patrickbdevaney/mia9 · Datasets at Hugging Face

text stringlengths 0 6.44k
Figure 1A) and with increasing understory plant community evenness (F1,68 = 8.83, p = 0.004; Figure 1B),
suggesting the maintenance of higher water levels and even plant understories on tree islands could
improve fungal diversity and therefore the variety of ecosystem services they provide (as supported
by the functional guild analysis below). Fungal community richness also increased with increasing
canopy openness (F1,65= 9.41, p = 0.003; Figure 1C) as well as with increasing understory evenness
(F1,65 = 7.66, p = 0.007; Figure 1D).
Diversity 2020, 12, x FOR PEER REVIEW 6 of 17
supported by the functional guild analysis below). Fungal community richness also increased with
increasing canopy openness (F1,65= 9.41, p = 0.003; Figure 1C) as well as with increasing understory
evenness (F1,65 = 7.66, p = 0.007; Figure 1D).
Figure 1. Factors explaining fungal community diversity (A,B) and richness (C,D). The results from
the model selection that indicated fungal Shannon diversity were explained by the relative water level
(A; F1,68 = 8.63, p = 0.004) and plant evenness (B; F1,68 =8.83, p = 0.004), and fungal richness was explained
by canopy openness (C; F1,65 = 9.41, p = 0.003) and understory plant evenness (D; F1,65 = 7.66, p = 0.007).
In all panels, each point represents a site in the experimental tree islands, and lines are based on a
linear model relationship between these variables with the shaded area indicating the 95% confidence
intervals around the line. In (A), note that relative water level values equal to zero represent sites
where the water level was on average at the soil surface. Values greater than zero represent the sites
that were on average inundated, and values less than zero represent sites that on average had water
below the soil surface.
3.3. Fungal Community Composition
In addition to its importance in fungal diversity, relative water level also explained a significant
amount of variation in fungal community composition (db-RDA, F1,68 = 2.95, p = 0.001; Figure 2). The
core type, one of the manipulative restoration treatments implemented during the construction of the
experimental tree islands, affected fungal community composition as well (F1,68 = 1.87, p = 0.002). We
found that limestone cores resulted in a significantly different community composition than peat
Figure 1. Factors explaining fungal community diversity (A,B) and richness (C,D). The results from
the model selection that indicated fungal Shannon diversity were explained by the relative water level
(A; F1,68 = 8.63, p = 0.004) and plant evenness (B; F1,68 = 8.83, p = 0.004), and fungal richness was
explained by canopy openness (C; F1,65 = 9.41, p = 0.003) and understory plant evenness (D; F1,65 = 7.66,
p = 0.007). In all panels, each point represents a site in the experimental tree islands, and lines are
based on a linear model relationship between these variables with the shaded area indicating the 95%
confidence intervals around the line. In (A), note that relative water level values equal to zero represent
sites where the water level was on average at the soil surface. Values greater than zero represent the
sites that were on average inundated, and values less than zero represent sites that on average had
water below the soil surface.
Diversity 2020, 12, 0324 7 of 17
3.3. Fungal Community Composition
In addition to its importance in fungal diversity, relative water level also explained a significant
amount of variation in fungal community composition (db-RDA, F1,68 = 2.95, p = 0.001; Figure 2).
The core type, one of the manipulative restoration treatments implemented during the construction of
the experimental tree islands, affected fungal community composition as well (F1,68 = 1.87, p = 0.002).
We found that limestone cores resulted in a significantly different community composition than
peat cores on both of the first two axes of fungal community composition (CAPs 1 and 2; Figure 2),
with a 218% higher value on CAP 1 (F1,76 = 3.83, p = 0.054) and a 185% smaller value on CAP2
(F1,76 = 61.55, p = 2.17 × 10−11) for limestone compared to peat core islands. To gain insight into the
difference between the communities in peat and limestone core islands, we performed an indicator
taxa analysis. After correcting for multiple comparisons, we determined that Thelephoraceae,
Eurotiomycetes, and Inocybe curvipes were indicative of peat communities, while Agaricomycetes and
Archaeorhizomyces were indicative of limestone communities (Table S1). We also found that there was a
significant relationship between the inter-site variations in the fungal and understory plant community
composition (Mantel test: R = 0.11, p = 0.003).
Diversity 2020, 12, x FOR PEER REVIEW 7 of 17
cores on both of the first two axes of fungal community composition (CAPs 1 and 2; Figure 2), with a
218% higher value on CAP 1 (F1,76 = 3.83, p = 0.054) and a 185% smaller value on CAP2 (F1,76 = 61.55, p
= 2.17 × 10−11) for limestone compared to peat core islands. To gain insight into the difference between
the communities in peat and limestone core islands, we performed an indicator taxa analysis. After
correcting for multiple comparisons, we determined that Thelephoraceae, Eurotiomycetes, and
Inocybe curvipes were indicative of peat communities, while Agaricomycetes and Archaeorhizomyces
were indicative of limestone communities (Table S1). We also found that there was a significant
relationship between the inter-site variations in the fungal and understory plant community
composition (Mantel test: R = 0.11, p = 0.003).
Figure 2. Ordination demonstrating the relationship between fungal community composition and
environmental variables. Each point represents the fungal community composition at a site on the
experimental tree islands. Points are colored by relative water level (in meters) from high water (light
colors) to low water (dark colors). The graphs on either side of the ordination indicate the mean and
standard error of tree density and island core type along the first two axes of community composition
(CAP1 and CAP2). Different lowercase letters denote significant differences. The table (in upper right
corner) details correlations between continuous environmental variables.
3.4. Distribution of Fungal Functional Guilds
In total, we found 15 fungal functional guilds in our dataset with each of the eight islands hosting
an average of approximately 5 guilds (±0.22) (Figure S3A). The most common functional guilds were
ectomycorrhizal fungi, dung-wood saprotroph, and plant pathogen-wood saprotroph, which
occurredin63%77%and67%ofthesites(respectively)andonallislands(FigureS3A)WhenFigure 2. Ordination demonstrating the relationship between fungal community composition and
environmental variables. Each point represents the fungal community composition at a site on the
experimental tree islands. Points are colored by relative water level (in meters) from high water
(light colors) to low water (dark colors). The graphs on either side of the ordination indicate the
mean and standard error of tree density and island core type along the first two axes of community
composition (CAP1 and CAP2). Different lowercase letters denote significant differences. The table
(in upper right corner) details correlations between continuous environmental variables.
Diversity 2020, 12, 0324 8 of 17
3.4. Distribution of Fungal Functional Guilds
In total, we found 15 fungal functional guilds in our dataset with each of the eight islands
hosting an average of approximately 5 guilds (±0.22) (Figure S3A). The most common functional
guilds were ectomycorrhizal fungi, dung-wood saprotroph, and plant pathogen-wood saprotroph,
which occurred in 63%, 77%, and 67% of the sites (respectively) and on all islands (Figure S3A).
When examining factors that influence guild richness, we found that the number of functional guilds
among fungi increased in sites that had a more even understory plant community (F1,63 = 10.12,
p = 0.0023; Figure 3A) with an approximately 70% increase in guilds in the most even compared to
the least even five sites. Core type and relative water level were also included in the best model for
the number of functional guilds (core type: F1,63 = 2.92, p = 0.1378; relative water level: F1,63 = 0.87,
p = 0.3521), with an 18% greater number of guilds in peat compared to limestone core islands and an
approximately 8% decrease in the driest compared to wettest sites. When we examined the factors
that influence abundances of the three most common guilds from our dataset, the abundance of only
one—plant pathogen-wood saprotroph—could be explained by any of our measured variables. As the
relative water level decreased, the abundance of the plant pathogen–wood saprotroph guild increased
(F1,68 = 5.47, p = 0.022; Figure 3B).
Diversity 2020, 12, x FOR PEER REVIEW 8 of 17
fungi increased in sites that had a more even understory plant community (F1,63 = 10.12, p = 0.0023;
Figure 3A) with an approximately 70% increase in guilds in the most even compared to the least even

Figure 1A) and with increasing understory plant community evenness (F1,68 = 8.83, p = 0.004; Figure 1B),

suggesting the maintenance of higher water levels and even plant understories on tree islands could

improve fungal diversity and therefore the variety of ecosystem services they provide (as supported

by the functional guild analysis below). Fungal community richness also increased with increasing

canopy openness (F1,65= 9.41, p = 0.003; Figure 1C) as well as with increasing understory evenness

(F1,65 = 7.66, p = 0.007; Figure 1D).

Diversity 2020, 12, x FOR PEER REVIEW 6 of 17

supported by the functional guild analysis below). Fungal community richness also increased with

increasing canopy openness (F1,65= 9.41, p = 0.003; Figure 1C) as well as with increasing understory

evenness (F1,65 = 7.66, p = 0.007; Figure 1D).

Figure 1. Factors explaining fungal community diversity (A,B) and richness (C,D). The results from

the model selection that indicated fungal Shannon diversity were explained by the relative water level

(A; F1,68 = 8.63, p = 0.004) and plant evenness (B; F1,68 =8.83, p = 0.004), and fungal richness was explained

by canopy openness (C; F1,65 = 9.41, p = 0.003) and understory plant evenness (D; F1,65 = 7.66, p = 0.007).

In all panels, each point represents a site in the experimental tree islands, and lines are based on a

linear model relationship between these variables with the shaded area indicating the 95% confidence

intervals around the line. In (A), note that relative water level values equal to zero represent sites

where the water level was on average at the soil surface. Values greater than zero represent the sites

that were on average inundated, and values less than zero represent sites that on average had water

below the soil surface.

3.3. Fungal Community Composition

In addition to its importance in fungal diversity, relative water level also explained a significant

amount of variation in fungal community composition (db-RDA, F1,68 = 2.95, p = 0.001; Figure 2). The

core type, one of the manipulative restoration treatments implemented during the construction of the

experimental tree islands, affected fungal community composition as well (F1,68 = 1.87, p = 0.002). We

found that limestone cores resulted in a significantly different community composition than peat

Figure 1. Factors explaining fungal community diversity (A,B) and richness (C,D). The results from

the model selection that indicated fungal Shannon diversity were explained by the relative water level

(A; F1,68 = 8.63, p = 0.004) and plant evenness (B; F1,68 = 8.83, p = 0.004), and fungal richness was

explained by canopy openness (C; F1,65 = 9.41, p = 0.003) and understory plant evenness (D; F1,65 = 7.66,

p = 0.007). In all panels, each point represents a site in the experimental tree islands, and lines are

based on a linear model relationship between these variables with the shaded area indicating the 95%

confidence intervals around the line. In (A), note that relative water level values equal to zero represent

sites where the water level was on average at the soil surface. Values greater than zero represent the

sites that were on average inundated, and values less than zero represent sites that on average had

water below the soil surface.

Diversity 2020, 12, 0324 7 of 17

3.3. Fungal Community Composition

In addition to its importance in fungal diversity, relative water level also explained a significant

amount of variation in fungal community composition (db-RDA, F1,68 = 2.95, p = 0.001; Figure 2).

The core type, one of the manipulative restoration treatments implemented during the construction of

the experimental tree islands, affected fungal community composition as well (F1,68 = 1.87, p = 0.002).

We found that limestone cores resulted in a significantly different community composition than

peat cores on both of the first two axes of fungal community composition (CAPs 1 and 2; Figure 2),

with a 218% higher value on CAP 1 (F1,76 = 3.83, p = 0.054) and a 185% smaller value on CAP2

(F1,76 = 61.55, p = 2.17 × 10−11) for limestone compared to peat core islands. To gain insight into the

difference between the communities in peat and limestone core islands, we performed an indicator

taxa analysis. After correcting for multiple comparisons, we determined that Thelephoraceae,

Eurotiomycetes, and Inocybe curvipes were indicative of peat communities, while Agaricomycetes and

Archaeorhizomyces were indicative of limestone communities (Table S1). We also found that there was a

significant relationship between the inter-site variations in the fungal and understory plant community

composition (Mantel test: R = 0.11, p = 0.003).

Diversity 2020, 12, x FOR PEER REVIEW 7 of 17

cores on both of the first two axes of fungal community composition (CAPs 1 and 2; Figure 2), with a

218% higher value on CAP 1 (F1,76 = 3.83, p = 0.054) and a 185% smaller value on CAP2 (F1,76 = 61.55, p

= 2.17 × 10−11) for limestone compared to peat core islands. To gain insight into the difference between

the communities in peat and limestone core islands, we performed an indicator taxa analysis. After

correcting for multiple comparisons, we determined that Thelephoraceae, Eurotiomycetes, and

Inocybe curvipes were indicative of peat communities, while Agaricomycetes and Archaeorhizomyces

were indicative of limestone communities (Table S1). We also found that there was a significant

relationship between the inter-site variations in the fungal and understory plant community

composition (Mantel test: R = 0.11, p = 0.003).

Figure 2. Ordination demonstrating the relationship between fungal community composition and

environmental variables. Each point represents the fungal community composition at a site on the

experimental tree islands. Points are colored by relative water level (in meters) from high water (light

colors) to low water (dark colors). The graphs on either side of the ordination indicate the mean and

standard error of tree density and island core type along the first two axes of community composition

(CAP1 and CAP2). Different lowercase letters denote significant differences. The table (in upper right

corner) details correlations between continuous environmental variables.

3.4. Distribution of Fungal Functional Guilds

In total, we found 15 fungal functional guilds in our dataset with each of the eight islands hosting

an average of approximately 5 guilds (±0.22) (Figure S3A). The most common functional guilds were

ectomycorrhizal fungi, dung-wood saprotroph, and plant pathogen-wood saprotroph, which

occurredin63%77%and67%ofthesites(respectively)andonallislands(FigureS3A)WhenFigure 2. Ordination demonstrating the relationship between fungal community composition and

environmental variables. Each point represents the fungal community composition at a site on the

experimental tree islands. Points are colored by relative water level (in meters) from high water

(light colors) to low water (dark colors). The graphs on either side of the ordination indicate the

mean and standard error of tree density and island core type along the first two axes of community

composition (CAP1 and CAP2). Different lowercase letters denote significant differences. The table

(in upper right corner) details correlations between continuous environmental variables.

Diversity 2020, 12, 0324 8 of 17

3.4. Distribution of Fungal Functional Guilds

In total, we found 15 fungal functional guilds in our dataset with each of the eight islands

hosting an average of approximately 5 guilds (±0.22) (Figure S3A). The most common functional

guilds were ectomycorrhizal fungi, dung-wood saprotroph, and plant pathogen-wood saprotroph,

which occurred in 63%, 77%, and 67% of the sites (respectively) and on all islands (Figure S3A).

When examining factors that influence guild richness, we found that the number of functional guilds

among fungi increased in sites that had a more even understory plant community (F1,63 = 10.12,

p = 0.0023; Figure 3A) with an approximately 70% increase in guilds in the most even compared to

the least even five sites. Core type and relative water level were also included in the best model for

the number of functional guilds (core type: F1,63 = 2.92, p = 0.1378; relative water level: F1,63 = 0.87,

p = 0.3521), with an 18% greater number of guilds in peat compared to limestone core islands and an

approximately 8% decrease in the driest compared to wettest sites. When we examined the factors

that influence abundances of the three most common guilds from our dataset, the abundance of only

one—plant pathogen-wood saprotroph—could be explained by any of our measured variables. As the

relative water level decreased, the abundance of the plant pathogen–wood saprotroph guild increased

(F1,68 = 5.47, p = 0.022; Figure 3B).

Diversity 2020, 12, x FOR PEER REVIEW 8 of 17

fungi increased in sites that had a more even understory plant community (F1,63 = 10.12, p = 0.0023;

Figure 3A) with an approximately 70% increase in guilds in the most even compared to the least even