Content uploaded by Aaron Mills
Author content
All content in this area was uploaded by Aaron Mills on Aug 18, 2020
Content may be subject to copyright.
Content uploaded by Derek H Lynch
Author content
All content in this area was uploaded by Derek H Lynch on Jul 10, 2019
Content may be subject to copyright.
Contents lists available at ScienceDirect
Applied Soil Ecology
journal homepage: www.elsevier.com/locate/apsoil
Relationships between field management, soil health, and microbial
community composition
☆
Carolyn Mann
a,b
, Derek Lynch
b
, Sherry Fillmore
c
, Aaron Mills
a,⁎
a
Charlottetown Research and Development Centre, Agriculture and Agri-Food Canada, 440 University Ave., Prince Edward Island C1A 4N6, Canada
b
Department of Plant, Food and Environmental Sciences, Dalhousie University, 50 Pictou Rd., Bible Hill, Nova Scotia B2N5E3, Canada
c
Kentville Research and Development Centre, Agriculture and Agri-Food Canada, 12 Main St., Nova Scotia B4N1J5, Canada
ARTICLE INFO
Keywords:
Soil health
Cornell Soil Health Assessment
Phospholipid fatty acid analysis
Agricultural intensity
ABSTRACT
More meaningful and useful soil health tests are needed to enable better on-farm soil management. Our objective
was to assess the relationship between field management, soil health, and soil microbial abundance and com-
position (phospholipid fatty acid analysis (PLFA)) in soil collected from two fields (farmer-designated ‘good’
versus ‘poor’) across 34 diverse (livestock, grain or vegetable cropping) farms in Maritime Canada. Soil health
was measured using soil texture, surface hardness, available water capacity, water stable aggregates, organic
matter, soil protein, soil respiration, active carbon, and standard nutrient analysis. All soils were medium to
coarse textured (< 8% clay). Mixed models analysis showed that both CSHA and PLFA were able to resolve
statistical differences between cropping systems, however conventional soil chemical analysis was the only
testing method to resolve statistical differences between farmer designated ‘good’and ‘poor’fields. Principle
component analyses determined management history (rotation over previous three years), but not ‘good’or
‘poor’field designation, to be an important determinant of soil health. Water-stable aggregates and soil re-
spiration were positively correlated with all PLFA microbial groups, and negatively correlated with sand, P, Cu
and Al. Lower-intensity management (perennial forage, mixed annual-perennial cropping), manure application
and low tillage were linked to higher soil respiration, water-stable aggregates, fungi, mycorrhizae, Gram ne-
gative bacteria, and lower soil available P. Correlations between CSHA and PLFA shows promise for integrating
these two tests for improved soil health assessment.
1. Introduction
Soil is a dynamic living system whose condition underpins agri-
cultural productivity and ecosystem function (Doran et al., 1996).
While healthy soils promote the provision of ecological services, soil
degradation can lead to environmental strain and loss of productivity
(Bennett et al., 2010;Pepper, 2013). Broadly, soil health is understood
to be the combination of physical, chemical and biological properties
that promote the ability of a soil to support human, plant and animal
needs while maintaining or enhancing environmental quality (Doran
et al., 1996;Moebius-Clune et al., 2016).
Assessing soil health requires a more comprehensive approach to
testing than conventional soil quality work that focuses on a few in-
dividual parameters; ideally, it must encompass chemical, biological
and physical indicators as well as trends and emergent properties
(Karlen et al., 1997). Methods for assessing soil health range from in-
field observational scorecards, such as the Wisconsin Soil Health
Scorecard (Romig et al., 1995) to comprehensive laboratory tests of a
minimum set of indicators, such as the Cornell Soil Health Assessment
(CSHA) (Moebius-Clune et al., 2016). Other soil health work has ex-
plored bio-indicator species, such as presence or abundance of nema-
todes, earthworms, collembola, or abundance and diversity of microbial
indicators (Pankhurst et al., 1995;van Bruggen and Semenov, 2000;
Griffiths et al., 2016). Phospholipid fatty acid (PLFA) profiling of soil
https://doi.org/10.1016/j.apsoil.2019.06.012
Received 20 December 2018; Received in revised form 9 June 2019; Accepted 19 June 2019
Abbreviations: G +ve, Gram positive; G −ve, Gram negative; ACE, Autoclaved-Citrate Extractable; AMF, arbuscular mycorrhizal fungi; AWC, available water
capacity; BSA, bovine serum albumin; CEC, cation exchange capacity; CSHA, Cornell Soil Health Assessment; ENVT, environmental and management data; F:B,
fungi:bacteria; FID, flame ionization detector; g, farmer-identified good field; KOH, potassium hydroxide; OSHA, Ontario Soil Health Assessment; p, farmer-identified
poor field; PCA, principal component analysis; PLFA, Phospholipid fatty acid analysis; REML, restricted maximum likelihood; SOC, soil organic carbon; SOM/OM,
soil organic matter; WSA, wet stable aggregates
☆
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
⁎
Corresponding author.
E-mail address: aaron.mills@canada.ca (A. Mills).
Applied Soil Ecology 144 (2019) 12–21
Available online 09 July 2019
0929-1393/ Crown Copyright © 2019 Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/BY/4.0/).
T
microbial groups is another possibility for soil health testing (Bossio
et al., 1998;Shestak and Busse, 2005;Zhang et al., 2014). Recent work
suggests that many biological indicators differentiate between site-level
differences, though most are not sensitive enough to management
practices on their own and should be integrated with a suite of other
indicators (Griffiths et al., 2016).
The CSHA was made available in 2006 as a cost-effective protocol
for assessing soil health in the New York region of the United States. It
incorporates key physical, chemical and biological indicators chosen for
their relevance, sensitivity, consistency and cost, and has been eval-
uated in multiple trials (Schindelbeck et al., 2008;Idowu et al., 2008;
Moebius-Clune et al., 2016). As part of the analysis, between 9 and 16
indicators are assessed. The standard suite includes soil texture, avail-
able water capacity (AWC), surface and subsurface hardness, wet ag-
gregate stability (WSA), organic matter (OM), Autoclaved-Citrate Ex-
tractable (ACE) soil protein, soil respiration, active C, and standard
nutrient analysis. Indicators are individually scored out of 100 and
mean values are used to generate an overall soil health score. These
scores are calculated within soil textural classes based on Cornell's soil
database, and thus are a comparison to a regional range of parameters.
This is a challenge for using the test outside of the New York area, as the
scoring is not calibrated to yield data, nor does it reflect real biological
or physical thresholds.
Recent research has worked to validate the CSHA for use in Canada.
In southwestern Ontario, the CSHA was used to assess soil health dif-
ferences between tillage management regimes in long-term (20 year)
field trials; the CSHA's assessment of soil health matched the assess-
ments of soil organic carbon (SOC) and total N (Van Eerd et al., 2014).
Congreves et al. (2015) compared soils from field trials of no-till, di-
verse rotational systems to conventionally tilled monocultures in the
same region. The authors compared the CSHA to an Ontario Soil Health
Assessment (OSHA), which assessed the same indicators as the CSHA
but used weighted averages based on principal component analysis
(PCA) (Congreves et al., 2015). PCA showed that root health, sand
content, Mn, and pH were less valuable as soil health indicators in
Ontario, and using weighted averages made the OSHA more sensitive to
management effects (Congreves et al., 2015).
Phospholipid fatty acids are present in all living cells and are found
in the cell membranes of microorganisms (Hill et al., 2000). Unique
fatty acids, or groups of fatty acids, have been linked to specific func-
tional groups of microorganisms and can be differentiated based on
chain length, branching and saturation (Willers et al., 2015). Biomarker
patterns or individual biomarkers can be used to identify microbial
groups, such as arbuscular mycorrhizal fungi (AMF) or Gram-negative
bacteria (Willers et al., 2015). The PLFA profile can be used to char-
acterise community composition, determine microbial biomass, and
provide an indication of the metabolic or functional state of the com-
munity (Frostegård et al., 2010).
PLFA profiles are detailed enough to demonstrate differences in the
microbial community affected by management practices and soil fac-
tors. Bossio et al. (1998) found significant differences between PLFA
profiles in organic and conventional field plots, and Bardgett et al.
(1997) found a clear shift in microbial community as grasslands shifted
between grazed and ungrazed management. The effect of pH on PLFA
profiles is also marked: PLFA profiles are sensitive to pH both in agri-
cultural soils (Rousk et al., 2010) and in forest soils which have re-
ceived lime (Frostegård et al., 1993). Compaction has a smaller effect
on PLFA profiles (Shestak and Busse, 2005). PLFA is useful for mea-
suring fungal:bacterial biomass ratios (Frostegård and Bååth, 1996;
Bardgett et al., 1996;Baath and Anderson, 2003), where a higher
fungal:bacterial biomass ratio is linked to increased C storage potential
in soils (Malik et al., 2016). Given PLFA's strengths, it would be useful
to explore its relationship to more comprehensive soil health assess-
ments.
Linking the CSHA to PLFA would provide valuable insights into the
relationship between a comprehensive soil health assessment and the
widely used soil microbial profiling tool. Evaluating and comparing
these two tests on a wide range of soil types and under diverse man-
agement regimes will contribute to improved understanding of soil
management effects on soil health and biological characteristics and
contribute valuable insights into the CSHA applicability in more diverse
regions. The costs and accessibility of these technologies are key factors
for the uptake of these methods with farmers. PLFA analysis is a highly
technical and relatively cost prohibitive approach to soil analysis that
requires a deeper understanding of the soil functioning and knowledge
of the existing literature for contextual interpretation. The CSHA, al-
though significantly more practical than the PLFA approach, is time
consuming and is cost prohibitive for farmers interested in character-
izing several fields. Therefore, it is important to evaluate the efficacy of
these emerging approaches relative to conventional soil analysis which
are low cost and accessible to all farmers. Based on these needs, the
objectives of this study were to evaluate the use of the CSHA in
Canadian Maritime agricultural fields by 1) exploring the relationship
between CSHA and PLFA profiles; and 2) relating changes in these soil
factors to field management history, conventional soil chemical ana-
lysis, and inherent soil characteristics.
2. Materials and methods
2.1. Site selection and soil sampling
Thirty-four farms in the Maritime region of Eastern Canada were
sampled between August 11 –September 16, 2016, with 12 farms in
Nova Scotia (NS), nine in New Brunswick (NB) and 13 in Prince Edward
Island (PEI) (Fig. 1). Farmer participation was initiated through an
online survey circulated through Maritime agricultural organisations.
An initial pool of 59 farmers provided contact and basic demographic
information to participate in the project, and responded to the question
“How do you know what is a healthy soil and what is a bad soil?”The
final 34 farm types varied widely: 14 farms were organic and 20 con-
ventional, including 17 vegetable, eight dairy, six field crops, and three
beef/sheep farms (Table 1).
Farmers were asked to select a ‘good’(g) and a ‘poor’(p) soil on their
farm based on their own perceptions and knowledge of the productivity
of each field. These gand pdesignations were used as subjective tags
only and were not used to drive analyses. In total, 68 fields were
sampled on 34 farms. Farmers provided basic background information
about the farm, including selected field history and management in-
tensity (current crop and 3-year crop history; intensity and frequency of
tillage; type, frequency and amount of inputs used, including manure,
compost, synthetic fertilizers and pesticides) (Table 2).
Soil samples were collected from both gand pfields based on the
methods described in the Comprehensive Assessment of Soil Health
Manual for the CSHA (Moebius-Clune et al., 2016). Samples were
randomly collected at 10–15 points per field, depending on field size, in
a“W”pattern. At each point, surface debris was removed, and a core
was collected 5 cm in diameter and 15 cm deep using a shovel. Penet-
rometer readings (Dickey-John Corporation, Auburn, IL) were taken
and maximum hardness recorded for the depths 0–15 cm and 15–46 cm
(surface and subsurface hardness, respectively). Samples were mixed
thoroughly in a bucket, and 2–3 kg of soil were bagged and placed in a
cooler.
Bulk soil samples were stored at 4 °C until further processing and
~50 g of soil was separated and stored at −20 °C for PLFA analysis. At
the end of the sampling period, approximately 1 kg of soil was sieved
(8 mm) and air dried to a constant weight. Approximately 250 g was
further subsampled for chemical analyses, and 100 g was bagged for
soil respiration analysis and Autoclave-Citrate Extractable (ACE) pro-
tein test (Cornell Soil Health Laboratory, 2016). The remainder of the
air-dried soil was further sieved (2 mm) and divided for the following
tests: active C and AWC (~200 g) and textural analysis and soil
moisture content (100–150 g). In addition, approximately 80 g of soil
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
13
macroaggregates (0.25 mm - 2 mm) were separated (W.S. Tyler Ro-Tap,
Mentor, OH) for WSA testing.
2.2. CSHA analyses
CSHA analyses (soil texture, surface and subsurface hardness, AWC,
WSA, OM, ACE soil protein, soil respiration, active C, and standard
nutrient analysis) were performed according to procedures outlined in
Cornell's Standard Operating Procedures (Cornell Soil Health
Laboratory, 2016). Nutrient analysis, pH and OM were conducted by
the Prince Edward Island Analytical Laboratory (Charlottetown, PE),
where OM was measured on a combustion analyzer (Hoogsteen et al.
2015) and nutrient analyses using Mehlich-3 extractions. Soil texture
was determined using the Buoycous Hydrometer method (Sheldrick and
Wang, 1993). Active C was measured by permanganate oxidation; WSA
(%) was measured using 0.25–2 mm aggregates placed under a rainfall
simulator. Soil respiration was measured via EC change to a KOH trap
caused by CO
2
respiration of re-wetted soil. ACE protein was measured
using sodium citrate as the extractant in conjunction with autoclaving,
and quantification was performed using bovine serum albumin (BSA)
standards on a microplate reader (Biotek™PowerWave XS2, Winooski,
VT). AWC was measured by calculating the difference in water content
between soils at field capacity and at permanent wilting point on a 5
Bar Pressure Plate Extractor and a 15 Bar Ceramic Plate Extractor.
2.3. Phospholipid fatty acid analysis
All samples to be used for PLFA analysis were bulked to form a
composite sample and stored at −20 °C until processing. Soil moisture
content was determined gravimetrically for each sample prior to the
extraction of fatty acids using a modified Bligh and Dyer technique
(Bardgett et al., 1996). Extracted fatty acids were methylated and
quantified using an HP 7890 gas chromatograph equipped with a flame
ionization detector (FID) (Agilent, Santa Clara, CA). Peaks were in-
tegrated and quantified using MIDI software (MIDI Corp, Newark, DE).
PLFA bioindicators included actinomycetes, Gram positive (G +ve) and
Fig. 1. Points indicate the 34 sampling sites in the Maritime provinces located in Eastern Canada.
Table 1
Participant farms by province and farm type.
New Brunswick Nova Scotia Prince Edward Island Total
Organic Conv. Organic Conv. Organic Conv.
Vegetable 2 5 3 4 3 17
Dairy 6 1 1 8
Field crops 1 2 3 6
Beef/sheep 1 2 3
Total 2 7 6 6 6 7 34
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
14
Gram negative (G −ve) bacteria, fungi, AMF, and eukaryotes. Ratios of
total fungal:bacteria (F:B) biomass, and G −ve stress indicator were
also computed automatically in the software. The G −ve stress in-
dicator is based on observations of increased G −ve PLFAs with stress
conditions (Frostegård et al., 1993;Willers et al., 2015). Soil quantifi-
cation of PLFAs are presented as nmol g
−1
of soil (Frostegård et al.,
1993;Frostegård and Bååth, 1996;Bardgett et al., 1999;Baath and
Anderson, 2003;Kelly et al., 2003;Kelly et al., 2007;Wardle and
Jonsson, 2014).
2.4. Statistical analysis
Field history data provided by the farmers were used to generate
correlations between soil factors and management practices.
Management factors were categorised according to three-year field
history based on the most common management practice over the past
three years, for example, a vegetable (veg) rotation was considered any
rotation that consisted of a majority of vegetable crops, whereas a field
that had a rotation of hay/hay/vegetable would be considered “mix”
(Table 2).
All categories of crop rotation (fallow, grain, grass, mix, veg) were
present in both good and poor fields, although only one good field was
in fallow (gfallow). Manure was applied more commonly on grass fields
under dairy production systems, compared to other field types. Organic
farms –which were mainly vegetable farms –tended to apply more
compost and other organic matter inputs including crab meal, bone
meal, diatomaceous earth, and Irish moss, compared to conventional
producers. The factors Tillage, Manure, Compost, Lime, and Synthetic
fertilizer were treated as categorical variables and were coded as 1–3
based on the frequency of their use in each field the previous three
years before the sampling period. The variable “Pesticides”was treated
as a categorical variable based on aggregated annual frequency of each
of herbicide, insecticide, and fungicide (0–9 with 0 being no pesticide
use and 9 being annual application of each of the three groups over
three years).
A residual maximum likelihood model (REML) was done on all
variates with a random effect for provinces and a fixed effect for rota-
tion and environment/management factors. The means from the REML
were used in all the principle component analysis (PCA) steps. PCA was
used for data reduction and data analysis: first, PCA was used as a data
reduction technique on the four main data sources: environmental and
management data (ENVT), soil chemical data from the Mehlich 3 ex-
tractible nutrients (“CHEM”), Cornell Soil Health Assessment data
(“CSHA”) and phospholipid fatty acid data (“PLFA”). The principal
components of these four PCAs were then used to further evaluate re-
lationships between factors explaining the greatest amount of varia-
bility in the dataset. A final PCA analysis, factors on score 1 from all of
the previous analyses which had the greatest relative importance (ei-
genvectors of ≥2.0) were then used in a subsequent PCA to explore
their relationship between each other under the context of the farmer
perceived “g”and “p”under each cropping system category.
Correlation matrices were generated between factor groups and visua-
lized using package “ggcorrplot”in R. All correlation plots had the
factors grouped based on the first principle component.
3. Results
Fields ranged in size from < 0.25 ha to > 40 ha. The soils ranged
widely and included Entisols, Inceptisols, Spodosols, Boralfs and Aqu-
suborders, but all fields were coarse- or medium-textured, with average
sand, silt and clay content at 53.6%, 39.1% and 7.3% respectively.
Samples were collected at the end of a summer in which most regions
had experienced significant drought, which negatively impacted the
reliability of penetrometer readings. For this reason, penetrometer
readings were not used in the final analyses.
3.1. Mixed models evaluations
Mixed models analyses (REML) showed significant differences be-
tween cropping practices (rotation) and environment/management
factors with regards to tillage, compost application and pesticides under
environmental and management factors. These analyses did not show a
significant difference between gvs. pfields, nor the interaction between
the two terms (Table 3). Similarly, mixed models analysis of the vari-
ables measured under the CSHA showed significant differences between
rotations for everything except subsurface hardness and available water
holding capacity. Of all of the CSH variates, only pH showed a sig-
nificant difference between farmer-designated gvs. pfields (6.29 vs.
5.91). Additionally, with the CSHA approach there were no significant
interactions between factors rotation and gvs. p. Mixed models analysis
performed on PLFA bioindicator data showed significant differences
between the rotations for all variables except for total fungi, and F:B
biomass (Table 3). No significant interactions were observed of PLFAs
between rotation and gvs. p, nor were there any significant interactions
between the two terms (Table 3). Mixed models analysis of soil chem-
istry showed the greatest resolution in detecting some statistical
Table 2
Description of factors and categories for field history data.
Factor Categories Meaning
g_p gFarmer-identified “good”field
pFarmer-identified “poor”field
Rotation type Grass Three years of perennial grass (hay, grass, native grass, triple mix, grass forage, pasture)
Grain Three years of majority annual grain (corn, wheat, oat, barley, soybean, canola, potato)
Veg Three years of majority vegetables
Mix Three years of a combination of annual (veg/grain) and perennial crops
Fallow Three years of fallow
Tillage intensity and frequency
a
1 No-till
2 Medium intensity/conservation till (conservation till, broadfork/hand till, s-tine, harrow, chisel plow, spading
machine)
3 High intensity (land-forming, sub-soiling, raised bed maker, conventional till, potato till, moldboard, 3-furrow, disc,
plow, till, rototill, etc.)
Manure Total amount applied Amount (Mg ha
−1
) of manure applied to each field over the duration of the previous three years.
Compost Total amount applied Amount (Mg ha
−1
) of organic inputs applied (including mussel shells, biowaste, Irish moss) over the previous three
years.
Lime Total amount applied Amount (Mg ha
−1
) of dolomitic or calcitic lime applied over the previous three years.
Synthetic fertilizer # of applications Total number of applications of synthetic fertilizers (N, P, K, S) was applied over the previous three years.
Pesticides # of products used Total number of pesticides applied over the previous three years including fungicides, insecticides and herbicides.
Sand, silt or clay % Percentage of each soil component
a
Tillage intensity is determined for each year then added together to give one measurement (i.e. low tillage for three years = 1 × 3 = 3).
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
15
differences between rotations, gvs. pfields, and the interaction of both
terms. Significant (p< 0.05) differences between rotations were ob-
served for P (83.7–205.3 ppm for grass and fallow), K
(71.79–151.3 ppm for grass and veg, Al (1215–1714 ppm for grass and
fallow), pH (5.92–6.41 for grass and veg), K (1.8–3.2% in grass and
veg), H+(5.7–17.2% in grain and fallow); highly significant
(p< 0.001) differences were observed in S (18.6–27.3 ppm in grass
and veg) and Ca (1075–1876 ppm in fallow and veg). Significant dif-
ferences between gvs. pfields were observed with Mg (172.1 vs.
138.2 ppm), B (0.73 vs. 0.56), Ca (1641 vs. 1201 ppm), pH (6.29 vs.
5.91), H+ (4.93 vs. 18.76%), and total base cations (89.2 vs. 77.1%).
There was a significant interaction between rotation and gvs. pfields
observed with Cu (ranging from 8.81 ppm in gfallow to 1.04 ppm in p
grass (Table 3).
Differences were observed between rotations and farmer designated
gand pfields regarding the various environmental and management
factors (Fig. 2). The use of tillage (frequency and intensity), compost,
fertilizer, and lime appeared to be higher in the gfields relative to the p
fields for most rotations. Manure application was higher in pfields for
all rotations except for grain and veg. The ‘Icides’category representing
pesticide application was higher in the fallow, grain and grass rotations,
but the opposite was observed for the mix and veg rotations (Fig. 2).
3.2. Multivariate analysis
Principal component analysis was used as a data reduction strategy
Table 3
Mixed models analysis output showing mean values and F-statistics for each factor tested.
Factor FRFGvs. PFR×G×P
Sand 0.71 54.48 0.21 56.09 1.35 53.27
Silt 0.70 38.35 0.38 36.69 1.56 39.51
Clay 0.38 7.23 0.12 7.28 0.38 7.29
Tillage 20.82 4.99
⁎⁎⁎
1.92 5.35 0.31 4.92
Manure
b
0.56 6.49 0 5.05 0.28 6.81
Compost
b
4.35 7.51
⁎⁎⁎
0.15 5.15 0.04 7.75
Lime
b
1.05 0.21 2.33 0.32 1.03 0.23
Synthetic fertilizer 0.58 1.11 1.34 0.22 0.25 1.07
Pesticides 5.48 0.91
⁎⁎⁎
0.51 1.16 1.91 0.81
Cornell Soil Health Assessment
Surface hardness 3.13 236.41
⁎
2.68 244.22 1.73 234.72
Subsurface hardness 0.93 299.82 2.01 299.82 1.07 299.84
AWC 1.32 0.21 0.16 0.23 1.08 0.22
WSA 3.29 59.49
⁎
0.17 55.72 1.45 60.59
Active carbon 2.73 553.94
⁎
0.07 526.31 1.07 559.44
Organic matter 3.39 4.04
⁎
0 3.84 0.95 4.12
ACE soil protein index 2.58 10.34
⁎
0.43 9.65 1.33 10.41
Soil respiration 4.72 0.89
⁎⁎
0.21 0.85 0.91 0.91
P 2.82 130.10
⁎
3.22 135.74 1.18 124.11
K 5.14 105.72
⁎⁎
0.67 102.21 0.65 106.01
pH 3.62 6.12
⁎
4.83 6.11
⁎
0.63 6.12
PLFA
AMF 5.92 0.49
⁎⁎⁎
0.07 0.45 0.17 0.49
G−ve bacteria 5.67 4.12
⁎⁎⁎
1.26 3.75 0.28 4.01
Eukaryotes 3.08 0.34
⁎
0 0.34 0.51 0.34
Non-mycorrhizal fungi 2.48 0.21 3.53 0.24 0.47 0.22
G +ve bacteria 5.33 2.32
⁎⁎
0.45 2.17 0.42 2.29
Anaerobe 3.91 0.11
⁎⁎
0.21 0.09 0.07 0.12
Actinomycetes 4.80 0.72
⁎⁎
2.76 0.66 2.23 0.73
F:B 2.25 0.01 1.04 0.12 1.98 0.01
G−ve stress 3.29 203.72
⁎
0.39 199.71 0.64 204.82
Soil chemistry analysis
a
P 3.19 130.12
⁎
2.77 135.73 1.12 124.12
K 5.23 105.72
⁎⁎
0.62 102.21 0.58 106.44
Mg 2.35 163.24 4.68 155.23
⁎
0.75 164.23
Fe 0.51 237.51 0.03 230.22 0.73 240.1
Mn 1.18 60.52 1.94 56.25 1.12 62.44
Zn 1.67 2.89 3.05 2.85 0.62 2.99
Na 0.94 24.88 0.38 24.29 1.24 24.93
B 1.95 0.72 4.05 0.65
⁎
1.31 0.74
Cu 0.81 2.66 1.95 3.02 2.85 2.41
⁎
S 5.68 21.66
⁎⁎⁎
0.02 21.08 1.06 21.66
Ca 5.13 1479.42
⁎⁎⁎
6.61 1421.12
⁎
0.35 1487
Al 2.65 1281.21
⁎
0.62 1342.09 0.55 1265
pH 3.62 6.12
⁎⁎
4.83 6.12
⁎
0.62 6.12
CEC 2.50 10.96 2.54 10.46 0.43 11.12
K (%) 2.85 2.63
⁎
0.52 2.67 0.52 2.59
Mg (%) 2.54 13.16 0.13 12.84 0.59 12.99
Ca (%) 1.34 67.33 3.85 67.65 0.64 66.49
H+ (%) 2.94 11.53
⁎
5.43 11.85
⁎
0.51 12.41
Na (%) 1.60 1.11 1.56 1.09 1.01 1.11
Total base cations (%) 1.64 83.14 4.36 83.17
⁎
0.9 82.09
⁎
p < 0.05.
⁎⁎
p < 0.01.
⁎⁎⁎
p < 0.001.
a
Unless otherwise stated, units for soil nutrient concentration is ppm.
b
Mg ha
−1
.
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
16
to infer correlations between the tested factors (Fig. 3a–d). For the
evaluation of environmental factors, 43% of variability in the data was
explained by loadings on the first score with tillage, synthetic fertilizer
and sand loading positively, and manure, silt and clay loading nega-
tively. Environmental factors on the second score loading positively
included lime and compost to a greater extent, and the loads on the
negative side were explained largely by pesticides. A PCA biplot of
environmental factors on rotation (Fig. 3a), shows associations between
ggrain rotations and increased use of synthetic fertilizer and tillage;
correlation between gveg rotations and higher levels of compost and
lime; both gfallow and g and pgrass rotations correlated with soil
textural components including sand, silt and clay. Pgrain rotations
correlated with increased pesticide use; pfallow was associated with
lower levels of pesticide and higher levels of manure application.
CSHA evaluation of soil health indicators through PCA explained
52% of the variation in the dataset under score one (positive loading:
AWC, WSA, active C, OM, ACE, soil respiration; negative loading: P).
Score 2 explained 26% of the variability in the data; it loaded positively
with K and pH, and negatively with surface and subsurface compaction
(Fig. 3b). Both gveg and ggrain were associated with increasing K and
pH values. Additionally pmix and fallow correlated with increasing
values of surface and sub-surface compaction. Both gand pgrass-based
rotations were associates with increasing soil respiration values. The g
mixed rotation correlated with increasing values of AWC and WSA.
Variation in the PLFA microbial community dataset (Fig. 3c) was
mainly explained in the first score (80%) which loaded positively with
all factors. Only 9% of variation was explained by the second score
which loaded positively with the eukaryote biomarker, and loaded
negatively with the fungi biomarker and F:B. Both gand pgrass rota-
tions as well as gand pmixed rotations correlated with increasing
concentrations of the G −ve bacteria biomarker. Finally, gfallow ro-
tation correlated with decreasing concentrations of actinomycete and
fungi biomarkers, as well as decreased G −ve stress indicator.
Principal components analysis of the soil chemistry parameters
Fig. 2. Mean values of gvs. pfarmer selected fields. Mean values for tillage, fertilizer and icides are based on categorical designations for frequency and intensity;
lime, manure and compost means are based on actual values of estimated amount applied in Mg ha
−1
; sand, silt and clay mean values are based on percentage of each
component in the soil sample.
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
17
(Fig. 3d) explained 37% of the variation on score 1 (loading positively
with most factors except OM, % K, % Mg and negatively with % H and
% Na); score 2 accounted for 30% of the variability in the dataset
(positive loading: OM, Na, Zn, Mg, Fe, Ca, Mn and CEC; negative
loading: P, Cu, and Al). PCA analysis of soil chemistry resulted in the
best resolution between gand pfields of any of the chosen analysis
methods. The majority of the protations were associated with in-
creasing values for % H and % Na as well as measured Fe and Na
concentrations in the soil gmix, gveg and pveg correlated positively
with increasing values of Mg, Zn, Ca, S, K, and B. Gfallow correlated
with increasing soil concentrations of Cu, P and Al.
As a visualization exercise, Pearson correlation coefficients for the
PLFA and CSHA were plotted and grouped based on the first principle
component (Fig. 4). These graphs show significant positive correlations
(p< 0.05) between most PLFA groups and soil respiration, water
stable aggregates, ACE, active C, percent organic matter and AWC.
There are also significant (p< 0.05) negative correlations between P
and G −ve bacteria, AMF, and actinomycetes biomarkers; pH is sig-
nificantly negatively correlated with WSA, AWC and surface compac-
tion. Management factors showing significant positive correlations be-
tween elements associated with intensive agriculture including
increased tillage, fertility, pesticides, lime application, P and K; sig-
nificant negative correlations between these elements and biologically
associated indicators including ACE, organic matter, WSA, and soil
respiration. Although compost addition showed significant negative
correlations with surface compaction, and positive correlations with
active C, ACE and percent organic matter, manure application did not
show a correlation with any of these factors (Fig. 4).
In order to interpret those factors which explained the majority of
the variation observed in the previous PCA analyses under the context
of both management factors and farmers perceived gvs. pfields, the
factors with eigenvalues on score 1 greater than or equal to 2.5 were
then used to evaluate all factor groups simultaneously (Fig. 5). There
was a clear separation between gand pfields and for the most part
cropping systems were grouped together on the biplot. For the first
score, positive loadings with longer eigenvectors (≥2.0) included soil
respiration, G −ve bacteria, fungi, WSA, AMF, AWC, silt, G +ve bac-
teria, anaerobic bacteria and G −ve stress ratio; negative loadings in-
cluded sand, synthetic fertilizer and tillage. Positive and negative
loadings combined accounted for 51% of the variability in the dataset.
On the second axis, positive loadings of similar eigenvector length
included Ca, pH, Mg, B, active C, % organic matter, % Ca (base cation),
total percent base cations; negative loadings included % H, % Na and %
clay. Positive and negative loadings combined accounted for 25% of the
variability (Fig. 5). The resulting biplot shows correlation between both
gand pgrass rotations and several PLFA biomarkers (G −ve bacteria,
fungi, G +ve bacteria, anaerobic bacteria) as well as increasing % silt
and soil respiration. The gmix rotation correlates with increasing va-
lues of organic matter, ACE, AMF biomarker and the G −ve stress in-
dicator, whereas both the pmix and pfallow rotations correlate with
increasing values for manure application, % silt as well as % Na and %
H. The gveg rotation correlates with increasing values for K, B, and pH;
pveg was in relatively close proximity with correlation with Mg. Fi-
nally, ggrain correlates with increased % sand, tillage and P whereas p
grain correlates with increasing values for synthetic fertilizer use. For
the fallow rotations, gfallow was closely associated with the negative
side of the first axis, and pfallow was closely associated with the ne-
gative side of the second axis (Fig. 5).
4. Discussion
Soil texture is a determinant for many other soil health indicators
such as OM, aggregation, nutrient available, water-holding capacity,
and compaction, which is why the CSHA accounts for texture in its
scoring functions. Clay content can play a large role in determining CEC
and the capacity of the soil to retain nutrients and OM, as well as ser-
ving as a component for the formation of aggregates, although the re-
latively low clay content overall for the soils in this study suggests
aggregate dynamics would be much more influenced by management.
Due to the inherently sandy nature of the soils in the region, cropping
systems tend to consist of a more intensive management approach –
undiversified vegetable and grains –with grass-based and mixed
cropping system categories tending to be found on soils slightly less
sandy in texture. Soil texture was an important factor in the CSHA and
PLFA analysis results: sand content had a high negative load on PCA
Score 1, and clay had a high negative load on PCA Score 2. Sand was
inversely related to soil respiration, AMF, G −ve, G −ve stress, non-
mycorrhizal fungi and WSA, but positively correlated with P, Cu and Al.
Along PCA Score 2, clay was inversely correlated with CEC, active C,
ACE protein and OM. However, because clay had a lower load on PCA
Score 1, it explained less variability through PCA. Congreves et al.
(2015) also found that sand was positively correlated with P and
Fig. 3. a–d: PCA biplot based on individual correlation matrices to evaluate a. management and environmental conditions; b. Cornell soil health assessments, c.
phospholipid fatty acid (PLFA) analysis; d. traditional soil chemistry analysis under different rotations in good vs. poor farmer selected fields.
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
18
inversely related to biological indicators like OM and Active C, al-
though they deemed sand less important for future OSHA work because
it occurred primarily on PC2.
Although all three analytical approaches were able to detect dif-
ferences in cropping systems and management, traditional soil chemical
analysis was the only approach which was able to resolve differences
between farmer perceived ‘good’and ‘poor’fields. Soils in the region
tend to be lighter, sandier soils with lower organic matter and pH
(Nyiraneza et al., 2017). There has been a trend, particularly in Prince
Edward Island, of increasing soil P and decreasing Mg levels through
Fig. 4. Visualization of a correlation matrix showing coefficients between CSHA, PLFA, and management and environmental factors. Circles indicate significant
(p< 0.05) correlations with positive relationships as oriented from the bottom left to the top right, and negative relationships oriented from the top left to the
bottom right. The degree of shading indicates the strength of the correlations. Factors are ordered according to the first principle component.
Fig. 5. Final principle component analysis biplot showing eigenvectors selected from previous analyses of a length > 2.5 on the first principle component score.
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
19
time (PEIDAF, 2012). In differentiating between good and poor fields,
conventional chemical analysis indicated significantly lower levels of
boron, calcium, magnesium and overall pH in the farmer-identified
poor fields. Boron is present in a pH-dependent equilibrium and is
highly mobile and easily leached from the soil under conditions of high
precipitation. It is essential for plant cell division and is a necessary
component of the cell (Gupta, 2016). Soil B levels are also quite low in
the region with typical concentrations below 2 ppm.
Both Ca and Mg are essential plant macronutrients. Ca is an es-
sential component of plant cell walls and is a key component in plant
cell signalling (White, 2015); Mg is a key nutrient for the control of
photosynthesis and nutrient partitioning among plant parts, and is es-
sential for N transport within the plant (Grzebisz, 2013).The range of
field types illustrated that differences in both selected CSHA parameters
and PLFA could be linked to management; most notably, perennial
grass fields had higher soil respiration, WSA, non-mycorrhizal fungi,
AMF, and G −ve bacteria, and lower P. Mixed perennial-annual
cropped fields were also higher in these measures than other field types,
though not as high as grass. Fields which were high in P were lower in
soil respiration, WSA, non-mycorrhizal fungi and AMF. These high P
fields generally reflected those that had been more intensively man-
aged, namely grain and veg fields. However, with the exception of pH,
there were no clear differences between farmer-identified good and
poor fields within any of the rotation types except for gfallow, which
contained only one field. Although pH is known to affect PLFA profiles
in agricultural soils (Rousk et al., 2010), PCA did not show clear shifts
in PLFA profiles between good and poor fields. The PLFA technique has
previously been shown to be a powerful tool to discern between dif-
ferent cropping systems (Duncan et al., 2016) when compared to mo-
lecular approaches.
Tillage and manure application were inversely related along PCA
Score 1, with tillage correlated with P, Cu, Al, sand and manure cor-
related with biological measures like soil respiration, AMF, G −ve
bacteria, G −ve stress indicator, non-mycorrhizal fungi and WSA. In
this study, manure was applied most often on dairy farm fields that
were in hay or pasture, which explains the inverse relationship between
tillage and manure application. Tillage is known to break down mac-
roaggregates and release SOC, thereby affecting biological soil health
indicators (Moebius et al., 2007;Acin-Carrera et al., 2013;Sağlam
et al., 2015). OM-inputs are known to improve a variety of soil physical
and biological indicators, and manure application in particular has
been shown to increase SOM, active C, AWC and WSA (Iqbal et al.,
2014). Gram negative stress indicator values are automatically included
as part of the Sherlock MIDI output and have been included in the
analyses. However, the interpretation of these values under the context
of environmental studies may not be relevant, as the initial work was
based on pure cultures in isolation (Willers et al., 2015).
The CSHA indicators WSA and soil respiration positively correlated
with all PLFA microbial groups, most notably AMF, G −ve bacteria, G
−ve stress indicator, and non-mycorrhizal fungi; these correlated ne-
gatively with P, Cu, and Al. It has been well established that AMF are
typically negatively correlated with soil-P (Hijri et al., 2006;Gosling
et al., 2013;Bainard et al., 2014;Bainard et al., 2015;Schneider et al.,
2015). AMF are also known to stabilize soil structure through physical
enmeshing of soil particles by mycorrhizal hyphae and the exudate
glomalin, by affecting plant root exudates, and by increasing overall
soil carbon (Rillig and Mummey, 2006;Daynes et al., 2013), thereby
confirming the positive relationship between WSA and AMF. Soil re-
spiration, an indicator of soil biological activity, was positively asso-
ciated with increases in all microbial groups. Heavy metals such as Cu
and Al have a negative effect on soil microbial biomass and activity, as
highlighted here by the inverse relationship between these metals and
all biological soil indicators from the CSHA and PLFA (Illmer et al.,
1995;Gillera et al., 1998;Vogeler et al., 2008).
Overall, the link between CSHA and PLFA indicators is a valuable
finding for the possibility of integrating these two measures. It may be
possible to use PLFA to explore more in-depth aspects of soil biology,
both structural and functional, in concert with the biological, physical
and chemical components measured in the CSHA. For example, PLFA is
efficient for rapid screening of the microbial community and is useful
for describing movement of substrates through the soil food web, for
determining bacterial:fungal ratios, for measuring microbial biomass
(both total, and by group) and for providing some indication of cell
activity and cell stress. Linking PLFA with the CSHA may develop a
deeper understanding of how microbial structure and function are af-
fected by a wider range of soil health factors.
5. Conclusions
PCA successfully related shifts in the PLFA profile to CSHA in-
dicators: in particular, WSA and soil respiration were positively corre-
lated with all PLFA microbial groups, and negatively correlated with P,
Cu, and Al. Management and environmental factors were linked to
PLFA and CSHA indicators. Notably, grass and mixed perennial-annual
cropped fields were higher in soil respiration, WSA, non-mycorrhizal
fungi, AMF and G −ve bacteria, and lower in soil available P. More
intensely managed fields like undiversified grain and vegetable rota-
tions showed the reverse trend, with higher P, sand content, and lower
levels of biological indicators (soil respiration, WSA, non-mycorrhizal
fungi, AMF and G −ve bacteria). Manure application was linked to
positive physical and biological indicators like respiration, AMF, non-
mycorrhizal fungi and WSA, and tillage and sand content correlated
with P, Cu and Al. Overall, the correlations between changes in CSHA
indicators and PLFA profiles shows promise for integrating these two
tests for stronger soil health assessment. With further research into the
possibilities for more in-depth integration with PLFA –for example, the
power to explore trophic cascades of C in conjunction with CSHA soil
health indicators –these two tests could provide powerful new insights
into soil health.
References
Acin-Carrera, M., Jose Marques, M., Carral, P., Alvarez, A.M., Lopez, C., Martin-Lopez, B.,
et al., 2013. Impacts of land-use intensity on soil organic carbon content, soil
structure and water-holding capacity. Soil Use Manag. 29 (4), 547–556. https://doi.
org/10.1111/sum.12064.
Baath, E., Anderson, T., 2003. Comparison of soil fungal/bacterial ratios in a pH gradient
using physiological and PLFA-based techniques. Soil Biol. Biochem. 35 (7), 955–963.
https://doi.org/10.1016/S0038-0717(03)00154-8.
Bainard, L.D., Bainard, J.D., Hamel, C., Gan, Y., 2014. Spatial and temporal structuring of
arbuscular mycorrhizal communities is differentially influenced by abiotic factors
and host crop in a semi-arid prairie agroecosystem. FEMS Microbiol. Ecol. 88 (2),
333–344. https://doi.org/10.1111/1574-6941.12300.
Bainard, L., Dai, M., Gomez, E., Torres-Arias, Y., Bainard, J., Sheng, M., et al., 2015.
Arbuscular mycorrhizal fungal communities are influenced by agricultural land use
and not soil type among the Chernozem great groups of the Canadian prairies. Plant
Soil 387 (1), 351–362. https://doi.org/10.1007/s11104-014-2288-1.
Bardgett, R.D., McAlister, E., 1999. The measurement of soil fungal:bacterial biomass
ratios as an indicator of ecosystem self-regulation in temperate meadow grasslands.
Biol. Fertil. Soils 29 (3), 282–290. https://doi.org/10.1007/s003740050554.
Bardgett, R.D., Hobbs, P.J., Frostegård, A., 1996. Changes in soil fungal:bacterial biomass
ratios following reductions in the intensity of management of an upland grassland.
Biol. Fertil. Soils 22, 261–264.
Bardgett, R.D., Leemans, D.K., Cook, R., Hobbs, P.J., 1997. Seasonality of the soil biota of
grazed and ungrazed hill grasslands. Soil Biol. Biochem. 29 (8), 1285–1294. https://
doi.org/10.1016/S0038-0717(97)00019-9.
Bennett, L.T., Mele, P.M., Annett, S., Kasel, S., 2010. Examining links between soil
management, soil health, and public benefits in agricultural landscapes: an Australian
perspective. Agric. Ecosyst. Environ. 139 (1–2), 1–12. https://doi.org/10.1016/j.
agee.2010.06.017.
Bossio, D.A., Scrow, K.M., Gunapala, N., Graham, K.J., 1998. Determinants of soil mi-
crobial communities: effects of agricultural management, season, and soil type on
phospholipid fatty acid profiles. Microb. Ecol. 36, 1–12.
Congreves, K.A., Hayes, A., Verhallen, E.A., Van Eerd, L.L., 2015. Long-term impact of
tillage and crop rotation on soil health at four temperate agroecosystems. Soil Tillage
Res. 152, 17–28. https://doi.org/10.1016/j.still.2015.03.012.
Cornell Soil Health Laboratory, 2016. Comprehensive Assessment of Soil Health Standard
Operating Procedures February 2016. Cornell University.
Daynes, C.N., Field, D.J., Saleeba, J.A., Cole, M.A., McGee, P.A., 2013. Development and
stabilisation of soil structure via interactions between organic matter, arbuscular
mycorrhizal fungi and plant roots. Soil Biol. Biochem. 57, 683–694.
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
20
Doran, J., Sarrantonio, M., Liebig, M., 1996. Soil health and sustainability. Adv. Agron.
56, 1–54. https://doi.org/10.1016/S0065-2113(08)60178-9.
Duncan, D.S., Jewell, K.A., Suen, G., Jackson, R.D., 2016. Detection of short-term crop-
ping system-induced changes to soil bacterial communities differs among four mo-
lecular characterization methods. Soil Biol. Biochem. 96, 160–168.
Frostegård, Å., Bååth, E., 1996. The use of phospholipid fatty acid analysis to estimate
bacterial and fungal biomass in soil. Biol. Fertil. Soils 22 (1–2), 59–65. https://doi.
org/10.1007/BF00384433.
Frostegård, Å., Bååth, E., Tunlid, A., 1993. Shifts in the structure of soil microbial com-
munities in limed forests as revealed by phospholipid fatty acid analysis. Soil Biol.
Biochem. 25 (6), 723–730. https://doi.org/10.1016/0038-0717(93)90113-P.
Frostegård, Å., Tunlid, A., Bååth, E., 2010. Use and misuse of PLFA measurements in soil.
Soil Biol. Biochem. 43 (8), 1621–1625.
Gillera, K.E., Witterb, E., McGrath, S.P., 1998. Toxicity of heavy metals to microorgan-
isms and microbial processes in soils: a review. Soil Biol. Biochem. 30 (10−11),
1389–1414.
Gosling, P., Mead, A., Proctor, M., Hammond, J.P., Bending, G.D., 2013. Contrasting
arbuscular mycorrhizal communities colonizing different host plants show a similar
response to a soil phosphorus concentration gradient. New Phytol. 198 (2), 546–556.
https://doi.org/10.1111/nph.12169.
Griffiths, B.S., Rombke, J., Schmelz, R.M., Scheffczyk, A., Faber, J.H., Bloem, J., Peres, G.,
et al., 2016. Selecting cost effective and policy-relevant biological indicators for
European monitoring of soil biodiversity and ecosystem function. Ecol. Indic. 69,
213–223.
Grzebisz, W., 2013. Crop response to magnesium fertilization as affected by nitrogen
supply. Plant Soil 368, 23–39. https://doi.org/10.1007/s11104-012-1574-z.
Gupta, U.C., 2016. Boron. In: Barker, A.V., Pilbeam, D.J. (Eds.), Handbook of Plant
Nutrition. CRC Press, Boca Raton, pp. 241–277.
Hijri, I., Sýkorová, Z., Oehl, F., Ineichen, K., Mäder, P., Wiemken, A., et al., 2006.
Communities of arbuscular mycorrhizal fungi in arable soils are not necessarily low
in diversity. Mol. Ecol. 15 (8), 2277–2289. https://doi.org/10.1111/j.1365-294X.
2006.02921.x.
Hill, G., Mitkowski, N., Aldrich-Wolfe, L., Emele, L., Jurkonie, D., Ficke, A., et al., 2000.
Methods for assessing the composition and diversity of soil microbial communities.
Appl. Soil Ecol. 15 (1), 25–36. https://doi.org/10.1016/S0929-1393(00)00069-X.
Hoogsteen, M.J.J., Lantinga, E.A., Bakker, E.J., Groot, J.C.J., Tittonell, P.A., 2015.
Estimating soil organic carbon through loss on ignition: effects of ignition conditions
and structural water loss. Eur. J. Soil Sci. 66, 320–328. https://doi.org/10.1111/ejss.
12224.
Idowu, O.J., van Es, H.M., Abawi, G.S., Wolfe, D.W., Ball, J.I., Gugino, B.K., et al., 2008.
Farmer-oriented assessment of soil quality using field, laboratory, and VNIR spec-
troscopy methods. Plant Soil 307 (1–2), 243–253. https://doi.org/10.1007/s11104-
007-9521-0.
Illmer, P., Marschall, K., Schinner, F., 1995. Influence of available aluminium on soil
micro-organisms. Lett. Appl. Microbiol. 21, 393–397.
Iqbal, M., van Es, H.M., Anwar-ul-Hassan, 2014. Soil health indicators as affected by long-
term application of farm manure and cropping patterns under semi-arid climates. Int.
J. Agric. Biol. 16 (2), 242–250.
Karlen, D.L., Mausbach, M.J., Doran, J.W., Cline, R.G., Harris, R.F., Schuman, G.E., 1997.
Soil quality: a concept, definition, and framework for evaluation. Soil Sci. Soc. Am. J.
61, 4–10.
Kelly, J.J., Häggblom, M.M., Tate, R.L., 2003. Effects of heavy metal contamination and
remediation on soil microbial communities in the vicinity of a zinc smelter as in-
dicated by analysis of microbial community phospholipid fatty acid profiles. Biol.
Fertil. Soils 38, 65–71. https://doi.org/10.1007/s00374-003-0642-1.
Kelly, J.J., Favila, E., Hundal, L.S., Marlin, J.C., 2007. Assessment of soil microbial
communities in surface applied mixtures of Illinois River sediments and biosolids.
Appl. Soil Ecol. 36, 176–183. https://doi.org/10.1016/j.apsoil.2007.01.006.
Malik, A.A., Chowdhury, S., Schlager, V., Oliver, A., Puissant, J., Vazquez, P.G.M., ...
Gleixner, G., 2016. Soil fungal:bacterial ratios are linked to altered carbon cycle.
Front. Microbiol. 1247, 1–11. https://doi.org/10.3389/fmicb.2016.01247. 7.
Moebius, B.N., Van Es, H.M., Schindelbeck, R.R., Idowu, O.J., Clune, D.J., Thies, J.E.,
2007. Evaluation of laboratory-measured soil properties as indicators of soil physical
quality. Soil Sci. 172 (11), 895–912.
Moebius-Clune, B.N., Moebius-Clune, D.J., Gugino, B.K., Idowu, O.J., Schindelbeck, R.R.,
Ristow, A.J., van Es, H.M., Thies, J.E., Shayler, H.A., McBride, M.B., Wolfe, D.W.,
Abawi, G.S., 2016. Comprehensive Assessment of Soil Health –The Cornell
Framework. Manual, Edition 3.0. Cornell University, Geneva, NY.
Nyiraneza, J., Thompson, B., Geng, X., He, J., Jiang, Y., Fillmore, S., Stiles, K., 2017.
Changes in soil organic matter over 18 yr in Prince Edward Island, Canada. Can. J.
Soil Sci. 97, 745–756.
Pankhurst, C.E., Hawke, B.G., McDonald, H.J., Kirkby, C.A., Buckerfield, J.C., Michelsen,
P., O'Brien, K.A., Gupta, V.V.S.R., Doube, B.M., 1995. Evaluation of soil biological
properties as potential bio-indicators of soil health. Aust. J. Exp. Agric. 35,
1015–1028.
Pepper, I.L., 2013. The soil health-human health nexus. Crit. Rev. Environ. Sci. Technol.
43 (24), 2617–2652. https://doi.org/10.1080/10643389.2012.694330.
Prince Edward Island Department of Agriculture and Forestry (PEIDAF), 2012. Prince
Edward Island Soil Quality Monitoring –2012. Soil Nutrient Status Over the Last 15
Years in Prince Edward Island.
Rillig, M.C., Mummey, D.L., 2006. Mycorrhizas and Soil Structure. Oxford, UK. https://
doi.org/10.1111/j.1469-8137.2006.01750.x.
Romig, D., Garlynd, M., Harris, R., Mcsweeney, K., 1995. How farmers assess soil health
and quality. J. Soil Water Conserv. 50 (3), 229–236.
Rousk, J., Brookes, P.C., Baath, E., 2010. The microbial PLFA composition as affected by
pH in an arable soil. Soil Biol. Biochem. 42 (3), 516–520. https://doi.org/10.1016/j.
soilbio.2009.11.026.
Sağlam, M., Selvi, K., Dengiz, O., Gürsoy, F., 2015. Effects of different tillage manage-
ments on soil physical quality in a clayey soil. Environ. Monit. Assess. 187 (1), 1–12.
https://doi.org/10.1007/s10661-014-4185-8.
Schindelbeck, R.R., van Es, H.M., Abawi, G.S., Wolfe, D.W., Whitlow, T.L., Gugino, B.K.,
et al., 2008. Comprehensive assessment of soil quality for landscape and urban
management. Landsc. Urban Plan. 88 (2–4), 73–80. https://doi.org/10.1016/j.
landurbplan.2008.08.006.
Schneider, K.D., Lynch, D.H., Dunfield, K., Khosla, K., Jansa, J., Voroney, R.P., 2015.
Farm system management affects community structure of arbuscular mycorrhizal
fungi. Appl. Soil Ecol. 96, 192–200.
Sheldrick, B.H., Wang, C., 1993. Particle size distribution. In: Carter, M.R. (Ed.), Soil
Sampling and Methods of Analysis. Canadian Society of Soil Science Lewis Publishers,
Ann Arbor, US, pp. 499–511.
Shestak, C., Busse, M., 2005. Compaction alters physical but not biological indices of soil
health. Soil Sci. Soc. Am. J. 69 (1), 236–246.
van Bruggen, A.H.C., Semenov, A.M., 2000. In search of biological indicators for soil
health and disease suppression. Appl. Soil Ecol. 15 (1), 13–24. https://doi.org/10.
1016/S0929-1393(00)00068-8.
Van Eerd, L.L., Congreves, K.A., Hayes, A., Verhallen, A., Hooker, D.C., 2014. Long-term
tillage and crop rotation effects on soil quality, organic carbon, and total nitrogen.
Can. J. Soil Sci. 94 (3), 303–315. https://doi.org/10.4141/CJSS2013-093.
Vogeler, I., Vachey, A., Deurer, M., Bolan, N., 2008. Impact of plants on the microbial
activity of soils with high and low levels of copper. Eur. J. Soil Biol. 44 (1), 92–100.
Wardle, D., Jonsson, M., 2014. Long-term resilience of above- and belowground eco-
system components among contrasting ecosystems. Ecol. https://doi.org/10.1890/
13-1666.1.
White, P.J., 2015. Calcium. In: Barker, A.V., Pilbeam, D.J. (Eds.), Handbook of Plant
Nutrition. CRC Press, Boca Raton, pp. 165–198.
Willers, C., van Rensburg, P.J.J., Claassens, S., 2015. Phospholipid fatty acid profiling of
microbial communities –a review of interpretations and recent applications. J. Appl.
Microbiol. 119, 1207–1218.
Zhang, W., Xu, J., Dong, F., Liu, X., Zhang, Y., Wu, X., et al., 2014. Effect of tetraconazole
application on the soil microbial community. Environ. Sci. Pollut. Res. 21 (13),
8323–8332. https://doi.org/10.1007/s11356-014-2844-5.
C. Mann, et al. Applied Soil Ecology 144 (2019) 12–21
21