potential microbial functional activity along a posidonia
TRANSCRIPT
University of Southern Denmark
Potential microbial functional activity along a Posidonia oceanica soil profile
Pineiro-Juncal, Nerea; Angel Mateo, Miguel; Holmer, Marianne; Martinez-Cortizas, Antonio
Published in:Aquatic Microbial Ecology
DOI:10.3354/ame01872
Publication date:2018
Document version:Accepted manuscript
Citation for pulished version (APA):Pineiro-Juncal, N., Angel Mateo, M., Holmer, M., & Martinez-Cortizas, A. (2018). Potential microbial functionalactivity along a Posidonia oceanica soil profile. Aquatic Microbial Ecology, 81(2), 189-200.https://doi.org/10.3354/ame01872
Go to publication entry in University of Southern Denmark's Research Portal
Terms of useThis work is brought to you by the University of Southern Denmark.Unless otherwise specified it has been shared according to the terms for self-archiving.If no other license is stated, these terms apply:
• You may download this work for personal use only. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying this open access versionIf you believe that this document breaches copyright please contact us providing details and we will investigate your claim.Please direct all enquiries to [email protected]
Download date: 08. Jan. 2022
1
Title: Potential microbial functional activity along a Posidonia oceanica (L) Delile soil 1
profile. 2
3
Running title: Posidonia soil microbial functional activity 4
5
Authors: Nerea Piñeiro-Juncal 1,2,*, Miguel Ángel Mateo 2, Marianne Holmer 3 and 6
Antonio Martínez-Cortizas 1. 7
8
1 Departamento de Edafoloxía e Química Agrícola, Facultade de Bioloxía, Campus Vida 9
s/n, 15782 Santiago de Compostela, Spain 10
2 Centro de Estudios Avanzados de Blanes (CEAB), c/ Accés a la Cala St. Francesc, 14, 11
17300 – Blanes, Girona, Spain 12
3 Institute of Biology, University of Southern Denmark, Campusvej 55, 5230 Odense M, 13
Denmark 14
15
Abstract: 16
17
Among the ecosystem services provided by the seagrass Posidonia oceanica, its 18
role as a carbon sink has received growing attention during the last decade. The 19
sequestration and stabilization of the carbon in the soils are tightly linked to the 20
activity of its microbial community. Ecoplates (Biolog) is a semi-quantitative tool for 21
the assessment of the microbial functional activity that generates profiles of microbial 22
carbon sources utilization. In this study the aerobic and anaerobic carbon consumption 23
* Corresponding author: [email protected]
2
in a healthy P. oceanica meadow was compared along the soil depth down to 130 cm. 24
After eleven days average well color development (AWCD), Shannon-Weaver diversity 25
index (H’), and kinetics of carbon consumption were assessed. Both the aerobic and 26
anaerobic metabolisms showed carbon consumption at all depths. Color observations 27
indicated that three carbon guilds were mainly consumed while three other presented 28
low or no consumption at all. Carbon consumption kinetics was similar for both 29
metabolisms. The results suggest a pronounced stratification of the microbial 30
community controlled by oxygen availability. Furthermore, the degradation of 31
carboxylic acids and amines was low. Despite of the higher aerobic metabolism in the 32
top 40 cm, the anaerobic metabolism was dominant, supporting high sequestration of 33
carbon in P. oceanica meadows. 34
35
Key words: Biolog® EcoPlatesTM, microbial activity, aerobic and anaerobic 36
metabolism, long term carbon sink, Posidonia oceanica. 37
38
39
3
40
1. Introduction: 41
42
Posidonia oceanica (L.) Delile is an aquatic macrophyte, endemic of the 43
Mediterranean Sea, forming extensive meadows down to 40 m depth (Ballesta et al. 44
2000), covering from 25.000 to 50.000 km2 (Pasqualini et al. 1998). The meadows grow 45
in height by accretion due to the sedimentation of particles, enhanced by the 46
macrophyte canopy (Gacia & Duarte 2001; Marbá et al. 2006). P. oceanica, as other 47
seagrasses, increases the carbon content of the substrate (Mateo & Romero 1997) and 48
its bacterial diversity (García-Martínez et al. 2009; Lopez et al. 1995). They have been 49
recognized as key coastal ecosystems (Campagne et al. 2014), acting as filter and sink 50
of natural particulates and pollutants (Richir et al. 2013; Serrano et al. 2013). Their 51
efficiency as filter of pollutants is much higher than for the average macrophytes 52
(Vassallo et al. 2013), and due to this capacity, P. oceanica is widely used as 53
bioindicator (Pergent-Martini et al. 2005) of water quality (Foden & Brazier 2007; 54
Gobert et al. 2009) and of human impact in coastal ecosystems (Balata et al. 2008; Luy 55
et al. 2012; Richir & Gobert 2014). 56
The different species of the Posidonia genus, distributed along the Mediterranean 57
and the Australian coasts, stimulate the deposition of sediments chronologically, 58
forming deep soils (or mats) with a high potential as environmental archives of the 59
Holocene (Ancora et al. 2004; Serrano et al. 2013; Tovar-Sánchez et al. 2010; Tranchina 60
et al. 2005; Serrano et al. 2016). The organic matter can be trapped on the soil over 61
millennia due to the anoxic conditions that prevent its fast remineralization (Canfield 62
4
1994), becoming the meadow a significant global carbon sink (De Falco et al. 2011; 63
Mateo et al. 2006). 64
Microorganisms play a major role in carbon stabilization and degradation, 65
determining how and how much carbon accumulates. Carbon degradation rates in 66
marine sediments are usually associated to redox conditions, i.e. the availability of 67
electron acceptors coupled with its correspondent microbial metabolism (Thamdrup et 68
al. 2000; Fenchel et al. 1998; Llobet-Brossa et al. 2002). Microbial metabolism is much 69
faster in aerobic than in anaerobic conditions (Kristensen & Holmer 2001), and leads to 70
a more complete mineralization of a wide range of carbon sources. Anaerobic 71
microorganisms usually need a more complex bacterial food web working stepwise 72
(Fenchel et al. 1998). Furthermore, they are inefficient in mineralizing structural 73
molecules such as lignin, being those molecules usually preserved in anoxic 74
environments (Kristensen et al. 1995). Bacterial activity in seagrass meadows has been 75
studied before, finding a positive correlation with seagrass production (Lopez et al. 76
1995; Danovaro et al. 1994; Glazebrook et al. 1996; Smith et al. 2004) that can shift to 77
a negative correlation when nutrients availability is low (Danovaro et al. 1994; Holmer 78
et al. 2003; López et al. 1998; Smith et al. 2004). Another study focusing in the 79
decomposition of the organic matter in a P. oceanica mat found overall low 80
remineralization rates and a decrease in nutrients (i.e. nitrogen) with depth in the mat 81
(Pedersen et al. 2011). 82
Genetic tools allow to study uncultivated microorganisms, but generate a large 83
amount of information, difficult to process, and not necessarily related with the carbon 84
consumption or the metabolism of the microbial community (Rodríguez-Valera 2004; 85
Sørensen et al. 2007). Biolog® EcoPlatesTM is a semi-quantitative method developed to 86
5
characterize the microbial functional activity by measuring the respiration of different 87
carbon sources divided in six guilds: carbohydrates, polymers, carboxylic acids, amino 88
acids, amines and miscellaneous compounds. Respiration of the microbial community 89
is revealed by the reduction of a tetrazolium dye that is included with the carbon 90
source. Ecoplates have been used to characterize microbial activity in Posidonia spp. 91
mats, in the water column above them (Richir et al. 2012) and in the soil (Säwström et 92
al. 2016), but only under aerobic conditions. As the mat is a highly reduced 93
environment, experiments comparing both aerobic and anaerobic activity may provide 94
important information on the role of the two metabolisms. 95
Ecoplates have also been used to measure aerobic activity in rivers, lakes and soils, 96
comparing differences in potential carbon consumption seasonally and geographically 97
(Pérez Rodríguez & Martínez Cortizas 2014; Freixa et al. 2015; Christian & Lind 2007; 98
Lyons & Dobbs 2012). Christian and Lind (2007), working in a seasonally stratified lake, 99
found a correlation between the consumed carbon sources, seasonal variations and 100
temperature, O2 diffusion, and redox potential. The same authors, successfully used 101
Ecoplates to measure anaerobic activity in water samples, being the anaerobic 102
communities also affected by seasonal variations (Christian & Lind 2006). 103
In the present study we used Ecoplates to assess the patterns of distribution of 104
bacterial activity within the mat. To this end we incubated mat samples, obtained 105
along a 1.30 m mat core of P. oceanica, under anaerobic and aerobic conditions. The 106
implications of this microbial patterns over carbon degradation and preservation in the 107
soil are discused. 108
109
110
6
2. Materials and Methods: 111
112
a. Sampling site and field works: 113
114
A core of P. oceanica soil was taken in Portlligat Bay, Girona, Spain (42o 17.454’ N, 115
3º 17.513’ E). Around 69% of the bay is covered by well-preserved P. oceanica 116
meadows growing in medium to coarse sand (Lo Iocano et al. 2008). The bay connects 117
to the sea by a 213 m wide opening to the North East. 118
Sampling was done by SCUBA diving by manually hammering and rotating a PVC 119
corer (160 cm long, i.d. 7.5 cm) into the soil. A core catcher was fitted at the bottom of 120
the corer to avoid the loss of material during the extraction. The corer penetrated 146 121
cm in the soil but the length of core retrieved was 91 cm. Assuming the core catcher 122
worked efficiently and that no material was left behind during retrieval, this yields a 123
core compression of 38%, a usual value for this coring method (Glew et al. 2001). 124
The core was sealed with stoppers, taped at both ends and transported to the 125
laboratory under cold conditions within the 2 hours after collection. In the laboratory 126
the core was cut longitudinally into two halves using a ceramic knife for the inner core 127
to avoid metal contamination. Both halves (referred to as hemi-cores) were then 128
sealed and stored at 5 oC in darkness for 2 days until processing. 129
130
b. Analysis of microbial functional activity: 131
132
The eight bottom centimeters were discarded resulting in a remaining core length 133
of 83 cm. Core-shortening (compression) was corrected applying a logarithmic function 134
7
to the length measured in the laboratory to approximate it to the field length, as in 135
Serrano et al. (2012), resulting in a 131 cm soil profile. The labels of the samples in this 136
study refer to the decompressed depth. 137
138
i. Ecoplates aerobic incubations: 139
140
In essence, the protocol performed for aerobic and anaerobic incubations of the 141
samples is a modification of the method applied by Pérez-Rodríguez and Martínez-142
Cortizas (2014) used for peat under aerobic conditions and adapted to marine 143
sediment samples and anaerobic conditions by the use of a nitrogen chamber and 144
artificial sea water. 145
Each hemi-core was cut into 2 cm sections, subsampled for microbial analysis in 15 146
layers (see Fig. 2 for sample distribution). Because Ca concentrations greater than 100 147
ppm can induce false positives in Ecoplates (Pierce et al. 2014), 1 sample was taken 148
from each half core to determine the content in Ca (see below). 149
All materials used were sterilized at 120oC for 1h, to avoid contamination between 150
samples. From each section a 4 cm3 sample was taken using a sterile polycarbonate 151
cylinder, and stored in 50 ml falcon tubes with 30 ml of sterilized artificial seawater. 152
The tubes were vigorously shaken at room temperature for 3 days to suspend and 153
activate the microorganisms after the cold storage period. The suspensions were then 154
filtered through 15-20 µm filters (Albet140) and used to inoculate the Ecoplates. Each 155
well was inoculated with 100 µl suspension using a multichannel micropipette and 156
read right away to get the blank (T0). The plates were then incubated at 26 oC for 11 157
8
days. The plates were read twice a day during the first 5 days and once a day after that 158
(T1 to T15). 159
Every plate has 96 wells containing 31 different carbon sources (2 amines, 3 160
miscellaneous compounds, 4 polymers, 7 carbohydrates, 9 carboxylic acids and 6 161
amino acids; see Fig. 1) in triplicate and a blank. Each well contains a minimum growth 162
medium and a tetrazolium dye that develops color when the microorganisms reduce 163
the substrate. The absorbance was read at 590 nm in a spectrophotometer (Model 164
680, Bio-Rad Laboratories, Hercules, CA). 165
Figure 1 166
The Ca concentration was analyzed after filtration of the two samples by atomic 167
absorption, and showed concentrations lower than the limit value (< 100 ppm Ca; data 168
not shown). 169
170
ii. Ecoplate anaerobic incubations: 171
172
Subsampling and replication was the same in the anaerobic and aerobic 173
incubations, except that the extraction, filtration and inoculation were performed in a 174
nitrogen atmosphere (chamber Cole Parmer MO: SL 34790-00). The artificial sea water 175
was bubbled with nitrogen before being used for the experiment within the chamber 176
(Christian & Lind 2006). The tubes and inoculated plates were sealed with parafilm 177
before removing them from the chamber. The incubation and the readings were as for 178
the aerobic incubations, keeping the anaerobic samples sealed with parafilm to avoid 179
the diffusion of oxygen into them. 180
181
9
c. Numerical procedures: 182
183
The data from the Ecoplates was processed using the software R to obtain the 184
mean of each source in each plate at each time, eliminating values that resulted in 185
coefficients of variation larger than 30. To all mean values, the respective mean 186
absorbance at T0 was subtracted, to obtain the variation due only to the consumption 187
of the carbon source, not to the initial color of the inoculum. After that, the average 188
well color development (AWCD) was calculated for each sample at T15 using the 189
following equation (Garland & Mills 1991): 190
AWCD=Σ(Ri-C)/N 191
192
where Ri is the color of each well at a given time (after blank correction), C is the 193
color development in the blank (A1) and N is the number of carbon sources (31). The 194
AWCD quantifies the metabolism in the aerobic and anaerobic incubations. 195
Shannon-Weaver diversity index (H’) was also calculated (Stefanowicz et al. 2010): 196
H’=-Σpi(lnpi) 197
198
where pi is the average corrected-color of each well divided by the sum of all the 199
wells in the plate. 200
To compare the metabolisms a ratio (AN/AE, anaerobic/aerobic) was calculated 201
with the absorbance values at T15 for each carbon guild. Values <10% of the maximum 202
absorbance value of the guild were excluded an considered as not consumed to avoid 203
extreme ratio values. 204
10
The change in AWCD with time showed two distinct phases characterized by a fast 205
(typically T1 to T4-7) and a slower color development (until T15). The slopes of each 206
phase were calculated to express the rate of microbial activity in the wells. 207
208
3. Results: 209
210
The core utilized in this study had a high content of macroscopic plant debris 211
(rhizomes, roots and leaf sheaths) in the upper part (25 cm) whereas limited debris 212
were present in the deeper part (visual observation). 213
214
d. Stratification: 215
216
The microbial metabolism, expressed as AWCD varied between microplates 217
incubated under aerobic and anaerobic conditions (Fig 2). There was consumption of 218
microbial substrates at all depths, but it was higher in the upper part of the mat (upper 219
34 cm) for the aerobic compared to the anaerobic incubation (Student test: P<0.001, 220
mean AWCD: aerobic 0.81±0.16; anaerobic 0.71±0.11). In the deeper samples (below 221
34 cm) anaerobic metabolism was higher than aerobic (Student test: P<0.001, mean 222
AWCD: aerobic 0.55±0.12; anaerobic 0.87±0.20). For the whole core anaerobic 223
metabolism was higher than the aerobic one (Student test: P<0.001, mean AWCD: 224
aerobic, 0.66±0.19; anaerobic, 0.80±0.18). 225
226
Figure 2 227
228
11
The development in AWCD showed two distinct phases in all incubations with a 229
fast rate in the beginning (until T4 or T7 in aerobic incubations and T4 to T12 in 230
anaerobic conditions) followed by a slower rate for the remaining part of the 231
incubation period. The aerobic bacteria responded faster in the upper parts of the 232
core, whereas the rates were similar for aerobic and anaerobic incubations in the 233
deeper parts. The rates of the second phase were similar for both metabolisms and 234
showed little variation with the depth of the core (Fig 3). 235
236
Figure 3 237
238
The H’ diversity index was very similar for both conditions and all the samples, 239
averaging 3.1±0.1. 240
241
e. Compounds consumption: 242
243
The carbon utilization was similar for aerobic and anaerobic metabolisms but there 244
were major differences between the substrates. A higher consumption (the average 245
color development of the carbon source in the whole core>1) was recorded in aerobic 246
conditions for B4, C1, C4, D2, D4, E1, E2, F1, G1 and H1 (3 amino acids, 4 247
carbohydrates, 2 polymers and 1 carboxylic acid; Fig. 1), and in anaerobic conditions 248
for B1, B4, C1, C4, D1, D2, E1, E2, E4, F1, F2, F4, G1 and H1 (1 miscellaneous 249
compound, 4 polymers, 2 amino acids, 4 carbohydrates and 1 carboxylic acid; Fig. 1). 250
There was no consumption (average color development <0.3) of compounds A2, C3, 251
D3, E3, G2, G3 and H2 (1 carbohydrate, 4 carboxylic acids and 2 miscellaneous; Fig. 1) 252
12
in the aerobic incubation. Similarly, in the anaerobic incubation compounds C3, D3, E3, 253
F3, G3 and H2 (5 carboxylic acids and 1 miscellaneous; Fig. 1) were not consumed. The 254
remaining compounds showed intermediate levels of consumption (aerobic: A3, A4, 255
B1, B2, B3, C2, D1, E4, F2, F3, F4, G1, G4, H3 and H4; anaerobic: A2, A3, A4, B2, B3, C2, 256
D4, G2, G4 H3 and H4). The carbon guilds with highest consumption, were polymers 257
(aerobic average: 1.32±0.54; anaerobic average: 1.23±0.34), carbohydrates (aerobic 258
average: 0.96±0.58; anaerobic average: 1.03±0.56) and amino acids (aerobic average: 259
0.79±0.58; anaerobic average: 1.17±0.42). Conversely, the carbon guilds with the 260
lowest consumption were carboxylic acids (aerobic average: 0.32±0.30; anaerobic 261
average: 0.39±0.42), amines (aerobic average: 0.45±0.26; anaerobic average: 0.55±0.5) 262
and miscellaneous compounds (aerobic average: 0.37±0.45; anaerobic average: 263
0.51±0.5). 264
The aerobic consumption of polymers was higher than the anaerobic in the first 40 265
cm (AN/AE < 1) whereas the opposite was the case below (AN/AE > 1; Fig. 4). The 266
carbohydrates consumption by both metabolisms followed a similar pattern than the 267
polymers, with most of the AN/AE ratios lower than 1 in the top 40 cm and most of 268
them above 1 below 40 cm, and higher heterogeneity among the compounds. Amino 269
acids were consumed at all depths, except for A4, E4 and F4 that showed no 270
consumption under aerobic conditions in several samples (Fig. 4). The distribution with 271
depth of the AN/AE ratio is more homogeneous in the first 40 cm (except for the top 272
sample) and more heterogeneous and higher below. The carboxylic acids showed a 273
low consumption for both metabolisms. The depth distribution of the AN/AE ratios is 274
homogeneous with lower data available below 60 cm due to the lack of consumption 275
of one of the metabolisms or both. The amines ratio showed a very heterogeneous 276
13
distribution with no clear depth pattern. The miscellaneous compounds G2 and H2 277
were almost not consumed under aerobic or anaerobic conditions. The miscellaneous 278
B1 was consumed by both metabolisms at all depths showing a homogeneous rate 279
distribution around 1. 280
281
Figure 4 282
f. AWCD Kinetics: 283
284
The AWCD value of each sample was calculated to the 15th time measurements 285
obtaining its change over time. The resulting curve fitted a logarithmic model where 286
two phases could be differentiated, a first fast consumption phase and a second slower 287
phase, varying the time of change between the two phases among samples. 288
The slope of the first phase was almost always higher than that of the slow 289
consumption phase (Fig. 3). Average slope for the faster aerobic phase is 4.9 10-3 2.6 290
10-3 and for the slow phase 1.5 10-3 0.4 10-3. Under anaerobic conditions the average 291
slopes are 4.5 10-3 2.3 10-3 and 1.0 10-3 0.4 10-3, respectively. On average, for the 292
first phase aerobic consumption rate (i.e. slope) is 3.3 times higher and the anaerobic 293
one is 4.5 times higher than the corresponding ones of the second phase. In the first 294
phase average rates are quite similar for both metabolisms, while the aerobic 295
metabolism is 1.5 higher than the anaerobic for the second phase. 296
The depth distribution of consumption rates is similar for the aerobic metabolism 297
(r= 0.67, p<0.01), with low values in the uppermost sample, the largest values in the 298
next two/three samples and a decreasing trend below. No trend was found for the 299
consumption rates of the two phases under anaerobic conditions. 300
14
301
4. Discussion: 302
303
Microbial functional activity was found for both aerobic and anaerobic 304
metabolisms along the P. oceanica 1.3 m soil core studied. The microbial consumption 305
of the substrates was higher under anaerobic conditions and showed large differences 306
between the carbon sources. Polymers, carbohydrates and amino acids were more 307
readily used than carboxylic acids, amines or miscellaneous compounds. 308
309
g. Stratification: 310
311
Substrates consumption evidences the presence of microorganisms able to 312
metabolize carbon sources with and without oxygen, facultative or strict, in all samples 313
studied. The degradation efficiency or the microbial density in the initial inoculum of 314
aerobic microorganisms was probably higher in the top most half of the core, where 315
the degradation of the carbon sources was higher (Fig. 2). Conversely, in the bottom 316
half of the core the efficiency/density of aerobic microorganisms was probably lower 317
than efficiency/density of anaerobic microorganism, as average AWCD values for the 318
whole core were 1.2 times higher for the anaerobic than for the aerobic experiment 319
(0.80 vs. 0.66), despite anaerobic metabolism being less efficient in carbon 320
degradation (Fenchel et al. 1998). This suggested that the cell number of anaerobic 321
microorganism must have been larger to account for the higher activity. P. oceanica 322
has adapted to the anoxic conditions of its soils by releasing oxygen through the roots 323
to the sediment to maintain an oxic rhizosphere and avoid the formation and uptake of 324
15
sulphides into the plant (Borum et al. 2006; Holmer et al. 2003). The average length of 325
Posidonia ssp. roots is ca. 43 cm (Duarte et al. 1998), justifying the occurrence of 326
aerobic activity in this part of the sediments. Bioturbation is also a common process 327
taking place in the upper layers of seagrass sediments resulting in an increased 328
oxygenation of the sediments (Kristensen 2000). Outside the influence of the 329
rhizosphere or the bioturbated areas, the carbon consumption in the soil occurs 330
through anaerobic metabolism. It seems then plausible to hypothesize that the main 331
process of degradation of organic matter in the mat is the anaerobic metabolism, 332
limiting the degradation of organic matter and thereby preserving plant remains 333
resulting in high accumulation of organic C, as previously observed (e.g., Serrano et al. 334
2012). 335
The H’ index was high and similar for both metabolisms along the depth of the core 336
suggesting a divers microbial community similar to previous findings in P. oceanica 337
meadows (García-Martínez et al. 2009; Säwström et al. 2016) and over terrestrial soils 338
(Liao et al. 2016; Thomas et al. 2016; Nurulita et al. 2016). 339
340
h. Substrate consumption: 341
342
The main difference between the two incubations related to consumption of 343
carbon sources is a larger number of compounds with high consumption (average 344
absorbance value through the whole core >1, Fig. 2) in the anaerobic incubations. Five 345
compounds were metabolized to a higher extent in the anaerobic compared to the 346
aerobic incubations suggesting a wider spectrum of potential carbon sources to the 347
anaerobic metabolism. 348
16
The polymers were metabolized easily by both metabolisms throughout the core. 349
In the first 40 cm the ratio between anaerobic and aerobic metabolism remains below 350
1, meaning a higher aerobic potential activity on polymers. Below this depth only F1 351
(Glycogen) stays < 1, whereas other compounds, e.g. C1 and D1 (Tween 40 and 80), 352
showed higher consumption under anaerobic conditions. 353
As the polymers, the carbohydrates show faster aerobic consumption in the first 40 354
cm, and shift to higher anaerobic consumption below. Some of the carbohydrates 355
differ from this behavior: A2 (β-Methyl-D-Glucoside) had limited aerobic consumption 356
in almost all depths but it was moderate consumed under anaerobic conditions (Fig. 2 357
and 4); C3 (2-Hydroxy Benzoic Acid) had low consumption under both conditions and 358
did not present any clear trend; B1 showed high aerobic consumption in the first 40 cm 359
and low in the remaining samples, but almost no anaerobic consumption. 360
All the amino acids were highly consumed under anaerobic conditions while under 361
aerobic conditions they were only consumed in the first 40 cm, with particular low 362
consumption of A4, E4 and F4 (L-Arginine, L-Treonine and Glycyl-L-Glutamic Acid). This 363
seems to point towards an adapted anaerobic community with a wider spectrum of 364
amino acids as substrates. The aerobic community in the upper sections would rely on 365
leached organic matter from the roots, while the anaerobic community, growing away 366
from the influence of the rhizosphere may have access to more aged organic matter, 367
developing metabolic pathways to degrade less labile carbon sources. 368
The carboxylic acids showed a very low consumption for both metabolisms all 369
along the core. The exception is F2 (D-Glucosaminic acid) that showed only low 370
consumption in the aerobic metabolism, while its consumption was one of the fastest 371
under anaerobic conditions. D-Glucosaminic acid is degraded by the Glucosaminic acid 372
17
dehydrase, which is produced by microorganisms highly adapted to this carbon source 373
(Balazs & Jeanloz 1966). According with our data, microorganisms with potential to 374
degrade glucosamine acids are primarily found under anaerobic conditions, with only 375
moderate potential activity in the most superficial aerobic samples. 376
Amines and miscellaneous compounds showed low consumption in both 377
metabolisms, except for B1 (Pyrubic Acid Methyl Ester) with high consumption under 378
both anaerobic and aerobic conditions, maintaining AN/AE values around 1 in all the 379
core. 380
The different behavior found for highly consumed carbon guilds (polymers, 381
carbohydrates and amino acids) between the first 40 cm and the rest of the core 382
seems to agree with the hypothesis of the rhizosphere influencing the microbial 383
community by changing oxygen availability around it, as mentioned above. The fact 384
that those differences are not shown for the remaining three guilds can be due to its 385
general lower potential consumption. As most of the dissolved organic matter, fresh 386
plant remains and seston organic matter would be deposited on the first layers of the 387
soil or on the rhizosphere, it seems reasonable to hypothesize that the resident 388
anaerobic community had developed the capacity to feed on a wider spectrum of 389
carbon sources (more amino acids, glucosaminic acids…) to compensate for the lack of 390
easy degradable ones. This hypothesis would be in agreement with the wider range of 391
carbon sources highly utilized by the anaerobic metabolism (15 vs 10). 392
Only a few compounds were not consumed in any of the two experiments (C3, D3, 393
E3, G3 and H2, 4 carboxylic acids and 1 miscellaneous compound). Furthermore, there 394
was no consumption of A2 and G2 (carbohydrate and miscellaneous) under aerobic 395
conditions and F3 (carboxylic acid) under anaerobic conditions. C3 and D3 (2 and 4-396
18
Hydroxy Benzoic Acid) are abundant in P. oceanica detritus, associated with roots and 397
sheaths, which are plant debris increasing down core (Kaal et al. 2016). The lack of 398
consumption of these carbon sources in our experiment supports the interpretation of 399
Kaal et al. (2016) that this compounds are part of a recalcitrant bigger compound. The 400
stability of this compound would prevent the microbial degradation associated with 401
the humified materials in the fine fraction of the sediment (Gadel & Bruchet 1987), 402
avoiding the development of a microbial consumer community. Compounds without a 403
community of consumers are good candidates to become blue carbon, even if the 404
mechanisms of stabilization are not yet clear. 405
In a similar work on a peat core, under aerobic conditions, Pérez-Rodríguez and 406
Martínez-Cortizas (2014) found that one of the carboxylic acids (E3, γ-Hydroxybutiric 407
Acid) was not consumed, and that a miscellaneous compound (H2, D,L-α-Glycerol 408
Phosphate) showed very low consumption. They did not test consumption under 409
anaerobic conditions. Also under aerobic conditions, Säwström et al. (2016) found that 410
amines, carboxylic acids and phenolic compounds (C3 and D3, here as carboxylic acids) 411
were the less consumed groups in seagrass sediment cores of estuarine and coastal 412
environments. Furthermore, in a work performed on larger sample sets (704 from the 413
Nederland Soil Monitoring Network, and 73 from a European-wide transect), Rutgers 414
et al. (2015) found a very low consumption of G3 and C3 (both carboxylic acids), and, 415
again, no consumption of E3, suggesting limited consumption of this carbon sources in 416
soils. 417
418
i. Kinetics: 419
420
19
Overall, the AWCD development with incubation time fitted an asymptotic curve. 421
Pérez-Rodríguez and Martínez-Cortizas (2014) obtained similar patterns, interpreting 422
them as sigmoidal, as usually found in the literature on the topic. In our study the lag 423
phase leading to that sigmoidal pattern was not evident, as we did not record data for 424
the first two days. The slope of the phases recorded, a fast growing phase and a slow 425
phase reaching the asymptote, was used as a proxy of carbon consumption rate. The 426
point of change between the two phases varies among the samples but was always 427
between 40-80 h under aerobic conditions and between 40 and 220 h under anaerobic 428
conditions (Fig. 3). This wider range for anaerobic incubations can be related to a 429
slower growth rate of the community (Fenchel et al. 1998; Kristensen & Holmer 2001). 430
The second phase rate is quite similar for both metabolisms with no major 431
variation along the depth of the core (Fig. 3). The first phase is more heterogeneous 432
without a particular trend for anaerobic metabolisms. Conversely, the aerobic 433
metabolism shows a maximum at the top most 40 cm (specially 6-9 and 12-15), with 434
lower values below. The higher rate can be related to a more efficient community or, 435
more likely, to a higher initial cell number. 436
437
5. Conclusions: 438
439
To our knowledge, Ecoplates have been used only once to test anaerobic 440
consumption in water samples (Christian and Lind (2006, 2007), but never in soil or 441
sediment samples. 442
Applying the same protocol under aerobic and anaerobic conditions allows to make 443
direct comparisons of the results. It must be taken into account that the information 444
20
about carbon sources consumption provided by the Ecoplates has to be regarded as 445
potential, because the conditions of the incubation are far from those in the field. 446
Furthermore, growing microorganisms in the laboratory can affect the community 447
structure promoting those more suited to growth in culture medium. Also, as 448
Ecoplates are intended to analyze aerobic metabolism, the final electron acceptor of 449
the respiration chain is O2, but when incubated under anaerobic conditions the 450
electron acceptors available to the microorganisms must come from the initial 451
inoculum. This can affect the results in two ways: the anaerobic acceptor may not be in 452
excess, as assumed, and the initial inoculum may not have one of the possible electron 453
acceptors (Mn+4 for example) present in the soil. This will result in a lower microbial 454
activity. Despite these methodological problems, we think that the following 455
conclusions can be inferred from this study: 456
- Carboxylic acids and amines do not seem to be easily degraded by the P. oceanica 457
soil community. Further research is needed to determine if these compounds are 458
preferentially preserved as recalcitrant forms. The anaerobic community seems to be 459
able to metabolize a wider spectrum of carbon sources than their aerobic 460
counterparts, and has adapted to feed on more recalcitrant compounds. 461
- Microbes showed a clear depth stratification, most likely resulting from oxygen 462
gradients associated to both the distance to the soil surface and to the roots of the 463
seagrass. Two zones were identified: one near the living plant, where the fresh plant 464
exudates and plant debris are actively decomposed, and a slow degradation zone, 465
where the buried carbon decays slowly. Further experiments will be necessary to know 466
if there is also a stratification within the anaerobic metabolism due to the other 467
electron acceptors. 468
21
- Overall, our results point to a clear dominance of the anaerobic metabolism in the 469
seagrass soil samples studied. Consumption stratification and kinetic patterns strongly 470
suggest that the community of anaerobes is larger (cell number) and more active 471
(higher potential activity) in P. oceanica soils than that of their aerobic counterparts. 472
This outcome adds knowledge to explain the low Corg turnover reported for P. oceanica 473
soils. 474
475
Acknowledgements: 476
The authors would like to thanks Marta Pérez Rodriguez and Dr. Xose Luis Otero 477
Perez for their valuable laboratory assistance and to Carmen Leiva Dueñas and Anna 478
Thoran for the samples collection. This work was funded by the projects SUMILEN 479
(CTM2013- 47728-R, MINECO) and PALEOPARK (1104/2014). This is a paper of the 480
Group of Benthic Ecology 2014 SGR 120. 481
482
22
References: 483
Ancora, S. et al., 2004. Posidonia oceanica as a biomonitor of trace elements in the 484
Gulf of Naples: temporal trends by lepidochronology. Environmental toxicology 485
and chemistry, 23(5), pp.1093–1099. 486
Balata, D. et al., 2008. Comparison between epiphyte assemblages of leaves and 487
rhizomes of the seagrass Posidonia oceanica subjected to different levels of 488
anthropogenic eutrophication. Estuarine, Coastal and Shelf Science, 79(3), 489
pp.533–540. Available at: 490
http://linkinghub.elsevier.com/retrieve/pii/S0272771408002187 [Accessed 491
September 25, 2015]. 492
Balazs, E.A. & Jeanloz, R.W., 1966. The Amino Sugars: The Chemistry and Biology of 493
Compounds Containing Amino Sugars United Kin., Berkeley Square House, 494
London: Academic Press Inc. 495
Ballesta, L. et al., 2000. Distribution and dynamics of Posidonia oceanica beds along the 496
Albères coastline. Comptes Rendus de l’Académie des Sciences - Series III - 497
Sciences de la Vie, 323(4), pp.407–414. Available at: 498
http://linkinghub.elsevier.com/retrieve/pii/S0764446900001359. 499
Borum, J. et al., 2006. Oxygen Movement in Seagrasses. In Seagrasses: Biology, Ecology 500
and Conservation. Springer, pp. 255–270. 501
Campagne, C.S. et al., 2014. The seagrass Posidonia oceanica: Ecosystem services 502
identification and economic evaluation of goods and benefits. Marine Pollution 503
Bulletin, 97(1–2), pp.391–400. Available at: 504
http://dx.doi.org/10.1016/j.marpolbul.2015.05.061. 505
Canfield, D.E., 1994. Factors Influencing Organic-Carbon Preservation in Marine- 506
23
Sediments. Chemical Geology, 114(3–4), pp.315–329. 507
Christian, B.W. & Lind, O.T., 2006. Key Issues Concerning Biolog Use for Aerobic and 508
Anaerobic Freshwater Bacterial Community-Level Physiological Profiling. 509
International Review of Hydrobiology, 91(3), pp.257–268. Available at: 510
http://doi.wiley.com/10.1002/iroh.200510838 [Accessed September 25, 2015]. 511
Christian, B.W. & Lind, O.T., 2007. Multiple carbon substrate utilization by bacteria at 512
the sediment-water interface: Seasonal patterns in a stratified eutrophic 513
reservoir. Hydrobiologia, 586, pp.43–56. 514
Danovaro, R., Fabiano, M. & Boyer, M., 1994. Seasonal changes of benthic bacteria in a 515
seagrass bed (Posidonia oceanica) of the Ligurian Sea in relation to origin, 516
composition and fate of the sediment organic matter. Marine Biology, 119(4), 517
pp.489–500. 518
Duarte, C.M. et al., 1998. Root production and belowground seagrass biomass. Marine 519
Ecology Progress Series, 171, pp.97–108. 520
De Falco, G. et al., 2011. Carbonate sedimentation and hydrodynamic pattern on a 521
modern temperate shelf: The strait of Bonifacio (western Mediterranean). 522
Estuarine, Coastal and Shelf Science, 93(1), pp.14–26. Available at: 523
http://linkinghub.elsevier.com/retrieve/pii/S0272771411001065 [Accessed 524
September 25, 2015]. 525
Fenchel, T., King, G.M. & Blackburn, H., 1998. Bacterial biogeochemistry: The 526
Ecophysiology of Mineral Cycling. 3rd ed., San Diego: Elsevier Ltd. Available at: 527
http://www.jstor.org/stable/2259699?origin=crossref. 528
Foden, J. & Brazier, D.P., 2007. Angiosperms (seagrass) within the EU water framework 529
directive: A UK perspective. Marine Pollution Bulletin, 55(1–6), pp.181–195. 530
24
Freixa, A. et al., 2015. The effects of sediment depth and oxygen concentration on the 531
use of organic matter : An experimental study using an in fi ltration sediment 532
tank. Science of the Total Environment, 540, pp.20–31. Available at: 533
http://dx.doi.org/10.1016/j.scitotenv.2015.04.007. 534
Gacia, E. & Duarte, C.M., 2001. Sediment retention by a Mediterranean Posidonia 535
oceanica meadow: The balance between deposition and resuspension. Estuarine, 536
Coastal and Shelf Science, 52(4), pp.505–514. Available at: 537
http://linkinghub.elsevier.com/retrieve/pii/S0272771400907534%5Cnhttp://linki538
nghub.elsevier.com/retrieve/pii/S0272771400907534%5Cnhttp://www.sciencedir539
ect.com/science/article/B6WDV-458W59D-540
45/2/d75ad39a721915d52b453aecc933f305. 541
Gadel, F. & Bruchet, A., 1987. Application of pyrolysis-gas chromatography-mass 542
spectrometry to the characterization of humic substances resulting from decay of 543
aquatic plants in sediments and waters. Water Research, 21(10), pp.1195–1206. 544
García-Martínez, M. et al., 2009. Bacterial community dynamics in a seagrass 545
(posidonia oceanica) meadow sediment. Estuaries and Coasts, 32(2), pp.276–286. 546
Garland, J.L. & Mills, a. L., 1991. Classification and characterization of heterotrophic 547
microbial communities on the basis of patterns of community-level sole-carbon-548
source utilization. Applied and Environmental Microbiology, 57(8), pp.2351–2359. 549
Glazebrook, P.W. et al., 1996. Seasonal Changes in Numbers and the Location of a 550
Particular Bacterial Strain of Alteromonas sp. in Seagrass Sediments EW. Microbial 551
Ecology, 31, pp.1–13. 552
Glew, J., Smol, J. & Last, W., 2001. Sediment core collection and extrusion. In Tracking 553
environmental change using lake sediments. pp. 73–105. Available at: 554
25
http://link.springer.com/chapter/10.1007/0-306-47669-X_5. 555
Gobert, S. et al., 2009. Assessment of the ecological status of Mediterranean French 556
coastal waters as required by the Water Framework Directive using the Posidonia 557
oceanica Rapid Easy Index: PREI. Marine pollution bulletin, 58(11), pp.1727–33. 558
Available at: http://www.ncbi.nlm.nih.gov/pubmed/19700176 [Accessed 559
September 25, 2015]. 560
Holmer, M., Duarte, C.M. & Marbá, N., 2003. Sulfur cycling and seagrass (Posidonia 561
oceanica) status in carbonate sediments. Biogeochemistry, 66(3), pp.223–239. 562
Lo Iocano, C. et al., 2008. Very high-resolution seismo-acoustic imaging of seagrass 563
meadows (Mediterranean Sea): Implications for carbon sink estimates. 564
Geophysical Research Letters, 35(18), pp.1–5. 565
Kaal, J. et al., 2016. Molecular composition of plant parts and sediment organic matter 566
in a Mediterranean seagrass (Posidonia oceanica) mat. Aquatic Botany, 133, 567
pp.50–61. Available at: 568
http://linkinghub.elsevier.com/retrieve/pii/S0304377016300596. 569
Kristensen, E., 2000. Organic matter diagenesis at the oxic/anoxic interfacein coastal 570
marine sediments, with emphasis on the role of burrowing animals. 571
Hydrobiologia, (426), pp.1–24. 572
Kristensen, E., Ahmed, S.I. & Devol, A.H., 1995. Aerobic and anaerobic decomposition 573
of organic matter in marine sediment : Which is fastest ? Limnology and 574
Oceanography, 40(8), pp.1430–1437. 575
Kristensen, E. & Holmer, M., 2001. Decomposition of plant materials in marine 576
sediment exposed to different electron acceptors (O2, NO3 and SO42), with 577
emphasis on substrate origin,degradation kinetics, and the role of bioturbation. 578
26
Geochimica et Cosmochimica Acta, 65(3), pp.419–433. 579
Liao, N. et al., 2016. Effects of biochar on soil microbial community composition and 580
activity in drip-irrigated desert soil. European Journal of Soil Biology, 72, pp.27–581
34. Available at: http://linkinghub.elsevier.com/retrieve/pii/S1164556315300406. 582
Llobet-Brossa, E. et al., 2002. Community structure and activity of sulfate-reducing 583
bacteria in an intertidal surface sediment: A multi-method approach. Aquatic 584
Microbial Ecology, 29(3), pp.211–226. 585
Lopez, N.I. et al., 1995. Bacterial activity in NW Mediterranean seagrass (posodonia 586
oceania) sediments. Journal of Experimental Marine Biology and Ecology, 187, 587
pp.39–49. 588
López, N.I. et al., 1998. The effect of nutrient additions on bacterial activity in seagrass 589
(Posidonia oceanica) sediments. Journal of Experimental Marine Biology and 590
Ecology, 224(2), pp.155–166. 591
Luy, N. et al., 2012. Chemical contamination along the Mediterranean French coast 592
using Posidonia oceanica (L.) Delile above-ground tissues: a multiple trace 593
element study. Ecological Indicators, 18, pp.269–277. Available at: 594
http://dx.doi.org/10.1016/j.ecolind.2011.11.005 [Accessed September 25, 2015]. 595
Lyons, M.M. & Dobbs, F.C., 2012. Differential utilization of carbon substrates by 596
aggregate-associated and water-associated heterotrophic bacterial communities. 597
Hydrobiologia, 686(1), pp.181–193. 598
Marbá, N. et al., 2006. Seagrass Beds and Coastal Biogeochemistry. In Seagrasses: 599
Biology, Ecology and Conservation. pp. 133–155. 600
Mateo, M.A. et al., 2006. Carbon flux in seagrass ecosystems. In Seagrasses: Biology, 601
Ecology and Conservation. pp. 159–192. 602
27
Mateo, M.A. & Romero, J., 1997. Detritus dynamics in the seagrass Posidonia oceanica: 603
Elements for an ecosystem carbon and nutrient budget. Marine Ecology Progress 604
Series, 151(1–3), pp.43–53. 605
Nurulita, Y. et al., 2016. Restoration of tropical peat soils: The application of soil 606
microbiology for monitoring the success of the restoration process. Agriculture, 607
Ecosystems & Environment, 216, pp.293–303. Available at: 608
http://linkinghub.elsevier.com/retrieve/pii/S0167880915300979. 609
Pasqualini, V. et al., 1998. Mapping of Posidonia oceanica using Aerial Photographs and 610
Side Scan Sonar: Application off the Island of Corsica (France). Estuarine, Coastal 611
and Shelf Science, 47, pp.359–367. Available at: 612
http://www.sciencedirect.com/science/article/pii/S0272771498903614. 613
Pedersen, M.Ø. et al., 2011. Temperature effects on decomposition of a Posidonia 614
oceanica mat. Aquatic Microbial Ecology, 65(2), pp.169–182. 615
Pérez Rodríguez, M. & Martínez Cortizas, A., 2014. Preliminary characterization of 616
microbial functional diversity using sole-C-source utilization profiles in Tremoal do 617
Pedrido mire (Galicia, NW Spain). Spanish Journal of Soils Science, 4(2), pp.158–618
178. 619
Pergent-Martini, C. et al., 2005. Descriptors of Posidonia oceanica meadows: Use and 620
application. Ecological Indicators, 5(3), pp.213–230. Available at: 621
http://linkinghub.elsevier.com/retrieve/pii/S1470160X05000178 [Accessed 622
September 25, 2015]. 623
Pierce, M.L., Ward, J.E. & Dobbs, F.C., 2014. False positives in Biolog EcoPlatesTM and 624
MT2 MicroPlatesTM caused by calcium. Journal of microbiological methods, 97, 625
pp.20–4. Available at: http://www.ncbi.nlm.nih.gov/pubmed/24333911 [Accessed 626
28
September 25, 2015]. 627
Richir, J. et al., 2013. Experimental in situ exposure of the seagrass Posidonia oceanica 628
(L.) Delile to 15 trace elements. Aquatic Toxicology, 140–141, pp.157–173. 629
Available at: http://dx.doi.org/10.1016/j.aquatox.2013.05.018. 630
Richir, J. et al., 2012. Use of semi-quantitative kit methods to study the heterotrophic 631
bacterial community of Posidonia oceanica meadows: Limits and possible 632
applications. Estuarine, Coastal and Shelf Science, 109, pp.20–29. Available at: 633
http://dx.doi.org/10.1016/j.ecss.2012.05.013 [Accessed September 25, 2015]. 634
Richir, J. & Gobert, S., 2014. A reassessment of the use of Posidonia oceanica and 635
Mytilus galloprovincialis to biomonitor the coastal pollution of trace elements: 636
New tools and tips. Marine Pollution Bulletin, 89(1–2), pp.390–406. Available at: 637
http://linkinghub.elsevier.com/retrieve/pii/S0025326X14005827. 638
Rodríguez-Valera, F., 2004. Environmental genomics, the big picture? FEMS 639
Microbiology Letters, 231(2), pp.153–158. 640
Rutgers, M. et al., 2015. Monitoring soil bacteria with community-level physiological 641
profiles using BiologTM ECO-plates in the Netherlands and Europe. Applied Soil 642
Ecology, 97, pp.23–35. Available at: 643
http://linkinghub.elsevier.com/retrieve/pii/S0929139315300238 [Accessed 644
September 1, 2015]. 645
Säwström, C. et al., 2016. Utilization of carbon substrates by heterotrophic bacteria 646
through vertical sediment profiles in coastal and estuarine seagrass meadows. 647
Environmental Microbiology Reports, 0. Available at: 648
http://doi.wiley.com/10.1111/1758-2229.12406. 649
Serrano, O. et al., 2012. Characterization of soils beneath a Posidonia oceanica 650
29
meadow. Geoderma, 185–186, pp.26–36. Available at: 651
http://linkinghub.elsevier.com/retrieve/pii/S0016706112001395 [Accessed 652
September 8, 2015]. 653
Serrano, O. et al., 2013. Millennial scale impact on the marine biogeochemical cycle of 654
mercury from early mining on the Iberian Peninsula. Global Biogeochemical 655
Cycles, 27(1), pp.21–30. Available at: 656
http://doi.wiley.com/10.1029/2012GB004296 [Accessed September 25, 2015]. 657
Serrano, O. et al., 2016. Reconstruction of centennial-scale fluxes of chemical elements 658
in the Australian coastal environment using seagrass archives. Science of The Total 659
Environment, 541(September 2015), pp.883–894. Available at: 660
http://linkinghub.elsevier.com/retrieve/pii/S0048969715306781. 661
Smith, A.C. et al., 2004. Seasonal composition and activity of sulfate-reducing 662
prokaryotic communities in seagrass bed sediments. Aquatic Microbial Ecology, 663
37(2), pp.183–195. 664
Sørensen, K.B. et al., 2007. Spatial structure of the microbial community in sandy 665
carbonate sediment. Marine Ecology Progress Series, 346, pp.61–74. 666
Stefanowicz, A.M. et al., 2010. Pine forest and grassland differently influence the 667
response of soil microbial communities to metal contamination. Science of the 668
Total Environment, 408(24), pp.6134–6141. 669
Thamdrup, B., Rossello-Mora, R. & Amann, R., 2000. Microbial Manganese and Sulfate 670
Reduction in Black Sea Shelf Sediments. Applied and Environmental Microbiology, 671
66(7), pp.2888–2897. Available at: 672
http://aem.asm.org/cgi/content/abstract/66/7/2888. 673
Thomas, F. et al., 2016. Isolation and substrate screening of polycyclic aromatic 674
30
hydrocarbon degrading bacteria from soil with long history of contamination. 675
International Biodeterioration & Biodegradation, 107, pp.1–9. Available at: 676
http://linkinghub.elsevier.com/retrieve/pii/S0964830515301268. 677
Tovar-Sánchez, A. et al., 2010. Long-term records of trace metal content of western 678
Mediterranean seagrass (Posidonia oceanica) meadows: Natural and 679
anthropogenic contributions. Journal of Geophysical Research, 115(G2), pp.1–10. 680
Tranchina, L. et al., 2005. Posidonia oceanica as a historical monitor device of lead 681
concentration in marine environment. Environmental science & technology, 39(9), 682
pp.3006–12. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15926545. 683
Vassallo, P. et al., 2013. The value of the seagrass Posidonia oceanica: a natural capital 684
assessment. Marine pollution bulletin, 75(1–2), pp.157–67. Available at: 685
http://www.ncbi.nlm.nih.gov/pubmed/23953894 [Accessed September 25, 2015]. 686
687
688
31
Figures: 689
690
Figure 1: Ecoplates carbon sources. Pol, polymers; Cbh, carbohydrates; Cb A, carboxylic 691
acids; Misc, miscellaneous compounds; Am A, amino acids and Amin, amines. Modified 692
from the technical document of Boilog®. 693
694
32
695
Figure 2: Top: consumption at day eleventh (T15) of each compound with depth in 696
aerobic conditions, average values and H’ and AWCD index. Bottom: consumption at 697
day eleventh (T15) of each compound with depth in anaerobic conditions, average 698
values and H’ and AWCD index. Color scale: white, without significant consumption 699
(lower than 0.5); yellow, low consumption (between 0.5 and 1); light red, high 700
consumption (between 1 and 1.5); dark red, very high consumption (upper 1.5); 701
purple, highly consumed compound at all depths (average higher than 1); green, 702
compound that did not show consumption (average lower than 0.3); A1: blank. 703
704
33
705
Figure 3: Left: examples of AWCD change with time (logarithmic Y axis) at three depths 706
(6-9, 18-22 and 50-53 cm). Black dot, phase change; grey line, logarithmic fitting curve; 707
dots line, moving average. Right: distribution of the slope (carbon consumption rates) 708
with depth (cm). 709
34
710
Figure 4: distribution of the ratio AN/AE of each Ecoplate carbon guild with depth at 711
T15. The values that appear on the “Y” axis mean there was no significant anaerobic 712
consumption; those on the “X” axis mean that no significant aerobic consumption was 713
recorded. The blue line indicates identity for anaerobic and anaerobic AWCD values. 714
715