapsed to calculate the bulk infiltration rate (inset diagram in Fig. 1C). 139

140

Phylogenetic sequencing. 141

Partial 16S rRNA gene (V3 and V4 variable regions) was amplified using primers(Klindworth et 142

al., 2013) modified to contain 5’ sequencing adapters (underlined) for barcoding and sequencing 143

using the Illumina MiSeq platform: 144

145

0341_16SV3V4-MS-F 146

(5’-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG-3’) 147

148

0785_16SV3V4-MS-R 149

(GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3’) 150

151

PCR methods were based on Klindworth et al. and were further optimized for Illumina MiSeq 152

sequencing platform.(Illumina Inc., n.d.; Klindworth et al., 2013) PCRs consisted of: 5 ng DNA 153

extract, 0.4 µM each forward and reverse primer, 0.5 µL of Titanium Taq DNA polymerase 154

(Clonetech), 3.5µL 25 mM dNTP mix (New England BioLabs), 5 µL 10X MasterAmp PCR 155

Enhancer (Illumina) adjusted to 50 µL final volume with nuclease free water. The following 156

thermocycler profile was used: 95˚C 5 min initial denaturation, 25 cycles of 95˚C 30 s, 55˚C 30 157

s, 72˚C 30 s, final extension at 72˚C for 5 min followed by a 4˚C hold step. Samples were 158

analyzed by agarose gel electrophoresis to confirm the presence of ~550 bp amplicons. A PCR 159

amplicon sequencing pipeline was adapted from the Illumina MiSeq platform sequencing 160

protocol for 16S metagenomic sequencing libraries(Illumina Inc., n.d.). The overall pipeline 161

included steps for the primary PCR using 16S rRNA gene primers, PCR clean-up, library 162

preparation (adding unique sequencing indices (i.e. “barcodes”) to each PCR amplicon), 163

normalizing DNA concentrations of each library, library pooling, and sequencing. 164

165

After the primary PCR described above, the amplicons were cleaned up using AMPure XP 166

magnetic beads (Agencourt) (0.8X final concentration). The Illumina sequencing indices were 167

added to the cleaned-up PCR amplicons using a PCR step similar to the one described above 168

except 5 µL of each Nextera XT index primer 1 and index primer 2 were used in place of the 169

forward and reverse 16S rRNA gene primers. The index primers anneal to the 5’ adapter 170

sequences. Each PCR amplicon received a unique combination of index primer 1 and 2. The 171

thermocycle protocol for Index PCR consisted of: 95˚C for 3 min, 8 cycles of 95˚C for 30 s, 172

55˚C for 30 s, and 72˚C for 30 s, and a final extension at 72˚C for 5 min with a 4˚C hold. 173

174

Indexed PCR amplicons were purified using AMPure XT magnetic beads (1.12X final 175

concentration) and eluted in a final volume of 25 µL. Library preparation was validated by 176

analyzing indexed PCR amplicons using a DNA 100 High Sensitivity Bioanalyzer chip. This 177

step verified the correct size and the removal of primers. The final steps before sequencing 178

included: quantify the DNA concentration for each library using real time PCR and the Illumina 179

Library Quant Kit (cat #: KK4824, KAPABiosystems); normalizing each library to 4 nM using 180

10 mM Tris pH 8.5, and pooling each normalized library (5 µL each) into one tube. A PhiX 181

control-spike (10%) was added to the pooled library as a technical control to improve base-182

calling during the sequencing run, as recommended by Illumina if the DNA libraries have a low 183

sequence diversity.(Illumina Inc., n.d.) The pooled library was sequenced on the Illumina MiSeq 184

(600 cycles v3 PE300 flow cell kit) at the University of California, Davis Genome Center. 185

186

References. 187

Ahmed, F., Nestingen, R., Nieber, J.L., Gulliver, J.S., Hozalski, R.M., 2014. A Modified Philip–188 Dunne Infiltrometer for Measuring the Field-Saturated Hydraulic Conductivity of Surface 189 Soil. Vadose Zone J. 13. https://doi.org/10.2136/vzj2014.01.0012 190

Alagna, V., Bagarello, V., Prima, S.D., Iovino, M., 2016. Determining hydraulic properties of a 191 loam soil by alternative infiltrometer techniques. Hydrol. Process. 30, 263–275. 192 https://doi.org/10.1002/hyp.10607 193

ASTM Standard D3385, 2009. Standard test method for infiltration rate of soils in field using 194 double-ring infiltrometer. ASTM International, West Conshohocken, PA, p. DOI: 195 10.1520/D3385-09. 196

Bagarello, V., Iovino, M., Lai, J., 2016. Testing steady-state analysis of single-ring and square 197 pressure infiltrometer data. Geoderma 261, 101–109. 198 https://doi.org/10.1016/j.geoderma.2015.07.002 199

Bouwer, H., Back, J., Oliver, J.M., 1999. Predicting infiltration and ground-water mounds for 200 artifical recharge. J. Hydrol. Eng. 4, 350–356. 201

Burgy, R.H., Luthin, J.N., 1956. A test of the single- and double-ring types of infiltrometers. 202 Trans Am Geophys Union 37, 189–192. 203

Fatehnia, M., Tawfiq, K., Ye, M., 2016. Estimation of saturated hydraulic conductivity from 204 double! ring infiltrometer measurements. Eur. J. Soil Sci. 67, 135–147. 205 https://doi.org/10.1111/ejss.12322 206

Hatch, C.E., Fisher, A.T., Revenaugh, J.S., Constantz, J., Ruehl, C., 2006. Quantifying surface 207 water-groundwater interactions using time series analysis of streambed thermal records: 208 Method development. Water Resour. Res. 42. https://doi.org/10.1029/2005WR004787 209

Heeren, D.M., Fox, G.A., Storm, D.E., 2015. Heterogeneity of infiltration rates in alluvial 210 floodplains as measured with a berm infiltration technique. Trans Am Soc Ag Biol Eng 211 58, 733–745. 212

Heeren, D.M., Fox, G.A., Storm, D.E., 2014. Berm Method for Quantification of Infiltration at 213 the Plot Scale in High Conductivity Soils. J. Hydrol. Eng. 19, 457–461. 214 https://doi.org/10.1061/(ASCE)HE.1943-5584.0000802 215

Illumina Inc., n.d. 16S Metagenomic Sequencing Library Preparation: Preparing 16S Ribosomal 216 RNA Gene Amplicons for the Illumina MiSeq System (No. part # 15044223 Rev. B). 217

Khodaverdiloo, H., Khani Cheraghabdal, H., Bagarello, V., Iovino, M., Asgarzadeh, H., 218 Ghorbani Dashtaki, S., 2017. Ring diameter effects on determination of field-saturated 219 hydraulic conductivity of different loam soils. Geoderma 303, 60–69. 220 https://doi.org/10.1016/j.geoderma.2017.04.031 221

Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., Glöckner, F.O., 2013. 222 Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-223 generation sequencing-based diversity studies. Nucleic Acids Res. 41. 224 https://doi.org/10.1093/nar/gks808 225

Lucke, T., Boogaard, F., van de Ven, F., 2014. Evaluation of a new experimental test procedure 226 to more accurately determine the surface infiltration rate of permeable pavement systems. 227 Urban Plan. Transp. Res. 2, 22–35. https://doi.org/10.1080/21650020.2014.893200 228

Marshall, J.T., Stirk, G.B., 1950. The effect of lateral movement of water in soil on infiltration 229 measurement. Aust J Agr Res 1, 253–257. 230

Racz, A.J., Fisher, A.T., Schmidt, C.M., Lockwood, B.S., Huertos, M.L., 2011. Spatial and 231 temporal infiltration dynamics during managed aquifer recharge. Groundwater 50, 562–232 570. https://doi.org/10.1111/j.1745-6584.2011.00875.x 233

Reynolds, W.D., Bowman, B.T., Brunke, R.R., Drury, C.F., Tan, C.S., 2000. Comparison of 234 Tension Infiltrometer, Pressure Infiltrometer, and Soil Core Estimates of Saturated 235 Hydraulic Conductivity. Soil Sci. Soc. Am. J. 64, 478–484. 236 https://doi.org/10.2136/sssaj2000.642478x 237

Ruggenthaler, R., Meißl, G., Geitner, C., Leitinger, G., Endstrasser, N., Schöberl, F., 2016. 238 Investigating the impact of initial soil moisture conditions on total infiltration by using an 239 adapted double-ring infiltrometer. Hydrol. Sci. J. 61, 1263–1279. 240 https://doi.org/10.1080/02626667.2015.1031758 241

Youngs, E.G., 1991. Infiltration measurements—a review. Hydrol. Process. 5, 309–319. 242 https://doi.org/10.1002/hyp.3360050311 243

244

Figure S1. Location of Harkins Slough MAR field site. The Pajaro Valley Groundwater Basin

is located along California’s Central Coast. Experimental plots for the present study were

installed in a part of the infiltration basin that was not instrumented in previous studies (Racz et

al., 2011).

Figure S2. The water level record for test NS2 indicates several wetting and drying cycles

over the 14-day test due to intermittent water supply. As saturated conditions were not

maintained in the shallow subsurface, this test provided limited useful hydrologic or geochemical

data.

Figure S3. Soil conditions were similar for all experimental plots. Average TN, TOC, d10

(grain size for which 10% of grains are finer) d50 (grain size for which 50% of grains are finer),

and d90 (grain size for which 90% of grains are finer) with depth for all plots. Error bars show

range among samples from the four experimental plots. For TN and TOC data, solid symbols

represent data from before the experiments and open symbols represent data from the end of the

experiment.

Figure S4. Vertical infiltration rates calculated from different thermal probes indicate that

infiltration was, at times, spatially variable. Vertical infiltration rates for tests NS1, PRB1, and

PRB2 were calculated from two thermal probes (one telemetering data in real time, the other

autonomous) at two different locations in the test plot. While the rates indicated by these two

probes often matched closely, they occasionally diverged, indicating spatial variability in

infiltration rates.

Figure S5. PCoA results for 16S rRNA gene sequencing from shallow soil samples

demonstrate that infiltration through woodchips significantly shifts soil microbial

communities. Samples from before and after infiltration in tests with native soil, as well as

samples from before infiltration through woodchips, are statistically very similar in terms of

microbial population distribution. Samples from after infiltration through woodchips show a

large deviation from initial conditions.

Table S1. Total organic carbon (TOC), total nitrogen (TN), and grain size data for all soil

samples collected.

Test #

Depth below

base of plot (cm)

TOC before

TOC after

TN before

TN after

d10 (µm)

d50 (µm)

d90 (µm)

1 0 0.04 0.04 0.003 0.005 149 273 505 1 20 0.03 0.04 0.003 0.003 151 269 501 1 40 0.04 0.03 0.003 0.004 152 287 534 1 60 0.03 0.03 0.002 0.006 162 276 523 1 80 0.03 0.03 0.003 0.004 157 279 493 1 100 0.03 0.03 0.004 0.004 159 311 554 1 120 0.03 0.03 0.003 0.004 152 283 500 1 140 0.05 0.03 0.003 0.006 154 299 556 2 0 0.03 — 0.003 — 154 281 453 2 20 0.03 0.05 0.003 0.007 140 274 446 2 40 0.03 0.04 0.003 0.006 130 249 462 2 60 0.02 0.03 0.003 0.007 158 269 465 2 80 0.07 0.03 0.003 0.006 149 253 421 2 100 0.03 0.03 0.004 0.002 153 254 413 3 0 0.05 0.03 0.007 0.004 115 274 449 3 10 0.04 0.03 0.006 0.004 127 270 438 3 20 0.04 0.03 0.006 0.004 148 269 454 3 30 0.04 0.03 0.006 0.004 139 260 439 3 50 0.04 0.05 0.006 0.006 149 279 480 3 70 0.04 0.04 0.004 0.002 161 282 458 3 90 0.06 0.02 0.005 0.002 163 267 459 3 110 0.05 0.03 0.004 0.004 162 322 538 3 130 0.05 0.03 0.005 0.005 155 293 488 4 0 0.03 — 0.003 — 149 247 407 4 10 0.06 — 0.004 — 144 272 462 4 20 0.04 — 0.005 — 151 265 445 4 40 0.03 — 0.003 — 137 252 423 4 60 0.03 — 0.003 — 146 269 471 4 80 0.04 — 0.004 — 134 264 428 4 100 0.04 — 0.004 — 102 258 427

Table S2. Water chemistry data for test NS1.

Date Day Piez. Depth (cm)

[NO3] (mg/L)

[NO2] (mg/L)

[NH4] (mg/L)

[DOC] (mg/L)

[Cl] (mg/L)

[SO4] (mg/L)

150817 4 0 22.0 0.19 0.26 27.7 192 198 150817 4 A 30 23.9 0.49 4.24 29.2 135 144 150817 4 B 30 23.5 0.38 2.45 28.1 179 222 150818 5 0 23.2 0.00 0.06 25.5 129 125 150818 5 A 30 24.0 0.00 0.28 27.0 150 152 150818 5 B 30 23.2 0.22 0.32 25.7 117 99.4 150820 7 0 22.3 0.00 0.09 26.6 127 135 150820 7 A 30 22.4 0.00 0.19 26.5 145 140 150820 7 B 30 22.4 0.00 0.12 26.5 124 92.7 150820 7 A 55 21.9 0.59 0.85 26.9 129 127 150820 7 B 55 22.7 0.00 0.78 26.2 142 162 150821 8 0 24.5 0.15 0.12 27.3 139 138 150821 8 A 30 23.1 0.11 5.69 27.4 118 122 150821 8 B 30 24.0 0.00 3.28 14.4 121 128 150823 10 0 23.1 0.00 0.25 25.4 150 169 150823 10 A 30 24.0 0.12 3.49 26.3 117 114 150823 10 B 30 24.2 1.26 2.34 26.3 122 147 150823 10 A 55 22.4 0.11 2.57 26.6 125 118 150823 10 B 55 23.1 0.87 3.51 27.2 132 134 150825 12 0 24.7 0.00 0.02 27.7 121 111 150825 12 A 30 24.7 0.32 0.03 27.8 119 108 150825 12 B 30 23.8 0.12 0.02 27.7 121 111 150825 12 A 55 23.6 0.65 0.07 27.2 133 144 150825 12 B 55 22.9 0.00 0.04 24.8 130 133 150826 13 0 23.3 0.00 0.00 25.2 136 137 150826 13 A 30 22.6 0.00 0.00 27.3 163 190 150826 13 B 30 23.7 0.00 0.00 27.4 143 150 150826 13 A 55 23.2 0.00 0.00 23.6 136 138 150826 13 B 55 22.6 0.00 0.03 26.1 114 122

Table S3. Water chemistry data for test NS2.

Date Day Piez. Depth (cm)

[NO3] (mg/L)

[NO2] (mg/L)

[NH4] (mg/L)

[DOC] (mg/L)

[Cl] (mg/L)

[SO4] (mg/L)

150628 6 0 21.4 0.00 0.03 28.6 94.9 68.5 150628 6 A 30 21.4 0.00 0.04 24.5 76.2 56.2 150628 6 B 30 21.1 0.00 0.04 24.5 91.2 65.7 150629 7 0 23.2 0.00 0.03 20.2 141 164 150629 7 A 30 23.6 0.00 0.08 22.0 127 133 150629 7 B 30 22.5 0.00 0.04 24.0 131 140 150630 8 0 21.9 0.00 0.05 28.3 137 150 150630 8 A 30 20.9 0.00 0.04 27.7 89.2 58.7 150630 8 B 30 21.0 0.00 0.03 25.8 85.3 65.6 150701 9 0 21.9 0.00 0.03 30.0 99.2 80.6 150701 9 A 30 21.4 0.00 0.02 26.6 138 117 150701 9 B 30 21.5 0.00 0.03 26.7 129 107 150702 10 0 23.3 0.00 0.04 29.5 121 118 150702 10 A 30 23.9 0.00 0.05 27.8 137 130 150702 10 B 30 23.3 0.00 0.03 29.2 640 602 150703 11 0 21.8 0.00 0.01 128.7 123 150703 11 A 30 21.7 0.00 0.03 82.2 64.3 150703 11 B 30 21.4 0.00 0.04 105.8 79.0

Table S4. Water chemistry data for test PRB1.

Date Day Piez. Depth (cm)

[NO3] (mg/L)

[NO2] (mg/L)

[NH4] (mg/L)

[DOC] (mg/L)

[Cl] (mg/L)

[SO4] (mg/L)

150828 2 0 23.5 0.00 0.23 29.5 167 181 150828 2 PRB 24.7 0.17 0.14 31.0 130 123 150828 2 A 30 24.5 0.29 3.70 32.1 168 194 150828 2 B 30 24.5 0.37 1.66 36.1 130 128 150828 2 A 55 23.6 0.00 5.86 35.7 140 146 150828 2 B 55 25.0 0.51 5.17 37.0 115 177 150830 4 0 23.5 0.00 0.00 27.5 127 116 150830 4 PRB 23.6 0.00 0.01 26.4 122 115 150830 4 A 30 23.5 0.00 0.28 28.4 121 118 150830 4 B 30 23.2 0.00 0.21 27.5 191 207 150830 4 A 55 23.8 0.00 0.08 26.6 127 119 150830 4 B 55 23.7 0.00 0.23 30.0 126 117 150831 5 0 24.6 0.00 0.00 28.8 175 200 150831 5 PRB 24.9 0.00 0.00 31.5 152 160 150831 5 A 30 24.6 0.14 0.37 32.9 129 126 150831 5 B 30 24.6 0.00 0.57 32.6 150 163 150831 5 A 55 22.5 0.24 0.05 34.9 160 184 150831 5 B 55 23.7 0.14 0.10 31.2 129 127 150831 5 A 80 19.8 0.27 0.03 32.9 113 116 150901 6 0 23.0 0.00 0.05 26.9 118 115 150901 6 PRB 23.6 0.00 0.05 27.6 139 126 150901 6 A 30 21.4 0.60 4.29 33.9 133 130 150901 6 B 30 23.1 0.27 2.47 30.5 111 120 150901 6 A 55 21.4 0.70 0.82 33.6 135 135 150901 6 B 55 23.4 0.27 4.06 30.8 144 138 150901 6 A 80 19.7 0.97 0.20 34.7 129 146 150902 7 0 23.0 0.00 0.00 25.1 127 111 150902 7 PRB 23.2 0.00 0.03 26.6 125 114 150902 7 A 30 22.7 0.00 0.94 27.9 117 116 150902 7 B 30 21.8 0.00 0.55 27.6 129 114 150902 7 A 55 20.2 0.25 0.78 29.2 116 109 150902 7 B 55 22.8 0.00 1.28 25.5 129 116 150903 8 0 23.0 0.00 0.08 26.8 127 116 150903 8 PRB 24.3 0.00 0.06 26.6 129 121 150903 8 A 30 21.5 0.20 4.75 32.9 142 134 150903 8 B 30 23.8 0.00 1.46 28.6 125 117 150903 8 A 55 21.5 0.17 3.08 33.2 119 126 150903 8 B 55 23.8 0.00 2.97 28.4 114 123 150903 8 A 80 21.1 0.19 1.76 34.4 118 117 150904 9 0 24.3 0.00 0.05 25.7 111 113 150904 9 PRB 24.1 0.00 0.00 25.8 107 130

Date Day Piez. Depth (cm)

[NO3] (mg/L)

[NO2] (mg/L)

[NH4] (mg/L)

[DOC] (mg/L)

[Cl] (mg/L)

[SO4] (mg/L)

150904 9 A 30 23.1 0.12 2.51 30.4 91.4 104 150904 9 B 30 24.5 0.00 0.76 27.1 138 137 150904 9 A 55 22.1 0.15 1.05 32.1 133 134 150904 9 B 55 24.9 0.00 1.45 27.0 125 121 150904 9 A 80 21.8 0.15 0.90 32.1 90.7 124 150907 12 0 25.6 0.00 0.00 26.8 115 132 150907 12 PRB 25.3 0.00 0.00 27.2 105 132 150907 12 A 30 24.8 0.00 2.52 30.3 116 125 150907 12 B 30 25.0 0.00 0.99 28.3 145 146 150907 12 A 55 24.9 0.00 2.97 27.4 138 139 150907 12 B 55 25.2 0.00 1.62 28.1 141 142 150907 12 A 80 24.1 0.13 3.47 32.7 135 138 150908 13 0 24.6 0.00 0.00 27.4 91.7 101 150908 13 PRB 23.8 0.00 0.00 26.7 105 132 150908 13 A 30 24.7 0.00 2.72 27.7 117 121 150908 13 B 30 24.6 0.00 1.00 27.3 91.2 120 150908 13 A 55 24.9 0.00 3.80 28.5 121 127 150908 13 B 55 24.9 0.00 1.89 26.7 138 135 150908 13 A 80 25.0 0.00 4.22 27.4 136 137 150909 14 0 23.5 0.00 0.00 27.0 117 112 150909 14 PRB 23.6 0.00 0.00 27.5 115 112 150909 14 A 30 23.8 0.00 0.94 26.4 120 120 150909 14 B 30 24.2 0.00 0.24 28.3 131 124 150909 14 A 55 24.1 0.00 1.81 24.7 116 122 150909 14 B 55 23.8 0.00 0.53 27.6 133 128 150909 14 A 80 24.0 0.00 2.22 28.4 113 122 150910 15 0 23.4 0.00 0.01 21.8 128 122 150910 15 PRB 22.4 0.00 0.02 27.4 119 117 150910 15 A 30 23.6 0.00 0.16 27.5 106 115 150910 15 B 30 23.4 0.00 0.07 26.7 126 122 150910 15 A 55 23.5 0.00 0.30 28.4 122 135 150910 15 B 55 23.8 0.00 0.07 28.0 129 123 150910 15 A 80 23.0 0.00 0.29 27.7 105 105

Table S5. Water chemistry data for test PRB2.

Date Day Piez. Depth (cm)

[NO3] (mg/L)

[NO2] (mg/L)

[NH4] (mg/L)

[DOC] (mg/L)

[Cl] (mg/L)

[SO4] (mg/L)

150708 2 0 23.1 0.00 0.03 101 78.9 150708 2 PRB 22.9 0.00 0.02 94.8 92.1 150708 2 A 30 23.5 0.00 0.13 119 116 150708 2 B 30 23.3 0.00 0.13 115 96.6 150709 3 0 22.2 0.00 0.02 85.6 57.1 150709 3 PRB 22.0 0.00 0.04 127 123 150709 3 A 30 22.2 0.00 0.12 81.0 72.8 150709 3 B 30 22.1 0.00 0.08 78.6 77.6 150709 3 A 60 22.2 0.00 0.07 109 94.5 150709 3 B 60 21.9 0.00 0.06 122 117 150709 3 A 90 22.2 0.00 0.10 121 117 150710 4 0 22.6 0.00 0.02 119 97.8 150710 4 PRB 21.8 0.00 0.04 124 118 150710 4 A 30 21.8 0.00 0.14 122 108 150710 4 B 30 22.2 0.00 0.05 121 108 150710 4 A 60 22.7 0.00 0.05 124 110 150710 4 B 60 22.2 0.00 0.05 133 124 150710 4 A 90 21.4 0.00 0.04 114 101 150710 4 B 90 22.0 0.00 0.03 123 110 150712 6 0 22.0 0.00 0.04 116 105 150712 6 PRB 22.2 0.00 0.04 146 140 150712 6 A 30 21.9 0.00 0.06 133 124 150712 6 B 30 21.8 0.00 0.08 109 88.2 150712 6 A 60 21.7 0.00 0.06 107 78.0 150712 6 B 60 22.1 0.00 0.00 143 134 150712 6 A 90 21.2 0.00 0.06 96.3 73.3 150713 7 0 22.7 0.00 0.03 180 170 150713 7 PRB 22.6 0.00 0.07 193 185 150713 7 A 30 23.9 0.00 0.06 199 186 150713 7 B 30 23.6 0.00 0.08 237 223 150713 7 A 60 22.9 0.16 0.96 195 185 150713 7 B 60 24.4 0.00 0.88 200 187 150713 7 A 90 24.9 0.00 0.48 192 176 150714 8 0 22.3 0.00 0.00 145 131 150714 8 PRB 22.5 0.00 0.05 148 131 150714 8 A 30 22.3 0.00 0.07 97.8 86.7 150714 8 B 30 22.2 0.00 0.04 106 93.7 150714 8 A 60 22.5 0.00 0.24 192 174 150714 8 B 60 22.2 0.00 0.43 179 164 150714 8 A 90 22.8 0.00 0.98 134 117 150715 9 0 24.2 0.00 0.06 193 174

Date Day Piez. Depth (cm)

[NO3] (mg/L)

[NO2] (mg/L)

[NH4] (mg/L)

[DOC] (mg/L)

[Cl] (mg/L)

[SO4] (mg/L)

150715 9 PRB 23.8 0.00 0.02 132 119 150715 9 A 30 23.8 0.11 0.04 145 132 150715 9 B 30 23.3 0.21 0.08 129 120 150715 9 A 60 23.7 0.00 0.45 140 139 150715 9 B 60 25.1 0.00 0.84 128 125 150715 9 A 90 24.1 0.75 0.53 151 165 150716 10 0 22.9 0.00 0.00 150716 10 PRB 23.2 0.00 0.02 123 110 150716 10 A 30 23.0 0.18 0.00 120 110 150716 10 B 30 23.2 0.00 0.00 120 108 150716 10 A 60 23.4 0.15 0.05 120 108 150716 10 B 60 20.5 0.24 0.12 115 106 150716 10 A 90 23.0 0.22 0.06 115 108 150717 11 0 23.5 0.00 0.02 137 142 150717 11 PRB 25.2 0.10 0.01 124 117 150717 11 A 30 24.5 0.00 0.05 144 158 150717 11 B 30 23.3 0.00 0.03 123 130 150717 11 A 60 24.0 0.34 2.36 144 170 150717 11 B 60 23.7 0.25 0.42 129 119 150717 11 A 90 23.4 0.52 3.72 133 134 150720 14 0 25.4 0.00 0.07 161 170 150720 14 PRB 25.3 0.00 0.03 165 154 150720 14 A 30 24.9 0.00 0.26 98.9 85.3 150720 14 B 30 22.5 0.40 0.79 100 112 150720 14 A 60 25.1 0.64 0.72 178 173 150720 14 B 60 23.6 0.75 2.55 127 117 150720 14 A 90 24.9 1.69 4.27 179 185 150721 15 0 22.7 0.00 0.28 137 152 150721 15 PRB 22.7 0.00 0.04 121 121 150721 15 A 30 23.8 0.00 0.15 120 120 150721 15 B 30 23.7 0.00 0.00 126 131 150721 15 A 60 24.7 0.00 0.06 129 138 150721 15 B 60 24.1 0.31 0.08 119 117 150721 15 A 90 23.0 0.62 0.09

Table S6. Nitrate isotope data for a subset of samples from tests NS1 and PRB1.

Day Piez. Depth (cm)

d15N Air (‰)

d18O VSMOW (‰)

Test NS1 6 0 6.45 5.56 6 A 30 6.57 5.21 7 0 5.54 5.32 7 B 30 5.56 5.36 7 B 55 5.49 5.31

10 0 5.62 5.18 10 A 30 5.82 5.50 10 A 55 5.65 5.41 13 0 5.66 5.41 13 A 30 5.50 5.27 13 A 55 5.59 5.49

Test PRB1 5 0 5.99 5.90 5 PRB 5.99 5.82 5 A 30 6.10 6.10 5 A 55 6.37 6.59 5 A 80 6.36 6.57 6 0 5.69 5.53 6 PRB 5.85 5.57 6 A 30 7.41 5.25 6 A 55 7.77 5.38 6 A 80 8.99 5.08 8 0 5.77 5.62 8 PRB 5.76 5.61 8 A 30 6.88 5.93 8 A 55 6.88 5.90 8 A 80 7.10 5.91 9 0 5.69 5.57 9 PRB 5.71 5.27 9 A 30 6.30 5.65 9 A 55 6.52 5.90 9 A 80 6.61 5.61

14 0 5.60 5.45 14 PRB 5.59 5.40 14 A 30 5.65 6.05 14 A 55 5.84 5.44 14 A 80 5.92 5.53

Table S7. Result of ANOSIM and adonis statistical testing of microbiological sequencing

data divided into different groups. A high p-value (p > 0.05) indicates that there is not a

statistically significant difference between the two groups; a low p-value (p < 0.05) indicates a

statistically significant difference between the two groups.

Sample group Number of

samples Divided by ANOSIM p-value

adonis p-value

All 16 PRB 0.004 0.002 All 16 Time 0.031 0.027

After infiltration 8 Depth 0.701 0.862 After infiltration 8 PRB 0.031 0.033

Before infiltration 8 Depth 0.477 0.215 Before infiltration 8 PRB 0.033 0.029

Native soil 8 Time 0.041 0.026 Native soil 8 Test 0.174 0.176

PRB 8 Time 0.023 0.029 PRB 8 Test 0.68 0.631

Table S8. List of OTUs present at >1% on average in any group of samples. The samples were grouped as follows: NSB (native soil, before

infiltration), NSA (native soil, after infiltration), PRBB (PRB, before infiltration), and PRBA (PRB, after infiltration).

Kingdom Phylum Class Order Family Genus Bacteria Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Rhodovulum Bacteria Proteobacteria Alphaproteobacteria Caulobacterales Caulobacteraceae Asticcacaulis Bacteria Verrucomicrobia Verrucomicrobiae Verrucomicrobiales Verrucomicrobiaceae Prosthecobacter Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales Sphingomonadaceae Novosphingobium Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales Sphingomonadaceae Sphingobium Bacteria Proteobacteria Alphaproteobacteria Caulobacterales Caulobacteraceae – Bacteria Proteobacteria Betaproteobacteria Burkholderiales Comamonadaceae Other Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales Erythrobacteraceae – Bacteria Proteobacteria Betaproteobacteria Burkholderiales Comamonadaceae – Bacteria Verrucomicrobia Verrucomicrobiae Verrucomicrobiales Verrucomicrobiaceae – Bacteria Proteobacteria Deltaproteobacteria Myxococcales OM27 – Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales Sphingomonadaceae – Bacteria Planctomycetes Planctomycetia Planctomycetales Planctomycetaceae Planctomyces Bacteria OD1 ZB2 – – – Bacteria Chlamydiae Chlamydiia Chlamydiales Rhabdochlamydiaceae Candidatus Rhabdochlamydia Bacteria Actinobacteria Actinobacteria Actinomycetales Micrococcaceae – Bacteria NC10 12-24 MIZ17 – – Bacteria Acidobacteria – – – – Bacteria Proteobacteria Alphaproteobacteria Rhizobiales Hyphomicrobiaceae Rhodoplanes Bacteria Proteobacteria Alphaproteobacteria Rhodospirillales Rhodospirillaceae – Bacteria Proteobacteria Betaproteobacteria – – – Unassigned Other Other Other Other Other Bacteria Proteobacteria Alphaproteobacteria Rhizobiales – – Bacteria Proteobacteria Gammaproteobacteria Xanthomonadales Sinobacteraceae – Bacteria WS3 PRR-12 Sediment-1 – – Bacteria Nitrospirae Nitrospira Nitrospirales Nitrospiraceae Nitrospira Bacteria Acidobacteria Acidobacteriia Acidobacteriales Koribacteraceae – Bacteria Acidobacteria DA052 Ellin6513 – – Bacteria Gemmatimonadetes Gemm-1 – – –

Kingdom Phylum Class Order Family Genus Bacteria Proteobacteria Betaproteobacteria MND1 – – Bacteria Actinobacteria MB-A2-108 – – – Bacteria Nitrospirae Nitrospira Nitrospirales 0319-6A21 – Bacteria Actinobacteria Thermoleophilia Gaiellales Gaiellaceae – Bacteria Proteobacteria Deltaproteobacteria Syntrophobacterales Syntrophobacteraceae – Bacteria Acidobacteria Acidobacteria-6 iii1-15 – – Bacteria Chloroflexi Ellin6529 – – – Bacteria Planctomycetes Planctomycetia Pirellulales Pirellulaceae – Bacteria Nitrospirae Nitrospira Nitrospirales Nitrospiraceae – Bacteria Actinobacteria MB-A2-108 0319-7L14 – – Bacteria Proteobacteria Betaproteobacteria IS-44 – – Bacteria Actinobacteria Acidimicrobiia Acidimicrobiales – – Archaea Euryarchaeota Thermoplasmata E2 Methanomassiliicoccaceae – Bacteria GAL15 – – – – Bacteria Firmicutes Bacilli Bacillales Bacillaceae Bacillus Bacteria Proteobacteria Gammaproteobacteria Chromatiales Ectothiorhodospiraceae Ectothiorhodospira Bacteria Proteobacteria Alphaproteobacteria Rhizobiales Methylobacteriaceae Methylobacterium Bacteria Proteobacteria Gammaproteobacteria Enterobacteriales Enterobacteriaceae –

Table S9. 16S rRNA gene sequencing results. The average relative abundance is calculated for groups NSB (native soil, before infiltration), NSA

(native soil, after infiltration), PRBB (PRB, before infiltration), and PRBA (PRB, after infiltration). Only OTUs with at least 1% in any of these

averages are shown (see complete taxonomy for each OTU in Table S8). The results of two-sample equal-variance two-tailed student’s t-tests

comparing NSA and PRBA samples highlight the difference between the two soil treatments (* = < 0.05, ** = < 0.01, *** = < 0.002). Log2 fold-

changes comparing NSB and NSA (column NS) and comparing PRBB and PRBA (column PRB) quantify enhancement of OTUs in native soil and

below the PRB.

Lowest assigned taxonomy Average relative abundance (%) Standard deviation (%) t-test: NSA vs. PRBA

log2 fold-change

NSB NSA PRBB PRBA NSB NSA PRBB PRBA NS PRB Rhodovulum (g) 0.001 0.002 0.000 5.9 0.001 0.001 0.000 2.3 0.002 ** 1.1 >18.6 Asticcacaulis (g) 0.001 0.000 0.000 1.3 0.001 0.001 0.000 1.4 0.12 -0.9 <-4.6 Prosthecobacter (g) 0.000 0.011 0.001 3.9 0.000 0.020 0.002 2.3 0.016 * >9.5 11.9 Novosphingobium (g) 0.003 0.51 0.007 14 0.003 0.80 0.007 4.5 0.001 *** 7.3 11.1 Sphingobium (g) 0.001 0.86 0.007 4.5 0.001 1.4 0.014 3.0 0.072 10.2 9.3 Caulobacteraceae (f) 0.003 0.21 0.003 1.9 0.004 0.21 0.003 1.4 0.059 6.2 9.1 Comamonadaceae(f); Other 0.078 0.17 0.004 1.9 0.15 0.27 0.003 1.1 0.024 * 1.1 8.9 Erythrobacteraceae (f) 0.027 0.011 0.022 2.0 0.026 0.021 0.042 1.2 0.017 * -1.3 6.5 Comamonadaceae (f) 0.068 0.48 0.063 5.2 0.053 0.22 0.055 1.6 0.001 ** 2.8 6.4 Verrucomicrobiaceae (f) 0.067 0.16 0.084 4.8 0.046 0.046 0.052 3.0 0.022 * 1.3 5.8 OM27 (f) 0.013 0.030 0.053 1.5 0.015 0.03 0.056 1.0 0.033 * 1.3 4.8 Sphingomonadaceae (f) 0.058 0.26 0.21 1.6 0.027 0.21 0.066 1.3 0.091 2.2 3.0 Planctomyces (g) 0.14 0.42 0.18 1.4 0.071 0.15 0.13 1.7 0.31 1.5 2.9 ZB2 (c) 0.006 1.3 0.025 0.18 0.011 1.7 0.045 0.27 0.25 7.8 2.8 Candidatus Rhabdochlamydia (g) 0.27 0.27 0.27 1.1 0.16 0.18 0.43 1.1 0.20 0.0 2.0 Micrococcaceae (f) 0.65 0.94 0.59 1.5 0.72 0.70 0.55 2.2 0.65 0.5 1.3 MIZ17 (o) 1.7 0.28 0.22 0.35 1.7 0.33 0.26 0.25 0.76 -2.6 0.6 Acidobacteria (p) 1.1 0.42 0.12 0.17 0.35 0.20 0.080 0.17 0.11 -1.4 0.5 Rhodoplanes (g) 1.6 1.6 0.89 0.89 0.18 0.19 0.31 0.40 0.017 * 0.0 0.0 Rhodospirillaceae (f) 2.0 2.3 1.7 1.1 0.20 0.45 0.81 0.55 0.013 * 0.2 -0.7 Betaproteobacteria (c) 3.8 3.6 3.4 2.0 0.78 0.97 1.7 1.2 0.094 -0.1 -0.7 Unassigned; Other 4.4 2.3 3.0 1.7 3.1 0.78 1.2 1.3 0.47 -0.9 -0.8

Rhizobiales (o) 2.0 2.6 2.4 1.4 0.37 0.50 1.3 0.38 0.008 ** 0.4 -0.8

Lowest assigned taxonomy Average relative abundance (%) Standard deviation (%) t-test: NSA vs. PRBA

log2 fold- change

NSB NSA PRBB PRBA NSB NSA PRBB PRBA NS PRB Sinobacteraceae (f) 2.0 2.4 2.1 1.1 0.44 0.34 0.38 0.90 0.039 * 0.3 -0.9 Sediment-1 (o) 1.4 0.98 0.55 0.25 0.34 0.14 0.26 0.19 0.001 *** -0.5 -1.1 Nitrospira (g) 1.8 2.3 1.7 0.70 0.44 1.2 0.93 0.40 0.041 * 0.3 -1.3 Koribacteraceae (f) 1.5 0.87 0.94 0.39 0.69 0.33 0.45 0.24 0.054 -0.8 -1.3 Ellin6513 (o) 0.96 1.1 0.44 0.17 0.59 0.58 0.23 0.12 0.021 * 0.2 -1.4 Gemm-1 (c) 7.6 6.2 4.1 1.6 0.85 1.4 1.7 0.91 0.001 ** -0.3 -1.4 MND1 (o) 4.0 4.3 3.0 1.2 0.27 1.4 0.75 0.70 0.007 ** 0.1 -1.4 MB-A2-108 (c) 3.5 1.8 0.56 0.21 2.3 0.58 0.37 0.16 0.002 ** -1.0 -1.4 0319-6A21 (f) 2.4 1.8 1.7 0.64 0.27 0.75 0.65 0.33 0.032 * -0.4 -1.4 Gaiellaceae (f) 1.8 1.2 1.3 0.49 0.21 0.36 0.60 0.31 0.019 * -0.5 -1.4 Syntrophobacteraceae (f) 2.1 1.9 1.9 0.69 0.61 0.40 0.83 0.38 0.004 ** -0.2 -1.5 iii1-15 (o) 3.7 4.1 3.9 1.3 0.58 0.58 0.84 0.73 0.001 ** 0.1 -1.5 Ellin6529 (c) 2.3 1.8 1.4 0.46 0.28 0.49 0.51 0.35 0.004 ** -0.3 -1.6 Pirellulaceae (f) 0.95 0.94 1.2 0.38 0.42 0.20 0.47 0.23 0.010 * 0.0 -1.7 Nitrospiraceae (f) 0.96 1.6 1.3 0.37 0.34 0.44 0.55 0.24 0.002 ** 0.7 -1.8 0319-7L14 (o) 1.6 1.4 1.6 0.44 0.68 0.36 0.79 0.27 0.006 ** -0.2 -1.9 IS-44 (o) 0.86 1.1 1.3 0.33 0.14 0.47 0.47 0.13 0.018 * 0.4 -1.9 Acidimicrobiales (o) 2.3 1.3 1.4 0.32 0.40 0.33 0.18 0.18 0.002 ** -0.9 -2.2 Methanomassiliicoccaceae (f) 0.70 2.5 1.1 0.21 0.19 1.1 0.84 0.16 0.008 ** 1.8 -2.4 GAL15 (p) 3.7 5.9 4.8 0.83 2.2 0.90 2.3 0.73 0.000 *** 0.7 -2.5 Bacillus (g) 0.84 0.36 1.0 0.15 0.56 0.19 0.22 0.10 0.099 -1.2 -2.8 Ectothiorhodospira (g) 2.9 2.3 9.2 1.1 2.1 1.8 9.3 1.3 0.28 -0.3 -3.1 Methylobacterium (g) 0.45 0.34 1.4 0.12 0.27 0.34 1.4 0.13 0.27 -0.4 -3.5 Enterobacteriaceae (f) 1.6 1.4 7.7 0.62 1.8 2.0 12 0.72 0.50 -0.2 -3.6