Introduction
Snowfall is common in mid-high and
high altitudes around the world. The insulating properties of snow protect soil
microorganisms against freezing (Hinkler et al.
2008), and the thickness of a snowpack influences the
subsurface soil temperature and the metabolic activity of the soil microbial
community (Wang et al. 2013; Freppaz
et al. 2014). It has been predicted
that local snowfall can decrease under global warming conditions (Kunkel
et al. 2009; IPCC 2013), leading to a
thinner snowpack in winter (Kunkel et al.
2009; Kapnick and Delworth 2013). Such a decrease in snow cover can affect the
soil microbial community composition and diversity, triggering changes in
essential ecosystem functions in unknown ways (Campbell et al. 2005; Zhang 2005). Allison et al. (2013) suggested that
characterizing soil microbial mechanisms would be critical for understanding
how ecosystem processes respond to changes in snowpack thickness under global
climate change. Though rarely investigated, during winter such responses would
affect soil carbon and nutrient cycles.
As an important contributor to soil ecological
processes, microbes play an essential role in litter decomposition, carbon and
nitrogen mineralization, and in soil nutrient conversion and circulation
(Garcia et al. 2002). Natural fluctuation of external factors strongly
affects soil microbial activity, but the existing research on these affects is
sparse. During winter, the relatively stable water and temperature conditions
that apply in soil covered by snow provide a suitable environment for growth
and metabolic activity of microbes (Bogoev et
al. 2002; Bombonato and Gerdon 2012). However, the alternate process of
snow formation and melting directly affects soil temperature and moisture (Yang
and Jin 2008; Wang et al. 2015), and
repeated freeze-thawing cycles may alter the microbial community structure and
function, possibly leading to microbial dormancy or decreased viability (Yan et al.
2018). Global warming has become an indisputable
fact, with local changes in snowfall patterns as an inevitable consequence
(Merino et al. 2014). Studies have
shown that the reduction of snow caused by climate warming directly reduces the
local soil temperature, enhances soil freezing, and increases the frequency of
freezing and thawing cycles (Price and Sowers 2004; Groffman et al. 2006; Monson et al. 2006). This causes mechanical disruption of animal and plant
residues and can change microbial communities present in the upper layer of the
soil. Under permissive temperatures nutrients released by cell rupture increase
substrate availability and promote microbial growth, enhancing soil microbial
activity (Wang et al. 2010). Soil
microbial biomass is a repository of nutrients and these provide an important
source of energy and nutrients for plant growth (Edwards et al. 2006). Microbial biomass represents the number of
microorganisms involved in soil organic matter conversion and nutrient cycling
and is sensitive to changes in soil environmental conditions (Drotz et al. 2010). Although the number of
cultivable microorganisms is less than 1% of the presumed total number of
microorganisms present in soil, the cultivable fraction can be studied
directly, providing reliable evidence for responses to soil environmental
changes. Therefore, to provide an in-depth understanding of the effects of
climate warming-related snow reduction on soil microbial biomass and the
cultivable microbial population, it is important to understand the response of
soil ecosystems to climate changes in local settings.
The Sanjiang Plain wetland in northeast China
plays an important role in regional climate regulation, water conservation, and
biodiversity conservation (Lipson et al. 2000).
The annual period during which the soil is frozen typically lasts 5–6 months (Larsen et al. 2007),
providing an excellent test opportunity to study the effect of snow cover.
Cycles of freezing and thawing have been shown to affect soil nutrient
characteristics, litter decomposition, soil biological activity and microbial
diversity in winter (Larsen et al.
2007; Edwards and Jefferies 2013), but changes in snow cover over time may
change the frequency and amplitude of soil freeze-thawing cycles. The soil
microbial community structure and diversity in the area have been
characterized, but changes in winter have so far not drawn much attention.
Therefore, this work concentrated on the Calamagrostis
angustifolia wetland in the Sanjiang Plain and investigated the soil
microbial structure and function under a natural snow cover and, following snow
removal, under artificially applied snow covers with defined thickness. This
provided insights into the ecological processes taking place in wetland soils
during winter and how these might response to changed snow coverage patterns as
a result of climate change.
Field site and experimental design
The study was conducted at the Sanjiang Wetland Experimental Station (47˚35’N, 133˚31’E), property of the Institute of Nature and Ecology of Heilongjiang Academy
of Sciences, China. Fig. 1 shows a map of the area. The local average temperature is 1.9°C and
the monthly temperature ranges from −21.6°C in January to 21.5°C in July. The average annual
precipitation in winter is about 200 mm, with approximately 80% occurring between November and March.
The experiments described here were performed in
January–March 2018. Snow was collected and reapplied with a cover thickness of
0, 20, 50 and 100 cm. Depending on the thickness of the original snow, these
covers were produced by means of reduction, or by repletion as necessary.
Absence of a snow cover (0 cm treatment) was obtained by full clearance of the
plots, and removal was repeated after fresh snowfall. For the other plots,
excess snow was evenly removed to obtain the desired snow coverage levels in
reduction treatments. Likewise, snow was uniformly deposited to the desired
thickness using an 8-mesh sieve in the repletion treatment plots, and its
thickness was regularly managed to correct for novel snowfall and windy weather
to ensure a constant thickness over time. Each treatment was performed with 3
replicate plots, each plot having an area of 5 m × 5 m. In order to prevent
different effects of melt water runoff, each treatment sample was sealed with a
5 mm thick, 50 cm high PVC plate that was buried to a depth of 30 cm to ensure
that any melted snow or rainwater only infiltrated the treated plot directly.
These enclosures had been put in place in the autumn of 2017.
Soil sample collection
At the end of March 2018, when the
snow started to melt, the snow cover was completely removed and the soil was
sampled at three points randomly selected per plot. For each sample point, 10
cm depth of soil was extracted and transported to the laboratory for analysis.
Plant roots and stones were removed, after which the fresh soil was sieved
through a 4 mm mesh and divided into two parts. One part was used for the
determination of soil bacteria, the other was air dried and passed through a 2
mm sieve to determine the soil pH and elemental content of carbon and of total
nitrogen. Total carbon and total nitrogen content were assessed with an
elemental analyzer (Vario Marcro Cube). For determination of the soil pH, 10 g
air dried soil was sieved (2 mm), mixed with 25 mL distilled water at room
temperature, stirred for 2 min, and allowed to stand for 30 min before
measuring the pH with a Sartorius PB-10 pH meter.
Extraction of total DNA and high-throughput analysis
A soil DNA extraction kit (MOBIO Power Soil ® DNA Isolation Kit, U.S.A.)
was used to extract total DNA for characterization of soil microbial genomic
DNA. DNA was quantified and its purity was checked using a nucleic acid
quantifier (NanoDrop ND-1000).
Specific primers with bar codes
(338F, 5'-actcctacgggaggcagca-3' and 806R,5'-ggactachvgggtwtctaat-3') were used
to amplify the V3-V4 region of 16S rDNA (Zhong
et al. 2010). The PCR
amplification was done with the TransGen AP221-02 kit and TransStart Fast-PfuDNA Polymerase (TransGen, China), on
an ABI GeneAmp ® 9700. All samples were amplified 3 times. The 20-μL PCR reaction contained 4 μL of 5×FastPfu Buffer, 2 μL of 2.5 mmol L-1
dNTPs, 0.4 μL of forward and
reverse primer (5 μmol L-1)
each, 2 μL of template (10 ng
DNA), 0.4 μL of Polymerase, and
10.8 μL of ddH2O. PCR
amplification steps were 95°C pre-denaturation for 2 min, then 30 cycles of 30
s at 90°C, 30 s at 50°C and 30 s at 72°, with a final extension time of 10 min
at 72°C. The PCR products of the three sample replicates were combined and
rechecked by 2% agarose gel electrophoresis. According to these results, PCR
products were quantified more precisely with the QuantiFluor™- ST blue
fluorescence quantitative system (Promega). The DNA concentration of the sample
amplicons was adjusted and the DNA was externally sequenced on a HiSeq Illumina
platform (Beijing Biomaker Company).
High-throughput sequencing data analysis
The DNA was pair-end sequenced. First, quality control of the raw data
was conducted, and the sequences were connected with the software Flash, while
unconnected sequences were discarded. Bases below the read value of the tail of
the read were filtered with a window set to 50 bp. When the average read value
within this window was below 20 bp, the base downstream of the window was cut
off, and the reads below 50 bp were filtered after quality control. The paired
reads were merged for which the minimum overlap length had to be 10 bp. The
maximum error ratio allowed in the overlapping region was 0.2, and the
non-conformance sequence was screened. The tag sequence at the end of the
sequence was detected, and the minimum mismatch number was 0. The sequence containing
tag at the beginning was inversely complemented, and the tag was removed. The
barcodes were used without mismatches, the maximum primer mismatches were set
at 2, and the final sequence for analysis was obtained as previously described
(Edgar et al. 2011).
With the application of QIIME
(Quantitative Insights into Microbial Ecology), the sequences were classified
into multiple OTUs (operational taxonomic units) according to similarities between the
sequences. The OTUs contained in each sample and the number of sequences contained in each OTU
was recorded. The uparse OTU (version 7.1; http://drive5.com/uparse/)
method was used for clustering, with the OTU sequence similarity setat 97% to
obtain representative OUT sequences (Edgar et al. 2011). Using uchime (version 4.2.40;
http://drive5.com/usearch/manual/uchime_algo.html), chimeric sequences were detected
and removed (Quast et al. 2013). Using the usearch_global
method, the map of the optimized sequence was compared back to the OTU
representative sequence, and an abundance statistics table of each sample
sequence of an OTU was generated.
To obtain each OTU corresponding
species classification information, as an RDP classifier the Bayesian algorithm
was use data 97% similarity level for at all levels (phylum, class, genus). For
the statistical community composition of each sample, we specified the dominant
population at the door, and steel level relative abundance had to be greater
than 10%, or for the genus level it had to be greater than 1%. The Silva
database (Release115 http://www.arb-silva.de)was used for comparison (Chao 1984). The OTUs with a similarity at or above 97% were selected to generate
the expected dilution curve, and mothur was used to calculate the abundance
indices Chao1 and ACE (for bacterial community richness) and the diversity
indices Shannon and Simpson (for community species richness). The Shannon index
reflects the degree of diversity of a sample while the Simpson index reflects
the dominance of species.
Results
After keeping snow cover levels of
isolated plots constant at various levels for 3 months (4 treatments with
different snow cover levels and 3 replicates for each treatment), the snow was
removed and from all 12 plots and soil samples were collected and analyzed. The
obtained physicochemical properties are summarized in Table 1. As expected, the
soil temperature was lower in absence of a snow cover. The thickness of the
snow cover correlated negatively with acidity (least acidic pH was found with
the thickest snow). Soil carbon content only significantly differed for the
uncovered soil. More variation was observed for the nitrogen content of the
soil, which was significantly higher in the uncovered soil, lowest for soil
covered by 10 and 100 cm of snow, and medium for the sample covered by 20 cm.
Sequencing results of soil samples
and validation of sampling depth
By sequencing amplicons of the bacterial 16S rDNA V3–V4 region, after
filtering out low quality sequences, a total of 210,383 valid sequences and
183,232 reads were obtained. The three replicate treatments were combined here,
as were the amplicon triplicates. The reads were clustered at 97% similarity to
classify the corresponding species into OTUs. A total of 25,843 different OTUs
were identified. As shown in Table 2, the highest number of reads were obtained
from uncovered soil, but the number of predicted OTUs varied only slightly from
1886 (0 cm) to 1896 (100 cm). This represented an increase of 0.9% of number of
OTUs, and this increase seemed to depend on snow thickness, as the numbers for
the 20 and 50
cm samples suggest.
Analysis of soil bacterial community richness and alpha diversity
Table 1: Physicochemical properties of the
soil after treatment (averages of triplicate samples)
Snow cover (cm) |
Soil temperature (ᵒC) |
pH |
Soil organic carbon (g kg-1) |
Soil nitrogen (g kg-1) |
0 |
-9.9 ± 0.02a |
5.62 ± 0.01b |
44.17 ± 2.52a |
4.62 ± 0.21a |
20 |
-5.4 ± 0.02b |
5.75±0.08ab |
42.64 ± 2.80b |
4.18 ± 0.18c |
50 |
-4.4 ± 0.01c |
5.83±0.03ab |
42.15 ± 1.58b |
4.42 ± 0.28b |
100 |
-4.1 ± 0.04c |
5.89 ± 0.07a |
42.37 ± 1.82b |
4.18 ± 0.23c |
Different letters in a column represent statistical significance (P < 0.05) as calculated by Least-significant difference (LSD), one-way ANOVA
Table 2: Number of sequence reads and OUT
sof soil bacterial sequences after the different treatments
Number of readsa |
Number of OTUsa |
|
53,967 |
1880 |
|
20 |
52,018 |
1886 |
50 |
51,889 |
1892 |
100 |
52,640 |
1896 |
aNo significant differences were
observed (P > 0.05)
Fig.
1: Map
of the research site in the Sanjiang Plain, China
Analysis of soil bacterial groups
At the phylum level, bacteria were distributed
into 15 known phyla, in addition to 11 candidate phyla and one unclassified
group. Members of the following phyla were detected (at >1%., in order of
decreasing abundance): Proteobacteria, Acidobacteria, Actinobacteria,
Chloroflexi, Bacteroidetes, Verrucomicrobia,
Gemmatimonadetes, Planctomycetes, Nitrospirae, Firmicutes, Candidatus,
Saccharibacteria, Elusimicrobia, Cyanobacteria, Ingavibacteriae, and Candidatus
Parcubacteria. The relative abundance of these 14 bacterial phyla is shown in
Fig. 2. The sum of the relative abundances of all detected phyla accounted for
more than 95% of the total amount of OTUs identified, in all the twelve soil
samples.
The most abundant bacterial
phyla in the four treatments were Proteobacteria, Acidobacteria, Actinobacteria, Chloroflexi and
Bacteroidetes, which in combination accounted for 85% in all treatments (Fig.
2). However, the relative abundance of these phyla varied with treatments,
although there was no common trend visible related to of snow-cover thickness.
The largest differences in relative phylum abundance were observed between
plots without snow cover versus those with a snow cover (P < 0.05). For instance, compared to uncovered soil, a snow
layer of 20 cm increased the relative abundance of Proteobacteria by 8% though
their abundance decreased again when the snow cover increased from 50 to 100
cm. Conversely, a decrease in abundance of Chloroflexi was observed between 0
cm and 20 cm, but the 100 cm sample contained more Chloriflexi than the 0 cm
sample did.
At the class level, a total of 41 known classes
of bacteria were obtained, in addition to unclassified classes (which accounted
for approximately 25% of the OTUs) and 40 candidate classes (lumped as ‘others’
in Fig. 3). At an abundance >5%, we identified (in decreasing order for
uncovered soil): Alphaproteobacteria, Betaproteobacteria, Deltaproteobacteria,
Solibacteres,
Actinobacteria, Thermoleophilia, Acidobacteria, Anaerolineae, Sphingobacteriia,
Holophagae, and Acidimicrobiia which in combination accounted for 63% (100 cm
samples) to 70% (20 cm samples) of the total amount of soil bacterial classes.
Table 3: Richness and diversity indices
of the soil bacterial communities in the different treatments
Snow cover (cm) |
ACE index |
Chao1 index |
Shannon index |
Simpson index |
0 |
1859.03 ± 16.00a |
1879.29 ± 15.64a |
6.23 ± 0.06b |
0.0053 ± 0.0005a |
20 |
1865.66 ± 13.28a |
1875.45 ± 10.41a |
6.31 ± 0.08ab |
0.0041 ± 0.0003a |
50 |
1869.05 ± 22.76a |
1880.43 ± 24.47a |
6.42 ± 0.02ab |
0.0039 ± 0.0004a |
100 |
1889.89 ± 13.75a |
1900.96 ± 10.71a |
6.45 ± 0.04a |
0.0046 ± 0.0005a |
Different letters in a column represent statistical significance (P < 0.05) as calculated by Least-significant difference
(LSD), one-way ANOVA
The most abundant bacterial classes were Alphaproteobacteria, Betaproteobacteria,
Gammaproteobacteria,
Thermolephilia
and Solibacteres, for all treatments, but with
variable proportions. Notably, the sum of these four was highest for the 20 cm
treatment, while uncovered soil contained relatively more members of the
Thermoleophilia class. Compared with the 0 cm, presence of a snow cover consistently
increased the relative abundance of Betaproteobacteria,
while Deltaproteobacteria were higher for the 20 and 50 cm samples.
The relative abundance level of
the top 20 bacterial genera was used to build a hierarchical clustering figure;
this is shown in Fig. 4 together with a heat map. The hierarchical clustering
is based on relative abundance of these genera was visualized by colour in the
heat map. The tree to the top clusters the 12 soil samples, with regard to
presence of these genera only. As can be seen, the different treatment
replicate were not grouping together. Rather, based on these representatives,
there were two groups of sample types, with the three uncovered soil samples
grouping together with two 100 cm and one 50 cm sample. Moreover, the
represented genera clustered according to their abundance (tree to the right)
that in part related to the phylum to which they belong. This indicates that
different bacterial species that are taxonomically related may respond similar
to the tested conditions, increasing the overall findings per phyla, as shown
in Fig. 2.
Fig. 2: Comparison of bacterial groups
at the phylum level after different snow pack treatments. The phyla are sorted
for relative abundance in the uncovered soil samples
Fig. 3: Comparison of bacteria groups
at the class level after different treatments
Discussion
Multiple studies
in alpine and arctic tundra ecosystems have reported active microbial
metabolism under snow-cover during winter (Schadt et al. 2003; Schmidt et al.
2007). However, until now little has been known about winter microbial
biogeochemical processes in temperate areas. This study determined the soil
microbial diversity and community composition in a variety of seasonally
snow-covered temperate wetland ecosystems. Following exposure to various
temporarily maintained levels of snow cover (from 0 cm to 100 cm) under
otherwise natural conditions, the Shannon diversity of naturally occurring soil
bacteria increased with snow height, indicating that snow thickness increases
soil microbial diversity, probably as a result of insulation and less severe
freezing soil temperatures (Brookes et al.
1985). While any water contained in the soil remains frozen, local water
shortness in combination with low temperatures can kill bacteria or induce
their dormancy, resulting in a decline in soil microbial biodiversity (Ross
1990) as observed here in the uncovered soil. We have demonstrated that with an
increase of snow cover thickness, the soil is better insulated and reaches less
low temperatures. This probably explains why snow cover thickness correlates
with soil bacterial diversity. All main bacteria phyla remained present in the
soil under the different snow cover conditions, so that snow cover (or absence
of it) did not change the soil microbial community composition completely, but
it did change the relative abundance of phyla and members therein, which is
consistent with previous studies (Fierer et
al. 2003).
The present study showed that the
soil bacterial community structure at the phylum level significantly differed
with snowpack changes observed in an alpine meadow. There, Acidobacteria were dominant (Barns 1999) while in the
wetland investigated here, Acidobacteria were only the second-most abundant
phylum. A previous study in this wetland reported that the local bacterial
community is affected by soil factors such as pH, organic carbon, C/N ratio,
soil temperature, and soil moisture (Naether et al. 2012). Those
studies were performed at temperatures above freezing, and here we complete
these findings for winter conditions. Acidobacteria in forest soil have been
shown to respond to these
environmental factors as well (Naether et al.
2012). We found that moderate snow cover decreased the relative abundance of Acidobacteria, compared to zero coverage
which was positively
correlated with soil temperature and pH. Our results further identified that different snow cover depth scaused
variation in relative abundance of Proteobacteria, but a general trend
correlating with cover thickness was not observed. This phylum covers very extensively different species with different metabolic
activities (Song et al. 2016).
Bacteria within the phylum Bacteroidetes are widely distributed across
ecological niches (Garrity and Holt 2001), and although they were not the most
abundant phylum, their abundance increased with moderate snow cover, to
decrease again at 100 cm snow coverage. Gram-positive Actinobacteria constitute
one of the largest phyla among bacteria and typical soil members have high
guanine and cytosine contents in their DNA (Ventura et al. 2007). Only limited variation was observed in their relative
abundance between the various treatments. More extensive variation was seen for
the Chloroflexi. In combination,
the relative abundance of all phyla in combination showed resemblance of the
100 cm covered soil with uncovered soil, while snow coverage of 20 and 50 cm
were most different. This demonstrates that relatively minor changes in snow
coverage can have severe effects on the soil microbial community. In conclusion
the diversity of the soil bacterial community varied inconsistently with the
snowpack gradient, with different changes observed for individual phyla of the
soil bacterial community, most likely resulting in functional differences in
the soil bacterial community.
Fig. 4: Hierarchical clustering diagram
of bacteria at the genus level after different soil treatments for the 12 individual samples.
The codes to the bottom start with ‘S’ to indicate snow cover, followed by
’10’, ’20’ or ‘100’ to indicate its thickness, and ‘1’, ‘2’ or ‘3’ for the
replicate samples. The tree to the left shows the hierarchical clustering based
on abundance of the genera identified to the right
Conclusion
Results of this
study proved that the change of snowpack caused significant changes in soil
physical-chemical parameters and the bacterial community in the temperate
wetland in winter. The relative abundances of Proteobacteria and Acidobacteria
were increased by the increasing of snowpack. In general, snowpack might
potentially affect soil bacterial structure and composition, but not changed
the alpha diversity in temperate wetland systems under global change scenarios.
Over the long term, investigating the snowpack change is important to
accurately track and predict the responses of global climate
change in wetland ecosystems.
Acknowledgements
This work was
funded by National Natural Sciences Foundation of China
(31470019, 41575153, 41330530).
References
Allison SD, Y Lu, C
Weihe, ML Goulden, AC Martiny, KK Treseder, JB Martiny (2013). Microbial
abundance and composition influence litter decomposition response to
environmental change. Ecology 94:714–725
Barns SM (1999). Wide
distribution and diversity of members of the bacterial kingdom Acidobacterium
in the environment. Appl Environ
Microbiol 65:1731–1737
Bogoev VM, AE Kenarova, VL
Vasilev (2002). Quantitative distribution of microbial biomass in the soil
profile of a high-mountain grassy ecosystem. Fol Microbiol 47:56–60
Bombonato L, R Gerdol (2012).
Manipulating snow cover in an alpine bog: effects on ecosystem respiration and
nutrient content in soil and microbes. Clim Chang 114:261–272
Brookes PC, A Landman, G
Pruden, DS Jenkinson (1985). Chloro form fumigation and
the release of soil nitrogen- A rapid direct extraction method to measure
microbial biomass nitrogen in soil. Soil Biol Biochem 17:837–842
Buckeridge KM, RL Jefferies (2007).
Vegetation loss alters soil nitrogen dynamics in an Arctic salt marsh. J Ecol 95:283–293
Campbell JL, MJ
Mitchell, PM Groffman, LM Christenson, JP Hardy (2005). Winter in northeastern
North America: A critical period for ecological processes. Front Ecol Environ 3:314–322
Chao A (1984).
Nonparametric estimation of the number of classes in a population. Scand J
Stat 11:265–270
Drotz SH, T Saparrman,
MB Nilsson, SJ Chleucher, MG Oquist (2010). Both catabolic and anabolic
heterotrophic microbial activity proceed in frozen soils. Proc Natl Acad Sci USA 107:21046–21051
Edgar RC, BJ Haas, JC Clemente, C Quince, R Knight (2011).
UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27:2194–2200
Edwards KA, RL
Jefferies (2013). Inter-annual and seasonal dynamics of soil microbial biomass
and nutrients in wet and dry low-Arctic sedge meadows. Soil Biol Biochem 57:83–90
Edwards KA, J McCulloch, GP
Kershaw (2006). Soil microbial and nutrient dynamics in a wet Arctic sedge
meadow in late winter and early spring. Soil Biol Biochem 38:2843–2851
Fierer N, JP
Schimel, PA Holden (2003). Variations in microbial community composition
through two soil depth profiles. Soil Biol Biochem 35:167–176
Freppaz M, SD
Aid-Pullicino, G Filippa, F Curtaz, L Celi, E Zanini, (2014). Winter-spring
transition induces changes in nutrients and microbial biomass in mid-alpine
forest soils. Soil Biol Biochem 78:54–57
Garcia CT, T Hernandez, A Roldan, A Martin (2002).
Effect
of plant cover decline on chemical and microbiological parameters under
Mediterranean climate. Soil Biol Biochem 34:635–642
Garrity GM, JG Holt
(2001). The road map to the manual. In: Bergey's Manual® of Systematic
Bacteriology, pp:119–166 Springer, New York, USA
Groffman PM, JP Hardy, CT
Driscoll, TJ Fahey (2006). Snow depth, soil freezing, and fluxes of carbon
dioxide, nitrous oxide and methane in a northern hardwood forest. Glob Chang Biol 12:1748–1760
Hinkler J, BU
Hansen, MP Tamstorf (2008). Snow and snow-cover in Central Northeast Greenland.
Adv Ecol Res 40:175–195
IPCC (2013). The Physical Science Basis. In: Contribution
of Working Group I to the Fifth Assessment Report of the Intergovernmental
Panel on Climate Change Stocker, Qin TF, D Plattner, GK Tignor, M
Allen, SK Boschung, J Nauels, A Xia, Y Bex, V Midgley (Eds.). Cambridge
University Press, Cambridge, United Kingdom and New York, USA
Kapnick SB, TL
Delworth (2013). Controls of global snow under a changed climate. J Clim 26:5537–5562
Kunkel KE, MA Palecki,
L Ensor, D Easterling, KG Hubbard, D Robinson, K Redmond (2009). Trends in
twentieth-century U.S. extreme snowfall seasons. J Clim 22:6204–6216
Larsen KS, P Grogan, S
Jonasson, A Michelsen (2007). Respiration and microbial dynamics in two
subarctic ecosystems during winter and spring thaw: effects of increased snow
depth. Arct Antarc Alp Res
39:268–276
Lipson DA, SK Schmidt, RK Monson (2000).
Carbon availability and temperature control the post-snowmelt decline in alpine
soil microbial biomass. Soil Biol Biochem 32:441–448
Merino A, S Fernández, L Hermida, L
Lopez, JL Sanchez, E Garcia-Ortega, E Gascon (2014). Snowfall in the northwest Iberian
peninsula: Synoptic circulation patterns and their influence on snow day
trends. Sci World J 48:1-14
Monson RK, DL Lipson, SP
Burns, AA Turnipseed, AC Delany (2006). Winter forest soil respiration
controlled by climate and microbial community composition. Nature 439:711–714
Naether A, BU Foesel, V
Naegele, PK Wüst, J Weinert, M Bonkowski, F Alt, Y Oelmann, A Polle, G Lohaus,
S Gockel, A Hemp, EKV Kalko, KE Linsenmair, S Pfeiffer, S Renner, I Schöning,
WW Weisser, K Wells, M Fischer, J Overmann, W Friedrich (2012). Environmental
factors affect Acidobacterial communities below the subgroup level in grassland
and forest soils. Appl Environ Microbiol
78:7398–7406
Price BP, T Sowers (2004).
Temperature dependence of metabolic rates for microbial growth, maintenance, and
survival. Proc Natl
Acad Sci USA 101:4631–4636
Quast C, E Pruesse, P Yilmaz, J Gerken, T Schweer, P
Yarza, J Peplies, FO Glockner (2013). The SILVA ribosomal RNA gene database
project: improved data processing and web-based tools. Nucleic Acids Res
41:D590–D596
Ross DJ (1990).
Estimation of soil microbial C by a fumigation-extraction method: influence of
seasons, soils and calibration with the fumigation incubation procedure. Soil Biol Biochem 22:295–300
Schadt CW, AP Martin, DA
Lipson, SK Schmidt (2003). Seasonal dynamics of previously unknown fungal
lineages in tundra soils. Science
301:1359–1361
Schmidt SK, EK Costello, DR
Nemergut, CC Cleveland, SC Reed, MN Weintraub, AF Meyer, AM Martin (2007).
Biogeochemical consequences of rapid microbial turnover and seasonal succession
in soil. Ecology 88:1379–1385
Song ZL, L Wang, XH Liu,
F Liang (2016). The diversities of Proteobacteria in four acidic hot springs in
Yunnan. J Henan Agric Univ 3:376–382
Ventura M, C
Canchaya, A Tauch, G Chandra, GF Fitzgerald, KF Chater, DV Sinderen (2007).
Genomics of Actinobacteria: tracing the evolutionary history of an ancient
phylum. Microbiol Mol Biol Rev
71:495–548
Wang CK, Y Han, J
Chen, X Wang, Q Zhang, B Bond-Lamberty (2013). Seasonality of soil CO2
efflux in a temperate forest: biophysical effects of snowpack and spring
freeze–thaw cycles. Agric For Meteorol 177:83–92
Wang W, SX Zhang, QF
Zhang (2015). Effects of freeze–thaw cycles on the soil nutrient balances,
infiltration, and stability of cyanobacterial soil crusts in northern China. Plant Soil 386:263–272
Wang W, S Peng, T Wang, J Fang (2010). Winter soil CO2 efflux and its
contribution to annual soil respiration in different ecosystems of a
forest-steppe ecotone, north China. Soil Biol Biochem 42:451–458
Yan G, Y Xing, J Wang, Z Zhang, L Xu, S Han, J Zhang, G Dai, Q Wang (2018). Effects of
winter snowpack and nitrogen addition on the soil microbial community in a
temperate forest in northeastern China. Ecol Indicat 93:602–611
Yang SZ, HJ Jin (2008).
Physiological and ecological effects of freezing and thawing processes on
microorganisms in seasonally froze ground and in permafrost. Acta Ecol Sin 28:5065–5074
Zhang T (2005).
Influence of the seasonal snow cover on the ground thermal regime: an overview.
Rev Geophys 43:1–26
Zhong W, T Gu, W Wang, B
Zhang, X Lin, Q Huang, W Shen (2010). The effects of mineral fertilizer and
organic manure on soil microbial community and diversity. Plant Soil 326:511–522