ECOLOGICAL RESTORATION OF BIOCRUSTS IN ALPINE TUNDRA BIOMES
by
Annie-Claude Letendre
B.Sc., University of Alberta, 2014

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
BIOLOGY

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
May 2018
© Annie-Claude Letendre, 2018

ii

Abstract
Biocrusts are complex communities of bryophytes, algae, fungi, lichens, and cyanobacteria living at
the uppermost surface of soils. They have a global distribution and commonly colonize early
successional and newly disturbed habitats, where they play important functional roles by
facilitating key ecosystem processes. While several studies have examined biocrust development
and function in arctic and alpine environments, the potential to use biocrusts in the restoration of
disturbed soils in alpine tundra biomes has rarely been examined. In a greenhouse trial, we
evaluated the restoration of biocrust through artificial inoculation of soils with mature biocrust.
Our results suggest that artificial inoculation with biocrusts increases soil surface nitrogen-fixation
rates. In a field study, we characterized alpine biocrust communities from cool mesic and xeric
environments and conducted an inoculation experiment to assess the recovery of biocrust
structure and function. Together these studies offer a comprehensive description of the functions
of biocrusts in alpine environments and provide key information regarding the efficacy of using
biocrusts for ecological restoration.

ii

Table of contents

Abstract ................................................................................................................................................. ii
Table of Contents ................................................................................................................................. iii
List of Tables......................................................................................................................................... vi
List of Figures .......................................................................................................................................vii
Acknowledgements .............................................................................................................................viii
1.

Introduction and literature review ............................................................................................... 1
Soil Surface Disturbance Ecology ...................................................................................................... 1
Ecological Restoration ....................................................................................................................... 2
Spatial Heterogeneity and Site Preparation.................................................................................. 3
Biological Soil Crust Communities..................................................................................................... 4
Abiotic Factors Influencing Biocrust Community Assembly and Function ....................................... 5
Biological Soil Crust Functions .......................................................................................................... 6
Extracellular polysaccharides ........................................................................................................ 7
Nutrient cycling ............................................................................................................................. 8
Restoration Techniques with Biocrusts......................................................................................... 8
Summary ........................................................................................................................................... 9

2. Inoculation promotes microbial community reassembly and restoration of ecosystem function
in alpine biocrusts – a microcosm experiment ................................................................................... 10
Introduction .................................................................................................................................... 10
Methods .......................................................................................................................................... 12
Site Description ........................................................................................................................... 12
Sample Collection........................................................................................................................ 13
Microcosms ................................................................................................................................. 14
Bacterial 16S rRNA and fungal ITS2 sequencing ......................................................................... 15
Biocrust establishment................................................................................................................ 17
Maximum photochemical quantum efficiency ........................................................................... 17
Nitrogenase activity .................................................................................................................... 18
nifH gene copy number ............................................................................................................... 19
Extracellular polysaccharide content .......................................................................................... 19
Soil properties ............................................................................................................................. 20

iii

Data Analysis ............................................................................................................................... 20
Results ............................................................................................................................................. 22
Bacterial 16S rRNA and fungal ITS2 amplicon sequencing.......................................................... 22
Biocrust establishment................................................................................................................ 25
Maximum photochemical quantum efficiency ........................................................................... 25
Nitrogenase activity and nifH gene copy number....................................................................... 26
Extracellular polysaccharide content .......................................................................................... 26
Soil Properties ............................................................................................................................. 27
Discussion ........................................................................................................................................ 27
Fungal and bacterial communities of inoculated surfaces differ from uninoculated surfaces and
from inoculant ............................................................................................................................. 27
Inoculation and microtopography facilitate biocrust establishment.......................................... 34
Inoculation and soil surface treatment promote function recovery .......................................... 35
Inoculation has limited effect on EPS and soil properties after 12 weeks .................................. 36
Conclusion ....................................................................................................................................... 38
3. Restoration of ecosystem function by soil surface inoculation of biocrust in mesic and xeric
alpine ecosystems ............................................................................................................................... 39
Introduction .................................................................................................................................... 39
Methods .......................................................................................................................................... 40
Study Areas ................................................................................................................................. 40
Microclimate Monitoring of Natural Biocrust ............................................................................. 41
Natural Biocrust Cover Assessment ............................................................................................ 42
Field Experiment Design.............................................................................................................. 42
Natural Biocrust Sampling ........................................................................................................... 44
Acetylene-Reduction Assays ....................................................................................................... 44
Extracellular polysaccharide content .......................................................................................... 45
Soil properties ............................................................................................................................. 45
Data Analysis ............................................................................................................................... 46
Results ............................................................................................................................................. 47
Characterization of alpine biocrust ............................................................................................. 48
Recovery of Inoculated Biocrust and Associated Ecosystem Functions ..................................... 51
Discussion ........................................................................................................................................ 53
Characterization of Alpine Biocrust ............................................................................................ 53
Indicators of Ecosystem Function Recovery in Experimental Plots ............................................ 57
iv

Conclusion ....................................................................................................................................... 61
4.

Implications ................................................................................................................................. 62

Appendix ............................................................................................................................................. 65

v

List of tables
Table 1. Analysis of variance (F values) for measurements taken at week 6 and 12. ........................ 79
Table 2.Properties of soils underlying experimental plots 12 weeks following inoculation. ............. 80
Table 3. Composition (bryophytes and lichens) of biocrust communities at Andesite and Mt
McIntyre. ............................................................................................................................................. 81
Table 4. ANOVA on mixed models of function parameters and linear models of soil properties and
biocrust cover of experimental plots at Andesite and Mt. McIntyre (Site) sampled 0, 12 and weeks
following inoculation........................................................................................................................... 83

vi

List of figures
Figure 1. Two-way cluster dendrogram of 62 bacterial OTUs with high relative abundances within
the six sample groups defined by treatment or control category…………………………………………………85
Figure 2. Relative abundance of bacterial 16S rRNA gene sequences. ............................................... 86
Figure 3. Two-way Cluster Analysis of 75 fungal OTUs averaged by sample units: Inoculant, biocrust
surface treatments (Flat, Pit & Mound, Microrills), and soils (Uninoculated controls and Stored).. . 87
Figure 4. Relative abundance of fungal ITS2 region sequences. “Stored” bar represents unincubated
soil that was sampled at the start of the experiment. ........................................................................ 88
Figure 5. a. Mean (±SE) biocrust percent cover at 6 and 12 weeks (light and dark bar, respectively).;
b. Proportional abundance of Fv/Fm at 6 and 12 weeks; c. Mean (±SE) acetylene reduced at 6 and
12 weeks (light and dark bar, respectively).; d. Mean (±SE) glucose content per biocrust at 6 and 12
weeks (light and dark bar, respectively). ............................................................................................ 89
Figure 6. Mean acetylene reduced as a function of nifH abundance at 12 weeks. ............................ 90
Figure 7. Microclimate data from mature biocrust monitoring station on Andesite Peak over the
duration of the experiment, expressed in Julian days. ....................................................................... 91
Figure 8. Microclimate data from mature biocrust monitoring station on Mt McIntyre over the
duration of the experiment, expressed in Julian days.. ...................................................................... 92
Figure 9. Biocrust cover 12 weeks following inoculation. ................................................................... 93
Figure 10. Acetylene reduced 0, 6, and 12 weeks following inoculation with biocrust at Andesite
(black bars) and at Mt McIntyre (grey bars). ...................................................................................... 94
Figure 11. Extracellular polysaccharide content (EPS) of soil surface 12 weeks following inoculation
at Andesite (black bars) and at Mt McIntyre (grey bars). ................................................................... 95
Figure 12. Soil properties 12 weeks following inoculation at Andesite (black bars) and at Mt
McIntyre (grey bars). ........................................................................................................................... 96

vii

Acknowledgements
I would first like to thank my thesis supervisors Dr. Darwyn Coxson of the Department of Ecosystem
Science and Management at University of Northern British Columbia and Dr. Katherine Stewart of
the Department of Soil Science at University of Saskatchewan. They were always available to
support me and offered tremendous help with the design, implementation, and realization of my
research and writing of my thesis. I would also like to thank Dr. Paul Sanborn of the Department of
Ecosystem Science and Management at University of Northern British Columbia for his valuable
input into my research design as well as for his eye for detail in the reviewing my thesis. I would
also like to thank Dr. Keith Egger and Dr. Nicole Sukdeo, the experts who were involved in the
validation and realization of the molecular biology aspect of my research project. Without their
participation and input, this research could not have been successfully conducted. I would also like
to acknowledge Bob Sagar for his assistance with data logger programming, Denise Gordon, Isobel
Ness, Krystal Isbister and Jessica Lowry for their precious help in the field and laboratory, Alfredo
Staffa (Génome Québec Innovation Centre) for sequencing the amplicon libraries as well as Pascale
Marquis and Mathieu Bourgey (Génome Québec Innovation Centre) for quality filtering and
clustering of amplicon sequencing data. Finally, I would like to acknowledge funding support
provided by Pacific Trail Pipelines Limited Partnership, through their Biodiversity Monitoring and
Assessment Program.

viii

1. Introduction and literature review
Soil surface disturbance ecology
Anthropogenic disturbances can have lasting effects on alpine plant communities, inducing
decreases in plant and microbial diversity, native species abundance, soil organic carbon and
nitrogen and mineralization rates [1]. The response of plant and soil communities to disturbance
depends on the nature of the disturbance, the environmental conditions (abiotic factors) and the
community composition (biotic factors).
Anthropogenic disturbances on alpine communities can be indirect, such as the impact of climate
change [2], or direct, such as excavation for pipelines, roads, or mining [3]. Pipeline rights-of-way
are a type of direct disturbance that can have profound and lasting impacts on alpine vegetation
and soils [4, 5]. Environmental characteristics common to most alpine environments (i.e. short
growing season, cold weather and nutrient poor soils) cause these biomes to have slow recovery
rates following disturbance [1]. The response of alpine communities to linear disturbances, such as
pipelines, will also depend on environmental factors such as climate, elevation, slope, aspect,
topography, etc. [6]. Alpine plant and soil microbe communities can be disturbed by physical
processes such as compression. Biocrusts are a key component of these alpine communities and
those from arid and semi-arid regions are particularly sensitive to disturbances by pipelines [7],
with factors such as compressional disturbances of biocrusts diminishing their resistance to erosion,
thereby altering their ecosystem function by reducing their C and N inputs [8]. The community
composition is also an important driver of its response to disturbance. The cyanobacterium
Microleus vaginatus was highly susceptible to disturbance on sandy soils in cold-desert
environments, whereas the lichen Collema tenax was less susceptible, but still exhibited decline in

1

nitrogenase activity after disturbance [9]. Nonetheless, alpine plant communities are resilient and
have the potential to recover following disturbances [1, 10].

Ecological restoration
Restoration can be defined as the process by which we aim to offset the impacts of contemporary
anthropogenic disturbances on natural systems. The overarching goal of ecological restoration is to
redirect a system towards its natural trajectory of successional processes, which can result in a
resilient and self-sustaining system. Ecosystem restoration that allows for self-sustaining and
resilient systems can be informed by naturally occurring diversity and its associated processes [11].
A function-based approach to ecological restoration seeks to facilitate the recovery of disturbed
sites by prompting ecosystem processes that would naturally occur. The use of predisturbance
community assembly as a guide relies on the premise that native local species are well adapted to
local site conditions, however, this assumption is increasingly being challenged under changing
climate conditions [12, 13]. For the most part, however, naturally occurring species remain those
which are best adapted to restoring natural successional processes and therefore constitute an
appropriate choice for restoration. This is echoed by the Society for Ecological Restoration
guidelines which state that genetic integrity and regional biodiversity are fundamental in
restoration projects [14, 15].
The use of non-native species is a common restoration approach used to facilitate the rapid
establishment of ground cover. This is thought to reduce erosion and enhance soil physical
properties so that native species can subsequently establish [16]. However, this approach has had
limited success in montane environments [17]. As a result, native species are generally preferred
for restoration protocols in ecologically sensitive alpine environments, where factors that limit
growth and reproduction of native species including soil properties, topography and climate. Hagen

2

et al. [20] postulate that species from alpine environments are particularly sensitive to competition
from non-native species and that the improvement of soil conditions may further increase their
competitive advantage [18]. Although using native species can maintain the ecological integrity of a
site, few studies have investigated the specific challenges associated with alpine restoration.
Various techniques have been used to restore native species in disturbed environments such as
rights-of-way. An example is the soil transfer technique, which is commonly used in pipeline rightsof-way construction. It consists in translocating soil and the associated biota (fauna, flora and
microbes) from a donor site to a disturbed site [5]. Another technique is hay transfer, the transfer
from a donor site to a restoration site of mowed hay stalks with the desired seeds ripe and still
attached, which can effectively restore pipeline rights-of-way located in grasslands [15]. In addition
to the system’s biodiversity, ecological restoration is also dependent on the environmental
conditions. Short and cool growing seasons, minimal precipitation, and low soil nutrient levels are
environmental conditions typical of alpine environments that pose specific challenges associated
with the restoration of alpine sites.

Spatial heterogeneity and site preparation
Alpine environments which are naturally highly variable provide a variety of niches for diverse plant
communities. Topography is a major driver of variability in alpine environments and is an important
element of successful ecological restoration which seeks to encompass the full range of a system’s
biodiversity. Site preparation techniques that create variation in topography and/or
microtopography are important drivers of success in restoration. Spatial heterogeneity increases
the number of niches in an ecosystem thereby enriching species diversity [19, 20].

3

Mounds have been widely used as a means to create spatial heterogeneity, hence promoting
successful restoration [21]. Mounds create varying moisture conditions, such that the leeward side
of the mound will be more consistently humid with less variation in moisture when compared with
the windward side. Li et al. (2010) [22] found that biocrusts had higher biomass and cover in the
mounds more humid hollows and on the leeward faces. At a smaller scale, microsites are
associated with increased moisture and soil stability which are characteristics that foster successful
restoration conditions [23]. The availability of microsites can be increased by augmenting the
spatial variability of resources [19, 20]. Another technique that can create spatial heterogeneity is
ripping which consists in mechanically disturbing the top few centimeters of soil [24]. This
technique that loosens the topsoil (and in some instances the subsoil) with minimal soil mixing has
been shown to promote establishment, increase water infiltration into the soil and reduce soil
erosion [25].

Biocrust communities
Biocrusts are communities of organisms forming a cohesive thin layer at the uppermost surface of
soils. They are composed of algae, lichens, mosses, liverworts, cyanobacteria, and other primary
successional species [26]. The accumulation of extracellular polysaccharide sheaths by
cyanobacteria, a response to wetting-drying cycles at the soil surface, binds soil particles together,
forming soil aggregates that stabilize soil surfaces and allow other species to colonize [27].
Biocrusts readily colonize disturbed soils, as well as, areas not occupied by vascular plant, where
they can take advantage of maximum sunlight and precipitations. Although biocrusts are global in
distribution, they are commonly found in alpine, boreal, and arctic ecosystems [28, 29].

4

Abiotic factors influencing biocrust community assembly and function
Temperature and moisture are abiotic factors that can have a profound impact on biocrust function
[30]. In a manipulative experiment, de Guevara et al. [31] demonstrated that soil warming caused a
reduction in photosynthesis rates of biocrusts. Biocrusts require moisture to be metabolically
active, hence moisture level has major influence on their recovery following disturbance. Biocrust
growth can also be limited by nutrients, such as phosphorus, which is often not readily available
since it is bound to unweathered minerals [32, 33].

The role of moisture availability in biocrust functioning is magnified by the poikilohydric nature of
biocrusts. Nitrogen-fixation rates are strongly correlated with availability of water [34-38]. Several
studies on boreal and arctic cyanolichens have found that limitations of moisture are a primary
constraint on nitrogen-fixation across all temperatures [39]. In a large-scale study, Raggio et al.
2017 [40] demonstrated that metabolic activity of biocrusts could be predicted from macroclimatic
data. The functions of biocrust are undoubtedly regulated by light and temperature although the
interactions between these two factors and water at the soil surface are not well studied.

Not only do microclimatic conditions influence biocrust functions but they also play a role in
regulating their community assembly. Differences in biocrust species composition are intrinsically
linked to different environmental and microclimatic conditions. In a comparison between alpine
and Antarctic biocrusts, Colesie et al. 2016 [41] found that compositional difference could be
correlated to climatic conditions and the adaptations necessary to thrive in extreme environments.
Accordingly, Li et al. 2010 [22]found that increasing soil moisture shifted species composition of
biocrust. In addition, increased species richness has been linked to milder climates in North
5

American montane environments [42]. Together, these studies indicate that biocrust community
composition and function are linked with environmental conditions.

Nutrient content is also a major driver of restoration success. Fertilization has been widely used in
biocrust restoration efforts and has been found to influence both species composition and functions.
Antoninka et al. 2015 [43] found that fertilization was correlated with an increase in nitrogen-fixation
under greenhouse conditions, and Maestre et al. 2006a [44] observed a positive relationship
between moderate fertilizer addition and nitrogenase activity under laboratory conditions. Elevated
levels of nitrogen also have the potential to suppress the production of EPS and the nitrogen fixing
activity of biocrusts [45, 46].

Biocrust functions
Biocrusts are important in the development of ecosystems because they facilitate the shift to later
successional seres by participating in several key ecosystem processes including nutrient cycling
and soil pedogenesis. Cyanobacteria and cyanolichens in biocrusts influence the nitrogen dynamics
of biomes through processes such as mineralization and fixation of organic nitrogen (R-NH2) and
atmospheric dinitrogen (N2), respectively, into ammonium (NH4+). The quantity of nitrogen fixed
by biocrusts that is released depends on several factors such as season, light, soil moisture, and
temperature and is estimated to be between 5 and 70% (e.g. Belnap et al. 2001 [47]). However, few
studies focus on temperate mesic and xeric ecosystems (but see Veluci et al. [48]) where nitrogen
inputs from atmospheric deposition and rates of nitrogen released may significantly differ from hot
environments. Biocrusts also influence ecosystem hydrology through altering runoff and infiltration
of water. The nature of their effect on water movement largely depends on biocrust

6

micromorphological features, as well as, associated soil characteristics [49]. The microtopography
of biocrusts also provides a habitat for soil microbes which in turn affect soil respiration. Through
these functions, biocrusts modify soils which affect vascular plants seed dispersal, germination,
establishment, survival and nutritional status [26, 44, 50-53]. Given the key roles of biocrusts in
ecosystem functioning, their restoration can promote the recovery of disturbed systems [54] by
prompting fundamental processes such as carbon and nitrogen cycling.

Extracellular polysaccharides
Extracellular polysaccharides (EPS) are compounds secreted by cyanobacteria and green algae in
biocrusts that form an envelope surrounding the cells that bind soil particles and retain water [55].
Therefore, soils colonized by biocrusts have increased stability and are more resistant to erosion
[56-58]. EPS attract and absorb water from their environment, which makes them particularly
important in biocrust communities where bryophytes and other poikilohydrous organisms require
externally provided water for growth and reproduction [30, 31]. Biocrusts are metabolically active
only when hydrated, therefore limitation of moisture availability hinders the growth of these
phototrophic communities [32, 59-61].
Biocrusts and the associated EPS can exert a strong influence on water movements. This influence
is determined by several factors, such as, soil texture and moisture, rainfall intensity, and biocrust
composition [62, 63]. EPS can modify the runoff-infiltration balance thereby increasing infiltration
in some zones while reducing it in others [26, 64]. EPS are also known to increase the hydraulic
conductivity of soils by increasing the amount of micropores present in biocrusts [62, 63].
EPS play an important role in carbon cycling in environments where soils are naturally carbon poor
and EPS may provide the main input source of carbon. By stabilizing the soil, EPS reduce the carbon
7

loss due to soil erosion [65]. Carbon is also the energy source cyanobacteria use to fix nitrogen [66].
Therefore, carbon fixation is intimately linked to nitrogen fixation, which is another important
ecosystem service provided by these organisms [62, 67].

Nutrient cycling
Nutrient availability is lower in colder systems because low temperatures inhibit both chemical
weathering and decomposition processes [68]. Therefore, alpine and arctic environments are
generally nutrient poor and often have low soil nitrogen mineralization rates [69]. In addition,
nitrogen typically limits primary production during early successional stages [32, 33]. Hence, the
nitrogen-fixing organisms that compose biocrusts play a fundamental role in fixing nitrogen,
thereby make it available to plants. Additionally, biocrusts may play a significant role in cycling of
trace gases such as CO2 [57, 61], as well as, retaining and concentrating nutrients such as sodium,
potassium, magnesium, calcium, manganese, iron, nickel, copper and zinc [58, 62].

Restoration techniques with biocrusts
It has been suggested that recovery of biocrust can be significantly accelerated by active
restoration methods such as inoculation or cultivation [70, 71]. Inoculation of soils with biocrusts
can speed up restoration processes [72] notably by increasing available nitrogen [51], a nutrient
that is known to limit plant productivity in high montane and arctic environments. Inoculation
approaches can range from placing discrete biocrust fragments onto the surface to spraying water
enriched with microbial organisms. Several studies have investigated the assisted recovery of
biocrusts in a plant community restoration context either through inoculation [73], pelletized
cyanobacterial amendment or direct transplants. Stewart and Siciliano (2015) [74] found that
8

biocrust nitrogenase activity could recover on Yukon mine tailings in a growth chamber inoculation
experiment. Buttars et al. (1998) [75] demonstrated that inoculation of soils by addition of
pelletized cyanobacteria could increase recovery rates of BSC. According to Maestre et al. (2006)
[44], inoculation in the form of slurry, as oppose to large discrete fragments, has the potential to
foster recovery of BSC communities. Chiquoine et al. (2016) [76] found that inoculation with
salvaged biocrusts dried for 2 years could hasten the biocrust recovery of disturbed drylands roads.
Cultured cyanobacteria have been successfully applied to the field in Chinese deserts [77] and fully
functional greenhouse-grown biocrust mosses and associated cyanobacteria have been produced
by Antoninka et al. 2016 [78] but survival in the field has not yet been tested.

Summary
Given the importance of biocrust establishment in restoration of soil surface properties (e.g. soil
stabilization, nitrogen and carbon enrichment) following disturbance, the artificial enhancement of
biocrust establishment on disturbed sites has been the subject of a number of studies, using
techniques such as cultivation and/or inoculation [26, 70, 71, 79]. While previous studies have
examined the efficacy of inoculation for establishing biocrust in grassland and desert sites [44, 73,
80], to our knowledge none have investigated the effects of assisted inoculation on alpine biocrust
establishment and function.
Our main objectives were 1) to compare biocrust community composition and ecological function
between inoculated and bare soil surfaces; 2) to evaluate the influence of microtopographic
features (Flat, Microrills, Pit and mound) on the development of biocrusts in the inoculated soils. ;
3) to compare the composition and nitrogen fixation potential of natural biocrust associated with
contrasting coastal and continental mountain ranges; 4) to assess the restoration potential of
alpine biocrust communities; and to 5) to evaluate the influence of biocrust inoculation treatments
9

on nitrogen fixation, extracellular polysaccharide content, soil mineralizable nitrogen and dissolved
organic carbon, as key indicators of soil ecological function.

2. Inoculation promotes microbial community reassembly and
restoration of ecosystem function in alpine biocrusts – a microcosm
experiment
Introduction
Biocrusts are soil surface assemblages comprised of diverse organisms such as algae, lichens,
mosses, liverworts, Cyanobacteria, as well as other primary successional bacterial, fungal [26, 81]
and archaeal species [82]. Biocrusts are highly heterogeneous, capable of thriving in a variety of
environments [83] and are commonly found in alpine ecosystems [29, 83] where they colonize
exposed soil surfaces, taking advantage of the conditions of high insolation exposure and greater
precipitation availability. Established biocrusts form a cohesive thin layer at the uppermost surface
of soils, where production of extracellular polysaccharides (EPS) augments soil organic matter
content, binds soil particles together [66], and generally exerts a strong influence on surface
hydrological processes [84]. These soil surface communities can also influence soil nitrogen
availability through their ability to fix atmospheric nitrogen [85]. Taken together these processes
have a major influence on soil carbon and nitrogen cycling [54, 86, 87].

The harsh environmental conditions found in most alpine environments, including nutrient poor
soils and short-cool growing seasons, can result in slow recovery rates after disturbance [68].
10

Anthropogenic disturbances, in particular, can have significant impacts on alpine vegetation and
soils [3-5, 68], resulting in accelerated soil loss through erosion, decreased microbial diversity and
abundance, and the alteration of soil carbon and nitrogen cycling, including mineralization [1, 75,
83].

Although biocrusts are sensitive to disturbances which can induce changes in their community
composition and function [1, 80, 88, 89], they are also well adapted to harsh alpine environments
[90]. The restoration of biocrust communities can therefore be a significant factor in the restoration
of disturbed systems [54], promoting recovery of fundamental ecosystem processes such as
nitrogen and carbon cycling [91-93]. It has been suggested that recovery of biocrusts can be
accelerated by the adoption of active restoration methods such as inoculation or cultivation [26,
70, 71, 79]. The effects of inoculation on recovery of ecosystem function are poorly described,
especially for alpine environments. Although previous studies have focused on separate aspects of
assisted recovery [44, 73, 75], ecological function [29, 94], and community composition [83] of
alpine biocrusts, few have investigated the combined effects of assisted inoculation on biocrust
composition and function. Using a microcosm experiment we investigated the development of
alpine biocrust communities on bare alpine soils over a 12-week period after inoculation with
biocrust inoculant prepared from prior collections made in alpine environments. We also examined
the influence of soil surface microtopography, which has previously been found to be an important
factor in promoting heterogeneity and associated niche diversity of soil surface biocrusts (26). Our
main objectives were 1) to compare biocrust community composition and ecological function
between inoculated and bare soil surfaces; and 2) to evaluate the influence of microtopographic
features (Flat, Microrills, Pit and mound) on the development of biocrusts in the inoculated soils.

11

Biocrust establishment after inoculation was assessed both directly, using soil surface cover
measurements, and assessments of biocrust composition from bacterial 16S rRNA gene and fungal
ITS2 amplicon sequencing, and indirectly, assessing ecological function from measurements of
nitrogen fixation, EPS content, chlorophyll fluorescence, and nifH gene abundance. Contextual data
or “controls” for the experiment were provided by identical assessments of ecological function in
the initial inoculant, and in soils exposed to the same environmental conditions as treatments but
with no biocrust inoculation, and soils that were collected then stored.

Methods

Site description
Samples were collected from two adjacent sites in the Coast Mountain Range of northern British
Columbia, Canada; Trapper Mountain (N 54° 30.683’, W 128° 27.317’, 1187 m elevation) and
Andesite peak (N 54° 13.868’, W 128° 01.499’, 1640 m elevation). Trapper Mountain occurs in the
transition zone between subalpine forest and alpine tundra, with the vegetation community
dominated by scattered clumps of Abies lasiocarpa (Hooker) Nuttall interspersed with Cassiope
mertensiana heath. The site was characterized by Podzolic soils with acidic pH. Vegetation
communities on Andesite peak were dominated by alpine tundra interspersed with rocky outcrops
and late-melt snow beds in depressions. Vegetation on dry ridge top habitats on Andesite peak was
dominated by Cassiope mertensiana and Stereocaulon alpinum heath, with Saxifraga tolmiei and
Ranunculus cooleyae dominated vegetation communities common in wetter mid-slope habitats.

12

Andesite peak had Podzolic and Regosolic soils, with substrates of volcanic origin (meta-basalts and
meta-andesites).

Sample collection
Biocrust samples were collected on August 16 and 17, 2014. Collections were taken from sites with
an intermediate soil moisture status, avoiding wet late-snowmelt depressions and xeric ridge-top
habitats. Biocrusts were stored in ventilated bins held between 1 and 4°C during shipment (48
hours). Samples were then frozen at -20°C until the start of the experimental period on December
5, 2014 (17 weeks).

The impact of a prior period of frozen storage was assessed in a pilot trial using biocrusts from
Trapper Mountain and Andesite Peak, where we compared acetylene reduction activity of freshly
sampled biocrusts with that of previously frozen biocrusts. We found that after a six-day thawing
period (moistened to field capacity daily and 16 hours photoperiod at 150 µmol/m2/s during light
hrs.; 20oC daytime, 10⁰C nighttime) rates of acetylene reduction were not significantly different
between fresh and previously frozen biocrusts (Welch’s t-test, t=0.921, p=0.4). Samples were
thawed, under the same environmental conditions, for a six-day period immediately before the
start of experiments.

Soils used in the experiments were collected from the Trapper Mountain site on August 16, 2014.
Soil samples were collected at depths of 2 to 10 cm, after removing surface biocrusts and other
vegetation.

13

Microcosms
Biocrust development was examined under four soil surface microtopography treatments: i) Flat:
soil with flat slightly compacted surface, ii) Microrills: soil with repeating microridges across the
surface (gullies separated by ridges 1 cm in height and width- simulating raking), iii) Pit and mound:
a single hummock-hollow complex 15 cm in height, iv) No top soil: coarse gravel without soil to
emulate an absence of top soil application (Plate 3). Nine replicates of each treatment were
established, each consisted of a shallow plastic tray (0.076 m2 surface area) filled with 3 cm of
crushed gravel topped with 2 cm of soil prior to establishing microtopography treatments. A control
set of untreated replicates (referred to as “Uninoculated” in amplicon sequencing results) (n=9) did
not receive any biocrust inoculant. In addition, excess soil used to make the microcosm treatments
was stored (referred to as “Stored” in amplicon sequencing results) at -20 °C and used to determine
the community composition of the soil prior to treatment.

Following soil surface treatments, each tray was wetted to field capacity and inoculated with
biocrust at 10% surface area (i.e. 0.0076 m2). The inoculant was obtained by homogenizing a
combined mixture of mature biocrust fragments from our two collection sites through a 4mm sieve.
After inoculation, each tray was watered with 300 mL DI water to ensure contact between the
biocrust fragments and the underlying substrate. Trays were subsequently watered 3 times a week
with 150 mL of DI water.

Microcosms were maintained under greenhouse conditions with a 16 hours photoperiod (150 µmol
m-2 s-1 PAR and 20 °C daytime and 0 µmol m-2 s-1 PAR and 10 °C nighttime conditions). Illumination
was provided by 168X-Pro Extreme LED Grow Lights (Hydrogrow, Sunrise, FL, USA). To maintain
14

relative humidity at approximately 70%, 4 large clear plastic tents were constructed, each overlying
11 or 12 trays. Under each tent, the trays were moved three times weekly to avoid location effect
and ensure moisture and light levels were uniform between microcosms. Microcosms were
maintained under experimental conditions for 12 weeks.

Two 14.5 cm2 biocrust samples were removed from each tray at 6 and 12 weeks after the start of
the experiment. The location of sampling within each tray was determined by overlaying a random
numbers grid and randomly choosing four different grid numbers for sampling (the first two at
week 6 and the others at week 12). For Pit and mound replicates, at each sampling period, three
sets of two samples were taken across the microtopographical gradient from mound, mid-slope to
pit. One soil sample was removed from each tray at week 12, after removing the top 0.5 cm
including any biocrusts present. No soil samples were removed from the trays with the No top soil
treatment.

Bacterial 16S rRNA and fungal ITS2 sequencing
We randomly chose samples from each Soil surface treatment, the Uninoculated control and the
Stored soil to be used for bacterial 16S rRNA gene (three replicates selected per Soil surface
treatment and Uninoculated control but two replicates for Stored soil) and fungal ITS2 (four
replicates selected per Soil surface treatment and Uninoculated control but three replicates for
Stored soil) amplicon sequencing.

15

Primer sequences used for targeting the V4 region of the bacterial 16S gene were the 515F/806R
sequences as described in [95]. 16S amplicons generated in this study were 450 bp in length
(including adapter and index sequences). Primer sequences used for targeting the fungal ITS2
regions were the ITS86F/ITS4 sequences as described in [95] . The ITS86F/ITS4 primer pair
(including adapter and index sequences) generated amplicons that were predominantly were 550
bp in length. Amplicons ranging in size from 500-600 bp were extracted from agarose gels for
purification.

Prior to PCR amplification 10 ng of template DNA from each sample was pre-incubated with 0.5 µg
BSA (New England Biolabs) for 10 minutes at 95 °C prior to the addition of master mix containing
primers, 5 PRIME Hot Master mix (2.5 X stock) and nuclease free water (IDT). The final
concentration of PCR components in a 25 µl reaction volume (including template and BSA) were 1X
5 PRIME Hot Master Mix and primers at 300 nM (bacterial 16S) or 200 nM (fungal ITS2).

The thermal cycle profile used for bacterial 16S amplification had an initial denaturation step at 94
°C for 3 minutes, followed by 30 cycles of 94 °C for 45 seconds (denaturation), 50 °C for 60 seconds
(annealing) and of 72 °C for 90 seconds. A final extension at of 72 °C for 10 minutes ended the
thermal cycle profile. The thermal cycle profile used for fungal ITS2 amplification had an initial
denaturation step at 94 °C for 3 minutes, followed by 30 cycles of 94 °C for 45 seconds
(denaturation), 55 °C for 45 seconds (annealing) and of 72 °C for 2 minutes. A final extension at of
72 °C for 7 minutes ended the thermal cycle profile. For each sample triplicate 25 µL PCR’s were
pooled and the amplicons were extracted from 1% agarose, 0.5X TBE gels after electrophoresis

16

using the GeneJET Gel Extraction and DNA Cleanup Micro Kit (ThermoFisher Scientific) and eluted in
20 µL of elution buffer.

DNA concentrations of the purified amplicons were determined with a Qubit Fluorometric assay
(ThermoFisher Scientific) using the dsDNA BR assay kit. DNA concentrations were manually
adjusted to 2 ng/ml and pooled and submitted for sequencing to the Génome Québec Innovation
Centre’s Massively Parallel Sequencing Services unit (McGill University, Montreal, QC, Canada). The
amplicon pools were sequenced using a 250 nt paired-end run for bacterial 16S and a 300 nt
paired-end run for fungal ITS2 on an Illumina MiSeq system.

Biocrust establishment
Percent biocrust cover was visually estimated for each tray at weeks 6 and 12. All assessments were
conducted by the same observer using a reference grid outlining 1% and 10% surface area
increments.

Maximum photochemical quantum efficiency
For every sample, taken at the height of day (between 10:00 and 14:00) at weeks 6 and 12, we
determined the maximum photochemical quantum efficiency (Fv/Fm) by dark-adapted chlorophyll
fluorescence using a pulse modulated fluorimeter ([96]; FMS2, Hansatech Instruments, Norfolk,
UK). Samples were dark-adapted, using leaf clips, for a minimum of 10 minutes prior to
fluorescence measurement.

17

Nitrogenase activity
Nitrogenase activity was assessed using acetylene reduction assays (ARAs) [97] . Following
chlorophyll fluorescence measurements, biocrust samples were placed in a 250 ml glass canning jar
with a serum stopper lid. Samples were misted with 1 mL DI water prior to closure of jars and
injection of 10% acetylene by volume. Jars were incubated 4 h at 150 µmol m-2 s-1 PAR and 20oC.
After incubations a 4-mL gas sample was removed from each incubation jar.

Each set of samples were processed along with two control samples: i) a biocrust sample not
injected with acetylene, which served both as a temperature control and to ensure that no natural
evolution of ethylene was occurring and ii) a jar with no biocrust that was injected with acetylene
to ensure the absence of ethylene contamination in the acetylene.

Gas samples were injected into a gas chromatograph (SRI 8610A, Wennick Scientific Corporation,
Ottawa, ON, Canada) fitted with a Porapak column (Alltech Canada, Guelph, ON, Canada) and a
flame ionization detector for detection of ethylene. Hydrogen, used as the carrier gas, was held at a
constant pressure of 32 psi while column temperature was held at 65⁰C. The detection limits for
acetylene reduction were equivalent to 3.7 µmol of ethylene hr-1 m-2.

After ARA assessments, biocrust samples were stored at -20⁰C for subsequent analysis; samples
from each tray were pooled and half was used for sequencing of bacterial 16S rRNA and fungal
ITS2, and qPCR of nifH gene the other half to measure EPS content.

18

nifH gene copy number
From each sample set aside for qPCR (one per replicate), we used a 0.25 g sub-sample to perform
DNA extractions which were carried out using PowerSoil DNA Isolation Kit (Mo Bio Laboratories,
Carlsbad, CA, USA) following the manufacturer’s instructions. Extracted DNA was stored at -20°C.
Primers used for quantification of nifH abundance were NifHF 5’-AAA GGY GGW ATC GGY AAR TCC
ACC AC-3’ and NifHR 5’-TTG TTS GCS GCR TAC ATS GCC ATC AT-3’ from [98]. The standard used for
the quantification of nifH was a synthetic double-stranded gene fragment generated at IDT Inc
(Coralville, IA, USA) consisting of nucleotides 192-649 from the Rhizobium meliloti nifH gene
sequence1, nifH gene copies were quantified against a standard calibration curve obtained by serial
dilution ranging from 102 to 108 nifH gene copies. Each 12.5 µL qPCR reaction was carried out in
triplicate (samples, standards and no template control) and contained 6.25 µL Power Syber Green
Mix (Applied Biosystems, Carlsbad, CA, USA), 0.25 µL nifHF primer (0.2µM) and 0.25µL nifHR primer
(0.2 µM) (Life Technologies, Carlsbad, CA, USA) (sequences as per [98]), as well as 3.75µL of DI
water and 2 µL of standards, nuclease free water (no template control) or DNA extract from
samples (diluted 1 µL to 100 µL). Following the nifH qPCR method from [99], a 7300 Real Time PCR
machine (Applied Biosystems, Germany) was used under the following thermal cycling conditions:
hot start (95°C for 10 min.); amplification (95°C for 45s, 55°C for 45s, and 72°C for 45s) for 40 reps;
dissociation (95°C for 15s, 60°C for 30s, and 95°C for 15s).

Extracellular polysaccharide content
EPS content was determined using the phenol-sulfuric acid method [100, 101]. Each reaction was
transferred to a well in a 96-well flat bottom microplate, with standards and template control

1

NCBI Nucleotide database accession. version V01215.1.
19

carried out in triplicate. EPS content was measured as glucose concentration quantified at 480, 486
and 490 nm using a microplate reader (Varioskan, Thermo Fisher Scientific, Walthman, MA).

Soil properties
Dissolved organic carbon and mineralizable nitrogen were assayed on soil samples collected at
week 12 (n=36), and on soils sampled prior to biocrust inoculation (n=2).

Dissolved organic carbon was measured from a soil extract (1:10 soil to water ratio shaken for 1 h
on a reciprocating shaker) filtered to 0.45 µm, with a total organic carbon analyzer (Formacs HT,
Skalar Analytical B.V., Breda, Netherlands). pH at 12 weeks was measured by a 1:5 biocrust to DI
water suspension with a hand-held pH meter (Oakton Instruments pH610, Vernon Hills, IL, USA).

Mineralizable nitrogen was measured by the anaerobic incubation method [102]. Soil samples were
incubated under water-logged conditions for two weeks at 30°C. Ammonium produced was
subsequently measured from a soil extract (1:10 soil to 1N KCl ratio shaken for 2 h on a
reciprocating shaker) with a discrete analyzer using the phenate method for colorimetric
determinations ([103]; SmartChem 200, Unity Scientific, Brookfield, CT).

Data analysis
The amplicon sequencing data analyses used to generate inventories of 97% operational taxonomic
units (OTU) was performed by the Bioinformatics unit at Génome Québec. For multivariate analysis
of fungal community composition, we sub-sampled the fungal ITS2 OTU inventory to include only
20

clusters with greater than 250 sequences detected across all samples that were taxonomically
classified to Class level. We also excluded four fungal OTUs classified as genus Vrystaatia that had
conflicting sequence identifications when searched against the NCBI nr database using BLASTn. A
total of 75 OTUs representing ~75% of the quality-filtered, clustered sequences were retained for
community comparisons by Two-way Cluster Analysis using PC-ORD. For comparisons of bacterial
community composition, we focused on exploring restoration treatments in re-establishing
dominant biocrust- and soil-associated taxa by selecting 62 highly abundant OTUs from an initial
inventory of 5,099 bacterial OTUs, included only if a minimum of 2,000 sequences total was
detected within the specific sample groups (i.e. Inoculant, Stored soil, Uninoculated soil, Flat, Pit
and mound, and Microrills treatments).

Two-way Cluster Analysis was performed in PC-ORD version 6 (MjM Software Design, Gleneden
Beach, OR, USA) using a Sorensen distance and the flexible beta (-0.250) linkage method. The
analysis with fungal and bacterial data was performed with an OTU table containing sequence
counts averaged and relativized to sequence total (for only the OTUs included in analysis) per
sample group.

Linear mixed models (LMM) were used to avoid pseudoreplication in analysis of the dataset created
by having samples collected from the same trays at weeks 6 and 12. LMM included tray number as
a random effect, week and measurement as fixed effects, and the interaction of week and
measurement as fixed effects. To meet assumptions of normality and homoscedasticity, biocrust
percent cover and acetylene reduction were log-transformed and EPS content Boxcox-

21

transformed2. Biocrust percent cover, acetylene reduction, and EPS content data were analysed
using R package lme4 [105] to generate LMM fitted by restricted maximum likelihood (REML) and
the R package lmerTest to calculate degrees of freedom through the Satterthwait approximation
[106]. As a post-hoc test, we used Tukey Contrasts from the R package multcomp [107].

Chlorophyll fluorescence data was analyzed using Kruskal-Wallis test and pairwise comparisons
using Tukey and Kramer (Nemenyi) test with Tukey-Distance approximation for independent
samples. nifH gene copy number and soil properties were Boxcox-transformed to meet
assumptions of normality and homoscedasticity of the subsequent analysis using ANOVA and Tukey
HSD post-hoc test. All statistical analyses were conducted in R (R Core Team, 2014).

Results

Bacterial 16S rRNA and fungal ITS2 amplicon sequencing
Two-way Cluster Analysis distinguished two main clusters (Fig. 1) of highly abundant bacterial OTUs
(i.e. those with > 2,000 sequences detected in each sample group) where Cluster A consists of OTUs
detected primarily in Inoculant, Flat, and Microrills sample groups. Cluster B consists of OTUs most
consistently and abundantly detected in the soil surface treatments (Flat, Pit and mound,
Microrills), Uninoculated soil, and the Stored soil sample groups (Fig. 1). Subcluster A1 contains
OTUs that were abundant in Inoculant yet weakly established in the soil surface treatments. This
subcluster included several Acidobacterial OTUs (in families Acidobacteriaceae and genus

𝑥 𝛌 −1

A power transformation such that ×′ =
, see 104.
Box GE, Cox DR (1964) An analysis of
𝛌
transformations. Journal of the Royal Statistical Society Series B (Methodological): 211-252.
2

22

Acidobacterium) OTUs, one OTU assigned to the Chloroflexi family Ktednonbacteracae and one
OTU assigned to the Alphaproteobacterial order Burkholderiales. OTUs in Subcluster A2 were most
abundant in the Flat and Microrills treatments, least abundant in Uninoculated and Stored soils,
and detectable with varying abundances in the Inoculant samples. Subcluster A2 is notable for the
association of several Bacteroidetes-associated OTUs (numbers 36, 38, and 39, Fig. 1) that appear
maximally established in the Microrills treatment. Subcluster B1 consists primarily of OTUs with
highest abundances in Uninoculated and/or Stored soils (Fig. 1), which are also consistently
detected in the soil surface treatments but not the Inoculant sample group. Within this subcluster
are 22 Acidobacterial OTUs that exhibit strong soil associations but also three OTUs from candidate
division WPS-2 that were maximally abundant in the Flat sample group. The Flat, Pit and mound,
and Microrills treatments appear mostly comparable for relative abundance and consistent
presence of subcluster B1 OTUs, which largely explained the clustering together of soil surface
treatments in Fig. 1. The subcluster B2 contains OTUs with highest average relative abundance
Inoculant sample group, and lower abundances in the Uninoculated, Stored, and soil surface
treatments sample groups. This subcluster contains two Acidobacterial OTUs (numbers 11, and 12),
two Chloroflexi-assigned OTUs (41 and 42), and three OTUs assigned to Gammaproteobacterial
family Sinobacteraceae, all of which are exhibit the highest average relative abundance in the
Inoculant sample group. The Flat and Microrills treatments exhibit subcluster B2 OTUs at greater
abundance compared to the Pit and mound treatment. The subcluster B3 consists of two
Alphaproteobacterial OTUs with maximal abundance of detected sequences in the Uninoculated
soils sample group. Overall, Acidobacteria and Proteobacteria had the highest relative abundance
across all treatments, the inoculant and the untreated control soil (Fig. 2). Although the relative
abundance of Cyanobacteria was generally low across our microcosm samples, Cyanobacteria

23

relative abundance was highest in the inoculant (5%) and represented only 1% in the untreated
control soils and 2-3% in the soil surface treatments at week 12 (Fig. 2).

Two-way Cluster Analysis distinguished three main fungal OTU clusters: Cluster A comprises isolates
that are most abundant in the Inoculant, Cluster B most abundant in the soil surface treatments
(Flat, Pit and mound, Microrills), and Cluster C most abundant in soils (Uninoculated and Stored)
(Fig. 3). Subcluster A1 comprises OTUs that were abundant in Inoculant but that failed to establish
in the soil surface treatments. This subcluster included several Zygomycota (Mortierella) OTUs,
Ascomycota (OTUs with affinities to Pleosporales and Lecanoromycetes), Agaricomycetes
(Polyporales, Galerina, Psilocybe, Serendipita) and the basidiomycetous yeast Leucosporidium (Fig.
3). Subclusters A2 and A3 were most abundant in the Inoculant but could also be found in
treatments (A2) or soils (A3) (Fig.3). These clusters were dominated by Serendipita that were able
to colonize the crust treatments but were not present in the soils, as well as several OTUs that were
widely distributed in soil surface treatments samples, including yeast forms (Saccharomycetales,
Cryptococcus), the chytrid Pateramyces, and several Mortierella (Fig. 3). Cluster B contained
subclusters with OTUs that had a highest abundance in the Flat (B1), Pit and mound (B2) or the
Microrills treatments (B3) (Fig.3). This cluster was dominated by Serendipita (18 of the 24
Serendipita OTUs occurred in Cluster B) the majority of which (16 OTUs) appeared to originate from
the Inoculant and only 2 from soil. Other OTUS included Mortierella and Eocronartium (Fig.3).
Cluster C showed highest OTU abundance in the Uninoculated soils (subcluster C1) or the Stored
soil (subcluster C3), with subcluster C2 showing relatively high abundance in both Uninoculated and
Stored soil (Fig. 3). This cluster was dominated by ectomycorrhizal (ECM) OTUs, including
Cortinarius, Inocybe, Sebacina, Laccaria, and Piloderma (Fig.3). Overall, the inoculant also had the
highest relative abundance of Lecanoromycetes in comparison to any other treatment (Fig. 4).
24

Biocrust establishment
Except for the No top soil treatment, biocrust established on all soil surface treatments (Flat,
Microrills, Pit and mound) where it was applied as an inoculant. Percent cover of biocrust was
significantly higher in the Flat, Microrills, and Pit and mound treatments than in the Uninoculated
soils and No top soil treatments at both week 6 and week 12 (Fig. 5a). Both treatment and duration
of incubation (week) had a significant effect on percent cover (F=366.25, p<0.01 and F=46.46,
p<0.01, respectively; Table 1). At week 6, there was no significant difference in biocrust cover
between the soil surface treatments (Flat, Microrills, Pit and mound), however at week 12 the Flat
and Microrills treatments had a significantly higher biocrust cover than the Pit and mound
treatment (Tukey Contrasts, p<0.01). From week 6 to week 12, the Microrills and Flat biocrust
cover increased from 37% ±1.7 to 52% ±1.6 and from 39% ± 3.2 to 52% ±4.3, respectively. Natural
colonization of Uninoculated soils resulted in a 0.6% ±0.3 biocrust cover at week 12.

Maximum photochemical quantum efficiency
The majority of soil surface treatment (Microrills, Flat, Pit and mound) Fv/Fm values increased from
undetectable (i.e. 0) at week 6 to above 0.8 at week 12. Most of Fv/Fm values for the No top soil
treatment and Uninoculated soils were undetectable at week 12 (both 89%). The Fv/Fm values
observed in soil surface treatments at week 12 were significantly higher than the No top soil and
Uninoculated soils (Kruskal-Wallis, and pairwise comparisons using Tukey and Kramer (Nemenyi)
post-hoc test (Chi-squared = 38.7, p< 0.02; Fig. 5b).

25

Nitrogenase activity and nifH gene copy number
The highest rates of nitrogenase activity (NA) were found in the inoculant prior to application (17 ±
3.7 mean µmol of ethylene hr-1 m-2 ± SE) and the lowest rate in the Uninoculated soils (3.5±0.4). We
observed recovery of NA where biocrust inoculant was applied with the exception of the No top soil
treatment (Microrills 11.4±2.2, Pit and mound 10.8±2.33, Flat 9.1±1.9, No top soil 4.9±0.4 mean
µmol of ethylene hr-1 m-2 ± SE) at week 12 (Fig. 5c). Although the linear mixed model fit by REML did
not detect significant differences in NA due to duration of incubation (week), NA increased from
week 6 to week 12 in soil surface treatments (Microrills, Flat, Pit and mound), whereas it declined
in the No top soil and stayed the same in the Uninoculated soils (Table 1).

The general patterns of nifH gene copy number were reflective of trends in NA. The nifH gene copy
number was highest in the inoculant (7.5X106 ± 2.3X106copies/g of soil), lowest in Uninoculated
soils (6.5X105± 1.0X105copies/g of soil) and found at intermediate levels in the soil surface
treatments (Flat 2.6 X106± 1.4X106, Pit and mound 2.0 X106 ± 4.6X105, Microrills 1.3 X106±
4.9X105copies/g of soil) at week 12 (Fig. 6). The inoculant had a significantly higher nifH abundance
than any of the soil surface treatments or Uninoculated soils (ANOVA, Tukey posthoc, F=7.91, all
p<0.04; Fig. 6). There was no significant difference in nifH copy numbers between soil surface
treatments (ANOVA, F=0.912, p=0.45).

Extracellular polysaccharide content
Between week 6 and week 12 EPS content increased markedly in the Microrills soil surface
treatment (16± 5.2 to 25±4.9 µg of glucose per g of biocrust) and Uninoculated soils (4.4±2.3 to
15±2.5 µg of glucose per g of biocrust) (Fig. 5d). A linear mixed model fit by REML showed that
26

week had a significant effect on EPS content, which was clearly driven by changes in the Microrills
and Uninoculated soils (F=9.98, p<0.01; Table 1). Overall, we did not detect significant treatment
effects on EPS content (F=1.63), however, EPS content in Microrills at week 12 (25 µg of glucose per
g of biocrust) was higher than any other treatment.

Soil Properties
Soil samples taken immediately below the surface biocrust were not significantly different across
our microcosm treatments in both dissolved organic carbon and pH following 12 weeks of
incubation (ANOVA, F=0.817, p=0.5 and F=0.409, p=0.676, respectively). Dissolved organic carbon
ranged from an average of 118 ppm to 162 ppm and average pH ranged from 4.6± 0.04 to 4.8± 0.04
with both the highest DOC and pH detected in soils from the Microrills surface treatment.
Mineralizable nitrogen was significantly higher in soils from the Pit and mound (12.1 ppm ±0.7)
compared with both the Flat surface treatment (10.9 ppm ±0.5) and Uninoculated soils (10.8 ppm
±0.5; ANOVA, F=4.36, p=0.01; Table 2). Mineralizable nitrogen was also high in Microrills treatment
(11.7ppm ±0.4) although it wasn’t significantly different from any treatment.

Discussion

Fungal and bacterial communities of inoculated surfaces differ from uninoculated surfaces
and from inoculant
The Inoculant was characterized by several OTUs that were annotated as fungi typically associated
with bryophytes. The Psilocybe OTU was closest (99% identity) to P. montana according to a
GenBank search; P. montana grows among mosses on alpine sites in North America. Galerina is also
typically moss-associated [108]. These species were presumably associated with mosses of the
27

intact soil crusts and were not able to establish in the crust regeneration treatments. Two OTUs
were annotated as Eocronartium and compared most closely to a GenBank isolate JX852332.1 that
was isolated from bryophytes in Antarctica [109]. Eocronatium OTUs were common in the crust
regeneration treatments and may have been associated with bryophytes within the developing
crusts.

OTUs that were annotated as saprotrophic fungi occurred in all samples and treatments. Important
saprotrophic OTUs included Mortierella, a fast-growing member of the Zygomycota that is often
isolated from environmental samples, and basidiomycetous yeasts such as Cryptococcus, and
Leucosporidiella, a member of the Leucosporidiales, an order of psychrophilic basidiomycetous
yeasts commonly found in polar and alpine habitats [110]. The ability to produce mycosporines,
UV-absorbing compounds, is found in a wide variety of basidiomyceteous yeasts [111]. The
presence of these yeasts in biocrusts from alpine environments, which have high UV radiation
exposure, is therefore not surprising. Their presence in the inoculant and the inoculated surfaces
may indicate that inoculation promoted functional recovery of mycosporine-producing yeasts.
Several abundant OTUs were annotated as Dothideomycetes with affinities to the Pleosporales.
Several studies have noted Pleosporales to be dominant members of biological soil crusts [112-114]
and Collins et al. [115] suggested that root endophytes in the Pleosporales may be involved in
nutrient transfer between crusts and associated plants. Our OTUs likely represent these common
crust inhabitants. Most saprotrophic OTUs showed little specificity for inoculant, treatments or soil.

Soils (Uninoculated and Stored) were dominated by typical soil-inhabiting ECM OTUs, particularly
Cortinarius, Inocybe, Laccaria, and Piloderma. In nearly all cases these showed their highest

28

abundance in the Stored soils, with reduced abundance in the Uninoculated soils and very low
abundance in the soil surface treatments, although Laccaria and Piloderma were detected in low
frequency in other treatments. We expect that these fungi were associated with plant roots in the
original soil and likely persisted on the dying root tissue in the soil used to construct the
microcosms. There is little evidence that they were actively establishing in the soil surface
treatments however, due to their declining abundance.

A novel finding of this study was the high proportion of Serendipita (Basidiomycota: Sebacinales)
OTUs in the alpine crust and soil surface treatments. Serendipita belongs to the hyper-diverse and
complex family Serendipitaceae, sister family to the equally diverse Sebacinaceae. While
Sebacinaceae are predominantly ectomycorrhizal and early-diverging saprotrophic fungi,
Serendipitaceae are mostly plant endophytes with derived ericoid, orchid and ectomycorrhizal
lineages [116]. To more closely ascertain the phylogenetic affiliations of our Serendipita OTUs we
generated a maximum-likelihood tree of our OTUs and representative Serendipita sequences
chosen from the major clades (c-h) in Weiß et al. [116]. All of our OTUs were interspersed with
members of subtrees c and g in their tree. Subtree c contains endophytic and ectomycorrhizal taxa
and subtree g liverwort-associated taxa [116]. Although we cannot unambiguously determine the
function of our Serendipita OTUs, the lack of ectomycorrhizal hosts in the Inoculant and restoration
surface treatments suggests that they were likely associated with bryophytes or they were
endophytes of vascular plants establishing in the soil crusts treatments.

Coleine et al. [83] reported that Pezoloma ericae, which forms root associations with plants in the
Ericales and with liverworts, was a dominant member of alpine biological soil crusts in northern

29

Sweden and suggested that its activity as a root symbiont may support the “fungal loop” hypothesis
of Collins et al. [115] which proposes that root-associated fungi, including endophytic species, may
mediate nutrient exchange between soil crusts and vegetation in arid landscapes. It is not
inconceivable that the root-associated Serendipita documented in this study could play such a role
in alpine biological soil crusts. In particular, the potential of root-associated Serendipita to transfer
nitrogen fixed by Cyanobacteria in the crusts to associated plants and liverworts deserves
investigation.

Lecanoromycetes, the main class of lichenized fungi [117], declined in abundance in all treatments,
compared to the initial inoculant, suggesting that this group might be particularly sensitive to
treatment effects or to the modification of environmental conditions. The symbiosis in lichenized
fungi is known to be quite sensitive to changes in environmental conditions [118]. Lichen thalli are
intolerant of prolonged exposure to conditions of saturating water content, which causes the
symbiosis to break down [119, 120]. The moist soils that were maintained under greenhouse
conditions may therefore have favoured growth of free-living Cyanobacteria, instead of lichenized
Cyanobacteria. Under field conditions, lichenized fungi may establish more readily due to naturally
fluctuating environmental conditions. This highlights the importance of water content fluctuation in
biocrust inoculation experiments in order to preserve this important functional group which is a
well-documented component of biocrusts (e.g. [81]). However, lichens are a later successional
species in biocrust [121], and the symbiosis may need a longer period to re-establish. Previous
reports suggest that lichen reestablishment in biocrusts can take several decades after disturbance
[88]. Lichen components are usually associated with thicker biocrusts and Wang et al. [122] found a
time after disturbance to biocrust thickness relationship. The time scale of this experiment
constrains the investigation of successional changes in biocrusts to early sere.
30

In contrast to our findings of marked differences in fungal taxa between treatments, we found less
variation in bacterial communities between restoration treatments. The two-way cluster analysis
did reveal that bacterial OTUs highly abundant in biocrust (Inoculant) were most effectively reestablished by the Flat and Microrills treatments, while soil-associated bacteria were detectable to
varying extents (abundances) in all of the soil surface treatments with biocrust after 12 weeks of
incubation. The prevalent bacterial phylum in all treatments and the inoculant was the
Acidobacteria, followed closely by Proteobacteria (data not shown). To our knowledge, no other
studies on biocrusts have reported a bacterial assemblage dominated by Acidobacteria; although
Kuske et al. [123] reported 51% of Acidobacteria-like sequences in arid soils and Nagy et al. [124]
Acidobacteria as being a dominant (relative abundance of 11%) and diverse component of biocrusts
of the Sonoran Desert, Arizona. The composition of the Acidobacteria clade is known to vary when
associated with biocrusts [125]. Although Acidobacterial ecological function remains poorly
defined, they are likely drivers of biocrust function, contributing to aggregate formation possibly
through EPS production, and tending towards being abundant community inhabitants in low
resource soils [126]. Our two-way cluster analysis also indicates that there might be distinguishable
soil versus biocrust associations of Acidobacterial taxa, which may indirectly be related to different
ecosystem services performed within this phylum along a vertical profile from biocrust to
underlying material.

While some studies [80, 127] report a higher relative abundance of Proteobacteria, they also report
higher Cyanobacterial abundance, in natural biocrusts when compared to below-crust soils. In
contrast, a study on biocrust of temperate climate. Cyanobacteria are often seen as being the

31

precursor to microbial diversity in biocrusts [128] as their morphology and physiology provide a
suitable environment for other microbes to proliferate [129]. Contrary to most studies reporting
Cyanobacteria as the dominant biocrusts bacterial phyla [82, 130], reportedly up to 40% [130],
representatives of this phylum were found in relatively low abundance in our samples (data not
shown). Cyanobacteria are likely limited by acidic pH since they lack the ability to control their
internal pH [131]. Although some found that Cyanobacteria growth was limited by pH in the range
of 5.5 to 6 [132] or 4 to 5 [131],Dominic and Madhusoodanan [133] found Cyanobacteria in peaty
bog lands at pH below 4. Nonetheless, they found decreasing Cyanobacteria abundance with
increasingly acidic pH. This suggests that the low Cyanobacterial abundance in our samples may be
a result of the acidic pH, 4.6 on average, of the experimental alpine soils. Also, the Cyanobacteria
family Nostocaceae-classified OTUs found in our biocrust-containing samples are almost completely
absent in the soils, suggesting that these taxa are specifically inoculant-associated, which is
consistent with the presence of this phylum of bacteria in other well-studied biocrust systems [26,
81].

Differences in bacterial composition and Cyanobacteria abundance between our study of an alpine
temperate environment and others of hot arid deserts may also be linked to differences in climate,
geochemistry and primary producers [134]. Most of the current work [43, 89, 135] is focused
towards Cyanobacteria, which are seen as the precursor to more diverse biocrusts mostly from arid
lands and used to start ecological processes in disturbed lands. However, there is growing evidence
that other bacterial species can also be important early colonizers depending on environmental
conditions and soil geochemistry. Diazotrophs other than Cyanobacteria were likely abundant in
our samples as evidenced by nifH detection using Rhizobium-specific primers, and due to the
abundance of Proteobacterial sequences. Yeager et al. [65] report up to 10% of nifH sequences
32

potentially from Proteobacteria in a study of biocrusts of the Colorado Plateau. Nagy et al. [124]
and Gundlapally and Garcia-Pichel [124, 136] also found Oxalobacteraceae (Betaproteobacteria) to
be common members of the diazotrophic communities of Sonoran and Colorado Plateau
respectively. Our results demonstrate that alpine biocrusts have a unique composition; with their
diazotrophic community dominated by taxa different from Cyanobacteria possibly consisting of
bacteria from the families Acetobacteraceae and Oxalobacteraceae (taxonomic affiliations for OTUs
in our dataset for Inoculant samples, which were not included in the two-way cluster analysis).
More research is needed to concretely identify members of and characterize the functioning of
microbial diazotrophs in alpine biocrusts.

In addition to Proteobacteria, other “novel” prokaryote groups [124, 136] were found in our
samples, including Chloroflexi, for which three Inoculant-associated OTUs were within our
subsample of highly abundant taxa (Fig. 1) for hierarchical clustering analysis. The exact role
Chloroflexi play in communities is unknown but they are potentially comprised of at least some
photoautotroph clades [137]. It has been suggested that their distribution may be associated with
Cyanobacteria [137]. Chloroflexi and Cyanobacteria have also been found in assemblages in hyper
saline mats and hot springs [138].It has been hypothesized they utilize different light wavelengths
to photosynthesise and the shorter wavelengths, with deeper penetrating power, could be utilized
by Chloroflexi deeper in soils [137]. Others suggest that the potential for anaerobic photosynthesis
from Chloroflexi is limited to the penetration depth (3mm) of incident light [127, 139]. While we
cannot infer that inoculant-associated Chloroflexi are functioning as photobionts in the biocrusts
we studied, the class-level taxonomy (Ktedonobacteria) of correlating OTUs is not associated with
photosynthetic activity [140]. More research is needed to understand the role of Chloroflexi play in
biocrusts and their rehabilitation.
33

Overall, bacterial and fungal communities of the inoculant were distinct from the Uninoculated
soils used in this microcosm experiment. Analysis of bacterial 16S rRNA and fungal ITS2 sequences
by hierarchical clustering analysis provided further confirmation that 12 weeks was insufficient to
establish a microbial community composition similar to that of the inoculant. However, the
observed changes in community composition following inoculation suggested that inoculation
exerted a strong influence on community composition. Whether these shifts are reflective of
natural restoration trajectories and whether they will lead to the similar late-seral soil crust
communities is unknown. Furthermore, even if the induced community composition differs from
that of natural communities, ecological function may remain similar. Further compositional
characterization of early seral biocrust communities in alpine environments and an improved
understanding of how changes in community composition drive ecological function are essential for
examining soil development and the efficacy of biocrust for restoration.

Inoculation and microtopography facilitate biocrust establishment
Soil surface treatment and inoculation facilitated biocrust establishment. Microtopography is
generally regarded as a factor enhancing the rehabilitation of biocrusts on disturbed sites [141]. We
found that biocrust cover was high in both the Flat and Microrills treatments. Nonetheless at a
small scale microtopography was still an important variable within our treatments. In the Microrills
treatment, for instance, we observed higher biocrust colonization in the gullies than in the ridges.
These results demonstrate the influence of moisture on biocrust development. Li et al. [22] found
that microtopography had a strong influence on water and material re-allocation and thus
influenced biocrust community composition notably by increasing pH which was correlated to

34

increased cyanobacterial abundance. Similarly, Davidson et al. [72] found that microtopography
had a strong influence on soil surface microclimate and nutrient availability, which, in-turn, was a
major factor influencing the development of lichen communities within biocrusts. Since Microrills
create an elevated moisture level in gullies, when combined with inoculation, they represent a
practical restoration tool to facilitate biocrust establishment and function recovery.

Inoculation and soil surface treatment promote function recovery
Overall, functional recovery, represented by nitrogen fixation potential, was highest in soil surface
treatments and lowest in the Uninoculated soils. Our results suggest that artificial inoculation with
biocrusts increased soil surface nitrogen-fixation rates. Similarly, Jeffries et al. [23] found that
algae-moss biocrust inoculation increased NA on copper mine tailings, and Maestre et al. [44] that
slurry increased NA in a microcosm experiment. Stewart and Siciliano 2015 [74] found that biocrust
could successfully establish on Yukon mine tailings and mining impacted soils. Others (e.g. [73,
142]) found that artificial inoculation of soils with biocrust was an effective way of increasing
diazotrophic community abundance.

Microrills were the most effective treatment in restoring NA, and this trend was also reflected in
the nifH gene copy number. The non-inoculated surfaces (Uninoculated and Stored) had the lowest
nifH abundance and NA rates, while the highest nifH quantity, NA values and 16S sequence counts
were observed in the inoculant. The lack of a significant difference between soil surface treatments
may be due to the high variability of nifH copy numbers. qPCR does not distinguish between viable
and non-viable genes, which can obscure differences in viable gene copy numbers [143].
Furthermore, expression of a gene does not unequivocally imply that the enzyme is active [144,
35

145]. Despite these limitations, nifH has been shown to be largely consistent with 16S rRNA gene
data and the study of nifH abundance can be used with nitrogen fixation data to understand
linkages between community structure and function [65, 143]. The soil alone hosts a reduced
population of nitrogen-fixers when compared to inoculated surfaces and therefore, the use of
inoculant may effectively increase the diazotroph community of restored soils. Our results
demonstrate that both nifH copy numbers and NA increased over time, suggesting that biocrust
inoculation is effective at restoring nitrogen-fixation capability.

Inoculation of soil surface clearly facilitated the recovery of chlorophyll fluorescence. Maximum
photochemical quantum efficiency of PSII (Fv/Fm) went from undetectable to higher than 0.8 in soil
surface treatments, whereas it stayed mostly undetectable in Uninoculated and Stored soils. Fv/Fm
values higher than 0.8 correspond to the maximal efficiency for plants. The values associated with
Cyanobacteria-dominated biocrusts are reportedly between 0.4 and 0.7 [146]. A limitation of Fv/Fm,
cyanobacterial chlorophyll fluorescence is masked by concurrent values from plants, which are
naturally higher [147]. Hence, it is uncertain if cyanobacterial chlorophyll fluorescence was present
in the inoculated surfaces but chlorophyll fluorescence from plants was undoubtedly present. The
chlorophyll fluorescence likely originated from the bryophytes colonizing the inoculated surfaces at
week 12, which was accompanied by an increase in percent cover.

Inoculation has limited effect on EPS and soil properties after 12 weeks
The increase in EPS content between weeks 6 and 12 was also significant in the Microrills
treatment. Together with significantly higher nitrogen rates, these results are consistent with
biocrust functional recovery following inoculation. EPS content is particularly relevant to measuring
36

early biocrust function since its compounds facilitate microbial life, notably by offering protection
against desiccation [58, 66, 148]. EPS content can therefore be associated with increased functional
recovery such as the higher NA we observed in the Microrills treatment. Similarly, in a study on
induced biocrusts in China, Colica et al. [149] found that EPS content at an 8 year old site was
significantly higher than at 3 and 5 year old sites and that the increase could be attributed to
recruitment of EPS producing organisms since it was correlated to higher microbial abundance
[150]. The short duration of our study could explain the apparent homogeneity between
treatments for the EPS content, as well as, for the dissolved organic carbon and the mineralizable
nitrogen levels. However, since we observed higher photosynthetic activity combined with higher
nitrogen-fixation rates in the Microrills we believe that after a longer period these additional
carbon and nitrogen inputs could be reflected in the EPS level and soil properties. The higher
mineralizable nitrogen level detected in our soil surface treatments when compared to the
Uninoculated soils may indicate that inoculation increases nitrogen availability in soils. This is
consistent with other studies that found biocrust to be positively correlated with mineralizable
nitrogen levels in soils [47, 85, 91, 151, 152].

The increase in EPS content over time may indicate an increase in EPS producing organisms on
inoculated surfaces. However, it cannot be excluded that these changes may have been caused by
changing rates of EPS production reflecting changing environmental conditions. The production of
EPS by cultured Cyanobacteria is known to be influenced by factors such as C: N ratio, temperature,
light intensity, pH, and nutrient availability, however the mechanisms by which these factors
influence EPS production varies between Cyanobacterial strains and are largely unknown [153]. The
moisture, temperature and illumination of our microcosms may have been more favorable to EPS
production than field conditions which could explain the sharp increase in EPS from week 6 to week
37

12 in the Uninoculated soils. However, it may also be the result of an increase in EPS producing
organisms present in the soil such as fungi or free-living bacteria.

Conclusion
We demonstrated that inoculation with biocrust was an effective means of promoting the recovery
of primary ecosystem function, notably by facilitating nitrogen fixation. Biocrusts can also support
soil stabilization by EPS secretion, contributing to the establishment of vascular plant communities.
Although bacterial communities were relatively unchanged after inoculation, relative contributions
from the fungal communities to biocrust formation reflect compositional differences between
Inoculant and soil samples. A novel finding was the high number of Serendipita OTUs associated
with biocrusts; further research is needed to determine their role. Non-cyanobacterial diazotroph
and other alpine pioneering species need to be the focus of research to develop a better
understanding of soil development and effective management of disturbed alpine sites.
Determining linkages between community changes and ecosystem functions at the soil surface is
paramount in understanding the role of biocrust in ecosystem recovery and development.

38

3. Restoration of ecosystem function by soil surface inoculation of
biocrust in mesic and xeric alpine ecosystems
Introduction
Anthropogenic disturbances can have severe impacts on alpine vegetation and soils [3-5] causing
decreases in plant and microbial diversity and the alteration of soil organic carbon and nitrogen
cycling processes. [1, 75, 83]. The recovery of alpine environments after disturbances is commonly
limited by nutrient poor-soils and short cool growing seasons [1]. One of the key components in
the recovery of alpine soils after disturbance are biocrust, ubiquitous soil surface communities
formed by a diversity of organisms such as algae, lichens, mosses, liverworts, Cyanobacteria, as well
as other primary successional species [12-14]. Established biocrust communities play a key role in
soil carbon and nitrogen cycling through processes such as the production of extracellular
polysaccharides (EPS) and nitrogen fixation [54, 86, 87]. Although biocrust are sensitive to
disturbances [80, 88, 89, 154] they are well-adapted to harsh alpine growing conditions [90] and
can facilitate the recovery of ecosystem services such as nitrogen and carbon cycling [91-93].
Given that biocrust establishment can initiate restoration of soil surface properties (e.g. soil
stabilization, nitrogen and carbon enrichment) following disturbance, the artificial enhancement of
biocrust establishment on disturbed sites has been the subject of a number of studies, using
techniques such as cultivation and/or inoculation [26, 70, 71, 79]. While previous studies have
examined the efficacy of inoculation for establishing biocrust in grassland and desert sites [44, 73,
80], to our knowledge none have investigated the effects of assisted inoculation on alpine biocrust
establishment and function.

39

We address this knowledge gap in a study of alpine biocrust in two contrasting sites: the mesic
coastal mountain ranges of northwestern British Columbia (Terrace, British Columbia, Canada), and
the xeric interior mountain ranges of the southern Yukon (Whitehorse, Yukon Territory, Canada).
At each location, we examined the development of soil biocrust after disturbance in controlled
experimental plots. Experimental treatments included plots with and without biocrust inoculation,
and plots with and without fertilizer addition. This allowed us to investigate site-specific potential
for biocrust restoration. Soil surface microclimate was monitored, and nitrogen fixation was
measured at each site to provide a better understanding of the operating conditions under which
physiological activity occurs in these contrasting biocrust communities. Our main objectives were 1)
to compare the composition and nitrogen fixation potential of natural biocrust associated with
contrasting coastal and continental mountain ranges; 2) to assess the restoration potential of
alpine biocrust communities; and to 3) to evaluate the influence of biocrust inoculation treatments
on nitrogen fixation, extracellular polysaccharide content, soil mineralizable nitrogen and dissolved
organic carbon, as key indicators of soil ecological function.

Methods

Study areas
Coastal alpine plots (mesic) were situated on Andesite Peak (lat 54° 13.868’N, long 128° 01.499’W,
1640 m elevation) in the Coast Mountain range near Terrace, British Columbia. Mean annual
precipitation in Terrace (at 706 m elevation) is 970 mm with roughly 40% falling as snowfall [155].
Predicted mean annual temperature and mean annual precipitation on Andesite Peak is -1 °C and
2019 mm respectively [156]. Vegetation communities on Andesite Peak were dominated by alpine
tundra with a mixture of rocky outcrops and late-melt snow beds in depressions. Vegetation on dry
40

ridge top habitats is dominated by Cassiope mertensiana (Bong.) G. Don and Stereocaulon alpinum
Laurer heath, while Saxifraga tolmiei Torr. & A. Gray and Ranunculus cooleyae Vasey & Rose
dominated vegetation communities in wetter mid-slope habitats. This site is characterized by
Podzolic and Regosolic soils [157], with substrates of volcanic origin (meta-basalts and metaandesites) [158].

Continental alpine plots (xeric) were located on Mt McIntyre (lat 60° 38’07.0” N, long 135° 11’59.8”
W, 1370 m elevation) on the Teslin Plateau near Whitehorse, Yukon, Canada. Mean annual
precipitation in Whitehorse (at 706 m elevation) is 262 mm with roughly half as snowfall [159].
Vegetation at the Mt McIntyre site is dominated by low growing shrubs (e.g. Salix spp., Vaccinium
spp., Dryas spp.), herbs (e.g. Chamerion angustifolium L., Anemone spp.) as well as bryophytes and
lichens (e.g. Stereocaulon spp., Cladina spp., Cladonia spp.). Located in the Mid-Cordilleran Alpine
ecoclimatic region of Yukon, the area is characterized by sedimentary rocks (clastic/limestone) and
Brunisolic soils [159].

Microclimate monitoring of natural biocrust
At each site, a microclimate monitoring station was installed. At Andesite Peak, a CR1000 data
logger (Campbell Scientific Inc., Logan, UT) was coupled to a AM416 multiplexer ( Campbell
Scientific Inc., Logan, UT), fitted with a HPM45 temperature probe (Campbell Scientific Inc., Logan,
UT), two EC-5 soil moisture/temperature probes (Decagon Devices Inc., Pullman, WA), two LiCor
Photosynthetically Active Radiation (PAR) sensors (LiCor Inc., Lincoln, Nebraska), twelve Omega
0.001 mm copper/constantan (Cu/Cn) thermocouples (Omega Engineering, Stamford, Connecticut)
inserted at the surface of mature biocrust (top 5 mm), as well as, twelve pairs of impedance clips
[160] attached to mature biocrust. At Mt McIntyre, a CR23 data logger (Campbell Scientific Inc.,

41

Logan, UT) and a CR1000 data-logger (Campbell Scientific Inc., Logan, UT) coupled to an AM25T
multiplexer (Campbell Scientific Inc., Logan, UT) were installed and fitted with the same probes and
sensors as at Andesite Peak. The Andesite Peak microclimate monitoring station was located in an
open alpine-tundra habitat, with no shrub cover. The Mt McIntyre microclimate monitoring station
was located in a sparse shrub- alpine tundra complex. Consequently, soil surface probes were
partially shaded from direct solar input at Mt McIntyre.

Natural biocrust cover assessment
Natural biocrust cover was estimated by conducting a survey along transects (60m) at Andesite
Peak and Mt McIntyre respectively. At 5m intervals, 1m2 plots were visually assessed for soil
surface cover. The locations of transects were selected to represent the various slope/aspect
combinations present at each site.

Field experiment design
Four treatments were applied to artificially disturbed plots in a randomized blocked design with 9
and 10 blocks at Andesite Peak and Mt McIntyre, respectively. Each block was located in an area
with uniform topography (slope / aspect) and where biocrust was present and vegetation was
sparse. Each treatment was applied within a 1 m2 area that was randomly assigned within each
block with a buffer of 0.5m between treatment plots. The 4 treatments were biocrust with fertilizer
(BF), biocrust only (B), fertilizer only (F), and a control (C) with no biocrust or fertilizer applied.

42

Biocrust used to prepare the inoculation mixture for restoration plots was collected from within the
areas delineated for disturbance treatment plots. All biocrust within each treatment plot was
removed to a depth of 2.5 cm, comprising ca. 0.5 to 2 cm of biocrust and 0.5 cm of underlying soil.
Additionally, any vegetation present in the treatment plots was removed. Collected biocrust was
sieved (to 4.5 mm) and homogenized. Once biocrust was removed from all the plots, the soil
surface was disturbed by raking and cultivating to a depth of 3 cm. Homogenized biocrust was then
applied within 1 m2 plots for B and BF treatments (n=18 and n=20 for Andesite Peak and Mt
McIntyre, respectively) at a rate of 10% surface area (i.e. 0.1 m2 per plot). The F and C treatments
(n=18 and n=20 for Andesite Peak and Mt McIntyre, respectively) did not receive biocrust
applications. Subsequently, 30-30-30 fertilizer was added to the BF and F treatments at a rate of
110kg/ha. This application rate was selected because it represents industry standard in restoration
work.

Soil surface samples were collected for analysis of biocrust recovery within experimental plots at
weeks 0, 6 and 12 after the start of the experiment. A composite sample (64cm20.5-1cm depth) was
collected from each treatment plot. Samples were packed in coolers for transport back to the
laboratory, where they were subsequently stored in a dark refrigerator at 2°C prior to Acetylene
Reduction Assays (ARAs). Stored biocrust was transferred to a growth chamber for pre-treatment
(average summer conditions:16 hr photoperiod at 250 µmol m-2 s-1 during light hours; 20oC
daytime, 10⁰C nighttime) for 24 hrs prior to ARAs.

43

Natural biocrust sampling
Naturally occurring biocrust in undisturbed alpine tundra was sampled to provide an assessment of
seasonal changes in mature biocrust nitrogenase activity. Mature biocrust was collected from
alpine tundra adjacent to the experimental plots at the same time points as the experimental plots
were sampled (i.e. 0, 6 and 12 weeks after the start of the experiment). At each site and at each
sampling time point, 10 samples of undisturbed mature biocrust (0.5cm to 2cm depth) were
collected from eight harvest locations. The samples were representative of biocrust composition at
each site. The nitrogen fixation potential of each mature biocrust sample was assessed through
ARAs following a pre-treatment of the biocrust for 24 hrs under optimal conditions (8hrs night at
15°C and 16hrs at 20°C and 350 µmol).

Acetylene-reduction assays
The nitrogen fixation potential of soil surface samples (from mature biocrust and experimental
restoration plots) was assessed using acetylene reduction assays (ARAs) [97]. Samples were placed
in 250 ml glass canning jars with a serum stopper lid. Samples were misted with 1mL of deionized
water prior to closure of jars and injection of 10% acetylene by volume. Jars were incubated 4 hr at
150 µmol m-2 s-1 PAR and 20oC. After incubation a 4-mL gas sample was removed from each
incubation jar for analysis of ethylene concentration. Each set of samples were processed with two
control samples: i) a biocrust sample not injected with acetylene, which served as a temperature
control and ensured that no natural evolution of ethylene occurred and ii) a jar with no biocrust
that was injected with acetylene to ensure the absence of ethylene contamination in the acetylene.
Gas samples were injected into a gas chromatograph (SRI 8610A, Wennick Scientific Corporation,
Ottawa, ON, Canada) fitted with a Porapak column (Alltech Canada, Guelph, ON, Canada) and a
flame ionization detector for detection of ethylene. Hydrogen, used as the carrier gas, was held at a
44

constant pressure of 32psi while column temperature was held at 65⁰C. After ARAs were
completed, samples from week 0 and 12 collections were homogenized and kept frozen at -20°C,
for subsequent EPS analysis.

Extracellular polysaccharide content
Extracellular polysaccharide (EPS) content was measured on samples from experimental plots taken
at week 0 and at week 12 (n=76 and 80 for Andesite Peak and Mt McIntyre respectively). EPS
content was determined using the phenol-sulfuric acid method [100, 101]. Reactions were
transferred to a 96-well flat bottom microplate, with standards and template control carried out in
triplicate. EPS content was measured as glucose concentration quantified at 480, 486 and 490 nm
using a microplate reader (Varioskan, Thermo Fisher Scientific, Walthman, MA).

Soil properties
Dissolved organic carbon and mineralizable nitrogen were measured on soil samples collected at
week 12 on composite samples obtained from each plot (n=36 and n=40 for Andesite Peak and Mt
McIntyre, respectively). Soil was sampled immediately below surface cover to a depth of 3 cm.

Dissolved organic carbon was measured from a soil extract (1:10 soil to water ratio shaken for 1 hr
on a reciprocating shaker) filtered to 0.45 µm, with a total organic carbon analyzer (Formacs HT,
Skalar Analytical B.V., Breda, Netherlands).

45

Mineralizable nitrogen was measured by the anaerobic incubation method [102]. Soil samples were
incubated under water-logged conditions for two weeks at 30°C. Ammonium produced was
subsequently measured from a soil extract (1:10 soil to 1N KCl ratio shaken for 2 hr on a
reciprocating shaker) with a discrete analyzer using the phenate method for colorimetric
determinations ([103]; SmartChem 200, Unity Scientific, Brookfield, CT).

Data analysis
Landscape level nitrogen fixation estimates were calculated for Andesite Peak data. These were
derived by multiplying estimates of biocrust cover by modelled values for ARA rates per m2 crust
cover during the field experimental period. The modelled values for ARA were developed from
incubations conducted under controlled conditions of light and temperature on mature biocrust
harvested from Andesite Peak. Acetylene reduction was measured at a combination of two light
levels (low-120 µmol and high- 450 µmol) and 3 temperatures (5, 15 and 20C°) for biocrust held at
optimal water content. A linear regression model of ARA as a function of temperature was
subsequently determined. ARA response to changes in crust water content was assessed under
optimal temperature and light conditions (20°C and 350 µmol), where crusts were sequentially
exposed to a desiccation series of moisture contents ranging from fully saturated to totally
desiccated. A regression model for ARA as a function of moisture content was subsequently
calculated. Field rates of ARA activity were estimated at hourly intervals by using microclimatic
data and ARA models. Each hourly record was classified as being a low or high light exposure for
crusts and the corresponding light/temperature regression equation was used to calculate potential
ARA rates. These estimates of potential ARA activity were then corrected for actual crust water
content, using the moisture content model. Application of the moisture content model assumed

46

that a similar moisture dependence of ARA activity would apply at other temperature/light
combinations (see discussion below). Finally, landscape-level ARA estimates were calculated based
on the summation of estimated ARA rates for each hourly period and landscape level biocrust cover
measurements.

Linear mixed models (LMM) were used for the analysis of ARAs and EPS content. Since samples
were collected from the same plots at multiple time points, pseudoreplication within plots was
avoided by including plot number as a random effect. LMM included plot number as a random
effect, week and measurement (i.e. ARAs and EPS content) as fixed effects, and the interaction of
week and measurement as fixed effects. To meet assumptions of normality and homoscedasticity,
all measurements except dissolved organic carbon content were log-transformed. Acetylene
reduction, and EPS content data were analysed using R package lme4 [105] to generate LMM fitted
by restricted maximum likelihood (REML) and the R package lmerTest to calculate degrees of
freedom through the Satterthwait approximation [106]. Post-hoc testing was carried out with
Tukey Contrasts from R package multcomp [107]. Soil properties and biocrust percent cover were
analyzed using ANOVA and Tukey HSD post-hoc tests. All statistical analyses were conducted in R (R
Core Team, 2014).

Results

47

Characterization of alpine biocrust
Community composition
On Andesite Peak, 40 lichen, 12 liverwort and 17 moss species were identified in mature biocrust.
Amygdalaria panaeola, Psoroma tenue var. boreale, Solorina crocea, as well as Stereocaulon
alpinum, botryosum, and vesuvianum formed the nitrogen fixing cyanolichen biocrust community.
On Mt McIntyre, 57 lichen, 3 liverwort and 21 moss species were identified in mature biocrust.
Nitrogen-fixing lichens were Fuscopannaria praetermissa, Nephroma arcticum, Peltigera spp.,
Psoroma tenue var. boreale, and Stereocaulon alpinum and tomentosum. Two nitrogen-fixing
lichens were found at both sites (Psoroma tenue var. boreale and Stereocaulon alpinum), as well as
11 other lichens, 2 liverworts (Anastrophyllum minutum var. minutum and Tritomaria
quinquedentata) and 3 mosses (Dicranoweisia crispula, Pogonatum contortum, Polytrichum
piliferum) species (Table 3).

Microclimate
At Andesite Peak, cumulative daily photosynthetically active radiation (PAR) showed a strong
seasonal pattern, declining from over 55 mol m-2 day-1 in early summer to less than 25 mol m-2 day-1
by mid-September (Fig.7). Mt McIntyre has a similar pattern of decline, from high early summer
PAR values (25 mol m-2 day-1) to low early fall (12 mol m-2 day-1) (Fig. 8). Overall lower solar input
was observed at the Mt McIntyre site.

Mean daily biocrust temperatures were consistently higher than mean daily air temperatures both
at Andesite Peak and Mt McIntyre (on average by 2.5 and 1.4°C respectively). At Andesite Peak,
biocrust temperatures reached a maximum of 40.6°C on 6 July and a minimum of -1.6°C on 14
48

September. At Mt McIntyre, maximum biocrust temperature was 36.3°C on 25 June, and minimum
biocrust temperature -4.0°C on 31 July. Biocrust temperatures below 0°C were recorded only once
at Andesite Peak, on 14 September. In contrast, at Mt McIntyre, sub-zero temperatures were
observed on 16 days over the summer. Overall, the ranges of biocrust temperatures were similar at
both sites with Mt McIntyre biocrust reaching lower temperature minima and Andesite Peak higher
temperature maxima despite a higher mean air and biocrust temperatures at Mt McIntyre.

Mean air temperature was higher at Mt McIntyre (9.1°C, 4 June – 24 August 2015) during the study
period compared to at Andesite Peak (6.8°C, 10 June – 14 September 2015). Typical summer
conditions at Andesite Peak included mean nighttime (21:00-8:00) air and biocrust temperature at
5 and 6°C respectively with mean daytime (9:00-20:00) temperatures increasing markedly in
biocrust (14°C up to a maximum of 25°C) compared to air (8°C up to a maximum of 13°C) (Fig. 8).
Fluctuating wet and dry periods were also typical of Andesite Peak microclimatic conditions (Fig.7).

The total duration of hydration episodes (defined as biocrust relative water content (RWC) >25%)
for biocrust at Mt McIntyre (907 hrs) was considerably shorter than that of Andesite (1589 hrs),
although the mean temperature of hydrated crusts was similar, at 9.4 and 12.1°C respectively.
Overall biocrust and surface soils (i.e. 5 cm depth) were drier at Mt. McIntyre compared to
Andesite Peak.

Mean volumetric soil water content was 26.6% at Andesite Peak and 21.0% at Mt McIntyre. RWC
was on average 62.9% at Andesite Peak and 30.6% at Mt McIntyre. At Mt McIntyre, biocrust water

49

content was lowest in June (on average 22.0% RWC) with no drought events occurring over the
monitoring period. At Andesite Peak, microclimatic data representing an extended period of
desiccation was observed between 8- 11 July followed by rewetting (Fig.7). The drought period had
daily high temperatures between 35 and 40°C and corresponding biocrust water content at or near
0%. Subsequently, daily high temperatures dropped to 15- 22°C and upon rewetting biocrust
relative water content increased up to 100%.

Nitrogen fixation
The response of acetylene reduction in biocrust from Andesite Peak was characterized against
changes in temperature, light, and moisture availability. Acetylene reduction rates in biocrust were
strongly correlated with gravimetric water content as shown by the regression model for acetylene
reduction as a function of moisture (R2=0.59) (Supp. Fig. 1). The highest acetylene reduction value
(9.8 µmol ethylene m-2 hr-1) was measured at 450% moisture content while minimal acetylene
reduction (<1 µmol ethylene m-2 hr-1) was detected at moisture contents below 100%.

Acetylene reduction was also strongly correlated with temperature, in both low and high light
linear regression models (R2=0.32 and 0.42, respectively) (Supp. Fig. 2). At all three temperatures,
acetylene reduction rates were consistently higher under high light than under low light.
Consequently, the maximum acetylene reduction rate (2.3 µmol ethylene m-2 hr-1) was obtained
under high light at 20°C.

50

Landscape nitrogen inputs
The landscape level cover of biocrust, estimated from the transect plots was 22.4±2.8% SE at
Andesite Peak. The application of the acetylene reduction models to the biocrust microclimate
data (PAR, biocrust moisture and temperature) yielded a modelled daily acetylene reduction rate of
2.1±0.1 X 106 mol of ethylene/ha. Multiplication of this modelled rate by the landscape percent
cover of biocrust yielded an estimated daily acetylene reduction rate of 4.8±0.3 X 105 mol of
ethylene/ha. A theoretical conversion ratio of three to one for ethylene reduced to nitrogen is
commonly used. However, we chose to leave our data as mol of ethylene because of the wide
range of conversion ratios reported for terrestrial species (0.1-6) along with a strong dependency
on environmental conditions [161]. Although landscape nitrogen inputs were not modelled for Mt
McIntyre because of time and budget constraints, the estimated landscape level cover of biocrust
was 23.0±3.4%SE.

Recovery of inoculated biocrust and associated ecosystem functions
Biocrust establishment
Treatment plots receiving biocrust inoculation (B and BF) had significantly higher biocrust cover
after 12 weeks (33% ±3 and 32% ±2 at Andesite Peak and 22% ±2 and 20% ±3 at Mt McIntyre; Table
4, Fig. 9). Addition of fertilizer with the biocrust inoculant did not significantly increase biocrust
cover at either site. Similarly, fertilization alone did not significantly increase natural colonization.
Fertilized plots (F) and control soils (C) had a 3% ±1 and 10% ±3 biocrust cover at Andesite Peak and
8% ±1 and 7% ±0.1 cover at Mt McIntyre. There was a significant interaction between site and
treatment (Table 4). Inoculated plots (B and BF) had higher cover at Andesite Peak compared to Mt
McIntyre, however biocrust cover of uninoculated plots (C and F) was not significantly different
between sites (Fig.9).
51

Site differences in acetylene reduction
Nitrogenase activity (NA) changed significantly over time at both sites; however, opposing trends
were observed (Table 4, Fig.10). At Mt McIntyre, NA declined significantly from week 0 to 12,
whereas at Andesite Peak, NA increased significantly from week 0 to 12 (Fig. 10). NA was
significantly higher at Andesite Peak compared to Mt McIntyre.

Effects of fertilization and inoculation on functional recovery
Extracellular polysaccharide (EPS) content was significantly higher in Mt McIntyre biocrust samples
(Fig. 11, Table 4). There were no significant treatment or time difference at Andesite Peak;
however, at Mt McIntyre, 12 weeks following inoculation, the fertilizer treatment (F) had a
significantly lower EPS (Fig.11, 76±19 µg of glucose per g of biocrust) compared to all other
treatments (Fig.11, B = 250±92, BF = 217±49, C = 170±42 µg of glucose per g of biocrust,
respectively).

Twelve weeks after inoculation, soil mineralizable nitrogen (Min N) measured in the top three cm
below crusts was significantly lower at Andesite Peak than Mt McIntyre, but not significantly
different between treatments at either site with the exception of BF being significantly higher than
B at Andesite Peak (Fig.12, Table 4). Conversely, dissolved organic carbon (DOC) was significantly
higher at Andesite Peak than at Mt McIntyre (Fig.12, Table 4). At Andesite Peak, DOC was
significantly higher in inoculated plots (B = 75±4.9 and BF = 122±14 ppm) than in uninoculated plots
(C=9.2±1.3 and F=45±5.8 ppm). Higher DOC was observed in both inoculated (BF>B) and

52

uninoculated plots (F>C) that received the fertilizer treatment. Fertilization increased DOC in both
the inoculated and uninoculated plots.

Discussion

Characterization of alpine biocrust
Community composition
On Andesite Peak, biocrust covered 22% of the alpine tundra surface. Liverworts such as
Marsupella brevissima, Pleurocladula albescens, and Ptilidium ciliare were abundant components of
biocrust communities on Andesite Peak. Nitrogen fixing lichens were also common, especially
Solorina crocea, Stereocaulon alpinum, S. botryosum, and S. vesuvianum. On Mt McIntyre biocrust
accounted for 23% of soil surface cover and were dominated by lichens, including nitrogen-fixing
species such as Stereocaulon alpinum, S. tomentosum, and Peltigera spp. Many of these species,
such as the liverworts Anastrophyllum minutum and Marsupella brevissima, and Tritomaria
quinquedentata, and lichen Psoroma tenue are found in both arctic and alpine soil crust
communities [162-164]. Others, such as the liverwort Cephaloziella divaricate, moss Dicranoweisia
crispula and lichen Placynthiella icmalea share affinities with grassland and desert soil crust
communities [165-167].

Microclimate
There was a clear contrast between the soil surface microclimate under the continental climate at
Mt McIntyre and the coastal climate at Andesite Peak. In the Coast Ranges, regular periods of
precipitation during the summer combined with abundant spring snowmelt resulted in prolonged
53

periods of biocrust hydration. In contrast, in the xeric continental site, seasonal moisture
availability for biocrust was derived mainly from spring snowmelt, with biocrust becoming
progressively drier in mid- to late-summer.

Contrary to the expectation that cumulative PAR exposure would be higher in biocrust communities
on Mt McIntyre, due to its higher latitude and more cloud-free days, daily biocrust PAR exposure
was consistently higher for the Andesite Peak crusts compared to Mt McIntyre. This discrepancy
was likely due to the more abundant shrub cover at Mt McIntyre when compared with Andesite
Peak. Although maximum recorded instantaneous PAR values were similar at Andesite and
McIntyre, daily PAR values were considerably lower at Mt McIntyre, due to the obstruction of sky
views at low sun angles by shrubs at this site. At Mt McIntyre biocrust were partially shaded by a
widely-dispersed shrub community (e.g. Betula glandulosa and Salix spp.), ca. 90 cm in height. In
contrast, biocrust on Andesite Peak were not shaded by taller shrubs, since the site was dominated
by dwarf shrubs (e.g. Cassiope mertensiana) that are closely appressed to the alpine tundra surface.
The spatial distribution of biocrust in relation to vascular plants within a given ecosystem can have
a significant impact not only on the microclimatic conditions under which the biocrust operates, but
also nitrogen and carbon cycling processes within the crust [168-170].

Nitrogen fixation
We found that increased moisture availability was correlated to an increase in nitrogenase activity,
which is in accordance with most studies of acetylene-reduction by biocrust from alpine and arctic
environments [34-38]. Using Andesite Peak biocrust, our low light model (150 umol m-2 s-1) was
correlated to a lower acetylene reduction rate than our high light (300 umol m-2 s-1) model. This is

54

contrary to Patova et al. (2016) [35] who found that at PAR above 100 umol m-2 s-1, there was no
effect on acetylene-reduction of cyanobacteria-dominated biocrust. Similarly, Zielke (2002) [37]
found that acetylene-reduction rates started to decrease only below 140umol m-2 s-1 for
cyanobacteria-dominated and moss-cyanobacteria biocrust. These results suggest that the light
saturation level of nitrogenase activity for our site might be higher than 150 umol m-2 s-1.
Zielke et al. (2005) found that only when moisture is not limiting does temperature become a driver
of nitrogen fixation [36]. This may introduce a potential flaw in our nitrogen fixation model since
the temperature and light responses were modelled under fully hydrated conditions, which may
not be the case in the field. Otherwise, moisture remains the main factor affecting nitrogen fixation
of the collective biocrust community. However, adaptations to water availability may be
dependent on cyanobacteria species’ ecology [36] and nitrogen-fixing difference between species
may be correlated to hydrological regimes [38].

We observed a positive linear relationship between acetylene reduction and temperature between
10 to 20°C. Although studies from arctic regions report optimal nitrogenase activity temperature
between 20 and 30°C (Zielke et al. [37] on Svlabard Island and Chapin et al. [34] on Devon Island
NWT), some also report nitrogen fixation at low temperatures (30% of maxima at 4°C [37]).

Our study worked with biocrust samples containing a composite of lichens, mosses, and liverworts,
as reported above. Nitrogen fixation in our biocrust was most likely primarily driven by
cyanolichens such as Stereocaulon, Solorina, and Peltigera spp. Previous studies have found
Peltigera spp. dominated biocrust in Sweden to fix up to 0.88 g N m-2 yearly [171]. Nitrogen-fixing
lichens such as Peltigera spp. can represent significant point sources for nitrogen enrichment of soil
55

surface horizons [172] however, the frequent association of mosses and liverworts with nitrogenfixing cyanobacteria also provides a significant source of fixed nitrogen in biocrust communities
[173]. Although association of cyanobacteria with liverworts are known to widely occur, many of
these associations have yet to be characterized. Marchantia, Porella, Blasia and Clavicularia are four
liverwort genera commonly known to host cyanobacterial associations [173]. None of these genera
were observed at our sites, however, it likely that some liverwort cyanobacteria associations were
present. Pleurozium schreberi, a moss species present at the Mt McIntyre site, commonly
associates with the cyanobacterium Nostoc sp. and reported rates of nitrogen fixation range from
of 1.5 – 2 kg N ha-1 yr-1 in boreal forests [174]. Cyanobacteria-moss associations supply most of the
combined nitrogen in some Arctic and Boreal regions [36, 37]. Furthermore, many mosses and
liverworts have adaptations that allow prolonged moisture retention, which creates hydric
microenvironments allowing for cyanobacteria to readily colonize [40]. The more constant moisture
availability at Andesite Peak, would have not only provided a more favourable operating
environment for nitrogen-fixing species, but may have also provided the conditions for greater
establishment and growth of nitrogen fixing components of the biocrust. Although we observed
similar rates of biocrust establishment in our inoculated plots at the two sites, declining versus
increasing rates of nitrogen fixation at Mt. McIntyre compared to Andesite Peak may indicate
better establishment of nitrogen fixing species inoculated biocrust at Andesite Peak. Comparison of
cyanobacterial colonization rates of recovering biocrust in contrasting environments would be
highly informative and should be considered in future studies.

56

Landscape-level nitrogen inputs
Several studies have assessed nitrogen fixation at a landscape level in low Arctic environments with
nitrogen inputs varying widely between hydric and xeric environments (10.89 and 0.73 kg N ha-1 yr1

, respectively) [175]. In a study of biocrust disturbance Belnap et al. [50] estimated the annual

input of nitrogen to 9 kg/ha and 1.4 kg/ha for biocrust of arid lands in Utah, US (dark cyanobacteria
dominated and disturbed, respectively). Our estimate of nitrogen fixation of 4.8±0.3 X 105 mol of
ethylene/ha is in the lower range of values reported for both low Arctic and hot arid and semiarid
lands. However, given the nitrogen-limitation often present in northern environments, including
alpine, even smaller inputs of nitrogen can have significant effects. Although exact calculation of
nitrogen inputs by alpine biocrust warrants a more detailed investigation, our study outlines the
significance of nitrogen fixation by biocrust in alpine environments. We expect nitrogen fixation at
a landscape level to be somewhat lower at Mt McIntyre than at Andesite Peak given the lower
moisture level which would induce a decline in nitrogenase activity, given a similar percent cover of
biocrust on both landscapes (23 and 22%, for Mt McIntyre and Andesite Peak, respectively). high

Indicators of ecosystem function recovery in experimental plots
Biocrust establishment
Mt McIntyre and Andesite Peak had similar biocrust cover 12 weeks after the start of the
experiment, however we observed significantly faster recovery of biocrust function (i.e. nitrogen
fixation and EPS) at Andesite Peak. The difference in cover observed between inoculated and
uninoculated was not reflected as clearly in the measurements of functions which may indicate a
shift in community composition at Mt McIntyre. Xiao and et al. (2008) [176] found that three years
after disturbance a full cover of moss-dominated biocrust was established. This study however did

57

not measure nitrogen fixation nor EPS content. Most studies [1, 80, 88, 89] have observed slow
nitrogenase activity recovery after disturbances. Although some have reported up to 60% biocrust
cover in 4 years with algae-moss inoculation [71]. These studies are in hot arid or semi-arid
environments which share the rugged character of alpine environments but have fundamentally
different precipitation regimes. Our findings indicate that although a biocrust cover can be
established within one growing season, recovery of key ecosystem functions, such as carbon and
nitrogen fixation may require longer.

Site differences in acetylene reduction pattern
In biocrust from experimental plots, opposing trends in acetylene-reduction were observed over
time. At Andesite Peak, the acetylene reduction rate increased as the experiment progressed
whereas at Mt McIntyre it decreased. The local climatic differences between both sites, notably the
contrasting moisture regimes, may play in important role in the opposing trends.
In a study of cyanobacteria dominated biocrust of subpolar Ural Mountains, Patova et al. (2016)
[35] reports hourly rates of acetylene reduction between 0.5 and 1.76 mg of ethylene/m2 (18-62
µmol of ethylene/m2)which is ten times higher than the rates measured in restored biocrust of
Andesite Peak.

The difference in activity level could explain the lower rehabilitation potential of dry cold
continental climate of Mt McIntyre when compared to moister and warmer coastal climate of
Andesite Peak. Additionally, the declining pattern in acetylene-reduction over time at Mt McIntyre
could be reflective of biocrust mortality linked to the uncharacteristically dry period at the
beginning of our experiment; while the normal precipitation for June is 32 mm, June 2015 only
58

received 16mm [155]. This result suggests that a limitation in water availability can considerably
slow restoration processes in montane environments. This is consistent with drylands where
biocrust succession and functions are limited by water availability [177]. In more xeric and cold
climates, different restoration protocols increasing surface moisture retention may be needed to
successfully restore biocrust communities.

Differences in biocrust composition that are intrinsically linked to different environmental and
microclimatic conditions between sites could also explain these opposing trends. In a comparison
between alpine and Antarctic biocrust, Colesie et al. (2016) [41] found that compositional
difference could be correlated to climatic conditions and the adaptation necessary to thrive in
extreme environments. Accordingly, in a manipulative experiment, Li et al. (2010) [22] found that
increasing soil moisture shifted the species composition to an increased proportion of mosses and
lichens and a decreased proportion of cyanobacteria. Furthermore, Raggio et al. (2016) [178] found
that lichens of dry and cold environments were less active than those from warmer moister
climates. In addition, increased species richness has been linked to milder climate regimes in North
American montane environments [42]. Collectively, these studies indicate that the community
composition of BSCs and the ecological functioning (e.g. nitrogen fixation) of species assemblages is
undoubtedly linked with environmental and operating conditions. Therefore, we expect that the
mesic climate of Andesite Peak will result in a more active community than those located in xeric
climate such as Mt McIntyre.

59

Effects of fertilization and inoculation on functional recovery
There was no clear effect of fertilization on acetylene-reduction rates in the experimental plots of
either site. Contrary to Maestre et al. (2006) who found a positive relationship between moderate
fertilizer addition (in the form of a composted sludge) and nitrogenase activity under laboratory
conditions [44]. In a recent study combining fertilizer addition and watering, Antoninka et al. (2015)
[43] successfully produced fully functional moss-dominated biocrust. While the effect of
fertilization on biocrust establishment remains unclear, these results suggest that hydration may be
the key limiting factor in biocrust nitrogen fixation.

Experimental plots at Andesite Peak that received fertilization and/or inoculation (F, B, BF) had
higher DOC in soils (3 cm depth). At Mt McIntyre, the DOC was extremely low in all soils and we did
not observe any treatment differences. These extremely low levels of DOC are typical of northern
latitudes, where primary productivity is limited and soil development slow [179]. However, it has
been suggested that phototrophic microbial communities may play an important role in carbon
uptake from the atmosphere in alpine environments where plant cover is limited [137]. Overall
DOC was higher at Andesite Peak, which is likely caused by the coastal climate’s association with
higher primary productivity which creates better developed soils richer in carbon [180]. The
difference in soil development level was also expected to be also reflected in the amount of
mineralizable nitrogen present. Surprisingly, mineralizable nitrogen levels were found to be lower
on Andesite peak than on Mt McIntyre. In Alaskan soils, nitrogen mineralization rates were found
to be insensitive to temperature between 3 and 9°C but increased two folds or more between 9
and 15°C [181]. Consequently, higher average temperatures at Mt McIntyre when compared with

60

Andesite Peak could explain its higher mineralizable nitrogen concentrations. At Andesite Peak,
DOC was approximately double in inoculated soils compared to uninoculated plots.

Conclusion
Our study provides a comprehensive examination of biocrust communities in two contrasting alpine
environments. Furthermore, our work on restoration of two sites with climatic and community
compositional differences are a first step in understanding mechanisms driving successful
restoration of biocrust in alpine environments. Although biocrust establishment was successful at
both sites, only Andesite showed recovery of nitrogen fixation, demonstrating a need for better
characterization of the early successional trajectories of nitrogen fixing communities in recovering
biocrust. The decline of nitrogen fixation following inoculation at Mt McIntyre likely indicates a
community shift, as well as, lower activity due to less optimal operating conditions. While
restoration of key ecosystem functions is highly desirable, biocrust establishment alone could help
mitigate physical impacts associated with soil surface disturbance, such as erosion. Furthermore,
our study suggests that depending on restoration goals, protocols for dry and cold climates may
need to include promoting soil surface moisture retention or providing additional water inputs for
successful re-establish of biocrust communities.

61

4. Implications
Most restoration work with biocrusts has been conducted in desert and grassland ecosystems; by
applying these restoration approaches in alpine tundra our study provides an important and unique
contribution. Furthermore, few studies combine the concurrent examination of biocrust structure
(species assemblages and functional groups), function (nitrogenase activity, eps production, carbon
and nitrogen cycling) and field microclimate. We addressed this knowledge gap by the study of
alpine biocrust in two contrasting sites: the mesic coastal mountain ranges of northwestern British
Columbia (Terrace, British Columbia, Canada), and the xeric interior mountain ranges of the
southern Yukon (Whitehorse, Yukon Territory, Canada). At each location, we examined the
development of soil biocrust after disturbance in controlled experimental plots. Experimental
treatments included plots with and without biocrust inoculation, and plots with and without
fertilizer addition. This allowed us to investigate site-specific potential for biocrust restoration. Soil
surface microclimate was monitored, and nitrogen fixation was measured at each site to provide a
better understanding of the operating conditions under which physiological activity occurs in these
contrasting biocrust communities.

We found that soil surface treatment and inoculation facilitated biocrust establishment. Since
Microrills create an elevated moisture level in gullies, when combined with inoculation, they
represent a practical restoration tool to facilitate biocrust establishment and function recovery. Our
results demonstrate that both nifH copy numbers and NA increased over time, suggesting that
biocrust inoculation is effective at restoring nitrogen-fixation capability.

62

We showed that inoculation with biocrust could facilitate the recovery of primary ecosystem
function. Specifically, processes associated with nitrogen fixation as demonstrated by increased
nifH copy numbers over time combined with increased nitrogenase activity. Although bacterial
communities were relatively unchanged after inoculation in a greenhouse setting, relative
contributions from the fungal communities to biocrust formation reflect compositional differences
between Inoculant and soil samples. A novel finding was the high number of Serendipita OTUs
associated with biocrusts; further research is needed to determine their role. Non-cyanobacterial
diazotroph and other alpine pioneering species need to be the focus of research to develop a better
understanding of soil development and inform the restoration practices for disturbed sites within
alpine tundra biomes.

Our study examined nitrogen fixation by soil surface communities in natural alpine environments
and is novel in that it quantifies the magnitude of nitrogen fixation at a landscape level. Our work
on restoration of two sites with climatic and community compositional differences is a first step in
understanding mechanisms driving successful restoration of biocrust in alpine environments.
Although biocrust establishment was successful at both sites, only Andesite Peak showed recovery
of nitrogen fixation, demonstrating a need for better characterization of the early successional
trajectories of nitrogen fixing communities in recovering biocrust. The decline of nitrogen fixation
level through time at Mt McIntyre likely indicates a community shift, as well as, lower activity due
to less optimal operating conditions. However, biocrust establishment alone could help mitigate
physical impacts associated with soil surface disturbance, such as erosion. Our study suggests that
depending on restoration goals, protocols for dry and cold climate may need to include increased
moisture retention or provide additional water inputs to successfully re-establish biocrust
communities with pre-disturbance composition and function.
63

Practical Implications for Restoration
•

In mesic climate conditions, inoculation of disturbed alpine environments with biocrusts
may be an effective restoration technique

•

In cold xeric climates, restoration protocols that increase surface moisture retention are
likely needed for successfully restoration of biocrust communities. Timing might be
particularly important in environments with high seasonal moisture variation.

•

Restoration of biocrust cover may not be indicative of recovery of ecological function, such
as soil surface nitrogen fixation. However, even if species assemblage has shifted some
ecological functions such as erosion control might still be present.

•

Alpine biocrusts appear to play a significant role in nitrogen and carbon input and
restoration of these communities following disturbance may support ecosystem recovery.

64

List of references
1.
2.
3.

4.

5.

6.
7.
8.

9.
10.
11.

12.
13.
14.
15.

16.
17.
18.

Capers RS, Taylor DW (2014) Slow Recovery In A Mount Washington, New Hampshire,
Alpine Plant Community Four Years After Disturbance. Rhodora 116: 1-24.
Olofsson J, Shams H (2007) Determinants of plant species richness in an alpine meadow.
Journal of Ecology 95: 916-925.
Desserud PA, Naeth MA (2013) Natural Recovery of Rough Fescue (Festuca hallii (Vasey)
Piper) Grassland after Disturbance by Pipeline Construction in Central Alberta, Canada.
Natural Areas Journal 33: 91-98.
Bayramov E, Buchroithner M, McGurty E (2012) Quantitative assessment of vegetation
cover and soil degradation factors within terrain units for planning, monitoring and
assessment of renaturation along oil and gas pipelines. Geocarto International 27: 535-555.
Bulot A, Provost E, Dutoit T (2014) A comparison of different soil transfer strategies for
restoring a Mediterranean steppe after a pipeline leak (La Crau plain, South-Eastern
France). Ecological Engineering 71: 690-702.
Chambers JC (1995) Disturbance, life history strategies, and seed fates in alpine herbfield
communities. American Journal of Botany: 421-433.
Xiao J, Shi P, Wang Y, Yang L (2016) The vegetation recovery pattern and affecting factors
after pipeline disturbance in northwest China. Journal for nature conservation 29: 114-122.
Belnap J, Gillette DA (1998) Vulnerability of desert biological soil crusts to wind erosion: the
influences of crust development, soil texture, and disturbance. Journal of arid
environments 39: 133-142.
Belnap J (1996) Soil surface disturbances in cold deserts: effects on nitrogenase activity in
cyanobacterial-lichen soil crusts. Biology and Fertility of Soils 23: 362-367.
Ebersole JJ (2002) Recovery of alpine vegetation on small, denuded plots, Niwot Ridge,
Colorado, USA. Arctic Antarctic and Alpine Research 34: 389-397. doi: 10.2307/1552196
Wright J, Symstad A, Bullock JM, Engelhardt K, Jackson L, Bernhardt E (2009) Restoring
biodiversity and ecosystem function: will an integrated approach improve results.
Biodiversity, ecosystem functioning, and human wellbeing: 167-177.
Harris JA, Hobbs RJ, Higgs E, Aronson J (2006) Ecological restoration and global climate
change. Restoration Ecology 14: 170-176. doi: 10.1111/j.1526-100X.2006.00136.x
Hobbs RJ, Higgs E, Harris JA (2009) Novel ecosystems: implications for conservation and
restoration. Trends in ecology & evolution 24: 599-605.
Clewell A, Rieger J, Munro J (2005) Guidelines for Developing and Managing Ecological
Restoration Projects. Washington, DC: Society for Ecological Restoration.
Coiffait-Gombault C, Buisson E, Dutoit T (2011) Hay Transfer Promotes Establishment of
Mediterranean Steppe Vegetation on Soil Disturbed by Pipeline Construction. Restoration
Ecology 19: 214-222. doi: 10.1111/j.1526-100X.2010.00706.x
Bishop SC, Chapin III FS (1989) Patterns of natural revegetation on abandoned gravel pads
in arctic Alaska. Journal of Applied Ecology: 1073-1081.
Walker D, Everett K (1987) Road dust and its environmental impact on Alaskan taiga and
tundra. Arctic and Alpine Research: 479-489.
Hagen D, Hansen T-I, Graae BJ, Rydgren K (2014) To seed or not to seed in alpine
restoration: introduced grass species outcompete rather than facilitate native species.
Ecological engineering 64: 255-261.

65

19.
20.

21.
22.

23.
24.

25.

26.

27.
28.

29.

30.

31.

32.

33.
34.
35.

36.

Tilman D (1993) Species richness of experimental productivity gradients - how important is
colonization limitation. Ecology 74: 2179-2191. doi: 10.2307/1939572
Baer SG, Blair JM, Collins SL, Knapp AK (2004) Plant community responses to resource
availability and heterogeneity during restoration. Oecologia 139: 617-629. doi:
10.1007/s00442-004-1541-3
Biederman LA, Whisenant SG (2011) Using mounds to create microtopography alters plant
community development early in restoration. Restoration Ecology 19: 53-61.
Li XR, He MZ, Zerbe S, Li XJ, Liu LC (2010) Micro-geomorphology determines community
structure of biological soil crusts at small scales. Earth Surface Processes and Landforms 35:
932-940. doi: 10.1002/esp.1963
Elmarsdottir A, Aradottir AL, Trlica MJ (2003) Microsite availability and establishment of
native species on degraded and reclaimed sites Journal of Applied Ecology 40: 9.
Ruthrof KX, Fontaine JB, Buizer M, Matusick G, McHenry MP, Hardy GESJ (2013) Linking
restoration outcomes with mechanism: the role of site preparation, fertilisation and
revegetation timing relative to soil density and water content. Plant ecology 214: 987-998.
Turner SR, Pearce B, Rokich DP, Dunn RR, Merritt DJ, Majer JD, Dixon KW (2006) Influence
of Polymer Seed Coatings, Soil Raking, and Time of Sowing on Seedling Performance in
Post‐Mining Restoration. Restoration Ecology 14: 267-277.
Bowker MA (2007) Biological soil crust rehabilitation in theory and practice: An
underexploited opportunity. Restoration Ecology 15: 13-23. doi: 10.1111/j.1526100X.2006.00185.x
Belnap J, Lange OLE (2003) Biological Soil Crusts: Structure, Function, and Management. In:
Belnap, J, Lange, OL (eds.). Springer-Verlag Berlin Heidelberg.
Castillo-Monroy AP, Maestre FT, Rey A, Soliveres S, García-Palacios P (2011) Biological soil
crust microsites are the main contributor to soil respiration in a semiarid ecosystem.
Ecosystems 14: 835-847.
Gold WG, Glew KA, Dickson LG (2001) Functional influences of cryptobiotic surface crusts in
an alpine tundra basin of the Olympic Mountains, Washington, USA. Northwest Science 75:
315-326.
Grote EE, Belnap J, Housman DC, Sparks JP (2010) Carbon exchange in biological soil crust
communities under differential temperatures and soil water contents: implications for
global change. Global Change Biology 16: 2763-2774.
de Guevara ML, Lazaro R, Quero JL, Ochoa V, Gozalo B, Berdugo M, Ucles O, Escolar C,
Maestre FT (2014) Simulated climate change reduced the capacity of lichen-dominated
biocrusts to act as carbon sinks in two semi-arid Mediterranean ecosystems. Biodiversity
and Conservation 23: 1787-1807. doi: 10.1007/s10531-014-0681-y
Schmidt SK, Nemergut DR, Todd BT, Lynch RC, Darcy JL, Cleveland CC, King AJ (2012) A
simple method for determining limiting nutrients for photosynthetic crusts. Plant Ecology &
Diversity 5: 513-519. doi: 10.1080/17550874.2012.738714
Walker T, Syers JK (1976) The fate of phosphorus during pedogenesis. Geoderma 15: 1-19.
Chapin DM, Bliss L, Bledsoe L (1991) Environmental regulation of nitrogen fixation in a high
arctic lowland ecosystem. Canadian Journal of Botany 69: 2744-2755.
Patova E, Sivkov M, Patova A (2016) Nitrogen fixation activity in biological soil crusts
dominated by cyanobacteria in the Subpolar Urals (European North-East Russia). Fems
Microbiology Ecology 92. doi: 10.1093/femsec/fiw131
Zielke M, Solheim B, Spjelkavik S, Olsen RA (2005) Nitrogen fixation in the high arctic: Role
of vegetation and environmental conditions. Arctic Antarctic and Alpine Research 37: 372378. doi: 10.1657/1523-0430(2005)037[0372:nfitha]2.0.co;2
66

37.

38.

39.
40.

41.

42.

43.

44.

45.
46.

47.
48.

49.
50.

51.

52.

53.
54.

Zielke M, Ekker AS, Olsen RA, Spjelkavik S, Solheim B (2002) The influence of abiotic factors
on biological nitrogen fixation in different types of vegetation in the High Arctic, Svalbard.
Arctic, Antarctic, and Alpine Research: 293-299.
Gavazov KS, Soudzilovskaia NA, van Logtestijn RSP, Braster M, Cornelissen JHC (2010)
Isotopic analysis of cyanobacterial nitrogen fixation associated with subarctic lichen and
bryophyte species. Plant and Soil 333: 507-517. doi: 10.1007/s11104-010-0374-6
Kershaw KA (1985) Physiological ecology of lichens. Cambridge University Press
Raggio J, Green TA, Sancho LG, Pintado A, Colesie C, Weber B, Büdel B (2017) Metabolic
activity duration can be effectively predicted from macroclimatic data for biological soil
crust habitats across Europe. Geoderma 306: 10-17.
Colesie C, Green TGA, Raggio J, Budel B (2016) Summer activity patterns of Antarctic and
high alpine lichen-dominated biological soil crusts-Similar but different? Arctic Antarctic
and Alpine Research 48: 449-460. doi: 10.1657/aaar0015-047
Kikvidze Z, Pugnaire FI, Brooker RW, Choler P, Lortie CJ, Michalet R, Callaway RM (2005)
Linking patterns and processes in alpine plant communities: A global study. Ecology 86:
1395-1400. doi: 10.1890/04-1926
Antoninka A, Bowker MA, Reed SC, Doherty K (2015) Production of greenhouse‐grown
biocrust mosses and associated cyanobacteria to rehabilitate dryland soil function.
Restoration Ecology.
Maestre FT, Martin N, Diez B, Lopez-Poma R, Santos F, Luque I, Cortina J (2006) Watering,
fertilization, and slurry inoculation promote recovery of biological crust function in
degraded soils. Microbial Ecology 52: 365-377. doi: 10.1007/s00248-006-9017-0
Mendrygal KE, González JE (2000) Environmental regulation of exopolysaccharide
production in Sinorhizobium meliloti. Journal of bacteriology 182: 599-606.
Janczarek M (2011) Environmental signals and regulatory pathways that influence
exopolysaccharide production in rhizobia. International journal of molecular sciences 12:
7898-7933.
Belnap J (2001) Factors influencing nitrogen fixation and nitrogen release in biological soil
crustsBiological soil crusts: structure, function, and management. Springer, pp. 241-261
Veluci RM, Neher DA, Weicht TR (2006) Nitrogen fixation and leaching of biological soil
crust communities in mesic temperate soils. Microbial Ecology 51: 189-196. doi:
10.1007/s00248-005-0121-3
Belnap J (2003) The world at your feet: desert biological soil crusts. Frontiers in Ecology and
the Environment 1: 181-189.
Belnap J (2002) Impacts of off road vehicles on nitrogen cycles in biological soil crusts:
resistance in different US deserts. Journal of Arid Environments 52: 155-165. doi:
10.1006/jare.2002.0991
Hawkes CV (2004) Effects of biological soil crusts on seed germination of four endangered
herbs in a xeric Florida shrubland during drought. Plant Ecology 170: 121-134. doi:
10.1023/b:vege.0000019035.56245.91
Bowker MA, Mau RL, Maestre FT, Escolar C, Castillo‐Monroy AP (2011) Functional profiles
reveal unique ecological roles of various biological soil crust organisms. Functional Ecology
25: 787-795.
Bu CF, Wu SF, Xie YS, Zhang XC (2013) The Study of Biological Soil Crusts: Hotspots and
Prospects. Clean-Soil Air Water 41: 899-906. doi: 10.1002/clen.201100675
Maestre FT, Bowker MA, Canton Y, Castillo-Monroy AP, Cortina J, Escolar C, Escudero A,
Lazaro R, Martinez I (2011) Ecology and functional roles of biological soil crusts in semi-arid

67

55.

56.

57.
58.

59.
60.

61.

62.

63.

64.

65.

66.

67.

68.
69.

70.

ecosystems of Spain. Journal of Arid Environments 75: 1282-1291. doi:
10.1016/j.jaridenv.2010.12.008
Li D, Xing W, Li G, Liu Y (2009) Cytochemical changes in the developmental process of
Nostoc sphaeroides (cyanobacterium). Journal of Applied Phycology 21: 119-125. doi:
10.1007/s10811-008-9340-6
Belnap J (1995) Recovery of nitrogenase activity in cyanobacterial-lichen soils crusts of the
Great Basin and the Colorado Plateau. Bulletin of the Ecological Society of America 76: 1819.
Zaady E, Kuhn U, Wilske B, Sandoval-Soto L, Kesselmeier J (2000) Patterns of CO 2 exchange
in biological soil crusts of successional age. Soil Biology and Biochemistry 32: 959-966.
Bowker MA, Belnap J, Chaudhary VB, Johnson NC (2008) Revisiting classic water erosion
models in drylands: The strong impact of biological soil crusts. Soil Biology & Biochemistry
40: 2309-2316. doi: 10.1016/j.soilbio.2008.05.008
Belnap J (2006) The potential roles of biological soil crusts in dryland hydrologic cycles.
Hydrological Processes 20: 3159-3178. doi: 10.1002/hyp.6325
Bowker MA, Belnap J, Davidson DW, Goldstein H (2006) Correlates of biological soil crust
abundance across a continuum of spatial scales: support for a hierarchical conceptual
model. Journal of Applied Ecology 43: 152-163.
Stark L, Brinda J, McLetchie D (2011) Effects of increased summer precipitation and N
deposition on Mojave Desert populations of the biological crust moss Syntrichia caninervis.
Journal of Arid Environments 75: 457-463.
Yu J, Glazer N, Steinberger Y (2014) Carbon utilization, microbial biomass, and respiration in
biological soil crusts in the Negev Desert. Biology and Fertility of Soils 50: 285-293. doi:
10.1007/s00374-013-0856-9
Rossi F, Potrafka RM, Pichel FG, De Philippis R (2012) The role of the exopolysaccharides in
enhancing hydraulic conductivity of biological soil crusts. Soil Biology and Biochemistry 46:
33-40.
Brotherson JD, Rushforth SR (1983) Influence of cryptogamic crusts on moisture
relationships of soils in Navajo National Monument, Arizona. The Great Basin Naturalist: 7378.
Yeager CM, Kornosky JL, Housman DC, Grote EE, Belnap J, Kuske CR (2004) Diazotrophic
community structure and function in two successional stages of biological soil crusts from
the Colorado Plateau and Chihuahuan Desert. Applied and Environmental Microbiology 70:
973-983.
Mager DM, Thomas AD (2011) Extracellular polysaccharides from cyanobacterial soil crusts
A review of their role in dryland soil processes. Journal of Arid Environments 75: 91-97. doi:
10.1016/j.jaridenv.2010.10.001
Belnap J, Phillips SL, Flint S, Money J, Caldwell M (2008) Global change and biological soil
crusts: effects of ultraviolet augmentation under altered precipitation regimes and nitrogen
additions. Global Change Biology 14: 670-686. doi: 10.1111/j.1365-2486.2007.01509.x
Jorgenson MT, Joyce MR (1994) Six strategies for rehabilitating land disturbed by oil
development in arctic Alaska. Arctic: 374-390.
Pearce I, Woodin SJ, Van der Wal R (2003) Physiological and growth responses of the
montane bryophyte Racomitrium lanuginosum to atmospheric nitrogen deposition. New
Phytologist 160: 145-155.
Xu S, Yin C, He M, Wang Y (2008) A Technology for Rapid Reconstruction of MossDominated Soil Crusts. Environmental Engineering Science 25: 1129-1137. doi:
10.1089/ees.2007.0272
68

71.

72.

73.
74.

75.

76.

77.

78.

79.

80.

81.
82.

83.
84.

85.
86.

87.

Liu W-q, Song Y-s, Wang B, Li J-t, Shu W-s (2012) Nitrogen fixation in biotic crusts and
vascular plant communities on a copper mine tailings. European Journal of Soil Biology 50:
15-20. doi: 10.1016/j.ejsobi.2011.11.009
Davidson DW, Bowker M, George D, Phillips SL, Belnap J (2002) Treatment effects on
performance of N-fixing lichens in disturbed soil crusts of the Colorado Plateau. Ecological
Applications 12: 1391-1405.
Belnap J (1993) Recovery rates of cryptobiotic crusts - Inoculant use and assessment
methods. Great Basin Naturalist 53: 89-95.
Stewart KJ, Siciliano SD (2015) Potential contribution of native herbs and biological soil
crusts to restoration of the biogeochemical nitrogen cycle in mining impacted sites in
Northern Canada. Ecological Restoration 33: 30-42.
Buttars SM, St Clair LL, Johansen JR, Sray JC, Payne MC, Webb BL, Terry RE, Pendleton BK,
Warren SD (1998) Pelletized cyanobacterial soil amendments: Laboratory testing for
survival, escapability, and nitrogen fixation. Arid Soil Research and Rehabilitation 12: 165178.
Chiquoine LP, Abella SR, Bowker MA (2016) Rapidly restoring biological soil crusts and
ecosystem functions in a severely disturbed desert ecosystem. Ecological applications 26:
1260-1272.
Lan S, Zhang Q, Wu L, Liu Y, Zhang D, Hu C (2013) Artificially accelerating the reversal of
desertification: cyanobacterial inoculation facilitates the succession of vegetation
communities. Environmental science & technology 48: 307-315.
Antoninka A, Bowker MA, Reed SC, Doherty K (2016) Production of greenhouse-grown
biocrust mosses and associated cyanobacteria to rehabilitate dryland soil function.
Restoration Ecology 24: 324-335. doi: 10.1111/rec.12311
Stewart KJ, Grogan P, Coxson DS, Siciliano SD (2014) Topography as a key factor driving
atmospheric nitrogen exchanges in arctic terrestrial ecosystems. Soil Biology &
Biochemistry 70: 96-112. doi: 10.1016/j.soilbio.2013.12.005
Steven B, Kuske CR, Gallegos-Graves LV, Reed SC, Belnap J (2015) Climate Change and
Physical Disturbance Manipulations Result in Distinct Biological Soil Crust Communities.
Applied and Environmental Microbiology 81: 7448-7459. doi: 10.1128/aem.01443-15
Belnap J, Lange OL (2001) Structure and functioning of biological soil crusts: a
synthesisBiological soil crusts: structure, function, and management. Springer, pp. 471-479
Soule T, Anderson IJ, Johnson SL, Bates ST, Garcia-Pichel F (2009) Archaeal populations in
biological soil crusts from arid lands in North America. Soil Biology & Biochemistry 41:
2069-2074. doi: 10.1016/j.soilbio.2009.07.023
Coleine C, Selbmann L, Ventura S, D’Acqui LP, Onofri S, Zucconi L (2015) Fungal Biodiversity
in the Alpine Tarfala Valley. Microorganisms 3: 612-624.
Kidron GJ, Yair A (2001) Runoff-induced sediment yield over dune slopes in the Negev
Desert. 1: Quantity and variability. Earth Surface Processes and Landforms 26: 461-474. doi:
10.1002/esp.191
Belnap J (2002) Nitrogen fixation in biological soil crusts from southeast Utah, USA. Biology
and Fertility of Soils 35: 128-135. doi: 10.1007/s00374-002-0452-x
Delgado-Baquerizo M, Maestre FT, Eldridge DJ, Bowker MA, Ochoa V, Gozalo B, Berdugo M,
Val J, Singh BK (2016) Biocrust-forming mosses mitigate the negative impacts of increasing
aridity on ecosystem multifunctionality in drylands. The New phytologist 209: 1540-1552.
doi: 10.1111/nph.13688
Bowker MA, Maestre FT, Eldridge D, Belnap J, Castillo-Monroy A, Escolar C, Soliveres S
(2014) Biological soil crusts (biocrusts) as a model system in community, landscape and
69

88.
89.

90.

91.
92.

93.

94.
95.

96.

97.
98.

99.

100.
101.

102.
103.
104.
105.

ecosystem ecology. Biodiversity and Conservation 23: 1619-1637. doi: 10.1007/s10531014-0658-x
Belnap J, Eldridge D (2001) Disturbance and recovery of biological soil crustsBiological soil
crusts: structure, function, and management. Springer, pp. 363-383
Kuske CR, Yeager CM, Johnson S, Ticknor LO, Belnap J (2012) Response and resilience of soil
biocrust bacterial communities to chronic physical disturbance in arid shrublands. Isme
Journal 6: 886-897. doi: 10.1038/ismej.2011.153
Čapková K, Hauer T, Řeháková K, Doležal J (2016) Some Like it High! Phylogenetic Diversity
of High-Elevation Cyanobacterial Community from Biological Soil Crusts of Western
Himalaya. Microbial ecology 71: 113-123.
Evans RD, Belnap J (1995) Nitrogen dynamics of disturbed and undisturbed arid grasslands
in Canyonlands National Park. Bulletin of the Ecological Society of America 76: 324-324.
Jarngerdur Gretarsdottir, Asa L. Aradottir, Vigdis Vandvik, Einar Heegaard, H. J. B. Birks
(2004) Long-Term Effects of Reclamation Treatments on Plant Succession in
Iceland Restoration Ecology 12: 11.
Jeffries D, Klopatek J, Link S, Bolton H (1992) Acetylene reduction by cryptogamic crusts
from a blackbrush community as related to resaturation and dehydration. Soil Biology and
Biochemistry 24: 1101-1105.
Peer T, Gruber JP, Tschaikner A, Türk R Alpine soil crusts, the biocoenosis which braves the
cold.
Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, Owens SM, Betley J,
Fraser L, Bauer M (2012) Ultra-high-throughput microbial community analysis on the
Illumina HiSeq and MiSeq platforms. The ISME journal 6: 1621-1624.
Bilger W, Schreiber U, Bock M (1995) Determination of the quantum efficiency of
photosystem II and of non-photochemical quenching of chlorophyll fluorescence in the
field. Oecologia 102: 425-432.
Stewart W, Fitzgerald G, Burris n (1967) In situ studies on N2 fixation using the acetylene
reduction technique. Proceedings of the National Academy of Sciences 58: 2071-2078.
Rosch C, Mergel A, Bothe H (2002) Biodiversity of denitrifying and dinitrogen-fixing bacteria
in an acid forest soil. Applied and Environmental Microbiology 68: 3818-3829. doi:
10.1128/aem.68.8.3818-3829.2002
Meyer A, Focks A, Radl V, Keil D, Welzl G, Schoening I, Boch S, Marhan S, Kandeler E,
Schloter M (2013) Different Land Use Intensities in Grassland Ecosystems Drive Ecology of
Microbial Communities Involved in Nitrogen Turnover in Soil. Plos One 8. doi:
10.1371/journal.pone.0073536
Crayton M (1982) A comparative cytochemical study of volvocacean matrix
polysaccharides. Journal of Phycology 18: 336-344.
Masuko T, Minami A, Iwasaki N, Majima T, Nishimura SI, Lee YC (2005) Carbohydrate
analysis by a phenol-sulfuric acid method in microplate format. Analytical Biochemistry
339: 69-72. doi: 10.1016/j.ab.2004.12.001
Waring S, Bremner J (1964) Ammonium production in soil under waterlogged conditions as
an index of nitrogen availability.
Solorzano L (1969) Determination of ammonia in natural waters by the phenol hypochlorite
method. Limnology and Oceanography 14: 799-801. doi: 10.4319/lo.1969.14.5.0799
Box GE, Cox DR (1964) An analysis of transformations. Journal of the Royal Statistical
Society Series B (Methodological): 211-252.
Bates D, Mächler M, Bolker B, Walker S (2015) Fitting linear mixed-effects models using
lme4. Journal of Statistical Software 67. doi: 10.18637/jss.v067.i01
70

106.
107.
108.
109.

110.
111.

112.

113.

114.
115.

116.
117.

118.

119.
120.

121.
122.

123.

Kuznetsova A, Brockhoff PB, Christensen RHB (2013) lmerTest: Tests for random and fixed
effects for linear mixed effect models (lmer objects of lme4 package). R package version 2.
Hothorn T, Bretz F, Westfall P (2008) Simultaneous inference in general parametric models.
Biometrical journal 50: 346-363.
Grzesiak B, Wolski GJ (2015) Bryophilous species of the genus Galerina in peat bogs of
Central Poland. Herzogia 28: 607-623.
Zhang T, Zhang Y-Q, Liu H-Y, Wei Y-Z, Li H-L, Su J, Zhao L-X, Yu L-Y (2013) Diversity and cold
adaptation of culturable endophytic fungi from bryophytes in the Fildes Region, King
George Island, maritime Antarctica. FEMS microbiology letters 341: 52-61.
Margesin R, Neuner G, Storey K (2007) Cold-loving microbes, plants, and animals—
fundamental and applied aspects. Naturwissenschaften 94: 77-99.
Libkind D, Moliné M, Sommaruga R, Sampaio JP, van Broock M (2011) Phylogenetic
distribution of fungal mycosporines within the Pucciniomycotina (Basidiomycota). Yeast 28:
619-627.
Bates ST, Garcia-Pichel F, Nash III T, Geiser L, McCune B, Triebel D, Tomescu A, Sanders W
(2010) Fungal components of biological soil crusts: insights from culture-dependent and
culture-independent studies. Biology of Lichens—Symbiosis, Ecology, Environmental
Monitoring, Systematics, Cyber Applications: 197-210.
Bates ST, Nash TH, III, Garcia-Pichel F (2012) Patterns of diversity for fungal assemblages of
biological soil crusts from the southwestern United States. Mycologia 104: 353-361. doi:
10.3852/11-232
Porras-Alfaro A, Herrera J, Natvig DO, Lipinski K, Sinsabaugh RL (2011) Diversity and
distribution of soil fungal communities in a semiarid grassland. Mycologia 103: 10-21.
Collins SL, Sinsabaugh RL, Crenshaw C, Green L, Porras-Alfaro A, Stursova M, Zeglin LH
(2008) Pulse dynamics and microbial processes in aridland ecosystems. Journal of Ecology
96: 413-420. doi: 10.1111/j.1365-2745.2008.01362.x
Weiß M, Waller F, Zuccaro A, Selosse MA (2016) Sebacinales–one thousand and one
interactions with land plants. New Phytologist 211: 20-40.
DePriest PT (2004) Early molecular investigations of lichen-forming symbionts: 1986-2001.
Annual Review of Microbiology 58: 273-301. doi:
10.1146/annurev.micro.58.030603.123730
MacFarlane J, Kershaw K (1982) Physiological-environmental interactions in lichens. XIV.
The environmental control of glucose movement from alga to fungus in Peltigera
polydactyla, P. rufescens and Collema furfuraceum. The New Phytologist 91: 93-101.
Dietz S, Hartung W (1999) The effect of abscisic acid on chlorophyll fluorescence in lichens
under extreme water regimes. New Phytologist 143: 495-501.
Farrar J (1976) Ecological physiology of the lichen Hypogymnia physodes. II. Effects of
wetting and drying cycles and the concept of'physiological buffering'. New Phytologist: 105113.
Williams AJ, Buck BJ, Beyene MA (2012) Biological soil crusts in the Mojave Desert, USA:
micromorphology and pedogenesis. Soil Science Society of America Journal 76: 1685-1695.
Wang W, Liu Y, Li D, Hu C, Rao B (2009) Feasibility of cyanobacterial inoculation for
biological soil crusts formation in desert area. Soil Biology & Biochemistry 41: 926-929. doi:
10.1016/j.soilbio.2008.07.001
Kuske CR, Barns SM, Busch JD (1997) Diverse uncultivated bacterial groups from soils of the
arid southwestern United States that are present in many geographic regions. Applied and
Environmental Microbiology 63: 3614-3621.

71

124.

125.

126.
127.

128.

129.
130.

131.

132.

133.
134.
135.

136.

137.

138.

139.

140.

Nagy ML, Perez A, Garcia-Pichel F (2005) The prokaryotic diversity of biological soil crusts in
the Sonoran Desert (Organ Pipe Cactus National Monument, AZ). Fems Microbiology
Ecology 54: 233-245. doi: 10.1016/j.femsec.2005.03.011
Kuske CR, Ticknor LO, Miller ME, Dunbar JM, Davis JA, Barns SM, Belnap J (2002)
Comparison of soil bacterial communities in rhizospheres of three plant species and the
interspaces in an arid grassland. Applied and Environmental Microbiology 68: 1854-1863.
Kielak AM, Barreto CC, Kowalchuk GA, Van Veen JA, Kuramae EE (2016) The ecology of
Acidobacteria: moving beyond genes and genomes. Frontiers in Microbiology 7: 744.
Steven B, Lionard M, Kuske CR, Vincent WF (2013) High Bacterial Diversity of Biological Soil
Crusts in Water Tracks over Permafrost in the High Arctic Polar Desert. Plos One 8. doi:
10.1371/journal.pone.0071489
Marusenko Y, Bates ST, Anderson I, Johnson SL, Soule T, Garcia-Pichel F (2013) Ammoniaoxidizing archaea and bacteria are structured by geography in biological soil crusts across
North American arid lands. Ecological Processes 2: 1-10.
Acea MJ, Diz N, Prieto-Fernandez A (2001) Microbial populations in heated soils inoculated
with cyanobacteria. Biology and Fertility of Soils 33: 118-125. doi: 10.1007/s003740000298
Bates ST, Garcia-Pichel F (2009) A culture-independent study of free-living fungi in
biological soil crusts of the Colorado Plateau: their diversity and relative contribution to
microbial biomass. Environmental Microbiology 11: 56-67. doi: 10.1111/j.14622920.2008.01738.x
Giraldez‐Ruiz N, Mateo P, Bonilla I, Fernandez‐Pinas F (1997) The relationship between
intracellular pH, growth characteristics and calcium in the cyanobacterium Anabaena sp.
strain PCC7120 exposed to low pH. New phytologist 137: 599-605.
Lederberg J, Alexander M, Bloom BR, Hopwood DA, Hull R, Iglewski BH, Laskin AI, Oliver SG,
Schaechter M, Summers WC (2000) Encyclopedia of Microbiology, Four-Volume Set.
Academic Press
Dominic T, Madhusoodanan P (1999) Cyanobacteria from extreme acidic environments.
Current science 77: 1021-1023.
Smith S, Abed RM, Gercia-Pichel F (2004) Biological soil crusts of sand dunes in Cape Cod
National Seashore, Massachusetts, USA. Microbial ecology 48: 200-208.
Bu C, Wu S, Yang Y, Zheng M (2014) Identification of Factors Influencing the Restoration of
Cyanobacteria-Dominated Biological Soil Crusts. Plos One 9. doi:
10.1371/journal.pone.0090049
Gundlapally SR, Garcia-Pichel F (2006) The community and phylogenetic diversity of
biological soil crusts in the Colorado Plateau studied by molecular fingerprinting and
intensive cultivation. Microbial Ecology 52: 345-357.
Freeman KR, Pescador MY, Reed SC, Costello EK, Robeson MS, Schmidt SK (2009) Soil CO2
flux and photoautotrophic community composition in high‐elevation,‘barren’soil.
Environmental microbiology 11: 674-686.
van der Meer MT, Schouten S, Damsté JSS, de Leeuw JW, Ward DM (2003) Compoundspecific isotopic fractionation patterns suggest different carbon metabolisms among
Chloroflexus-like bacteria in hot-spring microbial mats. Applied and environmental
microbiology 69: 6000-6006.
Steven B, Gallegos-Graves LV, Belnap J, Kuske CR (2013) Dryland soil microbial communities
display spatial biogeographic patterns associated with soil depth and soil parent material.
Fems Microbiology Ecology 86: 101-113. doi: 10.1111/1574-6941.12143
Gupta RS (2013) Molecular markers for photosynthetic bacteria and insights into the origin
and spread of photosynthesis. Adv Bot Res 66: 37-66.
72

141.

142.

143.

144.

145.

146.

147.

148.
149.

150.

151.

152.
153.

154.

155.
156.

Bowker MA, Belnap J, Davidson DW (2010) Microclimate and Propagule Availability are
Equally Important for Rehabilitation of Dryland N‐Fixing Lichens. Restoration Ecology 18:
30-33.
Acea MJ, Prieto-Fernandez A, Diz-Cid N (2003) Cyanobacterial inoculation of heated soils:
effect on microorganisms of C and N cycles and on chemical composition in soil surface. Soil
Biology & Biochemistry 35: 513-524. doi: 10.1016/s0038-0717(03)00005-1
Zehr JP, Mellon MT, Hiorns WD (1997) Phylogeny of cyanobacterial nifH genes:
evolutionary implications and potential applications to natural assemblages. Microbiology
143: 1443-1450.
Stewart KJ, Coxson D, Siciliano S (2011) Small-scale spatial patterns in N 2-fixation and
nutrient availability in an arctic hummock–hollow ecosystem. Soil Biology and Biochemistry
43: 133-140.
Zehr JP, Wyman M, Miller V, Duguay L, Capone DG (1993) Modification of the Fe protein of
nitrogenase in natural populations of Trichodesmium thiebautii. Applied and Environmental
Microbiology 59: 669-676.
Hui R, Li XR, Jia RL, Liu LC, Zhao RM, Zhao X, Wei YP (2014) Photosynthesis of two moss
crusts from the Tengger Desert with contrasting sensitivity to supplementary UV-B
radiation. Photosynthetica 52: 36-49. doi: 10.1007/s11099-014-0003-3
Krause GH, Weis E (1991) Chlorophyll fluorescence and photosynthesis - The basics. Annual
Review of Plant Physiology and Plant Molecular Biology 42: 313-349. doi:
10.1146/annurev.pp.42.060191.001525
Wynn-Williams D (2000) Cyanobacteria in deserts—life at the limit?The ecology of
cyanobacteria. Springer, pp. 341-366
Colica G, Li H, Rossi F, Li D, Liu Y, De Philippis R (2014) Microbial secreted
exopolysaccharides affect the hydrological behavior of induced biological soil crusts in
desert sandy soils. Soil Biology and Biochemistry 68: 62-70.
Chen L, Rossi F, Deng S, Liu Y, Wang G, Adessi A, De Philippis R (2014) Macromolecular and
chemical features of the excreted extracellular polysaccharides in induced biological soil
crusts of different ages. Soil Biology and Biochemistry 78: 1-9.
Bertrand I, Ehrhardt F, Alavoine G, Joulian C, Issa OM, Valentine C (2014) Regulation of
carbon and nitrogen exchange rates in biological soil crusts by intrinsic and land use factors
in the Sahel area. Soil Biology & Biochemistry 72: 133-144. doi:
10.1016/j.soilbio.2014.01.024
Halvorson JJ, Franz EH, Smith JL, Black RA (1992) Nitrogenase activity, nitrogen-fixation, and
nitrogen inputs by lupines at Mount St-Helens. Ecology 73: 87-98. doi: 10.2307/1938723
Pereira S, Zille A, Micheletti E, Moradas-Ferreira P, De Philippis R, Tamagnini P (2009)
Complexity of cyanobacterial exopolysaccharides: composition, structures, inducing factors
and putative genes involved in their biosynthesis and assembly. Fems Microbiology Reviews
33: 917-941. doi: 10.1111/j.1574-6976.2009.00183.x
Peer T, Turk R, Gruber JP, Tschaikner A (2010) Species composition and pedological
characteristics of biological soil crusts in a high alpine ecosystem, Hohe Tauern, Austria. Eco
Mont-Journal on Protected Mountain Areas Research 2: 23-30.
Canada E (2017). http://www.climatewna.com/climateBC_Map.aspx.
Wang T, Hamann A, Spittlehouse DL, Murdock TQ (2012) ClimateWNA—high-resolution
spatial climate data for western North America. Journal of Applied Meteorology and
Climatology 51: 16-29.

73

157.

158.
159.
160.
161.
162.
163.

164.
165.
166.

167.
168.

169.

170.

171.

172.

173.
174.
175.

Pojar J, Stewart A (1991) Alpine tundra zone. Ecosystems of British Columbia D Meidinger
and J Pojar (editors) Research Branch, BC Ministry of Forests, Victoria, BC Special Report
Series: 263-274.
Farley ALUoBC (2007) Figure 3.2.1 Soil map of British Columbia.
Strong WL (2013) Ecoclimatic zonation of Yukon (Canada) and ecoclinal variation in
vegetation. Arctic: 52-67.
Coxson DS (1991) Impedance measurement of thallus moisture-content in lichens.
Lichenologist 23: 77-84.
Evans R, Lange O (2001) Biological soil crusts and ecosystem nitrogen and carbon dynamics.
Biological soil crusts: structure, function, and management: 263-279.
Stewart KJ, Coxson D, Grogan P (2011) Nitrogen inputs by associative cyanobacteria across
a low arctic tundra landscape. Arctic, Antarctic, and Alpine Research 43: 267-278.
Lang SI, Cornelissen JH, Shaver GR, Ahrens M, Callaghan TV, Molau U, Ter Braak CJ, Hölzer
A, Aerts R (2012) Arctic warming on two continents has consistent negative effects on
lichen diversity and mixed effects on bryophyte diversity. Global Change Biology 18: 10961107.
Schuster RM (1958) Keys to the orders, families and genera of Hepaticae of America north
of Mexico. The Bryologist 61: 1-66.
Dettweiler-Robinson E, Ponzetti JM, Bakker JD (2013) Long-term changes in biological soil
crust cover and composition. Ecological Processes 2: 5.
Root HT, McCune B (2012) Regional patterns of biological soil crust lichen species
composition related to vegetation, soils, and climate in Oregon, USA. Journal of Arid
Environments 79: 93-100.
Stark LR, Whittemore AT (2000) Bryophytes from the northern Mojave Desert. The
Southwestern Naturalist 45: 226-232.
Gao S, Pan X, Cui Q, Hu Y, Ye X, Dong M (2014) Plant Interactions with Changes in Coverage
of Biological Soil Crusts and Water Regime in Mu Us Sandland, China. Plos One 9. doi:
10.1371/journal.pone.0087713
Belnap J, Prasse R, Harper K (2001) Influence of biological soil crusts on soil environments
and vascular plantsBiological soil crusts: structure, function, and management. Springer,
pp. 281-300
Zhang Y-M, Nie H-L (2011) Effects of biological soil crusts on seedling growth and element
uptake in five desert plants in Junggar Basin, western China. Chinese Journal of Plant
Ecology 35: 380-388. doi: 10.3724/sp.j.1258.2011.00380
Sorensen PL, Jonasson S, Michelsen A (2006) Nitrogen fixation, denitrification, and
ecosystem nitrogen pools in relation to vegetation development in the subarctic. Arctic
Antarctic and Alpine Research 38: 263-272. doi: 10.1657/15230430(2006)38[263:nfdaen]2.0.co;2
Knowles RD, Pastor J, Biesboer DD (2006) Increased soil nitrogen associated with
dinitrogen-fixing, terricolous lichens of the genus Peltigera in northern Minnesota. Oikos
114: 37-48. doi: 10.1111/j.2006.0030-1299.14382.x
Adams DG, Duggan PS (2008) Cyanobacteria-bryophyte symbioses. Journal of Experimental
Botany 59: 1047-1058. doi: 10.1093/jxb/ern005
DeLuca TH, Zackrisson O, Nilsson MC, Sellstedt A (2002) Quantifying nitrogen-fixation in
feather moss carpets of boreal forests. Nature 419: 917-920. doi: 10.1038/nature01051
Stewart KJ, Coxson D, Grogan P (2011) Nitrogen Inputs by Associative Cyanobacteria across
a Low Arctic Tundra Landscape. Arctic Antarctic and Alpine Research 43: 267-278. doi:
10.1657/1938-4246-43.2.267
74

176.

177.

178.

179.

180.
181.

Xiao B, Zhao Y, Shao M, Xiao B, Zhao YG, Shao MA (2008) Artificial cultivation of biological
soil crust and its effects on soil and water conservation in water-wind erosion crisscross
region of loess plateau, China. Acta Agrestia Sinica 16: 28-33.
Zaady E, Offer ZY (2010) Biogenic soil crusts and soil depth: a long-term case study from the
Central Negev desert highland. Sedimentology 57: 351-358. doi: 10.1111/j.13653091.2009.01081.x
Raggio J, Green TGA, Sancho LG (2016) In situ monitoring of microclimate and metabolic
activity in lichens from Antarctic extremes: a comparison between South Shetland Islands
and the McMurdo Dry Valleys. Polar Biology 39: 113-122. doi: 10.1007/s00300-015-1676-1
Buedel B, Colesie C, Allan Green TG, Grube M, Lazaro Suau R, Loewen-Schneider K, Maier S,
Peer T, Pintado A, Raggio J, Ruprecht U, Sancho LG, Schroeter B, Tuerk R, Weber B, Wedin
M, Westberg M, Williams L, Zheng L (2014) Improved appreciation of the functioning and
importance of biological soil crusts in Europe: the Soil Crust International Project (SCIN).
Biodiversity and Conservation 23: 1639-1658. doi: 10.1007/s10531-014-0645-2
Billings WD, Mooney HA (1968) The ecology of arctic and alpine plants. Biological reviews
43: 481-529.
Nadelhoffer KJ, Giblin AE, Shaver GR, Laundre JA (1991) Effects of temperature and
substrate quality on element mineralization in 6 arctic soils. ecology 72: 242-253. doi:
10.2307/1938918

75

Plates

Plate 1. A) Andesite Peak in the B.C. Coast Ranges is characterized by rolling topography, with
small wet depressions surrounding by drier rock outcrops. B) Biocrusts on Andesite peak are
comprised of a complex mix of lichens (orange thalli of Solorina crocea visible in image), mosses,
liverworts, and free-living algae and cyanobacteria.
76

Plate 2. A) Mt McIntyre in Yukon is characterized by rolling topography covered by shrubs with
biocrust in interstices. B) Biocrust on Mt McIntyre were lichen-dominated. C) A treatment block on
Mt McIntyre.

77

Plate 3. The control and four experimental treatments (flat, no top soil, microrills, and pit and mound)
in the greenhouse inoculation experiment.

78

Tables
Table 1. Analysis of variance (F values) for measurements taken at week 6 and 12. Fixed effects for
the mixed ANOVA model included treatment, week, and a two-way interaction of treatment and
week. Samples were taken from the same tray were treated as random variable to account for nonindependence between samples within trays. * denotes a significant interaction
Analysis

Parameters

DF

F

p

Biocrust cover

Treatment

4, 40

366.25

< 0.01*

Week

1, 40

46.46

< 0.01*

Treatment: Week

4, 40

2.63

< 0.05 *

Treatment

3, 78

1.63

=0.19

Week

1, 78

9.98

< 0.01*

Treatment: Week

3, 78

1.13

=0.34

Treatment

4, 114

4.41

<0.01*

Week

1, 114

1.91

=0.17

Treatment: Week

4, 114

0.23

=0.92

EPS content

ARAs

79

Table 2.Properties of soils underlying experimental plots 12 weeks following inoculation. Dissolved
organic carbon (DOC), potentially mineralizable nitrogen (Min N), and pHH20 values (±SE). Min N was
significantly higher in pit and mound than in the untreated and flat treatments (ANOVA with Tukey
HSD, p<0.02).

Untreated

Flat

Pit and Mound

Microrills

DOC (ppm)

154

(12.1)

119

(21.1)

133

(12.2)

163

(21.3)

Min N (ppm)

10.8A

(0.5)

10.9A

(0.5)

12.1B

(0.7)

11.7

(0.4)

pH

4.54

(0.13)

4.55

(0.04)

4.60

(0.15)

4.77

(0.04)

80

Table 3. Composition (bryophytes and lichens) of biocrust communities at Andesite and Mt
McIntyre. Nitrogen-fixing lichen species are greyed. Bolded species name indicate that the species
is present at both sites.
Species on Andesite
Amygdalaria panaeola
Psoroma tenue var. boreale
Solorina crocea
Stereocaulon alpinum
Stereocaulon botryosum
Stereocaulon vesuvianum
Anastrophyllum minutum var. minutum
Andreaea rupestris
Anthelia juratzkana
Arthrorhaphis sp.
Baeomyces carneus
Blepharostoma trichophyllum
Bucklandiella microcarpa
Bucklandiella sudetica
Cephaloziella divaricata
Cetraria commixta
Cetraria delisei
Cetraria islandica ssp. islandica
Cladina arbuscula
Cladina rangiferina
Cladonia bellidiflora
Cladonia borealis
Cladonia borealis
Cladonia cervicornis
Cladonia ecmocyna ssp. occidentalis
Cladonia gracilis group
Cladonia macrophyllodes
Cladonia pleurota
Cladonia singularis
Cladonia uncialis var. uncialis
Conostomum tetragonum
Cynodontium tenellum
Dicranoweisia crispula
Dicranum elongatum
Dicranum spadiceum
Diplophyllum taxifolium
Frutidella caesioatra
Gymnomitrion concinnatum

Group
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Liverwort
Moss
Moss
Lichen
Lichen
Liverwort
Moss
Moss
Liverwort
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Moss
Moss
Moss
Moss
Moss
Liverwort
Lichen
Liverwort

Species on Mt McIntyre
Fuscopannaria praetermissa
Nephroma arcticum
Peltigera cf. conspersa
Peltigera didactyla
Peltigera extenuata
Peltigera leucophlebia
Peltigera malacea
Peltigera ponojensis
Peltigera rufescens
Peltigera venosa
Psoroma tenue var. boreale
Stereocaulon alpinum
Stereocaulon tomentosum
Alectoria ochroleuca
Anastrophyllum minutum var. minutum
Anzina carneonivea var. carneonivea
Aulacomnium palustre
Barbilophozia hatcheri
Brachythecium frigidum
Bryocaulon divergens
Bryonora pruinosa
Buellia papillata
Caloplaca livida
Caloplaca stilliciodorum
Caloplaca tirolensis
Catapyrenium cinereum
Catapyrenium daedaleum
Cetraria islandica ssp. islandica
Cetraria odontella
Cladina arbuscula
Cladonia borealis
Cladonia coccifera
Cladonia macrophylla
Cladonia macrophyllodes
Cladonia phyllophora
Cladonia pleurota
Cladonia pocillum
Cladonia pyxidata
81

Group
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Liverw
Lichen
Moss
Liverw
Moss
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen

Kiaeria blyttii
Lecanora zosterae
Lecidea hypnorum
Lecidoma demissum
Lepraria alpina
Leprocaulon albicans
Marsupella brevissima
Marsupella ustulata
Moerckia blyttii
Nardia geoscyphus
Ochrolechia androgyna
Pertusaraia geminipara
Pertusaria oculata
Placidiopsis sp.
Placynthiella icmalea
Placynthiella uliginosa
Plagiothecium laetum
Pleurocladula albescens
Pogonatum contortum
Pohlia sp.
Polyblastia gelatinosa
Polytrichum piliferum
Porpidia contraponenda
Ptilidium ciliare
Ptychostomum sp.
Racomitrium lanuginosum
Rhizocarpon expallescens
Thamnolia vermicularis
Thelenella muscorum var. muscorum
Tritomaria quinquedentata
Umbilicaria hyperborea var. radicicola

Moss
Lichen
Lichen
Lichen
Lichen
Lichen
Liverwort
Liverwort
Liverwort
Liverwort
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Moss
Liverwort
Moss
Moss
Lichen
Moss
Lichen
Liverwort
Moss
Moss
Lichen
Lichen
Lichen
Liverwort
Lichen

Cladonia sulphurina
Cladonia uncialis var. uncialis
Coelocaulon aculeatum s. lat.
Dicranella heteromalla
Dicranoweisia crispula
Dicranum fuscescens
Distichium capillaceum
Ditrichum flexicaule
Flavocetraria cucullata
Flavocetraria nivalis
Hylocomium splendens
Hypnum bambergeri
Hypnum cupressiforme
Lecanora polytropa
Lecidea alpestris
Lecidoma demissum
Masonhalea richardsonii
Ochrolechia androgyna
Pertusaria panyrga
Placynthiella icmalea
Placynthiella oligotropha
Placynthiella uliginosa
Platydictya jungermannioides
Pleurozium schreberi
Pogonatum contortum
Pohlia crudoides
Pohlia nutans
Polyblastia gothica
Polytrichum juniperinum
Polytrichum piliferum
Protothelenella leucothelia
Psoroma sp.
Ptychostomum lonchocaulon
Ptychostomum pendulum
Ptychostomum pseudotriquetrum
Rinodina mniaraea var. mniaraea
Rinodina mniaraea var. mniaraeiza
Rinodina turfacea
Sanionia uncinata
Thamnolia vermicularis
Thelocarpon epibolum
Trapeliopsis granulosa
Tritomaria quinquedentata
82

Lichen
Lichen
Lichen
Moss
Moss
Moss
Moss
Moss
Lichen
Lichen
Moss
Moss
Moss
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Lichen
Moss
Moss
Moss
Moss
Moss

Lichen
Moss
Moss
Lichen
Lichen
Moss
Moss
Moss

Lichen
Lichen
Lichen
Moss
Lichen
Lichen
Lichen
Liverw

Table 4. ANOVA on mixed models of function parameters and linear models of soil properties and
biocrust cover of experimental plots at Andesite and Mt. McIntyre (Site) sampled 0, 12 and weeks
following inoculation (Time). TR = treatment (control, fertilizer only, biocrust only, and biocrust +
fertilizer). Min N = mineralizable nitrogen and DOC= dissolved organic carbon. F-values are not
available for the factor Time, for Cover, DOC and Min N that were only measured at 12 weeks.
Factor

TR
Time
Site
TR * Time
TR * Site
Time*Site
TR*Time*Site

Function Parameter
df
NA
EPS
----F-value ---(p-value)
0.96
3.67
3
(0.42)
(0.02)
16.21
0.70
2
(<0.01)
(0.41)
40.30
76.08
1
(<0.01) (<0.01)
1.18
0.50
6
(0.32)
(0.68)
0.54
2.86
3
(0.66)
(0.04)
81.64
1.71
2
(<0.01)
(0.19)
0.94
0.28
6
(0.47)
(0.84)

df

3

Soil Properties
DOC
Min N
----F-value ---(p-value)
40.4
2.03
(<0.01)
(0.12)

df

3

Cover
-F-value (p-value)
24.9
(<0.01)

1

-

-

-

-

1

157.1
(<0.01)

34.8
(<0.01)

1

1.64
(0.20)

3

-

-

-

-

3

33.2
(<0.01)

0.71
(0.55)

1

-

-

-

-

3

-

-

-

-

2.59
(0.06)

83

84

Figures

Figure 1. Two-way cluster dendrogram of 62 bacterial OTUs with high relative abundances within the six
sample groups (i.e. minimum 2,000 sequences total across all replicates in group) defined by treatment or
control category. Cluster A is comprised of OTUs detected mostly with high abundance in Inoculant, and
nearly absent in Uninoculated and Stored soil control groups. Cluster B is dominated by OTUs most
consistently detected in Uninoculated, Stored and soil surface treatment sample groups and least
consistently detected in the inoculant group. Highest assigned taxonomic identifiers are shown, and genuslevel assignments are indicated in italics.
85

Figure 2. Relative abundance of bacterial 16S rRNA gene sequences. Soil bar represents unincubated soil that
was sampled at the start of the experiment. Aciodobacteria were the most abundant across all treatments
and most bacterial phylum were represented evenly across treatments. Chloroflexi were most abundant in
the inoculant.

86

Figure 3. Two-way Cluster Analysis of 75 fungal OTUs averaged by sample units: Inoculant, biocrust surface
treatments (Flat, Pit & Mound, Microrills), and soils (Uninoculated controls and Stored). Cluster A are OTUs
that are most abundant in the Inoculant, Cluster B comprises OTUs that are most abundant in the soil surface
treatments, and Cluster C comprises OTUs that are most abundant in the soil samples.

87

Figure 4. Relative abundance of fungal ITS2 region sequences. “Stored” bar represents unincubated soil that
was sampled at the start of the experiment. Basidiomycota are shown in shades of green and Ascomycota in
shades of blue.

88

Figure 5. a. Mean (±SE) biocrust percent cover at 6 and 12 weeks (light and dark bar, respectively). Significant
differences are indicated by differences in symbols. Percent cover was significantly higher in the flat,
microrills, and pit and mound treatments than in the untreated and no top soil (ANOVA, Tukey HSD posthoc,
F=8.56, p<0.03); b. Proportional abundance of Fv/Fm at 6 and 12 weeks (left and right bar, respectively)
where no hash-line corresponds to no detected chlorophyll fluorescence (Fv/Fm =0), widely spaced hash-lines
to Fv/Fm between 0 and 0.70, and closely spaced hash-lines to Fv/Fm larger than 0.70. Twelve weeks after
inoculation, Fv/Fm was significantly higher in the flat, pit and mound, and microrills treatments than in the no
top soil treatment and untreated (Kruskal-Wallis, and pairwise comparisons using Tukey and Kramer
(Nemenyi) test with Tukey-Dist approximation for independent samples, chi-squared = 38.7, p< 0.02); c.
Mean (±SE) acetylene reduced at 6 and 12 weeks (light and dark bar, respectively). Microrills treatment had
significantly higher rates of nitrogen fixation than the untreated and no top soil treatments (ANOVA,
TukeyHSD posthoc, F= 9.01, p<0.01). The detection limit for acetylene reduction was 0.2 µmol ethylene m -2hr1
; d. Mean (±SE) glucose content per biocrust at 6 and 12 weeks (light and dark bar, respectively).
Extracellular polysaccharide content was significantly higher in the microrills treatment at 12 weeks than in
the untreated at 6 weeks (ANOVA, Tukey posthoc, F=9.23, p<0.03).

89

Figure 6. Mean acetylene reduced as a function of nifH abundance at 12 weeks. The inoculant had
significantly more nifH copies per gram of soil than the untreated and all treatments (ANOVA, Tukey posthoc,
F=7.91, all p<0.04). nifH gene copy number was not significantly different between surface microtopography
treatments (ANOVA, F=0.912, p=0.45). The detection limit for acetylene reduction was 0.2 µmol ethylene m2 -1
hr

90

Figure 7. Microclimate data from mature biocrust monitoring station on Andesite Peak over the duration of
the experiment, expressed in Julian days. Photosynthetically active radiation (PAR; mol m -2 day-1) (A).
Minimum, average, and maximum biocrust temperatures (°C; long, medium, and short-dash respectively) (B).
Average soil volumetric water content (%; solid line), as well as the minimum, average, and maximum
biocrust relative water content (%, long, medium, and short dashed lines respectively) (C). Modelled
estimation of nitrogenase activity (µmol of ethylene m-2 day-1) with error bars representing standard error for
each daily record (D). The three bars represent measured nitrogenase activity under optimal conditions in
mature biocrusts at three time points over the summer. ) The detection limit for acetylene reduction was 0.2
µmol ethylene m-2hr-1.

91

Figure 8. Microclimate data from mature biocrust monitoring station on Mt McIntyre over the duration of the
experiment, expressed in Julian days. Photosynthetically active radiation (PAR, mol m-2 day-1) (A). Average
air temperature (°C, solid line) and the minimum, average, and maximum biocrust temperatures (°C; long,
medium and short-dashed lines respectively) (B). Average soil volumetric water content (%; solid line) as well
as the minimum, average, and maximum biocrust relative water content (%; long, medium, and short-dashed
lines respectively) (C). Nitrogenase activity under optimal conditions in mature biocrust at three time points
over the summer (D). ) The detection limit for acetylene reduction was 0.2 µmol ethylene m-2hr-1.

92

Figure 9. Biocrust cover 12 weeks following inoculation. The horizontal line represents the initial inoculation
level of 10% biocrust (surface area based). At both Andesite (black bars) and Mt McIntyre (grey bars), plots
inoculated with biocrust (Biocrust and Biocrust & Fertilizer) had a significantly higher biocrust cover than
uninoculated plots (Control and Fertilizer). Different letters indicate significant differences. Bars represent
mean with standard error.

93

Figure 10. Acetylene reduced 0, 6, and 12 weeks following inoculation with biocrust at Andesite (black bars)
and at Mt McIntyre (grey bars). Acetylene reduction as a measure of nitrogen fixation increased at Andesite
and decreased at Mt McIntyre over the 12 weeks. Bars represent mean with standard error.) The detection
limit for acetylene reduction was 0.2 µmol ethylene m-2hr-1.

94

Figure 11. Extracellular polysaccharide content (EPS) of soil surface 12 weeks following inoculation at
Andesite (black bars) and at Mt McIntyre (grey bars). Overall, EPS was significantly lower at Andesite than at
Mt McIntyre. Different letters indicate significant differences. Bars represent means with standard error.

95

Figure 12. Soil properties 12 weeks following inoculation at Andesite (black bars) and at Mt McIntyre (grey
bars). Dissolved organic carbon (A) was higher in inoculated plots (Biocrust and Biocrust & Fertilizer) at
Andesite. Mineralizable nitrogen (B) was higher at Mt McIntyre than at Andesite but not significantly
different between treatments. Different letters indicate significant differences. Bars represent means with
standard error.

96

Acetylene Reduced (umol of ethylene*m2*hr-1)

10

8

6

4

2

0

0

100

200

300

400

500

Moisture (%)
Supplementary Figure 1.
Acetylene reduced by mature biocrust as a function of gravimetric water content. The adjusted R 2 for the
regression model is 0.59, p<0.05. ) The detection limit for acetylene reduction was 0.2 µmol ethylene m-2hr-1.

97

Acetylene Reduced (umol of ethylene*m2*hr-1)

4

Low Light

High Light

3

2

1

0
10

12

14

16

Temperature (°C)

18

20

10

12

14

16

18

20

Temperature (°C)

Supplementary Figure 2. Acetylene Reduction of mature biocrust under high light (left frame) and low light
(right frame) as a function of incubation temperature. Adjusted R 2 are 0.42 and 0.39 for the high and low
light regression models respectively, p<0.05). ) The detection limit for acetylene reduction was 0.2 µmol
ethylene m-2hr-1.

98