Available via license: CC BY 4.0
Content may be subject to copyright.
1
Only a minority of bacteria grow after wetting in both
1
natural and post-mining biocrusts in a hyperarid,
2
phosphate mine
3
4
Talia Gabay1,2, Eva Petrova3, Osnat Gillor2, Yaron Ziv1, and Roey Angel3*
5
6
1Department of Life Sciences, Ben Gurion University of the Negev, 8410501, Israel
7
2Zuckerberg Institute for Water Research, Blaustein Institutes for Desert Research, Ben-Gurion
8
University of the Negev, 8499000, Israel
9
3Institute of Soil Biology and Biogeochemistry, Biology Centre CAS, Na Sádkách 7, 370 05 České
10
Budějovice, Czech Republic
11
12
Correspondence: Roey Angel (roey.angel@bc.cas.cz ), Talia Gabay (taliajoann@gmail.com)
13
14
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
2
Abstract
15
Biological soil crusts (biocrusts) are key contributors to desert ecosystem functions; therefore,
16
biocrust restoration following mechanical disturbance is imperative. In the Negev Desert
17
hyperarid regions, phosphate mining has been practiced for over 60 years, destroying soil
18
habitats, and fragmenting the landscape. To understand the effects of mining activity on soil
19
health, we previously characterized the biocrust communities in four phosphate mining sites
20
over spatial (post-mining and natural plots) and temporal (2-10 years since restoration) scales.
21
We showed that bacterial abundance, richness, and diversity in natural plots were significantly
22
higher than in post-mining plots, regardless of temporal scale. In this study, we selected one
23
mining site and used DNA stable isotope probing (DNA-SIP) to identify which bacteria grow in
24
post-mining and natural biocrusts. Since biocrust communities activate only after wetting, we
25
incubated the biocrusts with H218O for 96 hours under ambient conditions. We then evaluated
26
the physicochemical soil properties, chlorophyll
a
concentrations, activation, and functional
27
potential of the biocrusts. The DNA-SIP assay revealed low bacterial activity in both plot types
28
and no significant differences in the proliferated communities’ composition when comparing
29
post-mining and natural biocrusts. We further found no significant differences in the microbial
30
functional potential, photosynthetic rates, or soil properties. Our results suggest that growth
31
of hyperarid biocrust bacteria after wetting is minimal. We hypothesize that due to the harsh
32
climatic conditions, during wetting bacteria devote their meager resources to prepare for the
33
coming drought, by focusing on damage repair, and organic compound synthesis and storage
34
rather than on growth. These low growth rates contribute to the sluggish recovery of desert
35
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
3
biocrusts following major disturbances such as mining. Therefore, our findings highlight the
36
need for implementing active restoration practices following mining.
37
38
Keywords
39
Biological soil crust; Biocrust restoration; Stable isotope probing; Hyperarid desert; Mining;
40
Restoration
41
42
43
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
4
1. Introduction
44
Phosphate mining in the Negev Desert, Israel, has been taking place since the 1960s in large
45
areas. ILC-Rotem mining company leads the phosphate mining activities and has been
46
practicing a reclamation-oriented mining protocol for the past 15 years. The mining protocol
47
entails the excavation of the top 50-70 cm of soil (which they consider to be topsoil) followed
48
by the overburden (the layer covering the phosphate), then storing the two soil layers in
49
separate piles. Following the excavation of the phosphate, the overburden is returned to the
50
mining pit followed by the topsoil. Finally, the terrain is leveled with heavy machinery. The
51
area is then considered a restored, post-mining site.
52
53
Open mining activities lead to the destruction of the local vegetation and seed bank, and the
54
fragmentation of the natural landscape (Sengupta, 2021). The consequences include land
55
degradation, erosion, soil and water pollution, and dust dispersion. In addition, mining
56
activity often leads to decreased biodiversity in and around mining sites (Bridge, 2004,
57
Sengupta, 2021). One of the ecosystem components being destroyed by mining activities in
58
the Negev Desert is the biological soil crust layer (biocrust). Biocrust is the topmost layer of
59
many arid soils and comprises primary-producing and heterotrophic microorganisms that
60
bind together soil particles using secreted extracellular polymeric substances (EPS), mainly
61
polysaccharides (Weber et al., 2022). Biocrusts provide many ecosystem services, including
62
fixing nitrogen and carbon, and soil stabilization (Belnap and Lange, 2003). While biocrust
63
microorganisms developed various adaptations to withstand the harsh desert environment
64
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
5
(Makhalanyane et al., 2015), biocrust structures are susceptible to mechanical disturbances.
65
Such a disturbance, especially over large scales (for example, mining activity), breaks and
66
buries biocrust organisms, often resulting in changed biocrust communities (Belnap and
67
Eldridge, 2003).
68
69
In a previous research, we evaluated the biocrust bacterial communities in phosphate
70
mining sites (Gabay et al., 2022). Briefly, we found that natural and post-mining biocrusts
71
differ in community composition and diversity. Following the biocrust community analysis,
72
we sought to identify which bacterial groups are actively growing in the biocrust and
73
whether the composition differs between natural and post-mining sites. To this end, we used
74
DNA-stable isotope probing (DNA-SIP): a culture-free approach that allows the detection of
75
actively growing microorganisms by labeling them with stable isotopes such as 15N, 14C, and
76
18O (Dumont and Hernández García, 2019). SIP has been widely applied in identifying
77
microbial groups that participate in carbon and nitrogen cycling, such as methanotrophs
78
(Sultana et al., 2019, Zhang et al., 2020), methylotrophs (Macey et al., 2020, Arslan et al.,
79
2022), and nitrogen fixers (Pepe-Ranney et al., 2016, Angel et al., 2018). Likewise, SIP can use
80
the incorporation of heavy water (H218O) into various biomarkers to study the growth and
81
function of microorganisms that become activated upon wetting (Schwartz et al., 2019).
82
Previous H218O SIP experiments measured microbial growth rates and dynamics following
83
hydration (Blazewicz et al., 2020). Desert biocrusts make an ideal study system for H218 O SIP
84
experiments, as they become active quickly following hydration (Angel and Conrad, 2013),
85
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
6
resuming growth, nutrient cycling, and excretion of extracellular organic materials (Garcia-
86
Pichel and Belnap, 1996, Belnap and Lange, 2003).
87
88
In this research, we investigated the proliferation of bacterial groups in biocrusts taken from
89
reference ('natural') areas and post-mining sites by incubating biocrust samples with
90
isotopically-labeled water (H218O). We hypothesized that growth patterns and taxonomic
91
identity of bacterial groups would differ significantly when comparing natural and post-
92
mining biocrusts. Specifically, we expected higher bacterial growth rates in natural
93
compared to post-mining biocrusts. Based on our previous findings, we specifically expected
94
higher activity of Cyanobacteria in the natural biocrusts (Gabay et al., 2022).
95
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
7
2. Materials and Methods
96
97
2.1. Study site and sample collection
98
Sampling was conducted during June 2020 at the Gov Mining Site, located in the Zin Valley
99
(30.84 °N, 35.09 °E, 98 m above sea level), where restoration was completed in 2007. The study
100
area was previously described in Gabay et al., (2022). Briefly, Zin Valley is a hyperarid region of
101
the Negev Desert, with 50 mm average annual rainfall (Zin factory meteorological data). The
102
soils are highly saline (average EC = 24 dS/m), composed of variable amounts of sand, silt, and
103
clay, and are classified as Solonchaks according to the World Reference Based soil
104
classification system.
105
106
Biocrusts were sampled either from the post-mining site or the adjacent natural area. In each
107
sampling site, we sampled along a 100 m strip at approximately 10 m intervals (Fig. 1). In total,
108
we sampled 20 biocrust samples (10 from each site). We collected the biocrusts using a
109
spatula, at an average depth of 2 mm. Biocrusts were placed in 100 mm x 15 mm petri dishes
110
lined with cotton. For the SIP assay, we chose 5 of the 10 samples from e ach site containing
111
the highest chlorophyll
a
concentrations as estimated in preliminary experiments (Table S1).
112
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
8
113
Figure 1: map of the research area. Map a shows the different post-mining sites around the
114
Zin factory. Map b shows the biocrust sampling points in gov mining site used for this
115
research. Green dots represent the natural biocrusts, and red dots represent the post-mining
116
biocrusts.
117
118
119
2.2. Soil properties
120
Five biocrust samples from each plot type (post-mining and natural) were sent for analysis of
121
soil properties (pH, EC, NO3- concentrations, and soil organic matter). The analysis was
122
performed at the Gilat Soil Laboratory (Gilat Research Center, Gilat, Israel).
123
124
2.3. chlorophyll
a
extraction
125
Chlorophyll
a
was extracted from biocrust samples using a protocol previously described in
126
Gabay et al., (2022). Briefly, chlorophyll
a
was extracted from 3 g soil of each biocrust sample
127
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
9
was diluted in 9 mL of methanol for 15 min at 65 °C. The soil solution was centrifuged at 2000
128
rpm for 5 minutes, supernatant was collected, and absorbance was measured in a
129
spectrophotometer at 665 nm. Concentrations were calculated according to Ritchie (2006)
130
and normalized to 1 g of soil. Extractions of the biocrusts were performed before (dry
131
biocrusts) and after 96 hr incubation with distilled water (DW) under identical conditions to
132
the incubation with H218O.
133
134
2.4. Stable isotope probing
135
2.4.1. Soil incubation
136
To test the incorporation of 18 O into biocrust samples, a microcosm was designed to control
137
for the incubation conditions. Each microcosm consisted of a 10 mL glass vial in which 1 g of
138
biocrust sample was placed. To achieve field water-holding capacity, 0.15 mL of H218O or
139
DNase-free water were added. The glass vials were then sealed with butyl rubber stoppers
140
(Sigma-Aldrich, St. Louis, Missouri, United States) to prevent evaporation. Both labeled and
141
unlabeled controls were incubated in duplicates, for a total of 40 vials. Samples were
142
incubated under a 12 hr photoperiod for 96 hr in an incubator (FOC 225 I, VELP Scientifica,
143
Usmate Velate MB, Italy) to allow the incorporation of 18O into the bacterial DNA. Following
144
incubation, the microcosms were sacrificed, and each biocrust sample was divided into 4
145
bead beating tubes (Qiagen, Hilden, Germany), each containing 0.25 g of soil, and stored at -
146
80 °C until further analysis.
147
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
10
Each labeled sample had a non-labeled control, incubated under identical conditions but
148
with DNase-free water instead of 18O water.
149
150
2.4.2. DNA extraction
151
DNA was extracted from all biocrust samples using DNeasy PowerSoil Pro Kit (Qiagen),
152
according to the manufacturer's instructions. The biomass in hyperarid biocrusts tends to be
153
very low, yielding only minute amounts of DNA. Therefore, each 1 g soil was extracted in
154
batches of 0.25 g, and the extracts were later consolidated to increase DNA yield.
155
156
2.4.3. SIP gradient preparation and fractionation
157
DNA (ca. 3.5 ng) was subjected to isopycnic gradient centrifugation in a solution of caesium
158
chloride (7.163 M; CsCl, Sigma Aldrich. St Loise, MI, USA) and buffer (0.1 M Tris-HCl at pH 8.0,
159
0.1 M KCl and 1 mM EDTA, all from Sigma Aldrich) to a final density of 1.725 g mL-1 as described
160
previously (Jia et al., 2019). The tubes were spun for 44 hr at 177,000 g and then fractionated
161
by water displacement using a syringe pump (NE-300 Just Infusion™ Syringe Pump, NewEra
162
Pump systems, Farmingdale, NY, USA). The refractive index was measured using an AR200
163
digital refractometer (Reichert, Depew, NY, USA) and then the DNA was precipitated using a
164
Polyethylene Glycol 6000 solution (30% PEG 8000 and 1.6 M NaCl) , and 30 µg of GlycoBlue
165
Coprecipitant (Thermo Fisher Scientific, Waltham, MS, USA). Copy numbers of the 16S rRNA
166
gene in each fraction were determined by qPCR using a probe-based approach. Primers 338F
167
and 805R (Yu et al., 2005) coupled with a 516P probe (FAM-BHQ1 dual labeled) were used for
168
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
11
the assay. Per one reaction 10 µL of TaqMan™ Fast Advanced Master Mix (Thermo Fisher
169
Scientific), 4 µL of Bovine Saline Albumin (BSA; Thermo Fisher Scientific), 1 µL of each primer
170
(10 µM), 0.4 µL of a probe (10 µM) and 2.2 µL of PCR water was combined and mixed with 5 µL
171
of DNA. After 5 min initial denaturation at 95 °C, cycling program: 40 cycles of 95 °C for 30 sec
172
followed by 62 °C for 1 min was applied. Gene copy numbers were established from a standard
173
curve of
Escherichia coli
16S rRNA gene.
174
175
2.4.4. PCR and sequencing
176
Following fractionation, all samples (labeled and unlabeled) were amplified using the 16S
177
rRNA primers 515F_mod and 806R_mod (Apprill et al. 2015, Parada et al. 2016). Each reaction
178
consisted of 2.5 µL Green Taq Buffer (Thermo Fisher Scientific), 2.5 µL of dNTP set
179
(Biotechrabbit, Berlin, Germany), 0.1 µL of BSA (Thermo Fisher Scientific), 0.625 µL of each
180
primer (10 µM), 0.125 µL DreamTaq Green DNA Polymerase (Thermo Fisher Scientific) and 17.5
181
µL of PCR water (Sigma). The PCR ran for 38 cycles using the following program: denaturation
182
at 94 °C for 45 sec, annealing at 52 °C for 45 sec, extension at 72 °C for 45 sec, and a final cycle
183
of extension at 72 °C for 10 min. The amplified fragments were sequenced using MiniSeq
184
(Illumina, San Diego, CA, USA) at the UIC sequencing core, University of Illinois, Chicago, Illinois
185
(https://rrc.uic.edu/cores/genome-research/genome-research-core/). DNA extraction and SIP
186
gradient controls, PCR negative controls and mock community (ZymoBIOMICS Microbial
187
Community Standard II Log Distribution; Zymo Research, Irvine, CA, USA) samples (2 of each)
188
were also sequenced to control for contaminants in the sequencing results.
189
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
12
190
2.5. Bioinformatic analysis
191
All the bioinformatic and statistical analyses were done in R V4.1.1 (R development core
192
team, 2013). Labeling of bacteria was detected using differential abundance analysis as
193
described in Angel (2019). Briefly, the sequences were processed using the DADA2 package
194
V8.8 (Callahan et al., 2016) for quality filtering, denoising, read-merging, chimera removal,
195
constructing amplicon sequence variants (ASV) tables, and taxonomic assignment. Detection
196
and removal of potential contaminant sequences were performed using the R package
197
decontam V.1.12.0 (Davis et al., 2017). Prevalence filtering of rare ASVs was done using the
198
Phyloseq package V1.36.0 (McMurdie and Holmes, 2013). ASVs that appeared in less than
199
2.5% of the samples were removed. A maximum-likelihood phylogenetic tree was calculated
200
using IQ-TREE2 V 2.1.1. (Minh et al., 2020). Finally, differential abundance analysis was
201
performed using DESeq2 V1.32.0 (Love et al., 2014) to compare the relative abundance of
202
each ASV in the heavy fractions of labeled DNA to the unlabeled heavy fractions (the negative
203
control samples), which allows identifying the bacterial groups that incorporated the water
204
isotope into their DNA. The results were filtered to include only ASVs with a 2-fold log change
205
and a significance value p < 0.1.
206
207
2.6. Predictions of genomic functions
208
Abundances of functional genes based on 16S rRNA gene abundances were performed using
209
Picrust2 (Douglas et al., 2019). Abundances of functional genes were predicted based on a
210
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
13
filtered ASV table containing only ASVs belonging to proliferated bacteria based on the
211
differential abundance modeling. The resulting output is functional identifications that were
212
annotated using the KEGG database to infer functional gene families. Each gene was then
213
classified into 10 function categories based on Meier et al. (2021). The abundance of genes
214
within each category was averaged.
215
216
2.7. Statistical analyses
217
chlorophyll
a
concentrations were visualized as an estimation plot using the dabestr
218
package V0.3.0 (Ho and Tumkaya, 2018). The effect size was calculated as a bootstrap 95%
219
confidence interval. Abundances of functional genes and soil properties were compared
220
between natural and post-mining biocrusts using Mann-Whitney tests. The community
221
composition of natural and post-mining biocrusts was assessed using only sequences
222
belonging to proliferated bacteria based on DESeq2 modeling. The weighted UniFrac
223
(Lozupone et al., 2011) was used to calculate the similarity between the natural and post-
224
mining communities, and an adonis model was used to assess whether communities differ
225
significantly from each other (package Vegan V2.6-2; Dixon, 2003).
226
227
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
14
3. Results
228
Sample wetting and greening
229
Most biocrust samples (both natural and post-mining) showed greening within 36 to 48 hr
230
into the 96 hr incubation. By the end, most samples displayed varying degrees of greening,
231
indicating cyanobacterial activity. Generally, post-mining biocrust showed less greening than
232
the natural biocrusts (Fig. 2).
233
234
235
236
Figure 2: Incubation setup. Top picture – biocrusts in sealed glass vials in the incubator.
237
Bottom picture – natural (a) and post-mining (b) biocrusts following the 96-hour incubation.
238
239
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
15
Soil properties
240
EC and NO3- were significantly higher in natural biocrusts compared to post-mining biocrusts
241
(EC: t = 2.89, p < 0.05; NO3-: t = 4, p < 0.01; Table 1). Soil organic matter was also significantly
242
higher in the natural biocrusts (t = 3.77, p < 0.01; Table 1). pH was slightly higher in natural
243
biocrusts; however, the differences were not statistically significant (pH: t = 1.41, p = 0.19;
244
Table 1).
245
246
Table 1: soil properties for natural and post-mining biocrusts. The numbers represent the
247
means for each property. Significant differences are marked with an asterisk (* = p < 0.05; ** =
248
p < 0.01).
249
250
Plot type/Soil property
Natural
Post-mining
pH
7.6
7.5
EC
26.22*
9.94
NO3
84.82**
14.75
Soil organic matter
1.2**
0.81
251
252
Chlorophyll
a
253
The estimation plot revealed an effect size estimate at 1.42 (95CI -0.432; 3.03; Fig. 3). In the
254
natural samples, there was no clear clustering according to the soil water content i.e., dry or
255
hydrated (following 96 hr incubation with water). In fact, there was a larger variance between
256
samples collected after incubation (Fig. 3). Hydrated post-mining biocrusts had consistently
257
higher chlorophyll
a
concentrations compared to dry biocrusts. It Is also apparent that the
258
variance between samples was smaller in the post-mining biocrusts (Fig. 3).
259
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
16
260
Figure 3: Estimation plots of chlorophyll a concentrations. Dots represent the biocrust
261
samples, and colors represent either dry or incubated soil.
262
263
Sequencing and differential abundance modeling
264
Sequencing resulted in 47,311 reads per sample on average (Table S2) and 10,275 ASVs
265
(Table S3). Following decontamination and filtering, 86% of the ASVs were removed (Table
266
S3). However, they accounted for only 16% of the total reads. Out of the remaining 1,404
267
ASVs, 1266 in total were labeled and used for the differential abundance modeling (Table S3).
268
Each sequence in the labeled samples was compared to its corresponding negative control,
269
and the Log2-fold change in labeled sequences was evaluated to determine whether an ASV
270
could be considered truly labeled (i.e., belonging to growing bacteria) based on the
271
significance threshold. One of the natural biocrust samples (no. 1; Fig. 4) displayed much
272
higher labeling than the other 4 samples (414 ASVs passed, out of a total of 1,093; Fig. 4).
273
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
17
Excluding sample 1, 38 out of 975 ASVs total passed the significance threshold for Log2 fold
274
change. In post-mining samples, the number of labeled reads was more consistent among
275
the different samples (Fig. 4). 68 ASVs out of 874 ASVs total passed the threshold for Log2 fold
276
change. Altogether, the number of labeled ASVs did not differ significantly between natural
277
and post-mining samples (natural sample 1 was excluded, natural community mean = 9.5,
278
post-mining community mean = 13.6, W= 9, p = 0.9).
279
280
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
18
281
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
19
Figure 4: community composition of proliferated bacteria in natural (a) and post-mining (b)
282
biocrusts. Each graph represents a different sample. Red dots indicate labeled ASVs, and
283
grey dots indicate unlabeled ASVs, based on Deseq2 modeling.
284
285
Composition of the proliferated bacterial community
286
Figure 5 depicts PCoA ordination based on weighted UniFrac metric showing that the
287
biocrust samples do not cluster according to plot type (natural sample 1 was excluded).
288
Furthermore, the adonis test revealed no significant differences in community composition
289
(Weighted UniFrac ~ Plot type; F = 1.23, R2 = 0.15, p = 0.21). Phylogenetic trees depict the
290
different bacterial groups indicating that, for the most part, sequences that appear in natural
291
and post-mining biocrusts belong to the same orders/classes. In the phylum Cyanobacteria,
292
labeled sequences belonged to two classes, and most sequences in both natural and post-
293
mining samples belonged to the class Cyanobacteria, with a slightly higher prevalence in the
294
post-mining samples (Fig. S1). The class Bacteroidia, belonging to the phylum Bacteroidota,
295
had a similar prevalence for natural and post-mining samples (Fig. S1). The trend was similar
296
in the class Bacilli, belonging to the phylum Firmicutes (Fig. S1). In the Alphaproteobacteria
297
phylum, the orders Rhodobacteriales, Rhizobiales and Sphingomonadales appeared in both
298
natural and post-mining samples (Fig. S1). The phylum Gammaproteobacteria appeared
299
only once in post-mining samples but was more prevalent in natural samples (Fig. S1). The
300
phylum Actinobacteria was more prevalent in post-mining samples, yet the orders
301
Frankiales, Micrococcales and Propionibacteriales appeared in both natural and post-mining
302
samples (Fig. S1). A Venn diagram of unique and overlapping sequences reveals that only 8
303
out of 88 labeled sequences appear both in natural and post-mining samples (Fig. S2).
304
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
20
305
Figure 5: PCoA ordination of community composition based on weighted UniFrac similarity
306
metric. Blue dots are natural samples and pink dots are post-mining samples. The ellipses
307
represent 95% confidence intervals.
308
309
Predictions of genomic functions
310
Abundances of 10 function categories (listed in Table S4) were compared between natural
311
and post-mining biocrust samples. Abundances were generally higher in post-mining
312
compared to natural biocrusts (Table S4). Also, the variance between samples was larger in
313
post-mining biocrust (Fig. 6). However, the differences between plot types were not
314
statistically significant in any of the function categories (Fig. 6, Table S4).
315
316
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
21
317
Figure 6: boxplot of functional predictions. The Y axis represents functional gene
318
abundances. The line represents the median, and the whiskers represent the range.
319
320
321
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
22
Discussion
322
In this study, we examined which groups of the biocrust bacterial communities grow after
323
hydration using a SIP assay and differential abundance modelling. We hydrated and
324
incubated the biocrusts for 96 hr expecting bacterial growth, yet very little growth was
325
detected. Only 3.9% of the natural and 7.7% of the post-mining biocrusts’ ASVs were
326
identified as truly labeled by the stable isotope. Post-mining biocrusts had a slightly higher
327
number of labeled ASVs compared to natural biocrusts but, the differences were not
328
significant. Also, the composition and taxonomic identity of the growing communities did
329
not significantly differ between natural and post-mining biocrusts.
330
331
Biocrust organisms are known to resume activity quickly following hydration, resuming
332
functions such as damage repair, germination, nutrient cycling, and growth (Harel et al.,
333
2004, Rajeev et al., 2013, Green and Proctor, 2016, Thomas et al., 2022). Hydration may also
334
lead to changes in biocrust bacterial communities (Angel and Conrad, 2013; Štovícek and
335
Gillor, 2022). In a H218O SIP assay using the Negev Desert biocrusts from arid and hyperarid
336
regions, samples were hydrated and incubated for three weeks at maximum water holding
337
capacity. Within days, changes in the labeled bacterial community composition and
338
abundance were observed, indicating growth (Angel and Conrad, 2013). Similarly, biocrusts
339
collected in the Negev Desert during a rain event and subsequent desiccation, demonstrated
340
an increase in Cyanobacteria and decrease in Actinobacteria abundance (Baubin et al.,
341
2021), implying selective proliferation of bacterial taxa in the hydrated biocrust.
342
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
23
In other H218O SIP assays on soil bacterial communities, a quick response to re-wetting was
343
observed, and bacterial growth was evident within 24 to 72 hours of incubation (Blazewicz et
344
al., 2014, Aanderud et al., 2015). Thus, we assumed that hydration and incubation of
345
hyperarid biocrusts under favorable conditions would result in growth. We expected distinct
346
proliferated communities when comparing natural and post-mining biocrusts, given our
347
previous results that demonstrated that the bacterial community differed between natural
348
and post-mining biocrusts (Gabay et al., 2022).
349
350
Our previous survey of biocrust bacterial communities in the Zin mining area revealed
351
significantly lower abundances of cyanobacteria and chlorophyll
a
concentrations (Gabay et
352
al., 2022). Out of the four mining sites surveyed, Gov (which was restored in 2007) showed the
353
most considerable shift in biocrust community following mining. However, in the current
354
study, we sampled post-mining biocrusts at a different location (~500 m away from the
355
original plot) due to technical constraints. In the new location we found that the
356
photosynthetic potential in the post-mining biocrust did not differ from the natural biocrust.
357
These results highlight the importance of microenvironments in shaping the functionality of
358
biocrusts (Garcia-Pichel and Belnap, 1996). The similarities in active communities and
359
photosynthetic potential could be due to more developed biocrusts in the new sampling
360
location compared to the previous ones.
361
362
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
24
Photosynthetic activity is usually observed in biocrusts within minutes to hours after
363
hydration by either dew or rain (Harel et al., 2004; Lange, 2003). In our experiment, we
364
hydrated the biocrusts to capacity and then incubated the samples for 96 hr. During the
365
incubation, most biocrust samples displayed some degree of greening, with more greening
366
in the natural biocrusts (Fig. 2). This indicates that the photosynthetic bacteria in the
367
biocrust were activated upon wetting. Yet, no significant differences were detected between
368
natural and post-mining biocrusts, chlorophyll
a
concentrations (Fig. 3) or functional
369
predictions (Fig. 6). This implies that similar abundances of photosynthetic bacteria were
370
activated upon wetting in both biocrusts, yet, they barely proliferated (Fig. 4).
371
372
The growth patterns of biocrust organisms are affected by local environmental conditions
373
(Kim and Or, 2017). Zin mining fields are in an hyperarid region, where extreme heat events
374
are frequent in the summer, and rains are scarce and unpredicted. Moreover, in recent years
375
there were only two or three rain events during each rainy season (Zin factory meteorological
376
data). Hydration is the most important factor affecting biocrust organisms’ growth rate while
377
long desiccation periods negatively affect growth (Zaady et al., 2016). Also, salinity levels in
378
Zin valley soils are high (Table 1; Levi et al., 2021). We suggest that due to these conditions,
379
the hyperarid biocrust communities prioritize activation and preparation for desiccation
380
over growth. It is known that in high stress environments, biocrust microorganisms resume
381
carbon and nitrogen fixation upon hydration. The resulting organic carbon and nitrogen
382
compounds can be metabolized during the long desiccation periods (Belnap, 2003; Colesie
383
et al., 2014).
384
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
25
Natural recovery of biocrusts has been long debated, and is generally estimated to be a slow
385
process, especially in sites that experience very short activity times for biocrust
386
development, such as the hyperarid Zin mining site (Kidron et al., 2020; Weber et al., 2016).
387
The time and trajectory of recovery depend on many factors relating to local climatic
388
conditions and site properties (Belnap and Lange, 2003). One such factor that greatly affects
389
establishment and restoration of biocrusts is the proximity, availability, and dispersal timing
390
of biocrust propagules (Bowker, 2007, Walker et al., 2007). Thus, the low proliferation rates
391
we observed, particularly in post-mining biocrusts, suggest that restoration processes might
392
be much slower than previously estimated. The topsoil from a stockpile is used to cover the
393
mining pits. This soil may not contain a rich biocrust seed bank that was probably destroyed
394
and buried during the mining processes. Further increase in bacterial biomass might highly
395
depend on the dispersal of biocrust propagules to the site from adjacent natural areas by
396
wind or water. Our results further emphasize the need for active restoration measures in Zin
397
mines. Such measures include soil inoculation with local cyanobacterial propagules (Acea,
398
2003, Wang et al., 2009, Zhao et al., 2016, Velasco Ayuso et al., 2017) and increased hydration
399
(Morillas & Gallardo, 2015, Zhang et al., 2018), which are effective methods in enhancing
400
biocrust establishment and recovery following disturbances (Antoninka et al., 2020).
401
402
403
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
26
Conclusions
404
Low proliferation of biocrust bacteria was detected after wetting suggesting prolonged
405
recovery times of biocrusts following major mechanical disturbance, such as mining.
406
Furthermore, recovery largely depends on site conditions and the ability of biocrust
407
propagules to disperse to post-mining sites. Further research is needed to confirm our
408
hypothesis of low proliferation and thus restoration rates in hyperarid biocrust bacterial
409
communities.
410
411
Code and data availability
412
All data produced in this study and scripts used for community analysis, functional
413
predictions and chlorophyll
a
estimation plot are available at
414
https://github.com/TaliaJoanne/SIP-experiment-Zin-mines.
415
The raw 16S sequences are available in the NCBI database under Bioproject ID
416
PRJNA906925, accession numbers SAMN31937891 – SAMN31937900.
417
418
Author’s contributions
419
Talia Gabay: Conception, Sample Collection, Incubation, Chlorophyll
a
and DNA extraction,
420
Statistical Analysis, Visualization, Writing-Original Draft Preparation, Writing-Reviewing and
421
Editing. Eva Petrova: DNA-SIP assay, DNA quantification and PCR amplification. Osnat Gillor
422
and Yaron Ziv: Conception, Writing-Reviewing and Editing, Investigation, Supervision. Roey
423
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
27
Angel: Conception, Statistical Analysis, Visualization, Supervision, Writing -Reviewing and
424
Editing.
425
All authors read and approved the manuscript.
426
427
Declaration of competing interest
428
The authors declare that they have no known competing financial interests or personal
429
relationships that could have appeared to influence the work reported in this paper.
430
431
Acknowledgements
432
We thank Matan Avital from ICL for coordinating sample collection and providing Zin factory
433
maps and meteorological data, Sharon Moscovitz for her assistance in sample collection and
434
Ofer Ovadia for suggestions on statistical analyses. Lastly, we thank ICL Rotem LTD for their
435
support and funding of this research.
436
437
Financial support
438
Funding for this research was provided by Rotem ICL LTD. RA was supported by the Czech
439
Science Foundation (Junior Grant No. 19-24309Y), EP was supported by the Czech Ministry of
440
Education Youth and Sport (EF16_013/0001782 - SoWa Ecosystems Research).
441
442
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
28
References
443
Aanderud, Z. T., Jones, S. E., Fierer, N., and Lennon, J. T.: Resuscitation of the rare biosphere contributes to
444
pulses of ecosystem activity, Front. Microbiol., 6, https://doi.org/10.3389/fmicb.2015.00024, 2015.
445
Angel, R.: Stable Isotope Probing Techniques and Methodological Considerations Using \textsuperscript15N, in:
446
Stable Isotope Probing: Methods and Protocols, edited by: Dumont, M. G. and Hernández García, M.,
447
Springer New York, New York, NY, 175–187, https://doi.org/10.1007/978-1-4939-9721-3_14, 2019.
448
Angel, R. and Conrad, R.: Elucidating the microbial resuscitation cascade in biological soil crusts following a
449
simulated rain event: Microbial resuscitation in biological soil crusts, Environ Microbiol, n/a-n/a,
450
https://doi.org/10.1111/1462-2920.12140, 2013.
451
Angel, R., Panhölzl, C., Gabriel, R., Herbold, C., Wanek, W., Richter, A., Eichorst, S. A., and Woebken, D.: Application
452
of stable-isotope labelling techniques for the detection of active diazotrophs: Detecting diazotrophs with
453
stable-isotope techniques, Environ Microbiol, 20, 44–61, https://doi.org/10.1111/1462-2920.13954, 2018.
454
Anon: Dabestr: Data Analysis Using Bootstrap-Coupled Estimation, 2020.
455
Antoninka, A., Faist, A., Rodriguez‐Caballero, E., Young, K. E., Chaudhary, V. B., Condon, L. A., and Pyke, D. A.:
456
Biological soil crusts in ecological restoration: emerging research and perspectives, Restor Ecol, 28,
457
https://doi.org/10.1111/rec.13201, 2020.
458
Arslan, M., Müller, J. A., and Gamal El-Din, M.: Aerobic naphthenic acid-degrading bacteria in petroleum-coke
459
improve oil sands process water remediation in biofilters: DNA-stable isotope probing reveals
460
methylotrophy in Schmutzdecke, Science of The Total Environment, 815, 151961,
461
https://doi.org/10.1016/j.scitotenv.2021.151961, 2022.
462
Baubin, C., Ran, N., Siebner, H., and Gillor, O.: The response of desert biocrust bacterial communities to
463
hydration-desiccation cycles, SOIL Discussions, 1–48, 2021.
464
Belnap, J.: The world at your feet: desert biological soil crusts, Frontiers in Ecology and the Environment, 1, 181–
465
189, https://doi.org/10.1890/1540-9295(2003)001[0181:TWAYFD]2.0.CO;2, 2003.
466
Belnap, J. and Eldridge, D.: Disturbance and Recovery of Biological Soil Crusts, in: Biological Soil Crusts:
467
Structure, Function, and Management, edited by: Belnap, J. and Lange, O. L., Springer Berlin Heidelberg,
468
Berlin, Heidelberg, 363–383, https://doi.org/10.1007/978-3-642-56475-8_27, 2003.
469
Belnap, J. and Lange, O. L. (Eds.)Baldwin, I. T., Caldwell, M. M., Heldmaier, G., Lange, O. L., Mooney, H. A., Schulze,
470
E.-D., and Sommer, U.: Biological Soil Crusts: Structure, Function, and Management, Springer Berlin
471
Heidelberg, Berlin, Heidelberg, https://doi.org/10.1007/978-3-642-56475-8, 2003.
472
Blazewicz, S. J., Schwartz, E., and Firestone, M. K.: Growth and death of bacteria and fungi underlie rainfall-
473
induced carbon dioxide pulses from seasonally dried soil, Ecology, 95, 1162–1172,
474
https://doi.org/10.1890/13-1031.1, 2014.
475
Blazewicz, S. J., Hungate, B. A., Koch, B. J., Nuccio, E. E., Morrissey, E., Brodie, E. L., Schwartz, E., Pett-Ridge, J.,
476
and Firestone, M. K.: Taxon-specific microbial growth and mortality patterns reveal distinct temporal
477
population responses to rewetting in a California grassland soil, ISME J, 14, 1520–1532,
478
https://doi.org/10.1038/s41396-020-0617-3, 2020.
479
Bridge, G.: CONTESTED TERRAIN: Mining and the Environment, Annu. Rev. Environ. Resour., 29, 205–259,
480
https://doi.org/10.1146/annurev.energy.28.011503.163434, 2004.
481
Callahan, B. J., McMurdie, P. J., Rosen, M. J., Han, A. W., Johnson, A. J. A., and Holmes, S. P.: DADA2: High-
482
resolution sample inference from Illumina amplicon data, Nat Methods, 13, 581–583,
483
https://doi.org/10.1038/nmeth.3869, 2016.
484
Colesie, C., Allan Green, T. G., Haferkamp, I., and Büdel, B.: Habitat stress initiates changes in composition, CO2
485
gas exchange and C-allocation as life traits in biological soil crusts, ISME J, 8, 2104–2115,
486
https://doi.org/10.1038/ismej.2014.47, 2014.
487
Davis, N. M., Proctor, D., Holmes, S. P., Relman, D. A., and Callahan, B. J.: Simple statistical identification and
488
removal of contaminant sequences in marker-gene and metagenomics data, bioRxiv, 221499,
489
https://doi.org/10.1101/221499, 2017.
490
Dixon, P.: VEGAN, a package of R functions for community ecology, Journal of Vegetation Science, 14, 927–930,
491
https://doi.org/10.1111/j.1654-1103.2003.tb02228.x, 2003.
492
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
29
Douglas, G. M., Maffei, V. J., Zaneveld, J., Yurgel, S. N., Brown, J. R., Taylor, C. M., Huttenhower, C., and Langille, M.
493
G. I.: PICRUSt2: An improved and customizable approach for metagenome inference, Bioinformatics,
494
https://doi.org/10.1101/672295, 2019.
495
Dumont, M. G. and Hernández García, M. (Eds.): Stable Isotope Probing: Methods and Protocols, Springer New
496
York, New York, NY, https://doi.org/10.1007/978-1-4939-9721-3, 2019.
497
Gabay, T., Rotem, G., Gillor, O., and Ziv, Y.: Understanding changes in biocrust communities following phosphate
498
mining in the Negev Desert, Environmental Research, 207, 112200,
499
https://doi.org/10.1016/j.envres.2021.112200, 2022.
500
Garcia-Pichel, F. and Belnap, J.: Microenvironments and Microscale Productivity of Cyanobacterial Desert Crusts,
501
J Phycol, 32, 774–782, https://doi.org/10.1111/j.0022-3646.1996.00774.x, 1996.
502
Green, T. G. A. and Proctor, M. C. F.: Physiology of Photosynthetic Organisms Within Biological Soil Crusts: Their
503
Adaptation, Flexibility, and Plasticity, in: Biological Soil Crusts: An Organizing Principle in Drylands, edited
504
by: Weber, B., Büdel, B., and Belnap, J., Springer International Publishing, Cham, 347–381,
505
https://doi.org/10.1007/978-3-319-30214-0_18, 2016.
506
Harel, Y., Ohad, I., and Kaplan, A.: Activation of Photosynthesis and Resistance to Photoinhibition in
507
Cyanobacteria within Biological Desert Crust, Plant Physiology, 136, 3070–3079,
508
https://doi.org/10.1104/pp.104.047712, 2004.
509
Jia, Z., Cao, W., and Hernández García, M.: DNA-based stable isotope probing, in: Stable Isotope Probing,
510
Springer, 17–29, 2019.
511
Kidron, G. J., Xiao, B., and Benenson, I.: Data variability or paradigm shift? Slow versus fast recovery of biological
512
soil crusts-a review, Science of The Total Environment, 721, 137683,
513
https://doi.org/10.1016/j.scitotenv.2020.137683, 2020.
514
Kim, M. and Or, D.: Hydration status and diurnal trophic interactions shape microbial community function in
515
desert biocrusts, Biogeosciences, 14, 5403–5424, https://doi.org/10.5194/bg-14-5403-2017, 2017.
516
Lange, O. L.: Photosynthesis of Soil-Crust Biota as Dependent on Environmental Factors, in: Biological Soil
517
Crusts: Structure, Function, and Management, edited by: Belnap, J. and Lange, O. L., Springer Berlin
518
Heidelberg, Berlin, Heidelberg, 217–240, https://doi.org/10.1007/978-3-642-56475-8_18, 2003.
519
Levi, N., Hillel, N., Zaady, E., Rotem, G., Ziv, Y., Karnieli, A., and Paz-Kagan, T.: Soil quality index for assessing
520
phosphate mining restoration in a hyper-arid environment, Ecological Indicators, 125, 107571,
521
https://doi.org/10.1016/j.ecolind.2021.107571, 2021.
522
Love, M. I., Huber, W., and Anders, S.: Moderated estimation of fold change and dispersion for RNA-seq data with
523
DESeq2, Genome Biol, 15, 550, https://doi.org/10.1186/s13059-014-0550-8, 2014.
524
Lozupone, C., Lladser, M. E., Knights, D., Stombaugh, J., and Knight, R.: UniFrac: an effective distance metric for
525
microbial community comparison, The ISME journal, 5, 169–172, 2011.
526
Macey, M. C., Pratscher, J., Crombie, A. T., and Murrell, J. C.: Impact of plants on the diversity and activity of
527
methylotrophs in soil, Microbiome, 8, 31, https://doi.org/10.1186/s40168-020-00801-4, 2020.
528
Makhalanyane, T. P., Valverde, A., Gunnigle, E., Frossard, A., Ramond, J.-B., and Cowan, D. A.: Microbial ecology of
529
hot desert edaphic systems, FEMS Microbiology Reviews, 39, 203–221,
530
https://doi.org/10.1093/femsre/fuu011, 2015.
531
McMurdie, P. J. and Holmes, S.: phyloseq: An R Package for Reproducible Interactive Analysis and Graphics of
532
Microbiome Census Data, PLOS ONE, 8, e61217, https://doi.org/10.1371/journal.pone.0061217, 2013.
533
Meier, D. V., Imminger, S., Gillor, O., and Woebken, D.: Distribution of Mixotrophy and Desiccation Survival
534
Mechanisms across Microbial Genomes in an Arid Biological Soil Crust Community, mSystems, 6, e00786-
535
20, https://doi.org/10.1128/mSystems.00786-20, 2021.
536
Minh, B. Q., Schmidt, H. A., Chernomor, O., Schrempf, D., Woodhams, M. D., von Haeseler, A., and Lanfear, R.: IQ-
537
TREE 2: New Models and Efficient Methods for Phylogenetic Inference in the Genomic Era, Molecular
538
Biology and Evolution, 37, 1530–1534, https://doi.org/10.1093/molbev/msaa015, 2020.
539
Pepe-Ranney, C., Koechli, C., Potrafka, R., Andam, C., Eggleston, E., Garcia-Pichel, F., and Buckley, D. H.: Non-
540
cyanobacterial diazotrophs mediate dinitrogen fixation in biological soil crusts during early crust
541
formation, ISME J, 10, 287–298, https://doi.org/10.1038/ismej.2015.106, 2016.
542
Rajeev, L., da Rocha, U. N., Klitgord, N., Luning, E. G., Fortney, J., Axen, S. D., Shih, P. M., Bouskill, N. J., Bowen, B.
543
P., Kerfeld, C. A., Garcia-Pichel, F., Brodie, E. L., Northen, T. R., and Mukhopadhyay, A.: Dynamic
544
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
30
cyanobacterial response to hydration and dehydration in a desert biological soil crust, ISME J, 7, 2178–
545
2191, https://doi.org/10.1038/ismej.2013.83, 2013.
546
Schwartz, E., Hayer, M., Hungate, B. A., and Mau, R. L.: Stable Isotope Probing of Microorganisms in
547
Environmental Samples with H 2 18 O, in: Stable Isotope Probing, Springer, 129–136, 2019.
548
Sengupta, M.: Environmental impacts of mining: monitoring, restoration, and control, Second edition., CRC
549
Press, Boca Raton, FL ; Abingdon, Oxon, 1 pp., 2021.
550
Štovícek, A. and Gillor, O.: The Response of Soil Microbial Communities to Hydration and Desiccation Cycles in
551
Hot Desert Ecosystems, in: Microbiology of Hot Deserts, edited by: Ramond, J.-B. and Cowan, D. A.,
552
Springer International Publishing, Cham, 319–339, https://doi.org/10.1007/978-3-030-98415-1_11, 2022.
553
Sultana, N., Zhao, J., Zheng, Y., Cai, Y., Faheem, M., Peng, X., Wang, W., and Jia, Z.: Stable isotope probing of
554
active methane oxidizers in rice field soils from cold regions, Biol Fertil Soils, 55, 243–250,
555
https://doi.org/10.1007/s00374-018-01334-7, 2019.
556
Team, R. C.: R: A language and environment for statistical computing, 2013.
557
Thomas, A. D., Elliott, D. R., Hardcastle, D., Strong, C. L., Bullard, J., Webster, R., and Lan, S.: Soil biocrusts affect
558
metabolic response to hydration on dunes in west Queensland, Australia, Geoderma, 405, 115464,
559
https://doi.org/10.1016/j.geoderma.2021.115464, 2022.
560
Weber, B., Büdel, B., and Belnap, J. (Eds.): Biological Soil Crusts: An Organizing Principle in Drylands, Springer
561
International Publishing, Cham, https://doi.org/10.1007/978-3-319-30214-0, 2016.
562
Weber, B., Belnap, J., Büdel, B., Antoninka, A. J., Barger, N. N., Chaudhary, V. B., Darrouzet‐Nardi, A., Eldridge, D.
563
J., Faist, A. M., Ferrenberg, S., Havrilla, C. A., Huber‐Sannwald, E., Malam Issa, O., Maestre, F. T., Reed, S. C.,
564
Rodriguez‐Caballero, E., Tucker, C., Young, K. E., Zhang, Y., Zhao, Y., Zhou, X., and Bowker, M. A.: What is a
565
biocrust? A refined, contemporary definition for a broadening research community, Biological Reviews,
566
97, 1768–1785, https://doi.org/10.1111/brv.12862, 2022.
567
Yu, Y., Lee, C., Kim, J., and Hwang, S.: Group-specific primer and probe sets to detect methanogenic communities
568
using quantitative real-time polymerase chain reaction, Biotechnology and bioengineering, 89, 670–679,
569
2005.
570
Zaady, E., Eldridge, D. J., and Bowker, M. A.: Effects of Local-Scale Disturbance on Biocrusts, in: Biological Soil
571
Crusts: An Organizing Principle in Drylands, edited by: Weber, B., Büdel, B., and Belnap, J., Springer
572
International Publishing, Cham, 429–449, https://doi.org/10.1007/978-3-319-30214-0_21, 2016.
573
Zhang, L., Dumont, M. G., Bodelier, P. L. E., Adams, J. M., He, D., and Chu, H.: DNA stable-isotope probing
574
highlights the effects of temperature on functionally active methanotrophs in natural wetlands, Soil
575
Biology and Biochemistry, 149, 107954, https://doi.org/10.1016/j.soilbio.2020.107954, 2020.
576
577
578
https://doi.org/10.5194/egusphere-2022-1510
Preprint. Discussion started: 9 January 2023
c
Author(s) 2023. CC BY 4.0 License.
