The Influence of Biological Crust Inoculum on Dryland Establishment in a Greenhouse Experiment

Rachael Merkt1

The University of Colorado Boulder

Dep. Ecology and Evolutionary Biology Honors Thesis 11-5-2019

Committee Members Dr. Nichole N. Barger1 Dr. Barbara Demmig-Adams1 Dr. Daniel Doak2

1Dep. Ecology and Evolutionary Biology 2Dep. Environmental Studies

1

Abstract Biocrusts influence vascular plant species performance by modifying soil stability, hydrology, and fertility. Multiple studies suggest that biocrusts have species-specific effects on plant species performance depending on plant characteristics and ecological context and that biocrust inoculum could be used in restoration. Therefore, by promoting the performance of some plant species, but inhibiting others, biocrusts may influence vascular plant community structure.

In this study, I compared emergence and establishment of C3 and C4 perennial grasses and annual and perennial forbs grown in biocrust-inoculated versus bare-soil mesocosms in a greenhouse. I hypothesized that biocrust inoculum will increase , growth and species richness of a plant community composed of multiple plant functional types based on increased resource availability of due to biocrusts soil binding effects that trap moisture and decrease evaporation as well as enhanced soil N availability through biocrusts fixing atmospheric N2. A seed mix containing five plant species was sown into 32 mesocosms containing full factorial crosses of two soil treatments (biocrust-inoculated versus bare-soil) and two watering levels (high- water versus low-water). To accept or refute my hypothesis, I tested parameters incorporating plant germination, growth, and species richness. Overall plant growth results indicate biocrusts did not benefit vascular pant species with neutral effects on germination (high-water) and plant species richness but negative effects on germination (low-water) and plant growth. However, the neutral effects that biocrust inoculum has on vascular plants shows that there may be few negative effects overall, allowing restoration to occur in a holistic way that restores both vascular plants and biocrusts.

2

Introduction

Biological Soil Crusts (biocrusts) are soil surface communities consisting of , , , green , fungi, and bacteria that reside in the top few millimeters (mm) of the soil surface and are widespread across dryland landscapes (Williams et al. 2012). Biocrusts coexist within a matrix of vascular plant species in dryland ecosystems worldwide making up approximately 12% of global terrestrial cover (Rodriguez-Caballero et al. 2018). These soil surface communities have mixed effects on plant germination and growth, that are highly dependent on plant species in addition to biocrust type (Zhang et al. 2016; Havrilla et al. 2019).

Recent research suggests that biocrusts may be beneficial to vascular plant species, increasing the performance of native dryland species while inhibiting (Havrilla et al. 2019), influencing soil hydrology (Warren 2001; Chamizo et al. 2016), and increasing soil nutrients

(Delgado-Baquerizo et al. 2013; Barger et al. 2016; Evans and Ehleringer 1993; Evans and Lange

2003) thus, making them a strong candidate for increasing restoration success.

While biocrust inoculum strategies have received interest as a means of restoration in dryland ecosystems (Zhang et al. 2016, Havrilla et al. 2019) the present knowledge of the effects of biocrust inoculum on native plant establishment is limited. Since the topsoil layer microbiome is important to plant growth by providing nutrients, its degradation poses challenges to restoration

(Bateman et al. 2018; Golos and Dixon et al 2014) as most restored regions can be depleted of soil nutrients. Thus, it is essential to restore the microbiome itself, not just plant species. Soil microbiota have the ability to increase root efficiency (amount of mineral uptake per root unit over time (Hunt 1973)) by improving nutrient uptake, increasing root surface area via the presence of fungal hyphae, and making soil nutrients available to plants in general (Mengual et al. 2014).

Biocrust inoculum has the potential to facilitate plant germination, growth, and establishment 3 while improving soil function in terms of soil structure and nutrients which makes biocrust inoculum useful in restoration (Munoz-Rojas et al. 2018) as it will benefit both the soil microbial community and fertility (Xiao and Maik 2017). The addition of cyanobacteria to soil used in restoration has shown a positive outcome without inhibiting the plant establishment with carbon content increasing in degraded 3-fold three months after the soil was inoculated (Munoz-

Rojas et al. 2018). All of these processes are an integral component of drylands and due to the millions of hectares of degraded drylands, restoration solutions need to support plant establishment and survival (Keesstra et al. 2018). Long term success is more important to restoration and the best solutions are those that will incorporate natural ecosystem functioning already present on the landscape such as interactions between the physical effects of biocrusts and vascular plant species.

Within dryland ecosystems, various physical characteristics of biocrusts create an environment beneficial to plant communities. Firstly, biocrusts can alter landscapes physically through their microtopography and effects on local hydrology (Pissolito et al. 2019; Zhang et al.

2016; Thiet et al 2014), and thereby affect rates of plant germination, growth, and survival. In natural landscapes, biocrusts can trap seeds via their physical microtopographies. It is assumed that pinnacled or rolling crusts are more effective at trapping seeds. However, increased roughness does not always benefit seed germination as it increases difficulty for seeds to root within the substrate beneath the biocrust (Zhang et al. 2016). As seed germination is dependent on the availability of water and the temperature of the soil, these microtopographies could have great effect on seedling germination (Zhang et al. 2016). Differences in the beneficial effects of a biocrust-plant relationship could result in various levels of diversity in arid landscapes (Zhang et al. 2016). These different levels of diversity can also be affected by biocrusts influence on soil moisture. 4

Biocrusts increase soil moisture in dryland ecosystems (Chamizo et al. 2016b). Since water availability is the most limiting resource in dryland ecosystems, biocrust effects on soil moisture could be key characteristics (Reynolds et al. 2004; Chamizo et al. 2016b). Studies show that 65% of biocrust research on soil hydrology found evaporation is reduced in biocrust compared to bare soil (Chamizo et al. 2016). When soil surface water content is high, there is a reduction in evaporation and soil moisture is retained longer due to biocrusts clogging soil pores from the establishment of cyanobacterial, , and communities (Chamizo et al. 2013). It is clear that biocrusts improve water resource availability in drylands and this benefits both biocrusts and vascular plants as biocrusts are only physically active in contributing to soil nutrients when wet

(Barger et al. 2016).

The soil nutrient effects of biocrust that pertain to this study are primarily those that alter soil nitrogen (N) content. Communities of biocrusts influence processes of N-fixation and mineralization rates (Delgado-Baquerizo et al. 2013) and contribute N inputs through biological fixation and capture, which they can then transform through N-fixation into N compounds that can be taken up by plants (Barger et al. 2016). Fixation of atmospheric N2 in biocrusts is a ubiquitous biogeochemical transformation, a major source of N nutrients in drylands (Evans and

Ehleringer 1993, 2003) and as N is a necessary plant nutrient its presence in soil is important to plant growth and survival. These soil nutrient resources are released after the process of N-fixation in biocrusts, known as extracellular release, in which 5-70% of fixed-N are released by N-fixing species (Magee and Burris 1954; Silvester et al 1996) including N compounds that can be taken up by plants. Biocrusts thus have a positive effect on soil nutrient availability, providing more fixed-N to propagating species. Due to this, I predict that biocrust inoculation will enhance soil N availability. 5

Research Questions and Hypothesis

In order to better understand how biocrust restoration strategies influence native plant establishment, I addressed the question of how biocrust inoculum influences plant germination, growth, and species richness. I hypothesized that biocrust inoculum will increase germination, growth and species richness of a plant community composed of multiple plant functional types based on increased resource availability of soil moisture due to biocrusts soil binding effects that trap moisture and decrease evaporation as well as enhanced soil N availability through biocrusts fixing atmospheric N2.

Methods

I conducted a greenhouse experiment in the University of Colorado Boulder 30th St.

Greenhouse using biocrust inoculum and soil sourced from Canyonlands National Park. For this experiment, I selected five plant species based on their plant functional types and their existence within the regional species pool of the semi-arid ecosystems of southeastern Utah, from where my biocrust was sourced. I conducted germination trials on 15 possible species to determine if the seeds were viable and which showed promising rates of germination. The finalized selections were made from these original species based on data provided by the USGS Plant Database and their availability with native seed companies that provide seeds that are not fertilized and harvested in the United States. The final five species used in my experiment were two perennial grasses, Leymus cinereus (C3) and Aristida purpurea (C4); two perennial forbs, Sphaeralcea grossulariifolia (C3) and Astragalus convallerius (C4); and one summer annual, Helianthus annuus (C3).

6

Experimental Design

To address the research questions, I devised four treatments in total, two soil treatments of biocrusts versus bare-soil at two watering levels of high water versus low water to ensure germination. These treatments were applied to 32 mesocosms (2x1’ plots) with 8 replicates per treatment. Soil moisture was recorded using soil moisture probes placed beneath the soil surface in four of the mesocosms in treatments A, B, C and D. Soil moisture probes took readings every four hours for the duration of the greenhouse stage of the experiment. These readings were transferred from the device into a spreadsheet. For readability in the figures, one reading was selected from every day to include in the mesocosm. In order to not include times directly after watering – as these would have a high biased towards high soil moisture content – the reading selected was in the late afternoon as watering occurred in the morning or early afternoon. These values were plotted on a line graph that included both bare-soil and biocrust-inoculated mesocosm values for the high-water treatment mesocosms over the course of one month per mesocosm.

Each mesocosm contained 75 seeds in total, 15 of each of the five species selected. Seeds were randomly placed in the mesocosm by hand. The inoculum and the seeds were placed on the soil within two days of each other with the inoculation of the selected soils occurring first.

Plant Responses

Plant Germination and Soil Moisture

Following the placement of seeds and inoculation of soils, germination (%) and survivorship was tracked by recording germination for the first month of experimentation and survivorship of seeds germinated for the duration of the experiment. The low-water mesocosms were watered on every fifth day and the high-water mesocosms on every third day beginning the 7 day of seeding. Once germination of a seed occurred, the date was recorded. If a seed died after germination, the date it died was also recorded. Once the germination term of one month (30 days) ended, misting was stopped, and the soil moisture probes were placed in each mesocosm (so that misting would not contribute to water treatment levels once germination occurred). Survivorship was monitored throughout the experiment but the official surviving species in each mesocosm were those alive at the end of the greenhouse experiment.

Plant Growth and Survival

Height (cm) was recorded every other week beginning the week the month-long germination period ended at the four-week, six-week, and eight-week point for each live individual plant using a ruler. Height was the measurement from the base of the plant at the soil surface to the tip of all plant organs (stem, leaves, or flowers).

The surviving plants at 10 weeks were harvested and separated into their roots and shoots. The roots and shoots were placed in individually labeled coin envelopes and these envelopes and their contents was dried in a Thermo Scientific model oven at 60 degrees Celsius for 48 hours in total. Once dried, roots and shoots were removed from their envelopes, weighed, and the root-to-shoot ratio was calculated as:

Root ÷ Shoot = R:S

Species richness was recorded after each surviving species was collected following the greenhouse stage of experimentation and calculated as:

Number of Species / Mesocosm = Species Richness 8

Total biomass (g) was determined for all surviving species per mesocosm. These weights include all plant material and are not split by root or shoot. Drying occurred after harvesting for

48 hours.

Plant density per mesocosm was determined for all surviving species at the end of the greenhouse stage of experimentation and calculated as:

Number of Individuals/Area = Density

Statistical Analysis

Statistical analysis was conducted using R Studio under the version of 3.5.2 “eggshell igloo.” For each figure, the mean and standard error were calculated for error bars using the dataSE function. Packages used to create these figures were ggmesocosm, ggmesocosm2, and plyr packages.

Germination results were analyzed using R and standard error was calculated using the dataSE function. Significance was calculated using a two-way ANOVA and Tukey test. Ending growth in terms of density, root-to-shoot ratio, and biomass was analyzed using R and standard error was calculated using the dataSE function. Significance for growth was calculated using a two-way ANOVA. Parameters associated with soil moisture were analyzed using R with standard error calculated using the dataSE function. Significance was calculated using a one-way ANOVA.

For ending mesocosm species richness, results were analyzed using R and statistical analysis were performed using a one-way ANOVA. Standard error was calculated using the dataSE function.

9

Results

Germination

In the high-water mesocosms, biocrust inoculation did not increase the percent of

germination relative to the bare-soil mesocosms (p = 0.366) and there was no significance within

species by treatment (p = 0.866) (Figure 1). Thus, biocrust inoculum did not increase percent of

germination in inoculated mesocosms. For the low water mesocosms, biocrust-inoculated

mesocosms were significantly lower in percent germination than bare-soil (p = 0.000737) and

there was a significant difference within species between treatments (p = 0.0061) with A.

purpurea showing lower germination rates in the biocrust-inoculated mesocosms than the bare-

soil (p = 0.0003).

Figure 1

pTreatment = 0.336

pSpecies :Treatment = 0.866

H. annuus

Percent germination per species for the high water mesocosms (percent germination mean per species ± SE) shows no significance between biocrust-inoculated vs. bare-soil mesocosms (p = 0.336) and no significance within species by treatment (p = 0.866). 10

Figure 2

pTreatment = 0.000737***

pSpecies :Treatment = 0.0061**

H. annuus

Percent germination per species for the low water mesocosms (percent germination mean per species ± SE) shows biocrust-

inoculated mesocosms being significantly lower in percent germination than bare-soil (p = 0.000737) and significance within species between treatments (p = 0.0061) specifically in A. purpurea (p = 0.0003)

Plant Growth and Survival

Plant height after eight weeks were significantly lower in the biocrust-inoculated versus

the bare-soil mesocosms (p = 1.58e-08). However, there is not a significant difference between

treatments within species (p = 0.055). 11

Figure 3

p = 1.58e-08*** Treatment p = 0.055 Species :Treatment

Eight week species height per treatment by species (height mean per treatment by species ± SE) shows a significant difference between biocrust-inoculated vs. bare-soil mesocosms (p = 1.58e-08) with lower heights in biocrust-inoculated mesocosms than bare-soil. However, there is no significant difference within species between treatments (p = 0.055)

Root-to-shoot ratio showed no significant difference between the biocrust-inoculated and

bare-soil mesocosms (p = 0.462) and no significant difference in root-to-shoot ratio between

treatments within species (p = 0.961) (Figure 4). 12

Figure 4

pTreatment = 0.462

pSpecies:Treatment = 0.961

Root to shoot ratio per treatment by species (root:shoot mean per treatment by species ± SE) shows no significant difference between the biocrust-inoculated vs. bare-soil mesocosms (p = 0.462) and no significant difference within species between treatments (p = 0.961).

13

Species richness are not significantly greater for the biocrust-inoculated versus the bare-

soil mesocosms with a p-value of p = 0.199 (Figure 5).

Figure 5

pTreatment = 0.199

Species richness (Number of Species / Mesocosm = Species Richness) by treatments (species richness mean per treatment ± SE) shows that there is no significant difference between biocrust-inoculated and bare-soil mesocosms (p = 0.199)

Total, root, and shoot biomass were all significantly lower biomass in biocrust

mesocosms in comparison to bare-soil mesocosms. Total biomass showed a significant

difference between biocrust-inoculated versus bare-soil mesocosms (p = 1.18e-06) and a

significant difference in total biomass within species between treatments (p = 1.98e-11) (Figure

6). H. annuus showed the largest difference in total biomass with biomass in the biocrust-

inoculated mesocosms being lower than the bare-soil (p < 0). Root biomass results showed a

significant difference between biocrust-inoculated and bare-soil mesocosms (p = 0.00144) and 14

between treatments within species (p = 0.036) (Figure 7). H. annuus again shows a significantly

lower root biomass in the biocrust-inoculated mesocosms than the bare-soil (p = 0.025). This

shows that biocrust inoculum does decrease root biomass (as this would correlate to a smaller

root-to-shoot ratio) but it does not in turn increase shoot biomass with shoot biomass results

showing that biocrust-inoculated mesocosms have significantly less shoot biomass than bare-soil

mesocosms (p = 1.34e-06). Additionally, there is a significant difference in shoot biomass

between treatments within species (p = 4.15e-13) (Figure 8) with H. annuus having significantly

lower biomass in the biocrust-inoculated mesocosms than bare-soil (p < 0). In relation to density,

there is not a negative correlation between biomass and density but a weak positive correlation

(R2 = 0.558) (Figure 10) and thus the negative effects seen are not a result of greater plant

density increasing competition.

Figure 6

pTreatment = 1.18e-06***

PSpecies:Treatment = 1.98e-11***

Total biomass by species per treatment (total biomass mean per species by treatment ±SE) shows that the biocrust-inoculated mesocosms have significantly lower total biomass than the bare-soil mesocosms (p = 1.18e-06) and a significant difference within species between treatments (p = 1.98e-11) specifically in H. annuus (p < 0). 15

Figure 7

pTreatment = 0.00144**

pSpecies:Treatment = 0.036*

Total root biomass by species per treatment (root biomass mean per species by treatment ±SE) shows that the biocrust-inoculated mesocosms have significantly lower root biomass than the bare-soil mesocosms (p = 0.00144) and a significant difference within species between treatments (p = 0.036) specifically in H. annuus (p = 0.025). 16

Figure 8

pTreatment = 1.34e-06***

pSpecies:Treatment = 4.15e-13***

Total shoot biomass by species per treatment (shoot biomass mean per species by treatment ±SE) shows that the biocrust-inoculated mesocosms have significantly lower shoot biomass than the bare-soil mesocosms (p = 1.34e-06) and a significant difference species within species between treatments (p = 4.15e-13) specifically in H. annuus (p < 0).

17

Plant density was not significantly different between the inoculated and bare-soil

mesocosms (p = 0.667) for density in plants per m2 but and shows that there is no significant

difference between treatments within species (p = 0.463) (Figure 9).

Figure 9

pTreatment = 0.667

pSpecies:Treatment = 0.463

Density (Number of Individuals/mesocosm area = Density) shows density by species per treatment (density mean by species per treatment ± SE) shows that there is no significant difference between the inoculated and bare-soil mesocosms (p = 0.667) and shows no significance between treatments within species (p = 0.463).

18

Figure 10 Correlation Between Biomass and Density 0.2 0.18 R² = 0.0558 0.16 0.14

0.12 Biomass 0.1 Linear (Biomass)

Biomass Linear (Biomass) 0.08 0.06 0.04 0.02 0 0 10 20 30 40 50 60 Density

Figure 10 shows that there is not a negative correlation between biomass and density

with density slightly increasing as biomass increases.

Soil Moisture

Results from soil moisture probes show that soil moisture was not higher in the biocrust-

inoculated mesocosms. In the first month, there was significantly higher soil moisture in the

bare-soil mesocosms than the inoculated mesocosms (0.00155) (Figure 11). In the second month

of soil moisture readings no significant differences were detected between the inoculated and

bare-soil mesocosms (p = 0.879) (Figure 12). Additionally, most species germinated around days

15-20 of the first month and it was after this that bare-soil and biocrust-inoculated soil moistures

were more similar continuing into month two where there was no significant difference between

treatments’ soil moistures. 19

Figure 11

pTreatment = 0.00155

Majority of Majority species begin germinate to

Soil moisture in the first month shows that the biocrust-inoculated mesocosms have lower soil moisture as the bare-soil mesocosms trend

higher in soil moisture throughout the first month (p = 0.00155). Differences between soil moisture decrease once a majority of plants begin to germinate around days 15-20 of the first month.

Figure 12

pTreatment = 0.879

Soil moisture in the second month of treatment, after the growth of the inoculum, shows that the biocrust-inoculated mesocosms and the bare-soil mesocosms now have no significant difference (p = 0.879). 20

Discussion

This experiment on the effects of biocrust inoculum on the germination, growth, and survival of five dryland plant species in a greenhouse experiment indicates that effects of biocrusts on plants may not be reproducible in a short term, controlled environment experiment.

However, the negative results seen could be context-dependent and result largely from various effects on controlled greenhouse conditions such as temperature variability and seedling viability. One could speculate that biocrust’s functional impacts on natural landscapes involve and ability to decrease the establishment of weedy or exotic species (Hernandez and Sndquist

2011; Li et al. 2006; Morgan 2006).

Germination

My results indicate that our hypothesis that germination is increased by the presence of a biocrust inoculum was rejected as there was not significantly higher percent of germination in the biocrust inoculate mesocosms in comparison to the bare-soil mesocosms regarding the high- water mesocosms (Figure 1). Additionally, our hypothesis is rejected because the low water treatment exhibited significantly lower rates of germination for biocrust-inoculated mesocosms

(Figure 2). This is consistent with past research which has shown that that the overall effects of a biological soil crust on germination were not significant (Havrilla et al. 2019). As observed, biocrusts can benefit vascular plant germination through increasing nutrient levels and water retention (Zhang et al. 2016) However, a reason for the lack of significance between treatments in my research may be the controlled greenhouse environment. Once seeding occurred, daily misting likely kept seeds moist regardless of whether or not a biocrust was present. The influence of biocrusts on vascular plant performance can depend on biocrust type, plant species, and local (Zhang et al. 2016). Within this small-scale environment, the biocrust inoculum, species 21 chosen, and climate were all controlled to exhibit baseline level interactions under such controlled conditions. However, it is possible that research conducted in natural environments could yield different results as there would be larger variation in climate and local heterogeneity in microtopography and plant species. Despite this, it is clear that germination plays a role in soil moisture as it is after a majority of species germinated (Figure 11) that there were fewer differences between the soil moisture levels of the biocrust-inoculated and bare-soil mesocosms and thus the no longer dormant germinating seeds were using more water.

The finding that percent germination was lower in the biocrust versus bare-soil in the low water treatment may have been due to water resources being too low as the high and low water treatments were part of experimental design to ensure survival in at least one set of inoculated mesocosms and bare-soil mesocosms respectively. Additionally, there are differences in biocrust’s soil hydrology effects due to seasonal variation in natural environments (Warren 2001) and when a biocrust is exposed to prolonged dry periods, crusts crack and the crusts lose their ability to bind soil particles (Warren 2001). This results in water retention decreasing as the lack of bound soils would leave them with evaporation rates similar to bare-soil. Moreover, as the biocrust inoculum propagated it required water and both the seeds and inoculum could have been taking up water during the misting period making competition for water high. Had there been more watering via irrigation during misting as there was in the high water mesocosms these effects may not have been seen but due to fewer days of irrigation watering in the low water treatment, the biocrust-inoculated mesocosms experienced significantly lower germination rates than the bare-soil. Additionally, there were lower rates of germination specifically in A. purpurea in the biocrust-inoculated mesocosms in comparison to the bare-soil mesocosm. As A. purpurea is a C4 grass species, this was not expected as in past research C4 grasses were benefited 22

compared to C3 by the presence of a biocrust (Havrilla et al. 2019). Again, this could largely be due to the biocrust inoculum requiring water to propagate at the same time as the seeds required water for germination and water levels in the low water treatment may have simply been too low.

Plant Growth and Survival

Measures of plant growth and survival includes root-to-shoot ratio, total root and shoot biomass, plant height, density, and species richness. For root-to-shoot ratio, there was no significant difference between the biocrust-inoculated and bare-soil mesocosms (Figure 4) and thus the hypothesis that biocrust inoculum decreases root-to-shoot ratio (smaller roots) in the selected species is rejected. Under different resource conditions, root-to-shoot ratio can increase or decrease. Since plants use their roots to take up water, root biomass can be an indicator of soil nutrient levels and water availability (Xu et al. 2015). The findings in my experiment between biocrust and bare-soil suggests that the inoculum did not increase competition. Within ecosystems, resource availability can alter competition. Higher resource availability within soils can result in greater competition with more species able to propagate and survive (Kardol et al.

2012). Furthermore, the role of competition in arid landscapes is relatively important, with competition leading to differences in plant density, growth, and survival with more species in a smaller area resulting in more competition (Fowler et al. 1986). Within arid ecosystems, the interactions between the many plants that may be on the landscape and biocrust correlate with productivity (Dettweiler-Robinson et al. 2018). With no difference between the inoculated and non-inoculated mesocosms, there is evidence that the biocrust inoculum is not increasing competition and impeding growth. More importantly, if the biocrust inoculum propagates with establishing plant communities, it can facilitate growth long term. Though this experiment was short term, literature indicates that biocrust could be a long-term solution to restoration success 23 as when in association with a biocrust, native species whose biocrust is removed experience a decline in growth and thus biomass (Dettweiler-Robinson 2018).

My results for total, root, and shoot biomass reject my hypothesis that biomass will be increased by the presence of a biocrust inoculum because the biocrust-inoculated mesocosms have significantly lower biomass than the bare-soil mesocosms (Figure 6). In correlation to density, there was no significant difference between species density in the biocrust-inoculated mesocosms versus the bare-soil mesocosms (Figure 6). If an increase in species density correlates to a decrease in biomass these results would be indicative of competition among plant species as the biocrust would support more plants and thus increase competition among them.

This would be consistent with a recent study by Kidron (2019) in which it was shown biocrusts can reduce individual plant mass while benefitting cover and plant survival through improving plant growth conditions in drylands (Chen and Duan 2015) However, as seen in Figure 10, there is no negative correlation between biomass and density. Thus, the biocrust itself may be the cause for increased competition. Additional support of this view can be seen in the root and shoot biomass as both were significantly smaller in the biocrust-inoculated mesocosms than the bare-soil mesocosms and H. annuus showed significantly lower biomass in the biocrust- inoculated than bare-soil mesocosms (Figure 7; Figure 8). As for root development, physical and chemical crusting on soils can inhibit root penetration (Chartres 1992) and a combination of physical and biological crust effects can be dependent on crust development with most results being intermediate (Belnap et al. 2001). Results indicating growth impediment by biocrusts were also seen by Havrilla et al. (2019) as biocrust presence decreased overall plant community growth by 42%. While the duration of greenhouse experimentation could be lengthened in future 24 experiments, this data is important in understanding how propagating inoculum affects the early growth stages of vascular plants.

Plant height was not greater in the biocrust-inoculated mesocosms and thus my hypothesis is rejected as results show heights in the biocrust-inoculated mesocosms being significantly lower than the bare-soil mesocosms (Figure 3). The observed significant difference between species in plant height is largely due to the difference in growth rates between species.

As for differences between biocrust versus bare-soil treatments, height is clearly impeded in the biocrust-inoculated mesocosms and this corresponds to the lower biomass in inoculated mesocosms. Again, this research further shows the decrease in plant growth as recorded by

Havrilla et al. (2019). While there have been multiple studies that test the relationship between biocrusts and vascular plants (Belnap et al. 2001; Zhang et al. 2016) some have shown that the removal of biocrusts can result in an increase in vascular plant growth in the field (Zhang et al.

2016; Beyschlag et al. 2008; Hernandez and Sandquist 2011; Langhans et al 2010; Li et al.

2006). This suggests that while there are benefits to biocrust, there may also be negative effects as biocrusts can slow growth processes and expose seedlings to environmental stress for longer as they require more time to establish themselves and ensure a greater chance of survival

(Escudero et al. 2007; Zamfir 2000). Thus, it is clear that negative effects to plant growth were seen as a result of more environmental stress on the establishing research species. Moreover, factors of stress were not alleviated by higher soil moisture as there was no significant difference between soil moisture between inoculated and bare soul mesocosms in the second month of experimentation after the inoculum had propagated (Figure 12).

My results show that overall plant density is not higher in the inoculated mesocosms

(Figure 9) versus the bare-soil mesocosms and thus, the hypothesis that density will be increased 25 by biocrust inoculum is rejected. Additionally, there was no difference within species between treatments (Figure 9). Again, this could be a factor of soil moisture (Figure 11; Figure 12) as the biocrust mesocosms (when the inoculum had propagated) are not significantly different in soil moisture. In addition to this, greenhouse conditions may have resulted in optimal seedling germination and survival and thus similar density in both treatments. As shown previously, germination rates from this experiment are not significantly different between treatments and therefore it is expected that density is not greater in correlation with germination in the high- water mesocosms as there was no difference between treatments. Furthermore, the lack of survival in the low water mesocosms and thus lack of cover is indicative of poor plant recruitment in low water conditions. As found by Moreira-Grez et al. (2019), few species performed well on all levels of plant fitness in a semi-arid restoration project using soil microbe inoculum. This suggests that inoculation procedures will need to consider climatological factors such as frequency at various times of year and or the use of an irrigation system in restoration to establish species.

Species Richness results indicate that the hypothesis that biocrust inoculum will increase the species richness in inoculated mesocosms is rejected as there is no significant difference between the biocrust-inoculated and bare-soil mesocosms (Figure 5). Species abundance and number of species is a mechanism of understanding a community and the drivers of ecosystem functioning (Castillo-Monroy et al. 2011). In addition, the effect of species richness on a given ecosystem’s multifunctionality can depend on the evenness and spatial pattern of species along with species composition (Maestre et al. 2012). While some studies show greater resource availability increases competition (Fowler 1986), it is most prominent when water availability is low. As these figures are based on the high-water mesocosms, water was present and thus may 26 have resulted in the lack of significant difference between soil moisture once the inoculum propagated (Figure 12).

While survival favored biocrust-inoculated mesocosms in the high-water treatment, there was no survival in the low-water mesocosms for either the biocrust-inoculated or bare-soil mesocosms thus species richness is based on the high-water mesocosms. However, biocrust propagation in low-water mesocosms continued despite water availability being too low for plant recruitment. This suggests, as past research has shown, that biocrusts react to low resource availability differently than vascular plant species (Pissolito et al 2019). In addition, the biocrust inoculum likely had a higher resistance to low-water conditions as it can be propagated from a dormant state as it was in this experiment.

Conclusion

This study produced results for a controlled environment and species selection. While overall plant growth results indicate biocrusts did not benefit vascular plant species, further research is needed to determine the best biocrust inoculum type, resource availability, and native vascular plant species to use in restoration in the field. Moreover, any negative effects of biocrusts on early vascular plant growth may be a component of their functionality on arid landscapes in decreasing establishment of weedy and exotic species, thus increasing overall resistance (Hernandez and Sndquist 2011; Li et al. 2006; Morgan 2006). However, those effects that were neutral for germination and species richness show that biocrust inoculum may not have a large number of negative effects overall.

The close association vascular plants in dryland ecosystems have with biological soil crusts is an indication that crust restoration is equally as important as vegetative restoration 27

(Roncero-Ramos et al 2019). Thus, biocrust recovery in the form of inoculation in correlation with vegetative plant recovery should be strongly considered as a method of restoration in arid lands. More importantly, current rates of restoration in drylands are low and innovative techniques involving plant restoration, soil rejuvenation, and restoration success are the most important to implement in the coming future (Munoz-Rojas 2018). In addition, if biocrusts are invaluable in ways such as these it is important to note they could decrease by 40% in the next few decades (Havrilla et al. 2019). While can influence vascular plants, it can also greatly affect biocrusts (Chen et al. 2019). Due to the important role biocrusts play in community assemblage in arid ecosystems, their loss due to climate change could greatly affect these ecosystems. Therefore, the most successful restoration techniques are those that incorporate the natural coexistence in drylands between soil microbial communities and plant species taking into account how species interact with one another depending on biocrust type, vascular plant species present, resource availability, and biocrust inoculum type used.

Acknowledgments

I would like to firstly thank Dr. Nichole Barger and Dr. Caroline Havrilla for their involvement in mentoring me through the duration of this research. I would also like to thank the

Biological Sciences Initiative Program and the Undergraduate Research Opportunities Program for awarding me two grants that funded me and my research as an undergraduate at the

University of Colorado Boulder. Additionally, I would like to thank the CU Boulder

Greenhouses staff who granted me space in the 30th St. Greenhouse to conduct this research.

Lastly, I would like to thank my committee members, Dr. Nichole Barger, Dr. Barbara

Demmig-Adams, and Dr. Dan Doak and the Ecology and Evolutionary Biology Honors Program for this opportunity. 28

Literature Cited

Barger NN, Weber B, Garcia-Pichel F, Zaady E, Belnap J. 2016. Patterns and Controls on

Nitrogen Cycling of Biological Soil Crusts. In: Weber B, Budel B, Belnap J. Biological

Soil Crusts: An Organizing Principle in Drylands. Ecol Studies (226). Switzerland:

Springer International Publishing 257-285.

Bateman A, Lewandrowski W, Stevens J, Munoz-Rojas M. 2018. Ecophysiological indicatorsto

assess drought responses of arid zone native seedlings in reconstructed soils. Land

Degrad Dev 29:984-993.

Belnap J, Prasse R, Harper KT. 2001. Influence of Biological Soil Crusts on Soil Environments

and Vascular Plants. In: Belnap J, Lange OL. Biological Soil Crusts: Structure, Function,

and Management. Ecol Studies (150). Berlin: Springer-Verlag Berlin Heidelberg 281-

300.

Beyschlag W, Wittland M, Jentsch A, Steinlan T. 2008. Soil crusts and benefit plant

germination, establishment, and growth on nutrient deficient sand. Basic Ap Ecol 9:243-

252.

Castillo-Monroy AP, Bowker MA, Maestre FT, Rodriguez-Echeverria S, Martinez I et al. 2011.

Relationship between biological soil crusts, bacterial diversity and abundance, and

ecosystem functioning: Insights from a semi-arid Mediterranean environment. J Veg Sci

22(1):165-174. doi: 10.1 1 1 1/j. 1654-1 103

Chamizo S, Belnap J, Eldridge DJ, Canton Y, Issa OM. 2016. The Role of Biocrusts in Arid

Land Hydrology. In: Weber B, Budel B, Belnap J. Biological Soil Crusts: An Organizing 29

Principle in Drylands. Ecol Studies (226). Switzerland: Springer International Publishing

321-346.

Chamizo S, Canton Y, Rodrigues-Caballero E, Domingo F. 2016b. biocrusts positively affect the

soil water balance in semiarid ecosystems. Ecohydrology 9:1208-1221. doi:

10.1002/eco.1719

Chamizo S, Canton Y, Lizaro R, Domingo F. 2013. The role of biological soil crusts in soil

moisture dynamics in two semiarid ecosystems with contrasting soil textures. J Hydrol

489:14-84.

Chen, X., Duan, Z., 2015. Impacts of soil crusts on soil physicochemical characteristics in

different rainfall zones of the arid and semi-arid desert regions of northern China.

Environ Earth Sci 73: 3335–3347.

Chen N, Liu X, Zheng K, Zhang C, Liu Y et al. 2019. Ecohydrological effects of biocrust type

on restoration dynamics in drylands. Science of the Total Environment 687:527-534. doi:

https://doi.org/10.1016/j.scitotenv.2019.06.007 0048-9697/ .

Delgado-Baquerizo M, Morillas L, Maestre FT, Gallardo A. 2013. Biocrusts control the nitrogen

dynamics and microbial functional diversity of semi-arid soils in response to nutrient

additions. Plant Soil 372(1-2):6443-654.

Dettweiler-Robinson E, Sinsabaugh RL, Rudgers JA. 2018. Biocrusts benefit from plant

removal. Am J Bot 105:1133-1141. doi:10.1002/ajb2.1120 30

Escudero A, Martinez I, de la Cruz A, Otalora MAG, Maestre FT. 2007. Soil lichens have

species specific effects on the seedling emergence of three gypsophile plant species. J

Arid Environ 70:18-28.

Evans RD, Ehleringer JR. 1993. A break in the in aridlands – evidence from delta

N-15 of soils. Oecologia 94:314-317.

Evans RD, Lange OL. 2003. Biological soil crusts and ecosystem carbon and nitrogen dynamics.

In: Belnap J, Lange OL. Biological Soil Crusts: Structure, Function, and Management.

Ecol Studies (150). Berlin: Springer-Verlag Berlin Heidelberg 263-280.

Fowler N. 1986. The role of competition in plant communities in arid and semiarid regions. An

Rev Ecol and Systematics 17:89-110. doi: 10.1146/annurev .es.17.110186.000513

Golos PJ, Dixon KW. 2014. Waterproofing topsoil stockpiles minimize variability decline in soil

seed bank in an arid environment. Rest Ecol 22:495-501.

Havrilla CA, Chaudhary VB, Ferrenberg S, Antoninka AJ, Belnap J et al. 2019. Towards a

predictive framework for biocrust mediation of plant performance: A meta-analysis. J

Ecol 107:2789-2807. doi:10.1111/1365-2745.13269

Hernandez RR, Sandquist DR. 2011. Disturbance of biological soil crusts increases emergence

of exotic vascular plants in California sage scrub. Plant Ecol 212:1709-1721.

Hunt R. 1973. A method of estimating root efficiency. J Ap Ecol 10(1):157-164.

Kardol P, Souza L, Classen AT. 2012. Resource availability mediates the importance of priority

effects in plant community assembly and ecosystem function. Oikos 122:84-94. doi:

10.1111/j.1600-0706.2012.20546.x 31

Keesstra S, Nunes J, Novara A, Finger AD, Kalantari Z, Cerdà A (2018) The superior effect of

nature based solutions in land management for enhancing ecosystem services. Sci Tot

Environ 610:997–1009. doi: https://doi.org/10.1016/j. scitotenv.2017.08.077

Kidron GJ. 2019. The dual effect of sand-covered biocrusts on annual plants: increasing cover

but reducing individual plant biomass and fecundity. Catena 192(104120). doi:

https://doi.org/10.1016/j.catena.2019.104120

Langhans TM, Storm C, Schwabe A. 2010. Regeneration processes of biological soil crusts,

macro-cryptogams and vascular plant species after fine-scale disturbance in a temperate

region: recolonization or successional replacement? Flora 205:46-60.

Li SZ, Xiao HL, Cheng GD, Luo F, Liu LC. 2006. Mechanical disturbance of microbiotic crusts

affects ecohydrological practices in a region of revegetation-fixed sand dunes. Arid Land

Res Manage 20:61-77

Maestre FT, Castillo-Monroy AP, Bowker MA, Ochoa-Hueso R. Species richness effects on

ecosystem multifunctionality depend on evenness, composition, and spatial pattern. J

Ecol 100(2):317-330. doi: 10.1111/j.11 365-2745.201 1.01918.x.

Magee WE, Burris RH. 1954. Fixation of N2 and utilization of combined nitrogen by Nostoc

muscorum. Am J Bot 41:777-782.

Mengual C, Schoebitz M, Azcon R, Roldan A. 2014. Microbial inoculants and organic

amendment improves plant establishment and soil rehabilitation under semiarid

conditions. J Environ Manage 134:1-7. doi:

http://dx.doi.org/10.1016/j.jenvman.2014.01.008 32

Moreira-Grez B, Munoz-Rojas M, Kariman K, Storer P, O’Donnell AG et al. 2019.

Reconditioning degraded mine site soils with exogenous soil microbes: plant fitness and

soil microbiome outcomes. Frontiers Microb 10:1617. doi: 10.3389/fmicb.2019.01617

Morgan JW. 2006. mats inhibit germination of non-native species in burnt temperate

native grassland environments remnants. Biol Invas 8:159-168.

Munoz-Rojas M, Romain JR, Roncero-Ramos B, Erickson TE, Merritt DJ et al. 2018.

Cyanobacteria inoculation enhances carbon sequestration in soil substrates used in

dryland restoration. Sci Tot Environ 636(1149-1154). doi:

https://doi.org/10.1016/j.scitotenv.2018.04.265 0048-9697/

Pissolito C, Gaibotti IA, Varela SA, Arana V, Gonzalez-Polo M. 2019. Water-mediated changes

in plant-plant and biological soil crust-plant interactions in a temperate forest ecosystem.

Web Ecol 19:27-38. doi: https://doi.org/10.5194/we-19-27-2019

Reynolds J, Kemp P, Ogle K, Fernandez R. 2004. Modifying the ‘pulse-reserve’ pardifm for

deserts of North America: precipitation pulses, soil water, and plant responses. Oecologia

141:194-201.

Rodriguez‐Caballero E, Belnap J, Büdel B, Crutzen PJ, Andreae MO, Pöschl U, Weber

B. 2018. Dryland photoautotrophic soil surface communities endangered by global

change. Nat Geosci 11: 185.

Roncero-Ramos B, Roman JR, Gomez-Serrano C, Canton Y, Acien FG. 2019. Production of a

biocrust-cyanobacteria strain (Nostoc commune) for large-scale restoration of dryland

soils. J Ap Phycol 31:2217-2230. doi: https://doi.org/10.1007/s10811-019-1749-6 33

Silvester WB, Parsons R, Watt PW. 1996. Direct measurement of realease and assimilation of

in the Gunnera-Nostoc . New Phytol 132:617-625.

Thiet R, Doshas A, Smith SM. 2014. Effects of biocrusts and lichen-moss mats on plant

productivity in a us sand dune ecosystem. Plants and Soil 377(1-2). doi:

http://dx.doi.org.colorado.idm.oclc.org/10.1007/s11104-013-2002-8

Warren SD. 2001. Synopsis: Influence of Biological Soil Crusts on Arid Land Hydrology and

Soil Stability. In: Belnap J, Lange OL. Biological Soil Crusts: Structure, Function, and

Management. Ecological Studies (150). Berlin: Springer-Verlag Berlin Heidelberg. 349-

360.

Williams A, Beyene M, Buck B. 2012. Biological soil crusts in the Mojave desert, USA:

micromorphology and pedogenesis. Soil Sci Soci Am J. doi: 10.2136/sssaj2012.0021

Xiao B, Maik V. 2017. Moss-dominated biocrusts increase soil microbial abundance and

community diversity and improve soil fertility in semi-arid on the Loess Plateau

of China. Ap Soil Ecol 117-118 (165-177). doi:

https://doi.org/10.1016/j.apsoil.2017.05.005

Xu W, Cui K, Xu A, Nie L, Huang J et al. 2015. Drought stress condition increases root-to-shoot

ratio via alteration of carbohydrate partitioning and enzymatic activity in rice seedings.

Acta Physiol Plant 37:9. doi: 10.1007/s11738-014-1760-0

Zamfir M. 2000. Effects of and lichens on seedling emergence of alvar plants:

evidence from greenhouse experiments. Oikos 88:603-611. 34

Zhang Y, Aradottir AL, Serpe M, Boeken B. 2016. In: Weber B, Budel B, Belnap J. Biological

Soil Crusts: An Organizing Principle in Drylands. Interactions of Biological Soil Crusts

with Vascular Plants. Ecological Studies (226). Switzerland: Springer International

Publishing 385-406.