PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KJProj/AccaPreleIRCIDateDue.indd

 

 

EXANIINATION OF THE TRANSPORT AND RETENTION AND
EXPLORATION OF THE SPATIAL DISTRIBUTION OF MICROBIAL
INDICATORS IN SOIL AGGREGATES

By

Mustafa A. Mazher

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the Degree of

MASTER OF SCIENCE

Fisheries and Wildlife

2010

E.‘

Run
sis?
[m

the

me

am

5X

L’)

(1.
,..,

at

 

 

 

 

ABSTRACT

EXAMINATION OF TRANSPORT, AND RETENTION AND EXPLORATION OF
THE SPATIAL DISTRIBUTION OF MICROBIAL INDICATORS IN SOIL
AGGREGATES

By

Mustafa A. Mazher

Runoff and infiltration of bacterial pathogens from agriculturally managed fields are of
significant public health concern. The objectives of this research were to I) examine the
transport and retention of the bacterial indicators E. coli and Ent. faecium and 2) explore
the spatial distribution of E. coli in soil aggregates. To conduct these experiments, a
bacterial extraction method was developed. The aggregates used came from fields with
three different soil treatments: conventionally tilled with chemical input (T1), non-tilled
with chemical input (T2) and native with no chemical input (T7). The bacterial extraction
method yielded 108% and 92% recoveries of E. coli in T1 and T7, respectively, and 97%
and 119% recoveries of Ent. faecium in T1 and T7, respectively. In transport
experiments, only T1 exhibited significantly less bacteria in the effluent from dry
aggregates compared to saturated aggregates, illustrating the importance of soil treatment
and moisture on bacteria transport. Similarly, soil treatment, and moisture had an effect
on E. coli spatial distribution. At air—dry conditions, three aggregate subsections
exhibited statistical differences across all soil treatments compared to one subsection at
30% moisture content. T 1 exhibited the highest variability as illustrated by statistical
differences in the bacterial concentration within three subsections. This study showed that

aggregates are useful models in understanding factors that influence bacterial transport.

I
h
C.

Cr.

TABLE OF CONTENTS

List of Tables ...................................................................................................................... v
List of Figures .................................................................................................................... vii

Chapter 1. Literature Review ............................................................................................... 1
1.1 Introduction ........................................................................................................ 2

1.2 Current knowledge on water quality indicators ................................................. 3

. 1.2.1 Escherichia coli as a fresh water quality indicator .......................... 5
1.2.2 Enterococcus spp. as marine water quality indicators ..................... 7

1.2.3 Use of fecal indicators in the agricultural environments ................. 8
, 1.3 Water quality monitoring deficiencies ............................................................. 10
T 1.4 Microbes in soil ............................................................................................... 12
} 1.4.1 Soil aggregates and microbial interactions ....................................... 13
‘ 1.4.2 Survival of bacterial pathogens in soil .............................................. 15
. 1.4.3 Factors of bacterial transport in soil .................................................. l6
" 1.5 Research Objectives ......................................................................................... 19
1.6 Reference list ................................................................................................... 20

Chapter 2. Material and Methods Development ................................................................ 33
2.1 Introduction ...................................................................................................... 34

2.2 Methods development ...................................................................................... 35

2.2.1 Soil aggregates ........................................................................................ 35

2.2.2 Bacterial extraction method .................................................................... 37
2.2.3 Direct E. coli and Ent. faecium spiking and immediate recovery ........... 39

2.2.4 Desiccation effect on E. coli recovery in whole and aggregate

subsections. 40

2.2.5 Soil aggregate rehydration and enrichment ............................................ 41
2.2.6 The effect of calcium chloride solution on E. coli recovery .................. 42

2.2.7 Examination of clumping in E. coli and Em. faecium cultures .............. 43

2.3 Flow chamber experiments .............................................................................. 45

2.4 Aggregate peeling for native bacterial enumeration ........................................ 47

2.5 Slicing and saturation experiments .................................................................. 49

2.6 Statistical analyses ........................................................................................... 51

' 2.7 Reference list ................................................................................................... 53
Chapter 3. Results and Discussion ..................................................................................... 55
3.1 Native bacterial concentrations ........................................................................ 56

3.1.1 Native bacterial concentrations in dry soil aggregates ........................... 56

3.1.2 Effect of settling on bacterial extraction ................................................. 57

3.2 Spiking and immediate recovery ..................................................................... 58
3.2.1 E. coli and Ent. faecium spiking and immediate recovery ...................... 58

; 3.2.2 Desiccation and E. coli recovery in whole aggregates ........................... 65

3 iii

1pm"!

 

3.3 Flow chamber experiments .............................................................................. 68

3.4 Bacterial spatial distribution ............................................................................ 75
3.4.1 Native bacterial concentrations in soil aggregate interior and exterior
layers ................................................................................................................ 75
3.4.2 Desiccation and E. coli recovery in aggregate subsections .................... 76
3.4.3 Slicing and saturation experiments ......................................................... 77

3.5 Discussion ........................................................................................................ 89
3.5.1 Bacterial extraction method from soil ..................................................... 89
3.5.2 Bacterial retention and transport in soil .................................................. 91
3.5.3 Spatial distribution of E. coli in soil ....................................................... 93

3.6 Conclusions ...................................................................................................... 94

3.7 Reference list ................................................................................................... 97

Appendix Raw Data for Analysis .................................................................................... 103

iv

LIST OF TABLES

Table 2.1. Average percentages of soil texture in aggregate treatments T1, T2 and T7...37

Table 2.2. C, N and bulk soil densities of soil aggregate treatments T1, T2 and T7 ......... 37

Table 3.1. Spiked CPU/aggregate counts and percentage of E. coli recovery in T1 soil
aggregates ............................................................................................................................ 60

Table 3.2. Spiked CFU/ aggregate counts percentage of E. coli recovery in T7 soil
aggregates .......................................................................................................................... 61

Table 3.3. Spiked CFU/ aggregate counts percentage of Ent. faecium recovery in T1 soil
aggregates .......................................................................................................................... 62

Table 3.4. Spiked CFU/ aggregate counts percentage of Ent. faecium recovery in T7
aggregates .......................................................................................................................... 63

Table 3.5. Comparison of the VMF method with the settling step and without the settling
step ..................................................................................................................................... 66

Table 3.6. Concentration of E. coli and Ent. faecium in the influent and T1, T2 and T7
aggregates and effluents ..................................................................................................... 72

Table 3.7. Concentration of E. coli and Ent. faecium in the influent and T1, T2 and T7
aggregates and effluents ..................................................................................................... 73

Table 3.8. The percent recovery of E. coli in effluents for T1, T2 and T7 aggregates
replicates at saturated and non-saturated conditions .......................................................... 73

Table 3.9. The percent recovery of Ent. faecium in effluents for T1, T2 and T7 aggregates
at saturated and non-saturated conditions .......................................................................... 73

Table 3.10. Averaged recovery of E. coli from sliced soil aggregate subsections ............ 77

The

1? .3 ‘
amt;
5% .

hher

lhkl
“1131.le

 

 

lhk}

labk.5

Table A.1. Data for heterotrophic plate counts for bacterial extraction from whole
aggregate experiments ..................................................................................................... 103

Table A2. Data for whole aggregate desiccation and E. coli recovery experiments ...... 104

Table A3. Data for heterotrophic plate counts for bacterial extraction from aggregate

interior and exterior layers ............................................................................................... 105
Table A4 Data for flow chamber experiments using saturated aggregates ................... 106
Table A.5.Data for flow chamber experiments using air-dry aggregates ........................ 107
Table A6. Data for slicing experiments using T1 aggregates ......................................... 108
Table A.7. Data for slicing experiments using T2 aggregates ......................................... 109
Table A8. Data for slicing experiments using T7 aggregates ......................................... 110

vi

figure

Figure

light

10 CF

httrs
limit}

 

Figure

Figurt

proc

LIST OF FIGURES

Figure 1.1. Soil pore types ................................................................................................. 14
Figure 2.1. KBS LTER soil sampling plots ....................................................................... 36
Figure 2.2. Differential interference contrast microscopy of E. coli cells (concentration of
10 CFU/ml) at 100X resolution ........................................................................................ 44

Figure 2.3. Differential interference contrast microscopy of Ent. faecium cells
(concentration of 107 CFU/ml) at 100X resolution ........................................................... 45

Figure 2.4 a). Glass beat matrix chamber design for investigation of bacterial transport.
b) glass bead matrix experimental setup ............................................................................ 46

Figure 2.5. SAE chamber and retainer base used for aggregate peeling and collection....49

Figure 2.6. Frontal view of the aggregate subsections ....................................................... 51

Figure 3.1. Native bacterial concentrations in aggregate treatments from T1, T2 and
T7 ....................................................................................................................................... 56

Figure 3.2. Comparison of aggregate weight from treatments T1, T2 and T7 to native
bacterial colony forming units per gram of soil with a trend line ...................................... 58

Figure 3.3. The averaged percent recoveries and standard deviations of E. coli in T1 and
T7 ....................................................................................................................................... 64

Figure 3.4. The averaged percent recoveries and standard deviation of But. faecium in T 1
and T7 soil treatment ......................................................................................................... 65

Figure 3.5. Recovery of spiked E. coli as a function soil moisture content when
processing the whole aggregate using the VMF method with settling step ....................... 66

vii

 

Figure
prom

 

Figure A

'1 .,
EHLULD

Ems
.; I
CUAJCC

Figure
tomb-.2;

‘Vnyj
ELM.
5

Uta-1136

 

Figure 3.6. Recovery of spiked E. coli as a function soil moisture content when
processing the whole aggregate using the VMF method without settling step .................. 67

Figure 3.7. Recovery of E. coli and Ent. faecium from T1, T2, T7 aggregates and
effluents at air-dry conditions ............................................................................................ 70

Figure 3.8. Recovery of E. coli and Ent. faecium from T1, T2, and T7 aggregates and
effluents at saturated conditions ......................................................................................... 71

Figure 3.9. Statistical analysis of the interaction between soil treatment and soil saturation
combining both bacterial spec1es74

Figure 3.10. Heterotrophic bacterial concentrations in interior and exterior layers of soil
treatments T1, T2 and T7 ................................................................................................... 76

Figure 3.11. E. coli concentrations in T1 aggregate slices at A) 0% moisture content, B)
15% moisture content and C) 30% moisture content ......................................................... 79

Figure 3.12. E. coli concentrations in T2 aggregate slices at A) 0% moisture content, B)
15% moisture content and C) 30% moisture content ......................................................... 80

Figure 3.13. E. coli concentrations in T7 aggregate slices at A) 0% moisture content, B)
15% moisture content and C) 30% moisture content ......................................................... 81

Figure 3.14. Statistical comparison of E. coli recoveries at 0, 15, and 30% moisture
content in T1 aggregate subsections .................................................................................. 84

Figure 3.15. Statistical comparison of E. coli recoveries at 0, 15, and 30% moisture
content in T2 aggregate subsections .................................................................................. 85

Figure 3.16. Statistical comparison of E. coli recoveries at 0, 15, and 30% moisture
content in T7 subsections ................................................................................................... 85

Figure 3.17. Comparison of E. coli recoveries in T1, T2, and T7 aggregates subsections at
0% moisture content .......................................................................................................... 86

viii

Figure 3.18. Comparison of E. coli recoveries in T1, T2, and T7 aggregates subsections at
15% moisture content ........................................................................................................ 86

Figure 3.19. Comparison of E. coli recoveries in T1, T2, and T7 aggregates subsections at
30% moisture content ........................................................................................................ 87

ix

Chapter 1.

Literature Review

 

 

 

5t".

20-
$10

‘
b

qua

imr

{Dill

 

 

9m

“H

assu
Drir.

Strin

1.1 Introduction

How do we keep our water safe? Although a seemingly simple and
unintimidating question, for the past few decades it has been one plaguing politicians,
scientists and global citizens alike. As it stands, the world population is approaching 7
billion people worldwide and within the span of 50 years, the population is predicted to
increase by 2 billion, mostly in developing countries (United States Census Bureau,
2009, United Nations, 2008). This rapid increase in population will no doubt place
significant strain on global resources, not the least of which is water quality and
quantity. It is already estimated that 1.1 billion people worldwide have no access to
improved sources of water with a further 2.6 billion having no access to improved
sanitation systems (WHO, 2000). Polluted water and lack of sanitation are the main
routes of human exposure to pathogenic microorganisms and has been reported to cause
9 million cases of gastrointestinal illnesses annually across the globe (Rose et al., 2001,
WHO, 2009)

However, we need not look beyond the borders of the United States for
assurance that this issue is worthy of attention. Although the passage of the Safe
Drinking Water Act (SDWA) and the Clean Water Act (CWA) have some provided
stringent guidelines, curbing point-source waterborne outbreaks in the past 30 years, the
threat of non-point source waterborne outbreaks has proven to be a more daunting task
(USEPA, 2004, Craun, 2006). Potable supplies and recreational waters such as rivers,
lakes and coastal waters are constantly receiving human and animal fecal discharges
through agricultural and storm water runoff, waste water utilities and septic tank systems
(USEPA, 2009). Pathogenic agents such as E. coli 0157:H7, Vibrio sp., Giardia,

C ryptosporidium, Hepatitis A and Noro viruses may be transmitted through contact with

2

 

 

(1's
C011

19'

Curr
path

pulu

 

 

strut?
SCIVE
(Hill;
trans
indit

Elli'ir

1.2 C

idemi

such untreated waters (Eaton, 2005, Crockett, 2007, Field, 2003, Gaffield et al., 2003,

Gerba and Smith 2005).

Agricultural non-point source pollution is the leading cause of microbial
impairment in rivers and lakes and a major contributor to groundwater contamination
(USEPA, 2004a). Much of the blame falls on agricultural management practices at
confined animal feeding and manure spreading operations (Jones, 1980, Thunegard,
1975, Goss and Richards, 2008, Thurston-Enriquez et al., 2005). Following intensive
irrigation, snowmelts or excessive rainfall, the manure-home pathogens from farm fields
can percolate through the soil and into the underlying water table or may runoff into
surface water (Goss and Richards, 2008, Gerba and Smith, 2005, Auld et al., 2004,
Curriero et al., 2001). Tillage practices also play a significant role in the degree of
pathogen percolation in the soil (Abu-Ashour et al., 1998). Aggressive grinding and
pulverizing by tillage machinery causes soil erosion and disrupts soil aggregate’s
structural integrity (Lal et al., 1997a). During these cultivation processes, pores, which
serve as microbial highways through the vadose zone are destroyed and/or restructured
(Hillel, 1998, Hattori and Hattori, 1976). Water quality pollution investigations of
transport mechanisms of well characterized model microorganisms (such as fecal
indicator bacteria) are needed to improve the understanding of the aforementioned

environmental and human factors in relation to soil-microbial interactions.
1.2 Current Knowledge on Water Quality Indicators

The use of microbial indicators may be traced back to 1880 following

identification Escherichia coli (initially termed Bacillus coli communis) and Klebsiella

 

 

 

3550‘

exam

SPetr
[he f0
Utility

indl'Ca

pneumonia by microbiologists Von Fritsch and Escherich respectively (Ashbolt et al.,
2001). Both microorganisms, characteristically found in human feces, were thought to
indicate the pollution of a water source and the possible presence of other fecal pathogens
(Buchanan and Gibbons, 1974). Because of the non-specificity of the detection culture
media at the time, the aforementioned bacterial genera as well as others (Enterobacrer
and Citrobacter) were assembled into the first indicator group, termed total coliforms

(Ashbolt et al., 2001).

Total coliforms were first adapted as indictors in 1914 by the United States Public
Health Service which was responsible for supervision of water safety until the formation
of the USEPA (Maier et al., 2000). The coliform group was defined as being composed
of Gram negative (i.e. have a thin, semi-permeable peptidoglycan layer), non-spore
forming, rod-shaped bacteria that are capable of growth in the presence of bile salts and
ferment lactose with the production of acid and gas at 3512°C within 24-48 hours
(Beveridge, 2000, Eaton, 2005). However, we now know that some of the bacteria
associated with the coliforms have no correlation with fecal pollution. Such is the
example with some Klebsiella which have been found in paper mill effluents, cotton mill
waste waters and textile plant effluents (Caplenas and Kanarek, 1984, Campbell et al.

1976, Dufour and Cabelli, 1976).

With the introduction of modern detection techniques and the need for more
specific indicators relating to fecal pollution, multiple indicators have been introduced in
the following decades with varying degrees of success. Classically, these indicators’
utility has been assessed based on the degree of their convergence with the “ideal

indicator” concept. From the perspective of water quality monitoring, the attributes of

4

 

 

[6

Cl.

3.1..

In I

beret

13.10

 

secii

retri
for i
d€lli
Hm

that

used

press

an ideal indicator should include: 1) presence in concurrence with pathogens 2) higher
resistance to disinfection than pathogens 3) higher in numbers as compared to pathogens
4) readily and easily detection in all types of waters 5) stable in the environment, i.e.
cannot multiply in the environment and 6) an indication of the degree of contamination
and hazard in relation to its density (Griffin et a1, 2001, Scott et al., 2002, Yates 2007).
In light of such rigorous criteria, it is quite obvious that an ideal indicator has yet to be
identified. In any case, the search for such a microorganism may be of secondary
importance. Instead, evaluation of indicators based on their intended usage has proven to
be more useful (Dufour, 1984, Ashbolt et al., 2001). E. coli and Enterococcus spp. have
been used as the primary bacteriological indicators for the past three decades and their
taxonomy and utility as water quality indicators are discussed in detail in the following

sections.

1.2.1 Escherichia coli as a Fresh Water Quality Indicator

With its abundance in animal and human feces as well as its relationship with
recreationally-acquired gastroenteritis, E. coli had been a good candidate to be an index
for fecal contamination. In 1984, Dufour showed a strong correlation between E. coli’s
density and incidence of swimming-associated gastroenteritis in fresh waters (1984). The
need for such a relationship was underscored by the findings that the previously reported
that fecal coliforms (a sub-group of the coliforms) lacked any significant association with
gastrointestinal illnesses (USEPA, 2004). Subsequently and ever since, E. coli has been
used as the primary indicator for fecal pollution in fresh waters by many states.

As with most indicator bacteria, detection of E. coli was based on a

presence/absence biochemical tests or agar based culture media that required incubation

 

 

 

she

(let

lfltl

 

cam
24 11
81165
Fur.
Pitt

{CSU

to re
COUt
Gab;
Ulilli

fesis

longer than 24 hours resulting in colony forming units (CFU). From a water quality
monitoring perspective, either choice was inadequate. While a presence/absence test may
indicate fecal pollution, information about the density or extent of pollution cannot be
ascertained without a most probable number format. And although the culture media
could resolve the quantification issue, it had two disadvantages: 1) the passage of 24
hours after sampling to search for evidence may not be useful since human contact with
impaired water could have already taken place and 2) selective agents in media were
shown to inhibit environmentally stressed organisms (Ashbolt, et al., 2001). The
development of the mTEC and then the modified mTEC agar are the most common
methods for enumerating E. coli. Utilizing the B-D-glucuronidase enzyme reaction to
catabolize glucuronic acid development, E. coli produces red or magenta colonies within
24 hours (USEPA, 2002). The method’s two hour resuscitation period induced growth of
stressed organisms while increasing the detection of E. coli within 90% accuracy.
Furthermore, the reliability, efficiency and technical ease at which an analyst may
perform the aforementioned method added another “ideal indicator” attribute to E. coli’s
resume.

Nevertheless, as with other indicators, E. coli has its limitations. It has been found
to replicate in tropical and subtropical soils (due to warm, humid climates that are
conducive to E. coli growth), thereby altering the true incidence of fecal pollution (Solo-
Gabrielle et al., 2000). Additionally, sensitivity to chlorination places doubts about its
utility in treatment plant facilities due to potential underestimation of more chlorine-

resistant pathogens (Miescier and Cabelli, 1982).

 

 

su
mi-

C06

Emu
relaiia
an eplt
Outbre:
fa? ( 1'10)
591(1)”, 5

(fawn.

 

1.2.2 Enterococcus spp. as Marine Water Quality Indicators

From the years of 1972-1979, Cabelli conducted a series of epidemiological
studies on fecal indicators in recreational beaches and lake water in the United States and
Egypt (1983). Regression analysis showed strong correlation between gastrointestinal
symptom of swimmers in these water bodies and E. coli and Enterococci over the other
microorganisms tested. However, Enterococcus spp. exhibited higher correlation
coefficients for their mean densities than E. coli in marine water. These results implicitly
illustrated the value of Entercocci as an indicator for pathogens that resist chlorination
and can survive in highly saline environments. Persistence of Enterococci under
conditions that are detrimental to microorganisms can be explained by Enterococci ’s cell
properties. The Enterococcus spp. are Gram positive (i.e. have a thick, relatively
impermeable peptidoglycan layer) cocci that optimally grow at 4410.50C, 6.5% Sodium
Chloride and an unusually high pH of 9 (Ashbolt et al., 2001, Eaton et al., 2005).

As a general rule, methods have not been developed to specifically identify
Enterococcus to the species level when assessing water quality. However, the
relationship of a species with specific source of fecal pollution is a great advantage from
an epidemiological standpoint, effectively narrowing down the list of suspects for
outbreak investigation. While there are no exclusive delineations, E. faecalis and Ent.
faecium have generally been associated with human fecal pollution, while E. bovis, E.
equinus and E. avium have been found to be indicative of presence of animal fecal matter
(Eaton, 2005).

Like most bacteria, Enterococci have been plagued with nomenclature

ambiguities and elusive detection techniques after they were first isolated in 1899.

 

 

h)d:

app;-

tec?
lan
The I

apprwl
by 3:;

(tsu-

 

appllt‘t
Sl'ller:

Smith

 

Typical qualitative detection techniques were biochemical examinations such as the
Lancefield classification scheme where the bacterium would react with a serological
antiserum (Murray, 1990). During the same time period that Cabelli was conducting the
study that would place Enterococci at the forefront of water quality indicators, Levin et
al. were devising a selective method for its enumeration using membrane filtration
(Levin, 1975). Since then, their method has been modified making it quicker and less
technically demanding. Method 1600, using mEI media for rapid enumeration through
membrane filtration is the most recently approved and recommended (USEPA, 2002a).
The mEI media contains indoxyl B-D-glucoside, which is a chromogenic agent
hydrolyzed by Enterococcus’s B-glucosidase enzyme resulting in a diffuse halo
appearance around the a positive colony (Messer and Dufour, 1998). Unlike previous
methods that required days, the detection for Enterococcus using mEI could be
enumerated within 24 hours. The mEI method has made Enterococci a readily utilizable

indicator for water quality assessment in marine and freshwater systems.

1.2.3 Use of fecal indicators in the agricultural environments

Findings in the National Water Quality report by the EPA showed that
approximately 18% and 14% of assessed river and lake miles respectively were impacted
by agricultural pollution, the leading contamination of which is pathogenic bacteria
(USEPA, 2004a). Agricultural pollution originates from runoff associated with manure
applied to fields, deposited by grazing livestock or leached from faulty septic tank
systems on agricultural fields (Simpson et al., 2002, Tyrrel and Quinton, 2003, Gerba and

Smith 2005). Pathogenic bacteria associated with agricultural fecal waste and manure

 

 

 

 

 

 

 

 

mt

55:.

res;

but

al.I

exu”

COHCJ

exar;
exam;
temp;
Obser
Funht
{Ohir
Smnph

Comp\i

 

include E. coli 0157:H7, Salmonella, Campylobacter jejuni, Listeria, Helicobacter and

Shigella (J amieson et al., 2002, Gerba and Smith, 2005).

While there are no standard guidelines for assessing fecal pollution on agricultural
fields, fecal coliforms (especially E. coli), Enterococci and Salmonella are frequently
used as indicators (Duffy, 2003, Benharn et al., 2006, Holley et al., 2008). When infected,
cattle may shed 102-105 and 102-107 CFU per g of feces of E. coli and Salmonella,
respectively (Himathongkhama, 1999). After applying the manure to the field, the
bacterial concentrations may decline 2-4 orders of magnitude within 9 weeks (Natvig et
al., 2002). However, indicator bacterial populations are known to survive in sediments for
extended periods of time and may persist for months or even years (Mallmann and
Litsky, 1951, Gerba and Smith, 2005, Anderson et al., 2005). After subsequent manure
application, indicator bacteria may regrow causing false positives and unwarranted

concern for fecal pollution (Natvig et al., 2002, Unc et al., 2006).

Best management practices (BMPs) have usually been based on research that
examined the mitigation of indicator survival in soil (USEPA, 2003, Benham, 2006). For
example, it is recommended to apply manure to agricultural fields during summer
temperatures because at cooler temperatures manure-home bacterial indicators have been
observed to survive longer (Himathongkham et al., 1999, Mannion et al., 2007).
Furthermore, storage practice and application of manure have shown to play an implicit
role in indicator viability (Unc and Goss 2003). Thunegard has shown that 35% of fecal
samples stored as liquid slurry contained Salmonella spp. as opposed to only 6% of

composted solid manure (1975). The process of composting, where the temperature of

 

Sit."

qu:

fill.

16C?

013

 

agen.
andl
“16:15
1991.
COHCE
Smite:
11110 p.
Dim-id

3010,]

OrCOnt

 

manure is raised 70°C prior to field application, is thought to be behind this pronounced

difference (Jones, 1980, Cools et al., 2001, Gerba and Smith, 2005).

Additional processes to mitigate pathogen load in agricultural soil include i) Lime
stabilization, where pH of manure is raised to 12 by adding hydrated lime (Ca(OH)2),
quicklime (CaO), or lime containing kiln dust or fly ash ii) anaerobic digestion, which
entails placing manure in an oxygen-free environment between 15-30 days at an elevated
temperature, iii) aerobic digestion, where the manure is frequently agitated with oxygen
or air for 20-60 days and iv) air-drying the manure for at least three months prior to land

application (USEPA, 2003, Gerba and Smith, 2005).
1.3 Water Quality Monitoring Deficiencies

The diffuse nature of agricultural runoff has been troublesome for regulatory
agencies to monitor (Wiebe, 2006). While regulations such as the Resource Conservation
and Recovery Act and SDWA have addressed point source pollution, regulatory
measures for non-point pollution due to agricultural runoff remains elusive (Nielsen,
1991, Wiebe, 2006). The passage of Section 319 of the CWA was meant to address this
concern by providing state and territories grant money to establish pollution control
strategies (Great Lakes Commission, 2004). Best management practices (BMPs) were put
into place by these entities to introduce suitable agricultural management practices and
provide barriers for water quality impairment (Michigan Department of Agriculture,

2010, Mackler and Merkle, 2000).

Although implementing BMPs was meant to curb microbial transport to drinking

or contact water, microbial outbreaks being traced back to livestock is still a public health

10

 

 

 

 

ii i.

sci.

seer.
beer.

alien

neC€\
SOll-rr
81 a]. ;

Predi-

 

Smud.
mal bt
[halite

problem (Cooley et al., 2007). A pertinent example of this is the E. coli 0157:H7 and
Campylobacterjejuni outbreak in Washington County Fair in New York. Following a
heavy rainfall event, runoff from a nearby farm leached into a well used by vendors to
supply attendees with drinking water and ice. Sixty five persons were hospitalized, two of

which died of hemolytic uremic syndrome (CDC, 2001).

Generally, there are no regulations concerning direct monitoring of agricultural
soils after manure application or if there are, the monitoring and surveillance strategies
are based on poor understanding of soil microbial transport (Cullen et al., 1995, Mackler
and Merkel, 2000, Gagliardi and Karns, 2000). Traditional surveillance has usually taken
place at sites that are suspected to be on the receiving end of the pathogenic load, i.e.
recreational or ground waters (Gagliardi and Karns, 2000). Although this approach is
seemingly logical because these bodies of waters are the points of human contact, it has
been criticized because evidence of contamination may come after the water source has

already been compromised (Cullent et al., 1995).

Even if surveillance is required, implementing guidelines for monitoring the soil
necessitates thorough understanding the region’s hydro- geological dynamics that effect
soil-microbial interactions (Steenhuis et al., 1995, Powelson and Gerba, 1995, J amieson
et a1, 2002). These dynamics can vary from region to region or even temporally, causing
predictions about microbial runoff or infiltration to be dubious (Unc and Goss, 2004,
Smucker et a1, 2007). For example, winter manure application in areas of extreme cold
may be a public health concern. The soil may become fractured due to freezing and
thawing cycles, creating pathways for bacterial movement to ground water (Jamieson et

a1, 2002, Rosa et al., 2009). Furthermore, the ability of the soil to filter pathogens under

11

 

 

 

 

——
>—
—

'—

l_i
‘—

1.4 .\

the r.
are ti.
and p}
fixatiu
LEG“;
I
(0110 ;

{Duff}.

 

 

Ofbafh

Chemit,‘

 

dry condition may be altered under extreme environmental changes such as excessive
rainfall leading to ground water contamination (Mahler et al., 2000, Curriero et al., 2001 ,

Stevik et al., 2004).

Aside from the regrettable cost of lives, there is also a hefty economic price tag
due to inadequate monitoring. The tragedy of Milwaukee’s C ryptosporidium outbreak,
which not only caused 403,000 cases of illness, cost the state of Wisconsin $96.2 million
in medical costs and productivity loss (Corso et al., 2003). Waterbome outbreak
prevention and curbing economic loss necessitates successful monitoring strategies
rooted in scientific understanding of the complex dynamics of microbial-soil interactions

(Nielsen, 1991).
1.4 Microbes in Soil

The soil environment houses approximately 109 bacteria per gram and is one of
the most microbially diverse terrestrial systems on earth (Torsvik et al., 1989). Bacteria
are the most numerous microorganisms in soil and play a significant role in soil processes
and plant physiologies such as carbon decomposition and mineralization, nitrogen
fixation, ammonification, nitrification and denitrification (Foster, 1988, Tan, 1994).
Likewise, soils in rural and agricultural areas may harbor zoonotic pathogens such as E.
coli 0157:H7, Camplyobaterjejuni, C ryptosporidium that pose notable human risks
(Duffy, 2003, Gerba and Smith, 2005). The specific mechanisms of survival and transport
of bacteria in such a diverse environment is influenced by the complex interplay of

chemical and physical properties of the soil as well as the biological processes (Hattori

12

 

 

al.

 

1.4

met

 

soil

“(11:

result
Smhi.
laIger
Stable

Cl alu‘

Chafac:

 

 

and Hattori, 1976, Tisdall and Oades, 1982, Ranjard and Richaume, 2001, Jamieson et

al., 2002).

1.4.1 Soil Aggregates and Microbial Interactions

Soil is defined as the heterogeneous outer layer of earth’s terrestrial surface that
influences the planet’s climatological and hydrological cycles and serves as a growth
medium for a community of living organisms (Hillel, 1998). Physically, the structure of
soil is a key factor in its ability to support plant and animal life and moderate nutrient and
water cycling (Lal et al., 1997a, Bronik and La], 2005). Soil’s structure can typically be
characterized based on the association of its structural units, i.e. its clay, silt and sand
content (Hillel, 1998). Soils with appreciable clay content tend to associate themselves
into composite structural sub-units called aggregates which may vary from several
millimeters to centimeters in size (Hillel, 1998). The stability of these associations is a
function of physical flocculation and biological cementing substances that holds these
clusters together (Bronik and La], 2005). The soil aggregate’s structural arrangement
results in the creation of niches or compartmentalized habitats for bacteria (Mummey and
Stahl, 2004, Mummey et al., 2006). Such an environment protects against intrusion of
larger predators, provides a buffer between competing microorganisms and allows for
stable moisture conditions (Vargas and Hattori, 1986, Ranjard and Richaume, 2001, Zhou

et al., 2002).

Reaching these microhabitats is facilitated by the soil aggregate’s pore structure
characteristics (Hattori, 1988, Foppen, 2005). Pores may be continuous allowing two way

movement or may have dead ends or become completely closed causing the bacterial

l3

 

in -'

Chi.

[:9
"Q
(11

 

attrib

b10103

maint
Charm
110m 5
batter
compc
Organic
mi‘ironl

Organic

 

 

, /’>72)>/77

Fig 1.1. Soil pore types: A) closed pore, B) dead-end pore and C) continuous, open-ended pore. Figure
from Hattori and Hattori, 1976.

entrapment (Fig 1.1). The pore size is equally as important in bacterial distribution in soil
aggregates (Stevik et al., 2004). It is usually within the smaller pores of approximately
2pm (but not under 0.8um) that bacteria may colonize (Ranjard and Richaume, 2001). In
larger pores, however, sorptive forces may not be enough to retain the bacteria following
a high hydraulic flow, subsequently flushing bacteria out of the aggregate (Hattori, 1988,

Stevik et al., 2004).

As previously mentioned, formation of the aggregate structure is usually
attributed to the spatial arrangements of the cemented subunits. Organic matter, the main
biological cementing substance, is thought to be the single most important factor in
maintaining aggregate structural stability which contributes to the soil’s pore
characteristics (Tisdall and Oades, 1982, La] et al., 1997b). Soil organic matter is formed
from sloughed plant cell components (soluble exudates, lysates, and decaying root hairs),
bacterial cell components (slime layer, capsule, and degraded metabolites) fungal
components (hyphae) or animal residue (Foster, 1988). Typically, the components of soil
organic matter can be classified into three categories based on their resistance to
environmental stress: A) transient binding agents which are rapidly decomposable

organic materials such as microbial, fungal or plant polysaccharides, B) temporary

14

 

CC

 

f.)

Slim

pore

of [h

(7,

nutri
Ritha
which
reachj
batter

lROlal

 

cementing agents such as plant roots, fungal hyphae and animal residues which a provide
a greater surface area for the binding of clay particles thereby affecting the stability of
larger groups of aggregates and C) persistent cements composed of complexes of clays,
polyvalent metals and organic matter-derived resistant fragments from plant roots, fungal
hyphae and degraded bacterial metabolites (Tisdall and Oades, 1982, Martens and

Frankenberger, 1992).

Physical disruption due to cultivation practices such as tillage compromise the
structural integrity of the aggregate by stimulating oxidation and loss of organic matter
(Tisdall and Oades, 1982). This causes redistribution of soil components and changes
pore size distribution and continuity (Hillel, 1998, Leij et al., 2002). However, studies
show conflicting results as to whether this variability results in a decrease or increase in
porosity (Lipiec et al., 2006). Nevertheless, this disturbance is thought to cause disruption

of the bacterial microhabitats and affect their distribution (Peixoto et al., 2006).
1.4.2 Survival of Bacterial Pathogens in soil

Aside from reinforcing the soil aggregate, the soil organic matter is the primary
nutrient cycling substrate for soil-colonizing bacteria (Foster, 1988, Ranjard and
Richaume, 2001). The availability of the organic matter is contingent upon the degree of
which it is decomposed and physical barriers in the soil which inhibit bacteria from
reaching it (Hattori and Hattori, 1976, Tan, 1994). Under these circumstances, soil
bacteria persist at a low metabolic rate to adapt to extreme nutrient limiting conditions

(Rozak and Colwell, 1987).

15

 

 

SUI

 

20f

 

Eiddi
than
prodi

Fires

passi
1991 ‘
‘Piele
Path“
micro

COPPO

Following manure application, however, soil is loaded with nutrients that enhance
survival of introduced zoonotic bacteria in their new environment (Himathongkhama,
1999, Natvig et al., 2002, Peacock 2004). The persistence of these pathogens in soil is
controlled by a myriad of factors including soil moisture content, texture, pH,
temperature, nutrient availability and predatory activities within soil and the capacity of
the microbe to avoid or resist these stresses (J amieson et al., 2002, Unc and Goss, 2003,
Unc and Goss, 2004, Goss and Richards, 2008). For example, viability of E. coli and
Enterococcus have shown marked difference when inoculated within sandy, loamy and
loamy sandy soil with 60, 80 and 100% moisture content at 5, 15 and 25°C (Cools et al.,
2001). While both microorganisms favored higher soil moisture and lower temperatures,
E. coli survived better in sandy soil and Enterococcus spp. preferred loamy soil.
Additionally, Enterococcus showed greater resistance to desiccation by surviving longer
than E. coli at lower moisture content. This may indicate Enterococci’s increased
production of exopolysaccharides in response to desiccation stress (Roberson and

Firestone, 1992, Hartke et al., 1998).
1.4.3 Factors of Bacterial Transport in Soil

Bacteria movement within soil can occur either actively through motility or
passively by being carried with water through soil pores (Lindqvist and Bengtsson,
1991). Passive transport includes an amalgam of physical processes that are termed
‘preferential flow’, signifying movement of bacteria as a particle across a defined
pathway (Coppola et al., 2009). Pathways that mediate preferential flow are created by
micro and macro-pores, fissures and cracks in the soil (Smucker and Hopmans, 2007,

Coppola et al., 2009). Active microbial motility on the other hand is a physiological

16

process that transports bacteria in response to an environmental stimulus (Hattori and
Hattori, 1976, Ford and Harvey, 2007). Bacteria may possess flagella that act as motors
to promote transport in liquid media (Manson, 1990, Eisenbach, 1990). The flagella drive
bacteria to preferentially swim towards chemical stimuli such as nutrients or oxygen
(Adler, 1966). Because soil alternates between unsaturated and saturated conditions, the
moisture content plays an integral role in bacterial motility in soil (Soby and Bergman,
1983). The ability of soil to retain moisture is usually a function of its pore size and
connectivity (Ranjard and Richaume, 2001). When these pore spaces are saturated, they
are thought to effectively act as microbial highways (Abu-Ashour et al., 1998, Zaval'skii

and Voloshin, 2003).

While passive and active transport govern microbial movement within soil, there
are two processes that retard their movement: the filtration capacity of soil and sorptive
interactions between the soil and the microorganisms (Lindqvist and Bengtsson, 1991,
Gannon et al., 1991, Unc and Goss, 2003). Filtration is a process by which particles are
physically trapped by colliding on the surface of a porous medium (Foppen et al., 2005).
This process is generally thought to be the main mechanism for mitigating microbial
percolation through the soil (J amieson et al., 2002). Microbes act as particles that collide
with the eluvial particles in soil by physical impediment (Foppen et al., 2005). Physical
straining or retention is influenced by the microorganisms’ size and morphology and
soil’s pore and grain sizes (Fontes et al., 1991, Huysman and Vestraete, 1993, Stevik et

al., 2004, Bolster et al., 2009).

Adsorption is an attachment process by which the chemical interactions are

dictated by varying properties of the soil’s substratum, the microorganisms’ cell wall

17

 

 

 

mar
limi
d6 \(

200

and

“it":

‘fir - 2': fl

characteristics, fluctuations in the soil environment and contact time between the
microorganism and the soil sediment (Powelson and Gerba, 1995, Hattori and Hattori,
1976, Foppen et al., 2005). Chemical forces (Van der Waals, electrostatic and ionic
bonding) are influenced by the soil particle size and texture, presence of cations (Ca2+,
Na+, Fe“), clay and soil organic matter cation exchange capacity (Unc and Goss, 2003,
Stevik et al., 2004). The variations in bacterial cell hydrophobicity, electrostatic charge
and extracellular polysaccharides also influence these chemical interactions (Stenstrom,
1989, Fontes et al., 1991, Bolster et al., 2006). After prolonged contact time, these factors
influence can change from reversible microbial adsorption to irreversible microbial
adhesion to soil (Powelson and Gerba, 1995). Ultimately, environmental factors such as
high hydraulic flow events (rainfall and snowmelts), pH change (lime stabilization,
manure loading and rainfall) and temperature fluctuations can reduce microbial contact
time with soil substratum and alter surface charges and metabolic physiology, causing
desorption (Powelson and Gerba, 1995, McEldowny and Fletcher, 1988, Guber et al.,

2005).

In reality the various mechanisms for transport and retardation are interdependent
and even may occur simultaneously or influence each other (Smucker and Hopsmans,
2007). For example, while motility itself facilitates microbial movement, it may also
increase adsorption due to the attachment of flagellar exopolysaccharides to the soil
substratum (McCaulou and Bales, 1995, Van Loosdrecht and Zehnder, 2005). Similarly,
the mechanism of adsorption can be a byproduct or influence filtration (Stevik, 2004).
After bacteria collide with soil particles, they may be able to form chemical bonds after

extended contact time (Huysman and Vestraete, 1993). Alternatively, microbes that are

18

 

 

pro

 

indi

ime

sorbed to fine clay particulates (>2um) may increase bacteria colloidal diameter thereby

increasing filtration (Mahler et al., 2000).

1.5 Research Objectives

Research has indicated that transport mechanisms have typically been characterized at
fields scale and laboratory bulk soil column models, which average localized
heterogeneities (Gagliardi and Karns, 2000, Unc and Goss, 2003, Bolster et al, 2006).
While only one study has been found to address the transport at the aggregate scale
(Guber, 2009), further research is needed to address relationships of bacterial transport
processes within aggregates. Examining transport at this scale using traditionally utilized
indicators may elucidate specific transport mechanisms that occur due to the microscale

interactions and the public health risk associated with these mechanisms.

The objectives of this research were to examine i) bacteria association with soil
aggregates from different agricultural management practices, and ii) transport, retention
and spatial distribution using well-characterized indicator bacteria as models. Achieving

these objectives entailed the following:

p—a

Development of methods for assessing bacterial extraction from aggregates.

2. Evaluation of bacterial retention and survival in aggregates.

3. Examination the retention and transport behavior of E. coli and Ent. faecium in
tilled and non-tilled soils at unsaturated and saturated conditions.

4. Exploration of spatial distribution of bacteria within inoculated soil aggregate

subsections.

19

1.6 Reference List

Abu-Ashour, J ., Joy, D.M., Lee, H., Whiteley, HR, and S. Zelin. 1998. Movement of
Bacteria in Unsaturated Soil Columns with Macropores. American Society for
Agricultural Engineers, Vol.41, No.4, pp.1043-1050.

Adler, J. 1966. Chemotaxis in Bacteria. Science, Vol.153, No.3737, pp. 708-716.

Anderson, K.L., Whitlock, IE, and V.J. Harwood. 2005. Persistence and Differential
Survival of Fecal Indicator Bacteria in Subtropical Waters and Sediments. Applied and
Environmental Microbiology, Vol.71, No.6, pp.3041—3048.

Ashbolt, N. J ., Grabow, W.O.K., and M. Snozzi.World Health Organization. 2001. Water
Quality: Guidelines, Standards and Health. Edited by Lorna Fewtrell and Jamie Bartram.
Published by IWA Publishing, London, UK.

Auld, H., Maclver, D., and J. Klaassen. 2004. Heavy Rainfall and Waterbome Disease
Outbreaks: The Walkerton Example. Journal of Toxicology and Environmental Health,
Part A Vol. 67, pp. 1897-1887.

Benham, B.L., Baffaut, C., Zeckoski, R.W., Mankin, K.R., Pachepsky, Y.A., Sadeghi,
A.M., Brannan, K.M., Soupir, M.L., and MJ. Habersack. 2006. Modeling Fate and
Transport in Watersheds to Support TMDLs. Transactions of ASABE, Vol.49, No.4,
pp.987-1002.

Beveridge, T.J. 2001. Use of the Gram Stain in Microbiology. Biotechnic and
Histochemistry, Vol.76, No.3, pp.111-118.

Bolster, C.H., Walker, S.L., and KL Cook. 2006. Comparison of Escherichia coli and
Camplyobacterjejuni Transport in Saturated Porous Media. Journal of Environmental
Quality, Vol.35, No.4, pp.1018-1025.

Bolster, C.H., Haznedaroglu, B.Z., and S.L. Walker. 2009. Diversity in Cell Properties
and Transport Behavior among 12 Different Environmental Escherichia coli Isolates.
Journal of Environmental Quality, Vol.38, No.2, pp.465-472.

Bronik, C.J., and R. Lal. 2005. Soil Structure and Management: A Review. Geoderrna,
Vol.124, No.1-2. PP.3-22.

20

Buchanan, R. E., and NE. Gibbons. 1974. Bergey’s Manual of Deterrninative
Bacteriology. 8th Edition. The Williams and Wilkins Company, Baltimore, MD.

Cabelli, V.J. 1983. Health Effects Criteria for Marine Recreational Waters. EPA-600/ 1-
80-031 US Environmental Protection Agency, Cincinnati, Ohio 45268.

Campbell, L.M., Michaels, G., Klein, RD, and LL. Roth. 1976. Isolation of Klebsiella
pneumoniae from Lake Water. Canadian Journal of Microbiology, Vol.22, No. 12
pp. 1762-1767.

Caplenas, N .R., and MS. Kanarek. 1984. Thermotolerant non-fecal source Klebsiella
pneumoniae: validity of the fecal coliform test in recreational waters. American Journal
of Public Health, Vol. 74, Issue 11, pp.1273-1275.

Center for Disease Control. 2001. Public Health Dispatch: Outbreak of Escherichia coli
0157:H7 and Campylobacter Among Attendees of the Washington County Fair —- New
York, 1999. Morbidity and Mortality Weekly Report Vol. 48, No. 36 pp.803.

Cooley, M. Carychao, D., Crawford-Miksza, L., Jay, M.T., Myers, C., Rose, C., Keys, C.,
Farrar, J ., and RE. Mandrell. 2007. Incidence and Tracking of Escherichia coli 0157:H7
in a Major Produce Production Region in California. PLoS ONE, Vol.2, No.11, c.1159.

Cools, D., Merkx, R., Vlassak, K., and J. Verhaegen. 2001. Survival of E. coli and
Enterococcus spp. Derived from Pig Slurry in Soils of Different Texture. Applied Soil
Ecology, Vol.17, No.1, pp.53-62.

Coppola, A., Kutflek, M., and ED. Frind. 2009. Transport in Preferential Flow Domains
of Soil Porous Systems: Measurement, Interpretation, Modeling, and Upscaling. Journal
of Contaminant Hydrology, Vol.104, No.1-4, pp.1-3.

Corso, P.S., Kramer, M.H., Blair, K.A, Addiss, D.G., Davis, J .P., and AC. Haddix. 2003.
Cost of Illness in 1993 Waterbome Cryptosporidium Outbreak, Milwaukee, Wisconsin.
Emerging and Infectious Disease Vol. 9, No. 4, pp. 426-431.

Craun, M.F., Craun, G.F., Calderon, R.L., and M.J. Beach. 2006. Waterbome Outbreaks
Reported in the United States. Journal of Water and Health, Vol. 4, Suppl.2, pp.19-30.

21

Crockett, C. S. 2007. The Role of Waste Water Treatment in Protecting Water Supplies
Against Emerging Pathogens. Water Environment Research Vol. 79, No. 3, pp. 221-232.

Cullen, S.J., Kramer, J H Everett, LG, and LA. Eccles. Is Our Groundwater
Monitoring Strategy Illogical? pp. 1-8. In Handbook of Vadose Zone Characterization and

Monitoring. CRC Press, Boca Raton, FL.

Curriero, F.C., Patz, J. A, Rose, J .B., and S. Lele. 2001. Association Between Extreme
Precipitation and Waterbome Disease Outbreaks in the United States, 1948-1994.
American Journal of Public Health Vol. 91, No. 8, pp. 1194-1199.

Duffy, G. 2003. Verocytoxigenic Escherichia coli in Animal Faeces, Manure and
Slurries. Journal of Applied Microbiology, Vol. 94, No. 1, pp. 94-103.

Dufour, AP. 1984. Health Effects Criteria for Fresh Recreational Water. EPA 600/1-84-
004, US Environmental Protection Agency, Cincinnati, Ohio 45268.

Dufour, AP, and V.J. Cabelli. 1976. Characteristics of Klebsiella from Textile Finishing
Plant Effluents. Journal of Water Pollution Control Federation, Vol.48, No. 5, pp.872-

879.

Eaton, A.D., Clesceri, L.S., Eugene, W. R., and AB. Greenberg. 2005. Membrane
Filtration Technique for Members of the Coliform Group. 9050C, 1a. In Standard
methods for the Examination of Water and Wastewater. 21st Ed. American Public

Health, Baltimore, MD.

Baton, A.D., Clesceri, L.S., Eugene, W. R., and A.E. Greenberg. 2005. Fecal
Streptococcus and Enterococcus Groups. 9230A. In Standard methods for the
Examination of Water and Wastewater. let Ed. American Public Health, Baltimore,

MD.

Eisenbach, E. 1990. Function of Flagellar Modes of Rotation in Bacterial Motility and
Chemotaxis. Molecular Microbiology, Vol. 4, No.2, pp.161—167.

Field, R. 1993. Storm and Combined Sewer Overflow: an Overview of EPA's Research
Program. In Integrated Storm Water Management. United State Environmental Protection

Agency Cincinnati, OH

22

 

 

 

 

Ger

Pp. i

Foppen, J .W.A., Mporokoso, A., and J .F. Schijven. 2005. Determining Straining of
Escherichia coli from Breakthrough Curves. Journal of Contaminant Hydrology, Vol.76,
No.3-4. pp.191-210.

Fontes, D.E., Mills, A.L., Homberger, G.M., and J.S. Herman. 1991. Physical and
Chemical Factors Influencing Transport of Microorganisms through Porous Media.
Applied and Environmental Microbiology, Vol.57, No.9, pp.2473-2481.

Ford, R.M., and R.W. Harvey. 2007. Role of Chemotaxis in the Transport of Bacteria
through Saturated Porous Media. Advances in Water Resources, Vol. 30, No.6-7,
pp.1608-1617.

Foster, RC. 1988. Microenvironments of Soil Microorganisms. Biology and Fertility of
Soils, Vol.6, No.3, pp.189-203.

Gaffield, S.J., Goo, R.L., Richards, LA, and R.J. Jackson. 2003. Public Health Effects of
Inadequately Managed Stormwater Runoff. American Journal of Public Health Vol. 93
No. 9 pp.1527-1533

Gagliardi, J .V., and J .S. Kams. 2000. Leaching of Escherichia coli 0157:H7 in Diverse
Soils under Various Agricultural Management Practices. Applied Environmental
Microbiology, Vol.66, No.3, pp.877-883.

Gannon, J .T., Manila], V.B., and M. Alexander. 1991. Relationship between Cell Surface
Properties and Transport of Bacteria through Soil. Applied and Environmental
Microbiology, Vol.57, No.1, pp.190-193.

Gerba, CR, and J. E. Smith, Jr. 2005. Sources of Pathogenic Microorganisms and Their
Fate during Land Application of Wastes. Journal of Environmental Quality Vol. 34 No. 1
pp. 42-48.

Goss, M., and C. Richards. 2008. Development of a risk-based index for Source Water
Protection Flaming, which Supports the Reduction of Pathogens from Agricultural
Activity Entering Water Resources. Journal of Environmental Management Vol. 87 No. 4
pp. 623-632

23

ll

 

 

 

(”5'23

Great Lakes Commission. 2004. Nonpoint Source Pollution in the Great Lakes Basin:
Important Federal Laws and Programs; Great Lakes Laws and Statutes; International
Agreements and Activities; and Key Studies. Accessed February 17, 2010 at URL:
http://www.glc.org/postpluarg/documents/nps_programs.pdf

Griffin, D.W, Lipp, E. K., McLaughlin, M. R., and J. B. Rose. 2001. Marine Recreation
and Public Health Microbiology: Quest for the Ideal Indicator. BioScience. Vol. 51,
No.10, pp.817-825.

Guber, A.K., Shelton, DR, and Ya. A.Pachepsky. 2005. Effect of Manure on
Escherichia coli Attachment to Soil. Journal of Environmental Quality, Vol. 34, No.6,
pp.2086-2090.

Guber, A.K., Pachepsky, Y.A., Shelton, DR, and 0. Yu. 2009. Association of Fecal
Coliforms with Soil Aggregates: Effect of Water Content and Bovine Manure
Application. Soil Science, Vol.174, No.10, pp.543-548.

Hardina C.M., and Fujioka. 1991. Soil: The environmental source of Escherichia coli and
Enterococci in Hawaii’s Streams. Environmental Toxicology and Water Quality: An
International Journal 6: 185-195.

Hartke, A., Giard, J-C., Laplace, J-M., and Y. Auffray. Survival of Enterococcusfaecalis
in an Oligotrophic Microcosm: Changes in Morphology, Development of General Stress

Resistance, and Analysis of Protein Synthesis. Applied and Environmental Microbiology,
Vol.64, No.11, pp.4238-4245.

Hattori, T., and R. Hattori. 1976. The Physical Environment in Soil Microbiology: An
Attempt to Extend Principles of Microbiology to Soil Microorganisms. Critical Reviews
in Microbiology, Vol.4, No.4, pp.423-461.

T. Hattori. 1988. Soil Aggregates as Microhabitats of Microorganisms. Reports of the
Institute for Agricultural Research, Tohoku University, Vol.37, pp.23-36.

Hillel, D. 1998. Humus: The Organic Constituent of Soil Colloids. pp.97-98. In
Environmental Soil Physics. Academic Press, San Diego, CA.

24

 

 

 

Hillel, D. 1998. Structure of Aggregate Soils. pp.106-110. In Environmental Soil Physics.
Academic Press, San Diego, CA.

Hillel, D. 1998. Classification of Soil Pores. pp. 1 18-120. In Environmental Soil Physics.
Academic Press, San Diego, CA.

Himathongkham, S., Bahari, S., Riemann, H., and D. Cliver. 1999. Survival of
Escherichia coli 0157:H7 and Salmonella Typhimurium in Cow Manure and Cow
Manure Slurry. FEMS Microbiology Letters, Vol.178, No.2, pp.251-257.

Holly, R., Walkty, J ., Blank, G., Tenuta, M., Ominski, K., and D. Kraus. 2008.
Examination of Salmonella and Escherichia coli Translocation from Hog Manure to

Forage, Soil, and Cattle Grazed on the Hog-Manure Treated Pasture. Journal of
Environmental Quality, Vol.36, No.6, pp.2083-2092.

Huysman, F., and W. Vestraete. 1993. Water-Facilitated Transport of Bacteria in
Unsaturared Soil Colums: Influence of Inoculation and Irrigation Methods. Soil Biology
and Biochemistry, Vol.25, No.1, pp.91-97.

J amieson, R.C., Gordon, R.J., Sharples, K.E., Stratton, G.W., and A. Madani. 2002.
Movement and persistence of fecal bacteria in agricultural soils and subsurface drainage
water: A review. Canadian Biosystems Engineering, Vol.44, pp.1.1-1.9.

Jones, P.W. 1980. Disease Hazards Associated with Slurry Disposal. British Veterinary
Journal, Vol. 136, No. 6, pp.529-542.

Lal, R., Kimble, J .M., Follett, R.L., and BA. Stewart. 1997a. Soil Processes and the
Carbon Cycle. In Soil Structure and Organic Carbon: A Review. CRC Press, Boca Raton,
FL.

Lal, R., Kimble, J.M., Follett, R.L., and BA. Stewart. 1997b. Soil Processes and the
Carbon Cycle. In Relationships between Soil Organic Carbon and Soil Quality in
Cropped Rangeland Soils: the Importance of Distribution, Composition, and Soil
Biological Activity . CRC Press, Boca Raton, FL.

Leij, F.J., Ghezzehei, TA, and D. Or. 2002. Analytic Models for Soil Pore-Size
Distribution After Tillage. Soil Science Society of America, Vol.66, No.4, pp. 1104-
1 114.

25

Lindqvist, R., and G. Bengtsson. 1991. Dispersal Dynamics of Groundwater Bacteria.
Microbial Ecology, Vol.21, No.1, pp.49-72.

Lipiec, J ., Kus, J ., Siowir’rska-Jurkiewicz, A., and A. Nosalewicz. 2006. Soil Porosity and
Water Infiltration as Influenced by Tillage Methods. Vol. 89, No.2, pp.210—220.

Levin, M.A., Fischer, J .R., and V.J. Cabelli. 1975. Membrane Filter Technique for
Enumeration of Enterococci in Marine Waters. Applied Microbiology, Vol. 30, No.1,
pp.66-71.

Mackler, B.A., and J .C. Merkle. 2000. Current Knowledge on Groundwater Microbial
Pathogens and their Control. Hydrogeology Journal, Vol. 8, No.1 , pp.29-40.

Mahler, B.J., Personne’, J .-C., Lods, G.F., and C. Drogue. 2000. Transport of Free and
Particulate-Associated Bacteria in Karst. Journal of Hydrology, Vol.23 8, No.3-4, pp.179—
193.

Maier, R.M., Pepper, LL, and C.P Gerba. 2000. Indicator Microorganisms. pp.49l. In
Environmental Microbiology. Academic Press, San Diego, CA.

Mallmann, W.L., and W. Litsky. 1951. Survival of Selected Enteric Organisms in
Various Types of Soil. American Journal of Public Health, Vol. 41, No.1, pp.38-44.

Mannion, C., Lynch, P.B., Egan, J ., and EC. Leonard. 2007. Seasonal Effects on the
Survival Characteristics of Salmonella Typhimurium and Salmonella Derby in Pig Slurry
During Storage. Journal of Applied Microbiology, Vol.103, No.5, pp.1386-1392.

Manson, MD. 1990. Introduction to Bacterial Motility and Chemotaxis. Journal of
Chemical Ecology. Vol.16, No.1, pp.107-113.

Martens, D.A., and WT, Frankenberger, Jr. 1992. Decomposition of Bacterial Polymers
in Soil and their Influence on Soil Structure. Biology and Fertility of Soils. Vol.13, No.2,
pp.65-73. .

Messer, J .M., and AP. Dufour. 1998. A Rapid, Specific Membrane Filtration Procedure
for Enumeration of Enterococci in Recreational Water. Applied and Environmental
Microbiology, Vol. 6, No. 2, pp. 678-680.

26

 

 

 

 

.\1'
ii

hr

 

 

.\lUr
Rev:

Nari
30!"
End
Niel.

Pp. 1.
MI.

McCaulou, DR, and RC. Bales. 1995. Effect of Temperature-controlled Motility on
Transport of Bacteria and Microspheres through Saturated Sediment. Water Resources
Research, Vol.31, No.2, pp.271-280.

McEldowney, S., and M. Fletcher. 1988. The Effect of Temprature and Relative
Humidity on the Survival of Bacteria Attached to Dry Solid Surfaces. Letters in Applied
Microbiology, Vol. 7, No.4, pp.83-86.

Michigan Department of Agriculture. 2010. Generally Accepted Agricultural and
Management Practices for Manure Management and Utilization: Michigan Commission
of Agriculture. Accessed April 13, 2010 at URL
http://www.michigan.gov/documents/MDA_Manure_GAAMP_129695_7.pdf

Miescier, J.J., and V.J. Cabelli. 1982. Enterococci and Other Microbial Indicators in
Municipal Waste Water Effluents. Journal (Water Pollution Control Federation), Vol.54,
No. 12, pp. 1599-1606.

Mummey, D.L., and PD. Stahl. 2004. Analysis of Whole- and Inner-Microaggregate
Bacterial Communities. Microbial Ecology, Vol.48, No.1, pp.41-50.

Mummey, D., Holben, W., Six, J ., and P. Stahl. 2006. Spatial Stratification of Soil
Bacterial Populations in Aggregates of Diverse Soils. Microbial Ecology, Vol.51, No.3.
pp.404-41 1.

Murray, BE. 1990. The Life and Times of Enterococcus. Clinical Microbiology
Reviews, Vol.3, No.1, pp.46-65.

Natvig, E., Ingham, S.C., Ingham, B.H., Cooperband, LR, and TR. Roper. 2002.
Salmonella enterica Serovar Typhimurium and Escherichia coli Contamination of Root
and Leaf Vegetables Grown in Soils with Incorporated Bovine Manure. Applied and
Environmental Microbiology, Vol.68, No.6, pp.2737-2744.

Nielsen, D.M. 1991. Regulatory Mandates for Controls on Ground-Water Monitoring.
pp. 1-16. In Practical Handbook of Ground-Water Monitoring. Lewis Publishers, Chelsea,
MI.

27

Payment, P., Waite, M., and A. Dufour. 2003. Introducing Parameters for the Assessment
of Drinking Water Quality. In Assessing Microbial Safety Of Drinking Water, Improving
Approaches and Methods. IWA Publishing: London, 47-77

Peacock, A.D., Mullen, M.D., Ringelberg, D.B., Tyler, D.D., Hendrick, D.B., Gale, P.M.,
and DC. White. 2004. Soil Microbial Community Responses to Dairy Manure or
Ammonium Nitrate Applications. Soil Biology and Biochemistry, Vol.33, No.7-8,
pp.1011-1019.

Peixoto, R.S., Coutinho, H.L.C., Madari, B., Machado, P.L.O.A., Rumjanek, N .G., Van
Elsas, J.D., Seldin, L., and AS. Rosado. 2006. Soil Aggregation and Bacterial
Community Structure as Affected by Tillage and Cover Cropping in the Brazilian
Cerrados. Soil and Tillage Research, Vol.90, No.1-2, pp.16-28.

Powelson, D.K., and C. P. Gerba. 1995. Fate and Transport of Microorganisms in the
Vadose Zone. pp. 123-135. In Handbook of Vadose Zone Characterization and
Monitoring. CRC Press, Boca Raton, FL.

Ranjard, L., and A, Richaume. 2001. Quantitative and Qualitative Microscale
Distribution of Bacteria in Soil. Research in Microbiology, Vol. 152, No. 8, pp.707-716.

Roberson, EB, and MK. Firestone. 1992. Relationship between Desiccation and
Exoploysaccharide Production in Soil Pseudamonas sp. Applied and Environmental
Microbiology, Vol.58, No.4, pp.1284-1291.

Rosa, B.A., Y, M.S., Burdenuk, L., Kjartanson, B.H., and KT. Leung. 2009. The
Transport of Escherchia coli through Freeze-Fractured Clay Soil. Water, Air and Soil

Pollution, online publication URL
http://www.springerlink.com/content/e388u3 8t07205574/

Rose, J .B., Epstein, P.R., Lipp, E.K., Sherman, B.H., Bernard, S.M., and J .A. Patz. 2001.
Climate Variability and Change in the United States: Potential Impacts on Water- and
Foodbome Diseases Caused by Microbiological Agents. Environmental Health
Perspectives, Vol.109, No.2, pp.211-220.

Rozak, DB, and RR. Colwell. 1989. Survival Strategies of Bacteria in the Natural
Environment. Micobiological Reviews, Vol.51, No.3, pp.365-379.

28

.l

 

 

 

Scott, T. M., Rose, J. B., Jenkins, T. M., Farrah, S. R, and J. Lukasik. 2002. Microbial
Source Tracking: Current Methodology and Future Directions. Appl. Environ Microbiol.
Vol. 68, No. 12, pp.5796-5803.

Smucker, A.J.M., Park, E.J., Domer, J., and R. Horn. 2007 Soil Micropore Development
and Contributions to Soluble Carbon Transport within Macroaggregates. Vadose Zone
Journal, Vol.6, No.2, pp.282-290.

Simpson, J .M., Santo Domingo, J .W., and DJ. Reasoner. 2002. Microbial Source
Tracking: State of the Science. Environmental Science and Technology, Vol. 36, No.24,
pp.5279-5288.

Smucker, A.J.M., and J .W. Hopmans. 2007. Soil Biophysical Contributions to
Hydrological Processes in the Vadose Zone. Vadose Zone Journal, Vol.6, No.2, pp.267-
268.

Soby, S., and K. Bergman. 1983. Motility and Chemotaxis of Rhizobium meliloti in Soil.
Applied and Environmental Microbiology, Vol.46, No.5, pp.995-998.

Solo-Gabrielle, H.M, Wolfert, M. A., Desmarais, TA, and C.J. Palmer. 2000. Sources of
Escherichia coli in 3 Coastal Subtropical Environment. Applied and Environmental
Microbiology, Vol.66, No.1 pp.230-237.

Steenhuis, T.S., Parlange, J .-Y, and SA. Aburime. 1995. Preferential Flow in Structured
and Sandy Soils: Consequence for Modeling and Monitoring. pp.61-77. In Handbook of
Vadose Zone Characterization and Monitoring. CRC Press, Boca Raton, FL.

Stenstrom, T. A. 1989. Bacterial Hydrophobicity, an Overall Parameter for the
Measurement of Adhesion Potential to Soil Particles. Applied and Environmental
Microbiology, Vol.55, No.1, pp.142-147.

Stevik, T.K., Aa, K., Ausland, K., and J .F. Hanssen. 2004. Retention and Removal of
Pathogenic Bacteria in Waste Water Percolating through Porous Media: A Review. Water
Research, Vol. 38, No.6, pp.1355-1367.

Tan, K.H. 1994. Beneficial Effect of Soil Microorganisms. pp.68-81. In Environmental
Soil Science. Marcel Dekker, New York, NY.

29

 

 

 

St

lt
r1

Ty
fro

Uni
afte
71

~I

 

line
“at

Thunegard, E. 1975. On the Persistence of Bacteria in Manure. A Field and Experimental
Study with Special Reference to Salmonella in Liquid Manure. Acta Veterinaria
Scandanavia, Suppl. 56, pp.5-85

Thurston-Enriquez, J. A., Gillery, J. E., and B. Eghball. 2005. Microbial Quality of
Runoff Following Land Application of Cattle Manure and Swine Slurry. Journal of Water
and Health Vol. 3, No.2 pp. 151-171.

Tisdall, J .M, and J.M. Oades. 1982. Organic Matter and Water-stable Aggregates in
Soils. Journal of Soil Sciences, Vol. 33, No. 2, pp.141-163.

Torsvik, V., Goksoyr, J., and F. L. Daae. 1989. High Diversity in Soil Bacteria. Applied
and Environmental Microbiology, Vol. 56, No.3, pp.782-787.

Tyrrel, SF, and IN. Quinton. 2003. Overland Flow Transport of Pathogenic Bacteria
from Agricultural Land Receiving Faecal Wasters. Journal of Applied Microbiology,
Vol.94, pp.87S-93S.

Unc, A., Gardner, J ., and S. Springthorpe. 2006. Recovery of Escherichia coli from Soil
after Addition of Sterile Organic Wastes. Applied and Environmental Microbiology, Vol.
72, No.3 , pp.2287-2289.

Unc, A., and M.J. Goss. 2003. Movement of Faecal Bacteria Through the Vadose Zone.
Water, Air, and Soil Pollution, Vol. 149, No. 1-4, pp. 327-337.

Unc, A., and M.J. Goss. 2004. Transport of Bacteria from Manure and Protection of
Water Resources. Applied Soil Ecology, Vol.25, No.1, pp.1-18.

United Nations. 2008. United Nations Department of Economic and Social
Affairs/Population Division. World Population Prospects: The 2008 Revision. Accessed
January 5, 2010 at URL:
http://www.un.org/esa/population/publications/wpp2008/wpp2008_highlights.pdf

United States Census Bureau. 2009. International Data Base. World Summary. Accessed
January 5, 2010 at URL: http://www.census.gov/ipc/www/idb/worldpopinfo.php

30

 

 

 

 

 

\\

 

 

 

 

United States Environmental Protection Agency (USEPA). 2002a. Method 1600:
Enterococci in Water by Membrane Filtration Using Membrane-Enterococcus Indoxyl-B
-D-Glucoside Agar (mEI). United States Environmental Protection Agency, Office of
Water, Washington DC.

United States Environmental Protection Agency (USEPA). 2002b. Method 1603:
Escherichia coli (E. coli) in Water by Membrane Filtration Using Modified membrane-
Thermotolerant Eschrichia coli Agar (Modified mTEC). United States Environmental
Protection Agency, Office of Water, Washington DC.

United States Environmental Protection Agency. 2003. Environmental Regulations and
Technology: Control of Pathogens and Vector Attraction in Sewage Sludge. Accessed
May 12, 2010 at URL: http://www.epa.gov/nrmrl/pubs/625r92013/625R92013.pdf

United States Environmental Protection Agency. 2004a. Monitoring and Assessing Water
Quality. National Water Quality Inventory: Report to Congress, 2004 Reporting Cycle.
Accessed January 18, 2010 at URL: http://www.epa.gov/owow/305b/2004report/

United States Environmental Protection Agency. 2004b. Implementation Guidance for
Ambient Water Quality Criteria for Bacteria. United States Environmental Protection
Agency, Office of Water, Washington DC.

United States Environmental Protection Agency. 2007. Microbiological Methods/ Online
Publications. Accessed September 16, 2009 at URL
http://www.epa.gov/nerlcwww/online.htm

United States Environmental Protection Agency. 2009. Clean Water Act (CWA).
Accessed November 2, 2009 at URL http://www.epa.gov/oecaagct/lcwa.html#Summary

Van Loosdrecht, M.C.M, and A.J.B. Zehnder. 2005. Energetics of Bacterial Adhesion.
Cellular and Molecular Life Sciences, Vol. 46, No.8, pp.817-822.

Vargas, R., and T. Hattori. 1986. Protozoan Predation of Bacterial Cells in Soil
Aggregates. FEMS Microbiology Letters, Vol.38, No. 4, pp.233-242.

31

Wiebe, K., and N. Gollehon. 2006. United States Department of Agriculture Economic
Research Service. Agricultural Resources and Environmental Indicators. In Water
Quality: Impacts of Agriculture. Accessed April 13, 2010 at URL
http://www.ers.usda.gov/publications/arei/eib16/

World Health Organization. 2000. Global Water Supply and Sanitation Assessment 2000
Report. Accessed on January 5, 2010 at URL
http://www.who.int/water_sanitation_health/monitoring/jmp2000.pdf

World Health Organization. 2009. Health through safe drinking water and basic
sanitation. Accessed on May 22, 2009 at URL
http://www.who.int/water_sanitation_health/mdg1/en/index.html

Yates, M. V. 2007.Classical Indicators in the 21St Century - Far and Beyond the
Coliform. Water Environment Research, Vol., 79, No.3 pp.279-286.

Zaval'skii, L.Y, and AG. Voloshin. 2003. Bacterial Motion in Porous Media.
Microbiology, Vol.72, No.3, pp.369-372.

Zhou, J ., Xia, B., Treves, D.S., Wu, L.Y., Marsh, T.L., O’Neill, R.V., Palumbo, A.V.,
and J .M. Teidje. 2002. Spatial and Resource Factors Influencing High Microbial
Diversity in Soil. Applied and Environmental Microbiology, Vol.68, No. 1, pp.326-334.

32

Chapter 2.

Material and Methods Development

33

2.1 Introduction

Transport of bacteria through agricultural soil and into the underlying ground
water or runoff to nearby surface water poses a public health risk. When humans come in
contact with receiving waters, they may be infected by pathogenic microorganisms.
Conventional agricultural management practices such as mechanical tillage modify the
soil structure and may alter pore structure. Because these pores act as conduits for
interaggregate and intraggregate microbial movement, it is hypothesized that such
practices may play a role in managing microbial risk. For example, when soil is tilled
microbial infiltration may be retarded due to the disruption of pore networks. Therefore,
the objective of this study was to examine the transport of the bacterial indicators, E. coli
and Ent. faecium, at the macroaggregate scale and investigate the correlation with
structurally modified soil aggregates through conventional tillage and soil aggregates that
have received no agricultural modification for 20 years, both sampled from the Kellogg

Biological Station Long Term Ecological Site.

Preceding the experimental analyses to address the aforementioned objectives, a novel
method was designed for bacterial extraction and obtaining optimal recovery of viable
spiked bacteria in soil aggregates. Following the methods optimization, three

experimental methods were designed for the following objectives

1. Evaluation of mechanisms controlling bacterial retention in aggregates.

2. Examination of the transport behavior of E. coli and Ent. faecium in tilled and non-
tilled soils at unsaturated and saturated conditions.

3. Exploration of spatial distributions of bacteria within subsections of inoculated soil
aggregates.

34

 

 

 

 

I0

lit

dl

2.2 Methods Development

2.2.1 Soil aggregates. Soil samples were collected at the Long Term Ecological Research
Site (LTER), Kellogg Biological Station (KBS) located in southwest Michigan (850 24‘
W longitude, 420 24‘ latitude) in November, 2008. Soil at the site are Typic Hapludalfs
(of the Alfisol order) made up of either fine loamy, mixed mesic Kalamzoo series or
coarse loamy, mixed mesic Oshtemo series. The 60 hectare site is subdivided into six
replicates of 1 hectare plots exposed to eight different agricultural management

treatments (Fig. 2.1).

For this study aggregates were collected from three different treatment plots: the
T1 treatment plot had conventional tillage (chisel-plowed) with a com-soybean-wheat
rotation field and was conventionally fertilized (3.35:0.3 kg N ha-l day-1). The T2
treatment plot received no tillage but had com-soybean-wheat rotation and was
conventionally fertilized. The T7 treatment was native successional plot and received no
tillage after spring 1989 (Robertson et al., 2000). From each replicated plot, sample soil
blocks, approximately 15 x 15 cm in size were extracted from 0-20 cm depths using a
sharp flat spade. Soil was air-dried and then manually sieved by gently shaking for 30
seconds into different aggregate sizes. The aggregates of 4-6.3 mm size fraction were
used for this study and stored at laboratory temperature in a plastic container for all
experiments. Aggregate level texture, C and N content and bulk soil densities are

described in Tables 2.1 and 2.2.

35

2008 KBS LTER Main Site

0 T1- conventional tillage with com-
soybean-wheat rotation and
fertilization

0 T2- non-tilled with com-soybean-
wheat rotation and fertilization

0 T7- native successional with no
tillage after 1989

 

T2

 

 

 

 

T7

 

 

 

T1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N T2
T1 T7
T1 T2 T2
T7 T2
T7 T1
T1
T2 T7
T7 T1

 

 

Fig 2.1. KBS LTER soil sampling plots.

36

 

Table 2.1. Average percentages of soil texture in aggregate treatments T 1, T2 and T7 (Chun, 2010,
Unpublished). Standard deviations are indicated in parentheses.

 

 

 

 

 

Soil Aggregate Treatment Texture
(0-15cm depth) (percent)

Sand Silt Clay
Tilled - T1 treatment 21 (18%) 35 (19%) 44 (2.5%)
Non-tilled - T2 treatment 35 (27%) 31 (18%) 34 (10%)
Native successional - T7 treatment 27 (17%) 34 (13%) 39 (5%)

 

 

 

 

 

 

Table 2.2. C, N and bulk soil densities of soil aggregate treatments Tl , T2 and T7 (Grandy and Robertson,
2007). Standard deviations are indicated in parentheses.

 

 

 

Soil Aggregate Treatment Total Average Bulk Density

(0-5cm depth) (g/mz) (g/cm3)
Carbon Nitrogen

Tilled - T1 treatment 621 (51.1) 57.3 (5.31) 1.37 (0.01)

 

Non-tilled - T2 treatment 885 (55.1) 81.0 (4.66) 1.36 (0.03)

 

Native successional - T7 1,1001 (38.6) 86.1 (3.54) 1.21 (0.02)
treatment

 

 

 

 

 

 

2.2.2 Bacterial extraction method. A vortexing and membrane filtration (VMF) method

was adapted from the technique devised by Singh to enumerate the native bacteria in the

37

whole, inner and outer layers of the T1, T2 and T7 soil aggregates (2007). First, each
whole soil aggregate was weighed to the nearest thousandth gram and placed into a 15ml
centrifuge tube (Difco, Franklin Lakes, NJ, #236940) containing 10ml of IX Phosphate
Buffered Water (PBVW (pH 7-7.2). The 1X PBW stock was made in accordance with
Standard Methods for Examination of Water and Waste Water guidelines (Eaton, 2005).
The centrifuge tubes containing the aggregate and PBW were votexed at full speed for 3
minutes, inverted 20 times, then vortexed again for another 3 minutes. In the initial
experiments investigating native bacterial concentrations, the soil sediment was allowed
to settle for 20 minutes before proceeding to membrane filtration. However, after
evaluation, the settling step was eliminated from the method due to underestimating the
viable bacterial counts (see section 3.1.3). After vortexing, the samples were serially
diluted ten-fold in 1X PBW. Dilutions were then membrane filtered through 0.45pm pore
sized membrane filters (Pall Corp., Ann Arbor, MI, #T914361) following Standard
Methods, Total Colifrom (TC) bacteria were enumerated using a membrane filter
procedure (Eaton, 2005). Total Heterotrophic (HPC), TC (and E. coli) and Enterococcus
bacterial concentrations per gram of soil aggregate were assayed using Tryptic Soy Agar,
m-Endo LES agar and mEI (BD and Co., Sparks, MD; #236920; #273620; #214881),
respectively. Each sample dilution was processed in triplicates for all media, then placed
bottom-side up in a 35:0.50C incubator for HPC and TC while mEI plates were
incubated at 41:0.50C. All media was incubated for 24 hours prior to enumeration.
Bacterial concentrations were calculated by manually counting colony forming units
(CFU) from dilution plates and back calculating to original concentration. The original

concentration was divided by the soil aggregate’s weight and reported at CFU/ g of soil.

38

2.2.3 Direct E. coli and Ent. faecium spiking and immediate recovery. Spiking
experiments were aimed at answering two questions: 1) what was the detection limit of
the assay, i.e., what was the lowest concentration of spiked bacteria recoverable and 2)
what was the total percentage recovered. Stocks of E. coli ATCC 15597 (designation C-
3000; derived from K-12 strain) and Ent. faecium ATCC 35667 kept at -80°C, were
thawed and then placed in 4ml of Tryptic Soy Broth (BD Diagnostics, Franklin Lakes,
#211768) and incubated overnight at 3510.50C in a shaking incubator. The overnight
grown E. coli and Ent. faecium cultures were serially diluted 1:10 in PBW, covering a
gradient of concentrations from approximately 1 CFU/soil aggregate to 1x107 CFU/ soil
aggregate for E. coli and l CFU/soil aggregate to 1x104 CFU/ soil aggregate for
Enterococci. A narrower gradient range for Ent. faecium was used after determination
that higher concentrations would not be used in the primary flow and spatial distribution
experiments so as not to over-saturate aggregates, thereby masking the effects of soil
structure on bacterial transport. Subsequently, each stock dilution used for spiking was
membrane filtered in triplicates onto m-Endo and anI. After the passage of 24 hours, the
CFUs were enumerated and each spiked dilution was back calculated to determine the

influent spiked concentration.

The following day, the soil aggregate was placed on sterile petridish (60 x15mm)
and weighed to the nearest thousandth gram. The volume of 50p] was observed to
saturate the soil aggregates, hence, each spiked stock dilution used to saturate the T1 and
T7 treatment aggregates with the aforementioned volume by gently touching the drops
formed at the end of the micropipette tip to the aggregate. To avoid desiccation (via

evaporation) and bacterial die-off in the aggregate, the spiked soil aggregates were placed

39

 

lg

 

in 10ml of 1X PBW and immediately assayed using the vortex and membrane filtration
method. Since only E. coli and Ent. faecium recovered concentrations were of interest,
only m-Endo LES and mEI agars were used as selective growth media. CFU counts were
obtained from countable dilution plates and back calculated to obtain recovered bacterial
concentrations from each aggregate. Calculation of percent recovery per aggregate was
obtained by dividing the recovered concentration after spiking by the total CFU (as
measured per dilutions). The data were statistically analyzed as a ratio of spiked bacterial

concentration to recovered bacterial concentrations.

2.2.4 Desiccation effect on E. coli recovery in whole and aggregate subsections. The
effect of desiccation on spiked bacteria survival was critical to assess the required
moisture content in the soil prior to designing the flow experiments (see next section).
The percent recovery of spiked E. coli as a function of soil moisture content by weight
was measured. E. coli stock was spiked into T7 aggregates at a concentration of
approximately 104 CFU/ aggregate using the same spiking procedure mentioned earlier.
The spiked concentrations were approximated based on calculations of E. coli stock
dilution concentrations from experiments 2.2.3. To affirm the approximated
concentration, the actual concentration was measured and calculated by membrane
filtering the stock dilution used and enumerated following 24 hour incubation. Next,
spiked aggregates were placed in petri dishes that were slightly open, allowing for
evaporation while avoiding condensation and the risk of contamination. Following
interval air-drying at 15 and 30 minutes, the spiked aggregates were processed using the
vortex and membrane filtration method both with and without the settling step. To

calculate moisture content at each time interval, aggregate weights were taken 1) prior to

40

Ll).

SC

 

 

axe

 

spiking, 2) immediately after spiking and 3) after interval air—drying. Moisture retained in
the aggregates after interval air-drying was determined by subtracting spiked aggregates’
weight after interval air-drying (3) from the aggregates’ weight prior to spiking (2). The
obtained value was divided by the aggregates’ weight prior to spiking, yielding the
moisture content of the aggregate after air-drying. While the recorded weight
immediately after spiking was not used in this equation, it was useful reference for the

amount of moisture loss.

The desiccation analysis was further extended to spiked soil aggregates that were
sliced into three subsections. The experiment aimed to ascertain the susceptibility of
spiked E. coli in the spatial regions of soil aggregate upon exposure to desiccation stress.
Preceding the soil aggregate spiking, the aggregate was cut into top, middle and bottom
subsections with a stainless steel razor blade, flame-ethanol sterilizing in between each
cut. The orientation of the subsections were in reference to the spatial location of the E.
coli spike, i.e., the top section was where the E. coli was directly added, the middle was
the section right below and the bottom was the lowest. Slices were left to dry for 0, 10, 40
and 60 minutes then processed via the vortexing and membrane filtration method without

settling.

2.2.5 Soil aggregate rehydration and enrichment. Another concern was resuscitating
native coliforms and Enterococci following prolonged hydration of soil aggregates during
the flow chamber and slicing and saturation experiments. Therefore, the re-growth of
both bacterial groups was assessed by suspending soil aggregates for an extended period
of time. First, the soil aggregate was weighed to the nearest thousandth gram and then

aseptically placed in a 15ml centrifuge tube containing either 5ml TSB or 10 ml sterilized

41

 

 

 

 

Ca
xe-

 

ho
ho

f

nano pure water. Because the brief time period that the aggregates were placed in PBW
during processing was inadequate to examine re- growth of native bacteria (see results),
the suspended soil aggregates were incubated for 24 hours. One set of aggregates
suspensions were incubated at 3510.50C to promote re- growth of native E. coli while
another set was incubated at 410C for Enterococci resuscitation. The aggregates were
processed as previously described. No growth on mEI and m-Endo indicated that there
were no viable coliforms or Enterococcus spp. from the aggregates suspended in
nanopure water and TSB. Therefore it was assumed that keeping the aggregates hydrated

for less than 24 hours posed no risk of resuscitating any native fecal bacteria.

2.2.6 The effect of calcium chloride solution on E. coli recovery. A calcium chloride
(CaClz) solution was used to prevent soil aggregates from collapsing upon application of
vacuum in the flow experiments. A study by Winslow and Falk had demonstrated that
with increasing CaClz concentrations, the percent live E. coli (at the time named
Bacterium coli) decreased after 24 hours even at a stable pH of 7 (Winslow and Falk,
1922). Furthermore, increasing the concentration of CaClz resulted in higher E. coli die-
off, ultimately reaching the maximum toxicity at 0.435M CaClz. Therefore, to ensure
that the E. coli strain used in the experiments did not die-off, the temperature and the time

that is required to keep spiked bacteria viable needed to be optimized.

Preliminary experiments were designed to examine E. coli recovery at 0, 2 and 4
hours at 40C and 280C suspended in a 0.5mM CaClz solution. High recoveries after 4
hours of immersion under all the variables indicated that the 0.5mM concentration of
CaClg (used to stabilize soil aggregates) was not detrimental to E. coli survival.

Nevertheless, it was determined that bacteria spiked in soil aggregates would not be

42

immersed in CaClz solution for longer than 30 minutes while aggregate processing was

completed within 3 hours in the flow chamber experiments.

2.2.7 Examination of clumping in E. coli and Ent. faecium cultures. Clumping of
bacterial cells could cause errors in CFU counts (Goldman and Green, 2009).
Microscopic examination of E. coli and Ent. faecium clumping were performed with a
Ziess Axioskop 2 Plus model (Gottingen, Germany) using the differential interference
contrast option at oil-immersion (100X) resolution. Approximately 20p.l of each bacterial
sample at approximately 107 CFU/ml were aliquoted onto glass slides and covered with a
glass cover slip and sealed by nail polish. Images were captured using the peripheral
camera Axiocam model MRc (Miinchen-Hallbergmoos, Germany) and stored in .jpeg

and .zvi format using AxioVision software Release 4.5 (Gottingen, Germany).

Initial observation of E. coli cells seemed to indicate that cells clumped as readily
as Ent. faecium. However, the aggregated appearance of the E. coli cells was showed that
the cells were in mid-division phase. Utilizing the zoom function in AxioVision software,
the images were examined more closely to observe clumping at higher magnification.
Individual Ent. faecium cells showed distinct delineation when clumping, only few E.
coli cells showed the same pattern (Figures 2.2 and 2.3). Most E. coli cells were observed
to be unassociated with other cells. Conversely, Ent. faecium frequently clumped and
sometimes exhibited formation of chains of three or more cells, a phenomenon not
observed while investigating E. coli cells. Although this analysis is qualitative, perhaps it
indicates nature of the Ent. faecium cell wall and its capacity to adhere to other cells more

readily than E. coli.

43

 

Fig 2.2. Differential interference contrast microscopy of E. coli cells (concentration of 10 CFU/ml) at
lOOX resolution. C indicates clumping of two cells.

 

Fig 2.3. Differential interference contrast microscopy of Ent. faecium cells (concentration of 10 CFU/ml)
at lOOX resolution. C indicates Ent. faecium chains longer than two cells.

2.3 Flow chamber experiments. A glass bead matrix chamber system was designed to
include a pore extraction chamber (PEC) to characterize E. coli and Ent. faecium
transport through soil aggregates (Fig 2.4.) (Hyen et al., manuscript in preparation).
Briefly, the system was composed of two chambers, the PEC and the collection chamber.
In the PEC chamber, 2cm of sterile glass beads, 1mm in diameter, were overlaid on a
single soil aggregate to avoid disruption of soil stability upon application of vacuum. The
glass/soil matrix was set on a porous cindered -glass filter with 25 um pore size to allow
bacterial leaching following vacuum extraction. The collection chamber, located at the
base of the PEC encased a sterile 2m] centrifuge tube designated for effluent collection

and was connected to a vacuum pump (model: RPC-R, Gast, USA) and vacuum

45

 

 

Vacuum

Fig 2.4 a) Glass beat matrix chamber design for investigation of bacterial transport. b) Glass bead matrix
experimental setup.

controller (model: CVC2, Vacuubrand, Germany). All flow experiments were conducted
at laboratory temperature to observe if motility affected transport (McCaulou and Bales,

1995).

For experiments examining bacterial transport at hydrated soil conditions, the
system was saturated through capillary action by placing the PEC in 0.5mM CaClz
overnight hydrating the soil. Experiments investigating flow at non-saturated conditions
did not include this pre-saturation step. Subsequently, - 100cm of water vacuum was

applied to the system for 30 minutes, to remove excess C3C12.

Spiking of E. coli and Ent. faecium was conducted as previously described (see

direct E. coli and Ent. faecium spiking and immediate recovery) with the exception that

46

25 ul of each stock was added to avoid potential for over saturating the aggregate. The
spiked bacteria were left for 10 minute to equilibrate in the aggregate, then 2ml of CaClz
was added to the upper chamber by slowly pipetting the solution on the side of the
chamber to avoid breaking the aggregate apart. Then, vacuum (-100cm) was applied for
another 30 minutes to collect the effluent volume. The soil aggregate and beads were
aseptically removed by forceps from the chamber and placed in 10m] and 9ml of PBW
respectively and processed via the VMF method. The extracted effluent was processed
without addition of PBW. All samples were processed within 2 hours after bacterial

spiking to maintain bacterial viability.

To compare the concentration of bacteria retained in the aggregates against the
bacterial concentration in the effluent, concentrations were calculated in the units of total
CFUs. For calculating bacteria retained in soil, the CFU enumerated from plate counts
were multiplied by the dilution factor and by volume of PBW the aggregate was stored
in. To obtain total CFU for the effluent samples, the CFU enumerated from plate counts
, were multiplied by the dilution factor and by the effluent volume. Because the influent
volume was different for E. coli and Ent. faecium, values had to be adjusted for statistical
analysis. This was achieved by converting bacterial concentrations retained in soil and
concentrations in the effluent to a ratio by dividing these concentrations by the total

influent concentrations.

2.4 Aggregate peeling for native bacteria enumeration. For analysis of the spatial
variation of native bacteria, the soil aggregates were peeled into three layers: exterior,
transitional and inner. Only the exterior and inner layers were analyzed for native

bacterial concentrations. The logic was to determine if there was a significant difference

47

between the two layers initially which may warrant further investigation and analysis of
bacterial concentration in the transitional layer. The aggregates were peeled using soil
erosion chambers (SAE) developed by Dr. Alvin Smucker and described in detail in Park
and Smucker (2005) (Fig. 2.5). At the onset of soil aggregate peeling a pre-weighed,
single aggregate was placed on a support screen and the top of the SAE chamber was
covered with heavy duty aluminum foil. The interior wall portion of the SAE chamber
was precisely machined into a uniformly knurled surface which eroded the rotating
aggregate. A 350 um-opening support screen was welded to the base of the chamber
through which finely eroded soil materials dropped into the retainer of the base of the
SAE. The entire SAE chamber was placed onto a spring mount onto a rotary shaker
platform (Innova, Model 2300, New Brunswick Scientific Inc., Edison, NJ) and rotated at
speeds ranging from 200 to 400 rpm. Rotational motion of the chamber generated
frictional forces at the surface interface of each aggregate. Sequentially, 1/3 and 2/3
(exterior and transitional layers, respectively) by weight of soil aggregates were peeled
and weighed. Because of the decrease of erosion rate with aggregate peeling, the peels
needed to be weighed several times to assure that 1/3 and 2/3 of soil aggregates were
peeled. Each concentric layer was placed in 15ml centrifuge tube containing 10ml of

PBW and processed using vortexing and membrane filtration as previously described.

48

 

Knurling inside wall for
effective friction

         
       

. - v 0 '- v 9' - ,

o‘g‘.:?o‘§‘pf’h.gf‘, 9".‘jV-‘u' g

'3 ’tr 9.5"? 93’0‘3‘ Q'Io‘ ‘
“ 0 x-.?9.€r.~.-.c' 4‘- ' I

Erosion Chamber
Soil Aggregate

350nm Sieve

Retainer Base Chamber

Eroded Soil Material

 

 

 

Fig 2.5. SAE chamber and retainer base used for aggregate peeling and collection (Park and Smucker,
2005).

2.5 Slicing and saturation experiments. These experiments were designed to
understand the spatial distribution of spiked E. coli added to aggregates that had different
soil moisture contents prior to added E. coli. Sterile DI water was applied to soil
treatments T1, T2 and T7 to add 15, and 30% soil moisture contents by weight. The 15%
moisture content aggregates had 16ul of DI water added and the 30% moisture content
aggregates had 331.11 of DI water added. The aggregates defined as 0% moisture content
aggregates were air—dry aggregates that did not have any DI water added prior to spiking
E. coli. E. coli is widely accepted as the primary indicator in agricultural soils and due to
the fact that this was an exploratory examination of spatial distribution, E. coli was used

for these experiments.

Calculating volumes that would be applied to the aggregates to add the specified

moisture contents was achieved by preliminary weighing experiments. Five replicates of

49

T1, T2 and T7 air-dry aggregates were placed in petri dishes and presumed to be “fully”
saturated with lml DI water. Excess DI water that did not saturate the aggregates was
pipetted and discarded and then the weight of each aggregate was measured again.
Subtracting the air-dried aggregate’s weight from the saturated aggregate’s weight
yielded the amount of water in grams for the “fully” saturated air-dry aggregates. The
average weight in grams of water to attain a presumed “full” saturation from each of the
five replicates per treatment was multiplied by 0.15 and 0.30 to achieve 15% and 30%
moisture contents for slicing experiments. Aggregates were allowed to equilibrate for 2
hours after adding the specified volume and kept inside closed containers within beakers
containing water to avoid evaporation. In these experiments, the aggregates were
hydrated to 0% (kept dry), 15% (using l6ul) and 30% (using 33 pl). Subsequently, E. coli
was added at a concentration of approximately 103 CFU/aggregate by seeding 50p.l of the
E. coli culture dilution and the aggregate was cut aseptically into seven pieces. Each
subsection was assigned a number in reference to where the E. coli was added.
Subsections 1, 2, 3, 4, and 5 correspond to the top, right, left, back and front of the
aggregate, respectively (Figure 2.6). The middle section was divided into two subsections
6 and 7 which correspond to the center-middle and the bottom-middle respectively (Fig

2.7).

After slicing, each subsection was weighed and immediately placed in lml of
PBW to prevent E. coli die-off. All samples were processed with the VMF method within
four hours to avoid variability; in the meantime they were stored at 40C. Triplicate

membrane filtered samples were placed on m-Endo agar and incubated at 3510.50C for

50

24 hours. E. coli CFU recovery per gram was calculated by the attained CFU value of

subsection divided by the subsection weight.

 

 

-----------------------------------------------

 

 

Fig. 2.6. Frontal view of the aggregate subsections. The arrow indicates the location where E. coli was
spiked. The numerical designation of the aggregate subsections correspond to the spatial position of
aggregate slicing. The larger font and darker toned numbers indicate that the sections are closer
dimensional depth.

2.6 Statistical analysis. Average and standard deviation values for all experimental
analyses were calculated using Microsoft Excel 2007. Two-way analysis of variance
(ANOVA) was conducted to evaluate statistical significance between native bacterial
concentrations in T1, T2 and T7 aggregate treatments (native bacterial extraction; section
2.2.2), and for E. coli and Ent. faecium recovery ratios in T1 and T7 aggregates (E. coli

and Ent. faecium spiking and immediate recovery experiments; section 2.2.3). Three-way

51

ANOVA was conducted on E. coli and Ent. faecium aggregate and effluent ratios in T1,
T2 and T7 aggregates (flow-chamber experiments; section 2.3) and log-transformed E.
coli recoveries from aggregate slices at 0, 15, 30% moisture contents in T1, T2 and T7
aggregates (slicing and saturation experiments; section 2.5). Because the data set
contained values of zero, a value of 1 was added to the slicing and saturation data to
perform log-transformations. Levene’s test was used in all ANOVA to check unequal
variances. Tukey’s test was used for pair-wise comparisons between variables in slicing
and saturation experiments and flow chamber experiments. The Akaike and Bayesion
criteria were used to determine the goodness of fit model for grouped data in slicing and
saturation experiments and flow chamber experiments. All statistical analyses were
performed with assistance from the College of Agriculture and Natural Resource

(CANR) Statistical Consulting Center using SAS Version 9.2 (SAS Institute, NC).

52

2.7 Reference List

Chun, H.C. 2010. Unpublished data.

Crum, J .R., and HP. Collins. 2009. The KBS Site Description. KBS Soils. Accessed
April 6, 2010 at URL: http://lter.kbs.msu.edu/about/site_description/soils.php

Eaton, A.D., Clesceri, L.S., Eugene, W. R. and A.E. Greenberg. 2005. Media
Specifications. Section 9050C, 1a. In Standard methods for the Examination of Water
and Wastewater. let Ed. American Public Health, Baltimore, MD.

Baton, A.D., Clesceri, L.S., Eugene, W. R. and A.E. Greenberg. 2005. Standard Coliform
Membrane Filter Procedure. Section 9222B, 2b. In Standard methods for the Examination
of Water and Wastewater. 21St Ed. American Public Health, Baltimore, MD.

Goldman, E., and L.H. Green. 2009. Practical Handbook of Microbiology. In
Quantitative Microbial Enumeration. 2nd Edition. CRC Press, Boca Raton, FL.

Grandy, AS, and GP. Robertson. 2007. Land-use Intensity Effects on Soil Organic
Carbon Accumulation Rates and Mechanisms. Ecosystems, Vol.10, No.1, pp.58-73.

Mahler, B.J., Personne’, J .-C., Lods, G.F., and C. Drogue. 2000. Transport of Free and
Particulate-Associated Bacteria in Karst. Journal of Hydrology, Vol.238, No.3-4, pp.179-
193.

McCaulou, DR, and RC. Bales. 1995. Effect of Temperature-controlled Motility on
Transport of Bacteria and Microspheres Through Saturated Sediment. Water Resources
Research, Vol.31, No.2, pp.271-280.

Park, B.J., and A.J.M. Smucker. 2005. Erosive Strenghts of Concentric Regions within
Soil Macroaggregates. Soil Science Society of America Journal, Vol.69, No.6, pp.1912-
1921.

Robertson, G.P., Paul, B.A., and RR. Harwood. 2000. Greenhouse Gases in Intensive
Agriculture: Contributions of Individual Gases to the Radiative Forcing of the
Atmosphere. Science, Vol.289, No.5486, pp.1922-1925.

53

Singh, S. 2007. Investigation of Bacterial Fecal Indicators and Coliphage Virus in
Sediment and Surface Water of Parks and Beaches along the Grand River (MI) and Lake
Michigan (MI). Master’s Thesis. Michigan State University.

Winslow, C.-E.A., and LS. Falk. 1929. Studies on Salt Action. The Influence of Calcium
and Sodium Salts At Various Hydrogen Ion Concentrations Upon the Variability of

Bacterium Coli. Proceedings of the Society for Experimental Biology and Medicine,
Vol.8, No.3, pp.215-236.

54

Chapter 3.

Results and Discussion

55

3.1 Native Bacterial Concentrations

3.1.1 Native bacterial concentrations in dry soil aggregates. Initial experiments were
designed to determine the levels of native bacterial populations and background levels of
E. coli and Ent. faecium prior to running the spiking and immediate recovery experiments
(section 3.2). Whole aggregates were evaluated for the enteric bacteria. No growth on m-
Endo media (n=5) and mEI media (n=3) with any aggregate from all soil treatments
indicated there were no viable coliforms or Enterococcus spp. in air-dried aggregates.
This may be attributed to the extended storage time of the soil aggregates at laboratory
conditions where the bacteria either died due to desiccation stress or underwent a
metabolic shift to a viable but non-culturable state (Rozak and Colwell, 1987). There
was, however, growth on TSA plates indicating presence of heterotrophic bacteria (HPC).
Treatments T1, T2 and T7 contained an average of 3.02x105, 3.05x105 and 3.76x105
CFU/ g (Fig. 3.1) with standard deviations of 8.67 x104, 1.22 x105 and 1.60x105

respectively. These differences were not significant (p>0.05).

 

 

 

 

 

 

 

8.00E+05
0.141 (0.035) g
6.00E+05 -
an 0.158 (0.037)g* 0'148(°-°3°)g
E 4.00E+05 ~
U
2.00E+05 -
0.00E+OO - . T ~
T1 T2 T7
Soil 'fi-eatment

 

 

 

Fig 3.1. Native bacterial concentrations in aggregate treatments from T1, T2 and T7 (n=5) treatments.
* The values above each bar indicate the average weight (g) of each aggregate for each soil treatment. The
values in parentheses indicate the standard deviations.

56

3.1.2 Effect of settling on bacterial extraction. Compiling native bacterial extraction
data from whole soil aggregate treatments T1, T2 and T7 indicated a negative correlation
between aggregate weight and bacterial extraction (n=15) (Fig. 3.2). It can be observed
that as the soil aggregate’s weight increased, the I-IPC/ g decreased. Arriving at such a
relationship may be explained by the sorption of native bacteria to the soil particulates,
the heavier of which settles much quicker in solution, thereby leading to a decreased
detection and underestimation of actual viable bacteria (Richaume et al., 1993, Mahler et
al., 2000). Thus an evaluation of the effect of settling on bacterial extraction and
enumeration was undertaken. The T7 soil treatment was used as a proxy for all the
treatments since it was shown that their bacterial concentrations were not significantly
different (see section 3.1.1). Processing of two replicates of soil aggregates yielded an
average of 2.83x106 CFU/ g of native heterotrophic bacteria when the aggregate was
dissolved in PBW, vortexed and assayed without settling; a l-log increase as compared to
extraction with a 20-minute settling step. This illustrated that viable bacteria adhered to
soil particulates very strongly after being stored at dry conditions, and even after

extensive vortexing, this attachment was not disrupted.

57

 

 

6.00E+05

 

5.00E+05

 

4.00E+05

 

 

CFU/g

y- --.——-— - .-., - ...- ..-.- ..._-» ...-. . --y. ~-+- -.'h t
. j ’ - i
, «...... -..“... ..-;. ...--. r _ _ -.._._ ..k. a _- .1..... y. _ _ .. ..- . -..;—
, 1 l I l
...... _‘ .. . .. . . 1 g . 4
. 1 . . I
I 1 L ' a
r l r ‘ . Y Y
o - . , ‘
, .' _,.l . ..., . ,. . ‘
, a ; . l . t
,.. “...... ._'.. . n I .4 , 1 ‘
- i
> ' ‘ t l t u L a t

 

 

2.00E+05

 

 

 

 

 

 

 

 

0.00300 ; _ 1.. i : .' ; 1.
0.00 0.05 0.10 0.15 0.20 0.25
Aggregate Weight (g)

 

 

 

Fig 3.2. Comparison of aggregate weight from treatments T1, T2 and T7 to native bacterial colony forming
units per gram of soil with a trend line (n=15).

3.2 Spiking and [mediate Recovery

3.2.1 E. coli and Ent. faecium spiking and immediate recovery. A gradient of
concentrations was used to determine the detection limit for the VMF method after
fully saturating the aggregate with both E. coli and Ent. faecium. T1 and T7
aggregates were spiked in triplicates and recovery rates and standard deviations
were calculated after processing. The combined average of all concentrations for
E. coli in the T1 treatment for was 108% (standard deviation: 37%) (Table 3.1).
This indicated very high recovery rates could be achieved with concentrations as
low as 6 CFU/ soil aggregate. Similarly, seeded T7 soil aggregates had very high
recovery rates. The total percent recovery for the combined average of all

concentrations was 92% (standard deviation: 31%) (Table 3.2).

58

Using the same procedure of spiking and processing for Ent. faecium, recovery
rates for T1 and T7 were 97% (standard deviation=41%) and 119% (standard
deviation=57%) (Tables 3.3 and 3.4). Unlike E. coli, average Ent. faecium recoveries at
the lowest concentrations for T1 and T7 were below 100%. In fact, two of the replicates
using T1 did not yield detectable CFUs at the low spike. This is possibly due to a lower
spike of Ent. faecium as compared to E. coli or a reflection of the harsher, selective media
conditions of mEI as compared to m—Endo (m-Endo being a more general growth media

for coliforms).

Furthermore, Ent. faecium yielded higher total standard deviation values
than E. coli. At lower concentrations this high variability may have been because
Ent. faecium was added at approximately 3 CFU/ aggregate as compared to E.
coli which was seeded at approximately 8 CFU/ aggregate. This is supported by
recovery rates of Ent. faecium in T1 soil, where the detection limit was reached at
the lowest concentration (Table 3.3). On the other hand, clumping and formation
of chains by Ent. faecium cells may have caused error in CFU counts (J ennison
and Wadsworth, 1940). Clumping is caused by cells attaching to each other
through interactions of the extracellular components of their cell wall (Wildy and
Anderson, 1964, Singh and Vincent, 1987, Hogt et al., 1986). Microscopic
observations (section 2.2.7) support this assumption, indicating that Ent. faecium

did indeed clump more readily than E. coli.

59

Table 3.1. Spiked CFU/ aggregate counts and percentage of E. coli recovery in T1 soil aggregates.

 

 

 

 

 

 

 

 

 

 

 

 

 

Replicate Bacteria Seeded Bacteria Recovered Percent
CFU/ soil CFU/ soil Aggregate Recovery
Aggregatea
9.50 1.33x10l 140%
7.17 1.33x10l 186%
7.67 6.67 87%
KT’ 8.11 138%
S c 1.23 50%
7.33x10l 5.33 x101 73%
7.67 x101 5.33 x101 70%
7.67E x10l 1.07 x102 139%
i— 7.56E x101 94%
S 1.92 39%
7.00x102 6.17x102 88%
6.17x102 6.77x102 110%
7.00x102 5.43x102 78%
x- 6.72x102 92%
S 4.81x101 16%
1.08x103 1.75x103 162%
6.67x103 5.83x103 88%
6.83x103 9.53x103 140%
' {- 4.86x103 130%
S 3.27x103 38%
9.83x104 6.77x104 69%
5.00x104 5.23x10" 105%
5.00x104 4.47x104 89%
5(— 6.61x104 88%
3 2.79x10" 18%
Total x 108%
Total S 37%

 

a. All aggregates were processed immediately at laboratory temperature, using PBW as a diluent. Averages

and standard deviation for each concentration and the recovery are calculated for all aggregates.

bf)? denotes the calculated mean for CFU/ aggregate and recovery percentages.

c. ‘5’ denotes the calculated standard deviation for CFU/ aggregate and recovery percentages.

 

Table 3.2. Spiked CFU/ aggregate counts percentage of E. coli recovery in T7 soil aggregates.

 

 

 

 

 

 

 

 

 

 

 

 

 

Replicate Bacteria Seeded Bacteria Recovered Percent Recovery
CFU/ soil CFU/ soil
Aggregate Aggregate
1 4.40 3.33 76%
2 5.50 3.33 61%
3 9.25 1.50x10l 162%
X— 6.38 100%
2.54 55%
1 5.70x10l 5.67x10l 99%
2 9.25x10l 5.67x10l 61%
3 9.50x10l 4.00x10l 42%
{- 8.15x10l 68%
2.13x10l 29%
1 5.70x102 4.83x102 85%
2 8.09x102 8.67x102 107%
3 7.83x102 8.00x102 102%
4 6.67x102 9.33x102 140%
x— 7.07x102 109%
1.11 x102 23%
1 3.73x103 4.57 x103 123%
2 3.73x103 4.53 x103 122%
3 9.50x103 7.33x103 77%
x— 5.65x103 107%
3.33 x103 26%
1 5.17x104 4.6Ox104 89%
2 6.75x10" 6.20x104 92%
3 9.33x104 6.87x104 74%
4 9.33x104 5.43x104 58%
x— 7.65x10" 78%
S 2.05x104 16%
Total x 92%
Total s 31%

 

61

 

Table 3.3. Spiked CFU/ aggregate counts percentage of Ent. faecium recovery in T1 soil aggregates.

 

 

 

 

 

 

 

 

 

 

 

 

 

Replicate Bacteria Seeded Bacteria Recovered Percent
CFU/ soil CFU/ soil Aggregate Recovery
Aggregate
1 1.00 1.00 100%
2 1.67 ND." -
3 5.67 N.D. -
X— 2.78 -
S 2.52 -
1 2.38x10l 2.00x10l 84%
2 3.03x10l 4.33x10l 143%
3 4.87x101 4.33x10l 89%
x— 3.43x101 105%
S 1.29x101 33%
1 2.70x102 1.98x102 73%
2 4.25x102 4.00x102 94%
3 3.27x102 3.63x102 111%
f 3.41x102 93%
S 7.84 x101 19%
1 2.67x103 2.83x103 106%
2 3.15x103 5.23x103 166%
3 3.4le03 3.88x103 114%
x— 3.08x103 129%
3 3.80x102 33%
1 2.58x104 3.67x104 142%
2 3.47x104 4.00x104 115%
3 3.85x104 4.50x104 117%
i- 3.30x104 125%
5 6.50 x103 15%
Total x 97%
Total 8 41%

 

* N.D. indicates that bacterial concentrations could not be detected.

62

 

Table 3.4. Spiked CFU/ aggregate counts percentage of Ent. faecium recovery in T7 aggregates.

 

 

 

 

 

 

 

 

 

 

 

 

 

Replicate Bacteria Seeded Bacteria Recovered Percent
CFU/ soil CFU/ soil Aggregate Recovery
Aggregate
1 1.67 1.00 60%
2 1.00 1.00 100%
_ 3 5.67 2.00 35%
X 2.78 65%
S 2.52 33%
1 2.38Ex101 4.67x10l 196%
2 3.03x10l 5.67x10l 187%
3 4.87x10l 3.00x10l 62%
i— 3.43x101 148%
S 1.29x10l 75%
1 2.70x102 3.55x102 131%
2 4.25x102 3.93x102 93%
3 3.27x102 4.67x102 143%
X— 3.41x102 122%
S 7.84x101 26%
1 2.67x103 4.90x103 184%
2 3.15x103 4.35x103 138%
3 3.42x103 4.05x103 119%
x— 3.08x103 147%
S 3.80 x102 33%
1 2.58x10" 4.33x10" 168%
2 5.00x10" 3.4711104 69%
3 3.85x10" 3.67x104 95%
{— 3.81x104 111%
S 1.21x104 51%
Total x 119%
Total 3 57%

 

63

 

At lower concentrations both bacterial species recovered from T1 and T7 treated
aggregates exhibited higher standard deviations (Fig. 3.3 and 3.4). This trend may be
explained from a methodological perspective. It has been established that in
microbiological plate counts, decreasing bacterial concentrations result in higher counting
error (Breed and Dotterrer, 1916). Additionally, at lower concentrations there was more
sediment present on the media. Olsen and Bakken have observed a similar decrease of
CFU/g counts as amounts of soil per plate increased (1987). From experimental
observations, the sediment altered the typical spherical morphology to an irregular shape
that, when colonies were in close proximity made it difficult to discern delineations
between them. It also may be possible that the high density of soil on the membrane filter

inhibited bacteria nutrient acquisition and growth.

 

 

200%
180%
160%
140%
120%
100%
80%
60%
40%
20%
0%

 

 

 

 

 

 

Percent E. coli Recovery

 

 

1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04

Bacterial Concentrations CFU/ soil aggregate

 

 

 

Fig 3.3. The averaged percent recoveries and standard deviations of E. coli in T1 (n=3) and T7
(n=3; n=4 at 1x10 and 1x104 concentrations).

64

 

 

250%

 

 

50%

>3

3

e 200%

U

Q

a:

E 150%

5

8

5% 100% eTl
E IT7
:

0

ii

fl-t

 

 

O
59.

1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04

Bacterial Concentration CFU/ soil aggregate

 

 

 

Fig 3.4. The averaged percent recoveries and standard deviations of Ent. faecium in T1 (n=3) and

T7 (n=3) soil treatment.

3.2.2 Desiccation and E. coli recovery in whole aggregates. After 30 minutes of air-
drying, high E. coli recovery persisted while the moisture content slightly decreased as
compared to non-air dried aggregates. The recovery rates after 30 minutes were not
significantly different for experiments processed with VMF method with settling and
without the settling steps (Table 3.5). After 60 minutes of air-drying, however, very
different recovery rates could be discerned. While the VMF method with settling yielded
no E. coli recovery, the VMF method without the settling step yielded an average
recovery of 49%. The moisture content was compared to the recovery rates, and a steep
decline of E. coli recovery could be ascertained from air-dried aggregates processed with
the VMF method with settling compared to the aggregates that were processed without

the settling step (Fig. 3.5 and 3.6).

65

Table 3.5. Comparison of the VMF method with the settling step and without the settlifl step.

 

 

 

 

 

 

 

Effect of Desiccation and Attachment on E. coli
Air-dying Time Recovery
30 minutes Moisture Content 15% (n=4)
Recovery with Settling Step 103% (n=2)
Recovery without Settling Step 89% (n=2)
Moisture Content 1.5 % (n=4)
Recovery with Settling Step 0% (n=2)
60 minutes
Recovery without Settling Step 49% (n=2)

 

 

 

 

 

* 50111 E. coli was spiked in all aggregates at a concentration of approximately 104 CFU/ aggregate. All
aggregates used for these experiments were T7.

 

 

120.00%
100.00% 4 ’ —+
80.00%
60.00%
40.00%
20.00%

0.00% H 1

0.

 

 

 

 

 

 

T I I l 7

5.00% 10.00% 15.00% 20.00% 25.00%

Percent Recovery CFU/ml

  

 

 

Moisture Content

 

Fig 3.5. Recovery of spiked E. coli as a function soil moisture content when processing the whole
aggregate using the VMF method with a settling step (n=13).

*The circled area highlights the steep decline of E. coli recovery at the lower moisture content while using
the VMF method with a settling step.

66

 

“Tm—”7

 

120.00%
100.00% .
80.00% +—
60.00% 41
40.00%
20.00%

0.00% l I I I I
0.00% 5.00% 10.00% 15.00% 20.00% 25.00%

 

 

 

 

 

 

Percent Recovery CFU/ml

Moisture Content

 

 

 

Fig 3.6. Recovery of spiked E. coli as a function soil moisture content when processing the whole
aggregate using the VMF method without settling step (n=8).

This converges with observed adsorption of native bacterial in air-dry soil
aggregates stored at room temperature which caused the underestimation of the actual
viable bacterial counts (see section 3.1.3). It may be that after extended air-drying
(desiccation stress) and extended contact with the soil, the spiked E. coli adsorbed to the
soil, thus making it difficult to extract them, even with vigorous shaking (Huysman and
Vestraete, 1993a). To avoid underestimation in further experiments, the settling step was

excluded from the VMF aggregate processing.

The experimental evaluations of native heterotrophic bacteria (section 3.1) and
spiking experiments with E. coli and E. faecium (section 3.2), indicated that the bacterial
extraction method developed did provide high and consistent recoveries of bacteria from
aggregates. The effects of desiccation, clumping and bacterial adhesion to soil particles
were addressed in order to explain the inherent methodological errors that may occur
while employing this technique. Other factors such as native bacterial resuscitation

67

and the effect of CaClz on E. coli viability have also been addressed (sections 2.2.5 and

2.2.6) and were shown not to have an impact on the method using dried soil aggregates.

3.3. Flow Chamber Experiments

Flow experiments were designed to examine retention and transport of E. coli and
Ent. faecium within T1, T2 and T7 aggregates. The investigation was conducted using
air-dry (unsaturated) aggregates and saturated aggregates to observe if moisture affected
the transport of either bacterial species. Transport of bacteria was evidenced by the
enumeration of bacterial species in the effluent. Conversely, retention was illustrated by
determining the bacterial concentrations remaining in the soil aggregate. Bacterial
concentrations in effluent and bacterial concentrations retained in the aggregate were

computed as log of total CFU to be able to conduct comparisons between samples.

The most apparent influence on bacterial retention and transport appeared to be
due to soil saturation. At air-dry conditions, retention was high for all soil treatments and
bacteria types (Figure 3.7). T1 had the highest bacterial retention capacity and no E. coli
or Ent. faecium were detected in the effluent. However, the difference in retention
between soil treatments was not shown to be statistically different (p<0.05). T2 and T7
aggregates did not retain E. coli and E. faecium as readily as T1, however the bacterial
concentrations in soil aggregates were approximately 3-log higher than the concentrations
in the effluent (Table 3.6). It is thought that at dry soil conditions, preferential flow
occurs within smaller pores carrying the bacteria with the solution, causing them to be
filtered and thus, more readily retained in the soil (Hattori, 1988, Stevik et al., 2004).
Furthermore, at lower soil moisture content the bacteria may adhere to the soil particles

due to reduced water-microbe interaction and thus increased contact with the solid soil

68

substratum (Huysman and Vestraete, 1993, Guber et al., 2009). These hypotheses are
supported when comparing bacterial effluents for all soil treatments using saturated
aggregates (Figure 3.8). For example, T1 treatment exhibited the complete opposite effect
when compared to air-dry conditions. Effluent concentrations for E. coli and Ent. faecium
were 2.7 and 1.6 log CFU/aggregate, respectively. Therefore, at higher moisture content,
bacteria may have had less contact with the solid—phase soil and transported more readily

when the soil was saturated.

69

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

E. coli concentration at air-dry condition

7

6
a 5
Q iiiEffluent

4 -
E lAggregate
a? 3 -
N)
3 2 -

1 -

0 .1

T1 T2 T7
Soil Treatment
E. faecium concentration at air-dry
condition

7

6
E 5
U faEffluent

4
E lAggregate
9°. 3 -
DD
.3 2 ~

1 ..

0 —1

T1 T2 T7
Soil Treatment

 

Fig 3.7. Recovery of E. coli and Ent. faecium from T1, T2, T7 aggregates and effluents at air—dry
conditions (n=5).

70

 

Log Total CFU
O .... N U) 45 LII O\ \l

E. coli concentration at saturated
condition

 

 

l Effluent

I Aggregate

     

T1 T2 T7
Soil 'IYeatment

 

 

LogTotalCFU
O H N DJ A U1 0\ \l

E. faecium concentration at saturated
condition

 

 

 

 

a Effluent
I Aggregate

     

T 1 T2 T7
Soil Treatment

 

 

Fig 3.8. Recovery of E. coli and Ent. faecium from T1, T2, and T7 aggregates and effluents at saturated
conditions (n=5).

The average effluent concentrations indicate that soil treatment was also another

variable that influenced the retention and transport of bacteria. While at air-dry

conditions there was no detection of bacterial concentrations in the T1 effluent, in T2 and

71

T7 soils, the concentrations in the effluents for E. coli were 0.9 and 0.8 log CFU/
aggregate (Table 3.6). On the other hand, under saturated aggregate conditions for T1
soil, higher E. coli effluent concentrations were detected compared to T2 and T7 (Table
3.7). It is likely that the difference in soil macroporosity may have caused the observed
variation in bacterial transport (Unc and Goss, 2003). When macroporosity of soil is
higher, the bacteria are filtered less and observed to leach out of soil in higher
concentratiOns then when soil is saturated (Abu-Ashour et al., 1998). This is supported by
X-ray microtomography analysis of the soil’s macropores that indicated T1 had
significantly more macropores than T7 (Wang et al., 2010). However, it should be noted
that when examining the individual effluent replicate results, T1 had considerable
variation in effluent concentrations for both bacteria types (Tables 3.8 and 3.9, first
columns). While T7 was also observed to have high variability, this was due to one
replicate which was considerably different than the rest (Tables 3.8 and 3.9, fifth
columns). This possibly indicates that the aggregate. structure for T7 is more uniform than

Tl’s aggregate structure.

Table 3.6. Concentration of E. coli and Ent. faecium in the influent and T1, T2 and T7 aggregates and
effluents (n=5).
E. coli Recoveries in Air-dry Soil

 

 

 

 

 

Sample \ Soil Treatment T1 T2 T7
Aggregatea 3.9 3.9 3.9
Effluent NDb 0.9 0.8
Influent 3.6 A 3.6 3.6

Ent. faecium Recoveries in Air-dry Soil
Sample\Soil Treatment T1 T2 T7
Aggregate 3.4 3.4 3.5
Effluent N.D. 0.5 0.7
Influent 3.1 3.1 3.1

 

a. Aggregate, effluent, and influent concentrations were calculated as log total CFU.
b. N.D. indicates that bacterial concentrations could not be detected.

72

 

Table 3.7. Concentration of E. coli and Ent. faecium in the influent and T1 , T2 and T7 aggregates and

effluents (n=5).

 

E. coli Recoveries in Saturated Soil

 

 

 

 

 

Sample\Soil Treatment T1 T2 T7
Aggregate 3.3 3.4 3.5
Effluent 2.7 1.8 1.1
Influent 3.6 3.6 3.6
Ent. faecium Recoveries in Saturated Soil

Sample\Soil Treatment T1 T2 T7
Aggregate 3.0 3.1 3.2
Effluent 1.6 0.4 0.5
Influent 3.1 3.1 3.1

 

Table 3.8. The percent recovery of E. coli in effluents for T1, T2 and T7 aggregates replicates at saturated and non-
saturated conditions (n=5).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T1 T2 T7
Replicate Saturated Non- Saturated N on- Saturated Non-
Saturated Saturated Saturated

1 23% N.D* ND ND ND N .D

2 41% ND ND ND N .D N .D

3 10% ND 4% N .D ND ND

4 2% N .D 16% ND 2% 1%

5 50% N .D 16% 1% 26% ND

x8 25% - 7.2% 0.2% 5.6% 0.2%

s 20% - 8.2% 0.5% 11% 0.5%

 

 

*N.D. indicates that there were no bacteria detected. N .D. values were treated as zeros to calculate
averages (xi) and standard deviations (5).

Table 3.9. The percent recovery of Ent. faecium in effluents for T1, T2 and T7 aggregates at saturated and non-
saturated conditions (n=5).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T1 T2 T7
Replicate Saturated Non- Saturated N on- Saturated N on-
Saturated Saturated Saturated

l 1% ND ND ND N .D 1%

2 47% ND ND ND ND 1%

3 1% ND N .D N .D N .D N .D

4 0% N .D N.D N .D ND ND

5 41 % ND 21% 1% 28% ND

xii 18% - 4.2% 0.2% 5.6% 0.4%

s 24% - 9.4% 0.5% 13% 0.5%

 

 

73

 

 

Three-way ANOVA - using the variables soil saturation, soil treatment, and
bacterial species — indicated that there was a two-way interaction between soil saturation
and soil treatment but not bacterial species. The response variable used was the ratio of
bacterial effluent concentration to total influent concentration to correct for the difference
between E. coli and Ent. faecium influent concentrations for the various experiments.
Grouping by soil saturation indicated the best goodness of fit criteria (using the Akaike
and Byesian criteria). Statistical significance was only observed in the T1 treatment
between the saturated and dry aggregate (Figure 3.9). This conforms to the graphically
illustrated results indicating that higher aggregate saturation and tillage increased the

average bacterial effluents.

 

 

 

 

 

 

 

Bacterial Effluent Concentrations in Dry and
Saturated Aggregates
0.20 l\ \
: b \\
\
\‘\
Effluent 0.15
to 2 \\\
Influent 0'10 i \.\
« \
Ratio ‘1 \ a a
0.05 ? \—— ,
l' a a a
0.00 f ,
T1 T2 T7
HSoil Treatment
D3} Saturated

 

 

Fig 3.9. Statistical analysis of the interaction between soil treatment and soil saturation combining both
bacterial species. Different letters in the effluent to influent ratio at different soil saturation indicate that
they are significantly different (p<0.05).

74

3.4 Bacterial Spatial Distribution
3.4.1 Native bacterial concentrations in soil aggregate interior and exterior layers.
Differences between native bacterial counts from the exterior and interior soil layers of
T1, T2 and T7 were minute. T1, T2 and T7 contained an average of 1.02x106, 3.78x105
and 9.45x105 CFU/ g in the exterior layer, while the interior soil layers contained
8.28x105, 3.26x105 and 8.43x10S CFU/g respectively. Although it is generally accepted
that the interior region of the soil harbors more bacteria than the exterior, our
experimental results indicated otherwise (Hattori and Hattori, 1976, Ranjard and
Richaume, 2001) (Figure 3.10). It could be that the after such an extended storage period,
moisture content even in the interior of the soil was too low to sustain the larger
concentration of viable bacteria (Stevik et al., 2004). It also possible that this is the actual
representation of natural bacterial distribution of aggregates collected. The soil used for
these experiments was sieved after sampling to separate different sized macroaggregates,
therefore, we may have processed the macroaggregate fractions that naturally contained a
homogenous distribution of native bacteria. The statistical difference of HPC in between
soil treatments could not be determined because only two replicates of each layer were

analyzed.

When comparing the concentrations of extracted native bacteria from whole T1
and T7 aggregates (see section 3.1.1) to the total native bacterial concentrations of the
inner and exterior concentric layers of same soil treatments, the separated aggregates
yield approximately l-log higher concentration. The process of separating the aggregate

is thought to be behind the observed increase where bacterial extraction improves

75

 

 

1 .20E+06

 

1 .00E+06

 

8.00E+05 -

 

6.00E+05

CFU/g

I Inner

 

4.00E+05
Outer

2.00E+05

l

 

 

0.00E+00

l

T 1 T2 T7
Soil Treatment

 

 

 

Fig 3.10. Heterotrophic bacterial concentrations in interior and exterior layers of soil treatments T1, T2 and
T7 (n=2).

homogeneous distribution of bacteria that were clumped in the aggregate (Richaume et
al., 1993). This may also be due to bacterial adsorption and particle interference
(discussed in section 3.1.2) that was overcome after the physical disruption of the soil

aggregate by separating it into layers.

3.4.2 Desiccation and E. coli recovery in aggregates subsections. To examine the
drying affect on spatially distributed bacteria in the aggregate, aggregates were separated
into three subsections in relation to where the E. coli was seeded (the SAE chamber was
not used for these experiments because the aggregates were moist after spiking and thus it
would be difficult to erode the aggregate accurately). The subsections were then allowed
to air-dry for 0, 10, 40 and 60 minutes. Aggregate subsections after desiccation showed a
decline in E. coli recovery (Table 3.10). It is noteworthy to point out that at all time
intervals, higher recoveries were always observed within the middle subsections. This is

especially true at 40 minutes of air drying, where the middle section showed a recovery

76

of 11.5% as compared to the 0.7% and 4.3% recoveries in the top and bottom
subsections. The observed results coincide with what might occur in nature, as bacteria
have been shown preferentially relocate to the center of soil aggregates for protection
(Ranjard and Richaume, 2001). At 60 minutes of air-drying, however, there was not
much difference between recovery yields in all three subsections due desiccation stress at

the lower moisture content.

Total E. coli recoveries from the aggregate subsections after 60 minutes of air-
drying were lower (7.9%) as compared to E. coli recoveries from whole aggregates
(49%) (Tables 3.5 and 3.8). While the aggregates were stable when wet (at times 0 and
10 minutes), once dried the aggregates were flaking (times 40 and 60 rrrinutes) during the
slicing procedure (flaking from the subsections). These pieces may have had E. coli cells
adsorbed to them, leading to the underestimations of actual E. coli recovery. To ensure
there was no underestimation due to flaking and to avoid desiccation stress due to low
moisture content, sliced aggregates were then processed immediately (results described in

below in section 3.4.3)

Table 3.10. Averaged recovery of E. coli from sliced soil aggregate subsections (n=2).

 

 

 

Time (minutes) Subsection Total Recovery
Top Middle Bottom
0 23.4% 30.2% 23.7% 77.3%
10 17.2% 29.5% 17.8% 64.5%
40 0.7% 11.5% 4.3% 16.5%
60 2.7% 3.2% 2.0% 7.9%

 

3.4.3 Slicing and saturation experiments. These experiments were designed to explore

the effect of soil saturation and soil treatment on bacterial distribution within the

77

 

aggregate. Aggregates that were air-dry and pre-saturated to 15% and 30% moisture
content were seeded with E. coli and then sliced into seven subsections to examine
bacterial translocation. Subsections 1, 2, 3, 4, and 5 corresponded to the top, right, left,
back and front of the aggregate respectively and the middle section was divided into two
subsections 6 and 7 which corresponded to the center-middle and the bottom-middle
respectively (see section 2.5). Calculation of E. coli concentrations were based on log

CFU/ g.

At air-dry condition (i.e. 0% moisture content), T 1 had high variability in E. coli
concentration in five of the seven slices indicating a non-uniform distribution of E. coli in
the three replicates (Figure 3.11A). T2 also showed variability at 0% moisture content,
but variability was confined to three subsections (Figure 3.12A). T7 showed the most
even distribution of E. coli as compared to T1 and T2, by exhibiting the least variability
in aggregate subsections at 0% moisture content (Figure 3.13A). These observations can
be explained by the variability in E. coli recovery replicates per subsection. While E. coli
could not be detected in at least one replicate per slice, T2 had only two replicates at
which E. coli could not be detected. E. coli in T7 subsections on the other hand, could be

detected in all replicates (Figures 3.13A, 3.13B, and 3.13 C).

When the aggregates were saturated, the dynamics of E. coli spatial distribution
were different in T1 and T2 treatments. At moisture contents of 15% only one slice in T2
treatment exhibited high variability while across all treatments at 30% moisture content
very minute variability was observed (Figures 3.11B, 3.11C, 3.12B and 3.12C). It’s
apparent that at increasing moisture content, E. coli spreads more evenly within the
aggregates T1.

78

 

7.00
6.00
5.00
4.00
3.00
2.00
1.00
0.00

Log CFU/g

 

1 2 3 4 5 6 7
Aggregate Subsection
7.00 B
6.00
5.00
4.00 ~
3.00 —
2.00 -
1.00 ~

 

 

 

 

Log CFU/g

 

 

1 2 3 4 5 6 7
Aggregate Subsection
7.00 C
6.00
5.00
4.00 ~
3.00 ~
2.00 -
1.00 ~
0.00 -

 

 

 

 

Log CFU/g

 

 

1 2 3 4 5 6 7

 

 

Aggregate Subsection
Fig 3.11. E. coli concentrations in T1 aggregate slices at A) 0% moisture content, B) 15% moisture content
and C) 30% moisture content (n=3).

 

79

 

 

7.00 A
6.00
5.00
4.00
3.00
2.00 -
1.00 -
0.00 ~

 

 

 

 

 

Log CFU/g

 

 

1 2 3 4 5 6 7

Aggregate Subsection
7.00 B
6.00

5.00
4.00 . - l t .

 

 

 

 

 

3.00
2.00
1.00
0.00

Log CFU/g

1 2 3 4 5 6 7
Aggregate Subsection
7.00 C
6.00
5.00
4.00 -
3.00 -
2.00 -
1.00 -
0.00 —

 

 

 

 

 

 

Log CFU/g

 

 

1 2 3 4 5 6 7

 

 

Aggregate Subsection

Fig 3.12. E. coli concentrations in T2 aggregate slices at A) 0% moisture content, B) 15% moisture content
and C) 30% moisture content (n=3).

80

 

 

7.00 A
6.00 I

5.00 —
4.00 -

 

 

CFU/g

Lo

2.00 -
1.00 -
0.00 -

 

 

1 2 3 4 5 6 7
Aggregate Subsection
7.00 B
6.00
5.00
4.00 -

 

 

 

 

CFU/g

Lo

2.00 -
1.00 -
0.00 -

 

 

1 2 3 4 5 6 7
Aggregate Subsection
7.00 C
6.00
5.00 T " "
4.00 -
3.00 -
2.00 -
1.00 ~
0.00 4

 

 

 

 

Log CFU/g

 

 

1 2 3 4 5 6 7

 

Aggregate Subsection
Fig 3.13. E. coli concentrations in T7 aggregate slices at A) 0% moisture content, B) 15% moisture content
and C) 30% moisture content (n=3).

 

 

81

To understand the difference in E. coli distribution, both the effect of the soil
aggregate treatment and moisture content should be assessed. It is of interest that E. coli
distribution in T7 treatment did not seem to change between any of the moisture contents
and was evenly distributed among the slices. This is in contrast to T1 soil which exhibited
entirely opposite distribution when comparing air-dry aggregates and aggregates with
increasing moisture content. The only known difference between treatments is tillage and
fertilization. Tillage could have altered the soil’s pores structure as reflected in its higher
bulk density (i.e. lower porosity) of T1 in comparison to T7 (Table 2.3). As a
consequence it is possible that that the pores that conduct flow are more uniform in T7
aggregate, therefore allowing bacteria to spread evenly throughout the aggregate when it
was air-dry. Results from section 3.4, where bacterial effluent concentrations in T1 were
more variable than T7, similarly suggest that T7 had more uniform pore characteristics
(Tables 3.8 and 3.9).

T2 on the other hand seemed to lie in between the extremes of E. coli distribution
at 0% moisture content. T2 aggregates did not receive any tillage, but did receive N -
fertilization and had agricultural com-wheat and soybean crop growth. The addition of
fertilizer and growth of non-native agricultural plants are known to alter the soil organic
matter (Grandy and Robertson, 2007, Liebig, 2002). This alteration can impact the soil’s
structural stability and therefore alter its preferential transport pathways (Blazier et al.,
2008). From these observations, it seems that the impact of fertilization has less of an
effect as opposed to tillage on bacterial distribution and thus the aggregate pore

continuity.

82

Statistical analyses confirmed the graphically illustrated results. Three-way
ANOVA using the variables of soil saturation, soil treatment and subsection number was
used. The response variable in the experiment was the natural log of CFU/ g. Levene’s
test for unequal variance was significant indicating unequal variances in all variables.
Variances were highest in the aggregate saturation and subsection variables. Grouping by
aggregate saturation indicated the best goodness of fit criteria (using the Akaike and
Byesian criteria). Interestingly, by grouping the data using the aggregate saturation, the
0% moisture content indicated the highest variability estimate (6.08) as compared to T2
(1.81) and T7 (1.14). Next, a repeated measure was run to reduce individual differences
between aggregate subsections using a compound symmetry model structure. The

ANOVA was found to have a three-way interaction between all variables.

Using two combinations of two-way interactions allowed the investigation of the
influencing factors in E. coli recoveries as affected by the subsection location. The first
two-way interaction examined was between the soil saturation and the subsection
number. Similar to Figure 3.11, the statistical analysis showed that three subsections, 3, 6
and 7 (i.e. the left, center-middle and the bottom-middle section) exhibited statistical
significance in between aggregate saturations of 0, 15 and 30% in T1 soils (Figure 3.14).
Only two subsections in T2 exhibited statistical differences and none in T7 soil (Figures
3.15 and 3.16). It is noteworthy, that at all moisture contents, the center-bottom

subsection (7) in T7 exhibited the least concentration of E. coli.

83

 

E. coli Recovery in T1 Aggregate Subsections

 

 

 

 

 

Aggregate Subsection
m 0 M015 on 30

 

 

 

Fig. 3.14. Statistical comparison of E. coli recoveries at 0, 15, and 30% moisture content in T1 aggregate
subsections. Moisture contents at specific subsections with different alphabets indicate that E. coli
concentrations are significantly different.

The second two-way interaction examined was between the soil treatment and the
subsection number. At the 0% moisture content, the greatest numbers of subsections with
statistical differences between all soil treatments were observed (Figure 3.17). It’s
noteworthy to point out the statistical difference in subsection 6 (center-middle) between
T1 and T2 as well as T1 and T7. At the 15% moisture content, there were only two
subsections exhibiting statistical significant results (Figure 3.18). Finally, at the 30%
moisture content, there was only one subsection (7; the center-bottom) that exhibited

statistical difference between T1 and T2 as well as T1 and T7 aggregates (3.19).

84

 

E. coli Recovery in T2 Aggregate Subsections

 

 

 

 

 

1 2 3 4 5 6 7
Aggregate Subsection
m 0 r” 15 000 30

Fig. 3.15. Statistical comparison of E. coli recoveries at 0, 15, and 30% moisture content in T2 subsections.

 

 

 

E. coli Recovery in T7 Aggregate Subsections

 

Log
CFU/ g

 

 

 

6.0

 

.. - .t1w—v—r t—rr'r‘lT

llIl t l 'll’Il It I

1 2 3 4 5 6 7
Aggregate Subsection

Fig. 3.16. Statistical comparison of E. coli recoveries at 0, 15, and 30% moisture content in T7 subsections.

 

 

85

 

 

 

E. coli Recovery in Aggregate Subsections at
0 % Moisture Content

 

12.0;
Imam-c /\ ‘b\
100; ~.
9.0;
8.05
7.0%
6.0%
5.01
4.0?
3.0%
2.03

 

 

 

 

1 2 3 4 5 6 7

Aggregate Subsection
H-OTl MT2 MT7

 

 

 

Fig. 3.17. Comparison of E. coli recoveries in T1 , T2, and T7 aggregates subsections at 0% moisture
content. Moisture contents at specific subsections with different alphabets indicate that E. coli
concentrations are significantly different.

 

E. coli Recovery in Aggregate Subsections at
15% Moisture Content

 

11.0:

A.
10.0: // \4
9.0:

Log 8.0;
CFU/g 7.0j
6.0
5.0:
4.0:
30$
20‘

1 2 3 4 5 6 7
Aggregate Subsection

 

 

 

 

 

 

 

Fig. 3.18. Comparison of E. coli recoveries in T1, T2, and T7 aggregates subsections at 15% moisture
content.

86

 

E. coli Recovery in Aggregate Subsections at
30% Moisture Content

 

12.0:

 

 

 

 

1 2 3 4 5 6 7
Aggregate Subsection
"-0 T 1 m T2 mT’]

Fig. 3.19. Comparison of E. coli recoveries in T1, T2, and T7aggregates subsections at 30% moisture
content.

 

 

Statistical analyses indicated that all three variables (soil treatment, moisture
content and subsection location) had an effect on the bacterial spatial distribution of E.
coli in the aggregates. It was observed that T1 had the highest variability between
subsections and most often exhibited significant differences when compared to other soil
treatments at 0% moisture content (Figure 3.14). It also appeared that when the moisture
content is high, the soil treatment does not seem to impact E. coli distribution as much.
At 0% moisture content there were three subsections that exhibited significant differences
between the soil treatments as compared to only one subsection at 30% moisture content.
This is probably because as more water was added to the aggregates at increasing

moisture contents, more pores were filled creating a continuous pathway for E. coli

87

 

movement (Hillel, 1998). E. coli’s flagellar motility could have utilized the water

continuity to disperse within the aggregate (Soby and Bergman, 1983).

Of interest was the impact the aggregate subsection had across all treatments. As
mentioned earlier, T1 had the least E. coli concentration in subsection 6 across all
moisture contents (Figure 3.14) and this subsection had significantly lower E. coli
concentrations in T1 compared to T2 or T7 (Figure 3.17). T2 and T7 did not exhibit any
significant differences in subsection 6 at any moisture content. This is of importance
because subsection 6 is the center-middle, possibly indicating inaccessibility of that area
in T1 soil. Also interesting is that subsection 7 (center-bottom) in the T7 aggregates
always had less E. coli concentrations (Figures 3.16 and 3.19). The method of E. coli
addition to the aggregates could have influenced this. E. coli was added to the top
subsection, therefore it may not have reached the bottom subsection before slicing the
aggregate.

It is important to note that the second replicate for T1 aggregate at 0% moisture
content had no recovery in five of the seven subsections and very low bacterial
concentrations in the other two subsections (See Appendix, Table A6). This may indicate
that the E. coli died-off during the slicing process whereby the aggregates may have been
left to air—dry for an extended period of time. This puts into question the results of the T1
aggregate at 0% moisture content, because the statistical analysis may have been skewed
due to the lower concentration in the second replicate. Nonetheless, the slicing and
hydration experiments will be performed again in future work and statistical analysis will

be subsequently conducted to compare with these results.

88

3.5 Discussion

3.5.1 Bacterial extraction method from soil. Over the last 60 years, many approaches to
extract and detect bacteria in soil have been developed, each with their respective
advantages and limitations (Ranjard and Richaume, 2001). Studies that have addressed
bacterial extraction have usually utilized a variation of an agitation (i.e. vortexing or
blending), sonication or fractionation method (Hattori, 1988, Mahler et al., 2000,
Holdaway, 2003, Singh, 2007, Boehm et al., 2009). To detect bacteria, culturing methods
using agar have been the most prevalent (Fontes et al., 1991, Gannon et al., 1991, Mahler
et al., 2000, Guber et al., 2005, Bolster et a1. 2006). However, other studies have used
most probable numbers (MPN) liquid based assays, immunogenic assays, microscopic
evaluations and genetic analyses (Bakken and Olsen, 1987, Richaume et al., 1993,

Mummey et al., 2006, Zimmerman et al., 2009).

In this study, the vortexing extraction and membrane filtration culture method
seemed to underestimate native bacterial population but proved to be suitable for
detecting spiked bacteria. Underestimation was most likely due-strong adhesion of native
bacteria to the soil particles because of the very dry soil conditions. The vortexing
agitation was not enough to detach the bacteria from the soil particles. But in
combination with peeling or fractioning the aggregate, more native bacteria appeared to
become detached from soil. Thus it is recommended to keep soils moist after sampling to
avoid bacterial desiccation and strong adhesion to soil particles. It also may be useful to

employ a combination of methods to increase the accuracy of native bacterial extraction.

89

Culture media and growth conditions were most likely limiting factors in
detecting native bacteria as well. Many species of native soil bacteria have slower growth
rates, so the 24 hour incubation period may have not been enough for most of the bacteria
to grow (Rozak and Colwell, 1989). Therefore, a longer incubation period on low
nutrient media has been recommended for enumerating native soil bacteria (Olsen and

Bakken, 1987, Davis et al., 2005).

The very high recovery rates (section 3.2.1) of E. coli and Ent. faecium indicate
that the vortexing and membrane filtration method was optimal for aggregate spiked
bacteria. This high recovery was evident even at low bacterial concentrations. However,
similar to native bacteria, spiked bacteria may succumb to desiccation stress, so it is
important to ensure that the soil does not dry out. It is recommended to process the spiked

aggregate within 2 hours to avoid bacterial growth or die-off.

In regards to detection of spiked bacteria, the method had its limitations. Ent.
faecium seemed to clump and attach to soil particles more readily than E. coli, causing
high variability and underestimation of bacterial influent concentrations. A possible
improvement to this method would be to introduce a series of washing and centrifugation
steps when the adding stock dilutions of bacteria to soil aggregates (Bolster et al., 2006).
In preparing the stock one could minimize clumping by removing nutrients in the stock
culture. It also may be beneficial to add detergents such as Tween 80 to the washed stock
culture and/or during the extraction with vigorous vortexing to detach the bacteria and to

decrease the appearance of clumps (McConville et al., 1974).

90

3.5.2 Bacterial retention and transport in soil. When bacteria are applied to soil they
may be retained by adsorbing to soil surfaces or become trapped in soil micropores
(Stevik et al, 2004). Our experiments illustrate that this high capacity to retain bacteria
can be discerned even at the macroaggregate scale. Perhaps the most important factor in
bacterial transport was the soil moisture content. This is of interest because after rainfall,
soil is saturated and thus pathogenic bacteria in soil may runoff into adjacent water
bodies or be carried into ground water, posing a public health risk (Unc and Goss, 2004,
Muirhead et al., 2006).

When the soil was dry, both indicator bacteria were found to be retained in
aggregates in high concentrations even after flushing with solution (Figure 3.7). It is
thought that when the soil is at low moisture content, capacity of bacteria to adsorb to soil
increases (J amieson et al., 2002). This is supported by our experiments that have
indicated that native bacteria and spiked bacteria can strongly adhere to soil particles
when soil is dry (sections 3.1.2, 3.2.2 and 3.4.2). Furthermore, Guber et al. have shown
that fecal coliforms adhered to dry soil 2.5 times more readily than water saturated
aggregates (2009). It is also possible that the addition of CaClz enhanced bacterial
retentions in soil aggregates. Ca2+ ions in the CaClz solution may form ionic bridges
between the bacteria and the soil, thereby increasing attachment (Stevik et al., 2004).
Although the micropores (sizes <2um) in our soil aggregates have not been characterized,
filtration could have also increased retention. Hattori has shown that when soil is dry,
bacteria is passively carried with the solution in small pores due to capillary force (1988).
There they could get stuck and may not be able to exit the soil aggregate even when

flushed (Powelson and Gerba, 1995).

91

Soil structure seems to be important in bacteria retention, however, it is not well
understood. Some studies indicate that physical re-structuring of soil can retard bacterial
movement from soil surface to groundwater, thereby suggesting tillage as a suitable
agricultural management practice (Abu-Ashour et al., 1998, McMurry et al., 1998).
However, other studies have indicated tillage can have a variable effect on bacterial
transport into groundwater and may even enhance runoff (Stoddard et al., 1998, Gagliardi
and Karns, 2000, Jenkins et al., 2008). Results from the flow chamber experiments
described in Chapter 3, corresponded to the latter studies. Although the average leached
bacterial concentration for T1 (tilled) aggregates was higher as compared to non-tilled
aggregates, the effluent concentrations were highly variable (Figure 3.8). This
inconsistency most likely suggests that the tillage causes the aggregates to have non-
uniform structures, making measurement of bacterial transport even at the
macroaggregate scale more uncertain.

The type of bacterial species seemed to contribute less to the retention than soil
moisture or soil management practice. This was not anticipated because Enterococci are
known to exhibit higher retention in soil when compared to E. coli (Mahler et al., 2000).
This is because they can adsorb to clay particles and clump together, making extracting
them from soil more difficult (Stenstrom, 1989, Huysman and Vestraete, 1993).
Microscopic evaluations (section 2.2.7) support this, because unlike E. coli, Ent. faecium
were observed to clump more readily and form chains of three or more cells as it grows.
Therefore, it is tempting to conclude that Ent. faecium was retained more readily than E.
coli as evidenced by higher recovery of the latter in effluent of T1 treatment at saturated

conditions (Table 3.7). However, these results are not statistically significant. Therefore,

92

at the aggregate-scale, the different bacterial species, cell properties and motility did not
have a substantial affect on retention. At larger scales, such as in field conditions, this
affect could be magnified due to increased contact with soil increasing chances of
filtration and/or adsorption. More work is needed to address the observed trends to
determine significance.

Finally, more than one flushing regime could be done to simulate multiple rainfall
events. Other experiments have illustrated that the bulk of bacteria concentrations are
drained from the soil after more solution has passed through the soil column (Fontes et
al., 1991, Foppen et al., 2005). Therefore, it is recommended to conduct aggregate-scale
flow experiments with multiple or continuous flushes to better understand the transport of
bacteria under extensive or continuous rainfall.

3.5.3 Spatial distribution of E. coli in soil. The importance of understanding spatial
distribution of spiked bacteria may help us ascertain the movement of pathogens loaded
in soil after addition of manure slurry (Unc and Goss, 2004). Pathogens carried in liquid
manure can percolate and distribute through the soil (Cools et al., 2001). To our
knowledge, there have been no studies investigating spiked bacterial spatial distribution
by subsectioning soil aggregates.

Similar to bacterial retention and transport, bacterial spatial distribution was
largely influenced by saturation of the aggregate. We observed that soil structure plays a
vital, yet secondary role. When the soil was saturated (at 30% moisture content by
weight), the E. coli distributed throughout the aggregates regardless of soil treatment (i.e.

tilled or non-tilled soil). This is possibly because most of the pore spaces where the

93

l we...»

motile E. coli could disperse were filled with water allowing it to move freely within the
aggregate.

Tillage is known to affect the aggregate’s structure by compressing the soil and
destroying the organic binding agents that the soil keep the soil aggregate stable (Tisdall
and Oades, 1982, Hillel, 1998). The variable distribution of E. coli in tilled aggregates
confirms the unpredictable effect tillage aggregate structure. This is further supported by
the results from the flow chamber experiments where the effluent concentrations in
saturated T1 aggregates were highly variable, suggesting the non-uniformity of flow
conducting pores. Furthermore, the lower recoveries in the center-most subsection as
compared to the non-tilled aggregates indicate that the tillage may alter the natural
diffusion of bacteria to the inner part of the aggregate where they could colonize.

Due to the fact that these experiments were exploratory, only E. coli was used to
simulate spatial distribution in soil aggregates. It would be of interest to examine the
translocation of other bacterial species that are non-motile such as Enterococci. This
would confine the spatial distribution due only to preferential flow, and thus help

understand if motility of the bacteria has an affect or alters distribution.

3.6 Conclusion

Perhaps the most interesting aspect of the research is the high attachment rate of
bacteria to soil. As a survival mechanism against stress, bacteria are thought to increase
exopolysaccharide production for protection which increases adsorption to soil due to
attachment to soil particles (Gerba and Mcleod, 1978, Wilkinson, 1958, Roberson and

Firestone, 1992, Abu Lail et al., 2007). From a water quality perspective this ideal,

94

because it suggests that bacteria would be retained more readily in soil and would not
runoff in surface or transport into groundwater. However, as we have observed, there are
factors can reduce or alter the soil capacity. The increasing of soil saturation was the most
important variable in enhancing bacterial transport and spatial distribution. This
highlights the concern when applying manure-fertilizers followed by extensive irrigation
or prior to excessive rainfall where the bacteria may percolate through the soil and
contaminate the ground water (Stoddard et al., 1998). Studies that have examined the soil
saturation’s affect on bacterial transport on field-scale and column studies have reported
similar results (J amieson et al., 2002). Therefore, as an agricultural best management
practice, application of manure followed by intensive irrigation or during seasons with
increased precipitation is discouraged.

While disturbing the soil has been reported to increase the soil’s filtration capacity
and inhibit bacterial transport, the aggregate experiments indicated otherwise (Abu-
Ashour et al., 1998). Tillage increased bacterial transport in flow experiments and altered
E. coli distribution within aggregates. This illustrates negative and often confounding
effect tillage has on filtration because of its impact on the aggregate’s pore structure.
Studies that observe bacterial transport in large scale experiments to provide information
on effect of tillage may not discern differences due to the averaging of the soil
heterogeneities (Guber et al., 2009).

The elucidation of these interactions at such a small scale indicates the usefulness
of utilizing aggregates for modeling soil-microbial interactions. It even suggests that the
major factors, such as tillage and soil moisture, that influence bacterial retention,

transport and spatial distribution can be explained. Because of this, it is proposed that

95

more research for characterizing bacterial transport in soil on the aggregate-scale would
be of great importance.

In addition to follow up experiments suggested in the discussion section, it would
be of interest to observe the retention and transport of the bacterial indicator Clostridium
prefringens. This is because it is utilized in warmer tropical climates and has the
advantage over E. coli and Enterococci in that it does not replicate in the soil
environment (Fujioka and Byappanahalli, 2001, Desmarais et al., 2002). It also may be of
interest to compare bacterial transport with an indicator virus such as coliphage. Viruses
do not replicate, have entirely different properties and are a fraction of the size of bacteria

(Powelson and Gerba, 1995).

96

3.7 Reference List

Abu-Ashour, J ., Joy, D.M., Lee, H., Whiteley, HR, and S. Zelin. 1998. Movement of
Bacteria in Unsaturated Soil Columns with Macropores. American Society for
Agricultural Engineers, Vol.41, No.4, pp.1043-1050.

Abu Lail, L.I., Lui, Y., Atabek, A., and TA. Camesano. Quantifying the Adhesion and
Interaction Forces Between Pseudamonas aeruginosa and Natural Organic Matter.
Environmental Science and Technology, Vol.41, No.23, pp.8031-8037.

Bakken, LR, and RA. Olsen. 1987. The Relationship between Cell Size and Viability of
Soil Bacteria. Microbial Ecology, Vol.13, No.2, pp.103-114.

Blazier, M.A., Patterson, W.B., and S. L. Hotard. 2008. Straw Harvesting, Fertilization,
and Fertilizer Type Alter Soil Microbiological and Physical Properties in a Loblolly Pine
Plantation in the Mid-South USA. Biology and Fertility of Soils, Vol.45, No.2, pp.145-
153.

Bloem, A.B., Griffith, J ., McGee, C., Edge, T.A., Solo-Gabriele, H.M., Whitman, R.,
Cao, Y., Getrich, M., Jay, J.A., Ferguson, D., Goodwin, K.D., Lee, C.M., Madison, M.,
and SB. Weisberg. 2009. Faecal Indicator Bacteria Enumeration in Beach Sand: A
Comparison Study of Extraction Methods in Medium to Coarse Sands. Journal of
Applied Microbiology, Vol. 107, No.5, pp.1740-1750.

Bolster, C.H., Walker, S.L., and KL Cook. 2006. Comparison of Escherichia coli and
Camplyobacterjejuni Transport in Saturated Porous Media. Journal of Environmental
Quality, Vol.35, No.4, pp.1018-1025.

Breed, RS, and W.D. Dotterrer. 1916. The Number of Colonies Allowable on
Satisfactory Agar Plates. Journal of Bacteriology, Vol.1, No.3, pp.321-331.

H.C. Chun. Unpublished data.

Davis, K.E.R., Joseph, S.J., and RH. Janssen. 2005. Effects of Growth Medium,
Inoculum Size, and Incubation Time on Culturability and Isolation of Soil Bacteria.
Applied and Environmental Microbiology, Vol.71, No.2, pp.826-834.

97

Desmarais, T.R., Solo-Gabriele, H.M., and C.J. Palmer. 2002. Influence of Soil on Fecal
Indicator Organisms in a Tidally Influenced Subtropical Environment. Applied and
Environmental Microbiology, Vol.68, No.3, pp.1165-1172.

Fontes, D.E., Mills, A.L., Homberger, G.M., and J .S. Herman. 1991. Physical and
Chemical Factors Influencing Transport of Microorganisms through Porous Media.
Applied and Environmental Microbiology, Vol.57, No.9, pp.2473-2481.

Fujioka, RS, and MN. Byappanahalli. 2001. Microbial Ecology Controls the
Establishment of Fecal Bacteria in Tropical Soil Environment. pp.273-283. In Advances
in Water and Waste Water Treatment Technology. Elsevier Science B.V., Amsterdam,
Netherlands.

Gagliardi, J.V., and J.S. Kams. 2000. Leaching of Escherichia coli 0157:H7 in Diverse
Soils under Various Agricultural Management Practices. Applied Environmental
Microbiology, Vol.66, No.3, pp.877-883.

Gannon, J .T., Manilal, V.B., and M. Alexander. 1991. Relationship between Cell Surface
Properties and Transport of Bacteria through Soil. Applied and Environmental
Microbiology, Vol.57, No.1, pp.190-193.

Gerba, CR, and J.S. Mcleod. 1976. Effect of Sediments on the Survival of Escherichia
coli in Marine Waters. Applied and Environmental Microbiology, Vol.32, No.1, pp.114-
120.

Grandy, AS, and GP. Robertson. 2007. Land-use Intensity Effects on Soil Organic
Carbon Accumulation Rates and Mechanisms. Ecosystems, Vol.10, No.1, pp.58-73.

Guber, A.K., Shelton, DR, and Ya. A.Pachepsky. 2005. Effect of Manure on
Escherichia coli Attachment to Soil. Journal of Environmental Quality, Vol. 34, No.6,
pp.2086-2090.

Guber, A.K., Pachepsky, Y.A., Shelton, DR, and 0. Yu. 2009. Association of Fecal
Coliforms with Soil Aggregates: Effect of Water Content and Bovine Manure
Application. Soil Science, Vol.174, No.10, pp.543-548.

98

 

Hattori, T. and R. Hattori. 1976. The Physical Environment in Soil Microbiology: An
Attempt to Extend Principles of Microbiology to Soil Microorganisms. Critical Reviews
in Microbiology, Vol.4, No.4, pp.423-461.

T. Hattori. 1988. Soil Aggregates as Microhabitats of Microorganisms. Reports of the
Institute for Agricultural Research, Tohoku University, Vol.37, pp.23-36.

Hillel, D. 1998. Soil Compaction in the Field. pp.365-369. In Environmental Soil
Physics. Academic Press, San Diego, CA.

Hogt, A.H., Dankert, J ., Hulstaert, CE, and J. Feijen. 1986. Cell Surface Characteristics
of Coagulase-Negative Staphylococci and Their Adherence to Fluorinated
Poly(Ethylenepropylene). Infection and Immunity, Vol.51, No.1, pp.294-301.

Holdaway, HA. 2003. Changes in Microbial Community Responses to Gradients of
Carbon, Nitrogen and Wetting Cycles in Concentric Layers of Soil Macro-Aggregates.
Master’s Thesis. Michigan State University.

Huysman, F., and W. Vestraete. 1993.Water-Facilitated Transport of Bacteria in
Unsaturated Soil Colums: Influence of Inoculation and Irrigation Methods. Soil Biology
and Biochemistry, Vol.25, No.1, pp.91-97.

J amieson, R.C., Gordon, R.J., Sharples, K.E., Stratton, G.W., and A. Madani. 2002.
Movement and persistence of fecal bacteria in agricultural soils and subsurface drainage
water: A review. Canadian Biosystems Engineering, Vol.44, pp.1.1-1.9.

J ennison, M.W., and GP. Wadsworth. 1940. Evaluation Errors Involved in Estimating
Bacterial Numbers by the Plating Method. Journal of Bacteriology, Vol.39, No.4, pp.389-
397.

Jenkins, M.B., Truman C.C., Siragusa, G., Line, E., Bailey, J .S., Frye, J., Endale, D.M.,
Franklin, D.H., Schomberg, H.H., Fisher, D.S., and SS. Sharpe. 2008. Rainfall and
Tillage Effects on Transport of Fecal Bacteria and Sex Hormones l7beta-estradiol and
Testosterone from Broiler Litter Applications to a Georgia Piedmont Ultisol. Science of
Total Environment, Vol.403, No.1-3 pp.154-163.

99

Liebig, M.A., Varvel, G.E., Doran, J .W., and B]. Weinhold. 2002. Crop Sequence and
Nitrogen Fertilization Effects on Soil Properties in the Western Corn Belt. Soil Science
Society of America Journal, Vol.66, No.2, pp.596-601.

Mahler, B.J., Personne’, J .-C., Lods, G.F., and C. Drogue. 2000. Transport of Free and
Particulate-Associated Bacteria in Karst. Journal of Hydrology, Vol.238, No.3-4, pp.179-
193.

McConville, J .F., Anger, CB, and W.A. Anderson, Jr. Method for Performing Aerobic
Plate Counts of Anhydrous Cosmetics Utilizing Tween 60 and Arlacel 80 as Dispersing
Agents. Applied Microbiology, Vol.27, No.1, pp.5-7.

McMurry, S.W., Coyne, M.S., and E. Perfect. 1998. Fecal Coliform Transport through
Intact Soil Blocks Amended with Poultry Manure. Journal of Environmental Quality,
Vol.27, No.1, pp.86-92.

Muirhead R.W., Collins, RP, and P.J . Bremer. 2006. Interaction of Escherichia coli and
Soil Particles in Runoff. Applied and Environmental Microbiology, Vol.72, No.5,
pp.3406-3411.

Mummey, D., Holben, W., Six, J ., and P. Stahl. 2006. Spatial Stratification of Soil
Bacterial Populations in Aggregates of Diverse Soils. Microbial Ecology, Vol.51, No.3,
pp.404-41 1.

Olsen, RA, and LR. Bakken. 1987. Viability of Soil Bacteria: Optimization of Plate-
Counting Technique and Comparison between Total Counts and Plate Counts within
Different Size Groups. Microbial Ecology, Vol.13, No.1, pp.59-74.

Powelson, D.K., and C.P. Gerba. 1995. Fate and Transport of Microorganisms in the
Vadose Zone. pp.123-135. In Handbook of the Vadose Zone Characterization and
Monitoring.

Ranjard, L., and A, Richaume. 2001. Quantitative and Qualitative Microscale
Distribution of Bacteria in Soil. Research in Microbiology, Vol. 152, No. 8, pp.707-716.

100

Richaume, A., Steinberg, C., Jocteur-Monrozier, L., and G. Faurie. 1993. Differences
Between Direct and Indirect Enumeration of Soil Bacteria: The Influence of Soil
Structure and Cell Location. Soil Biology and Biochemistry, Vol. 25, No.5, pp.641-643.

Roberson, EB, and MK Firestone. 1992. Relationship Between Desiccation and
Exopolysaccharide Production in Soil Pseudamonas sp. Applied and Environmental
Microbiology, Vol.58, No.4, pp.1284-l291.

Rozak, DB, and RR. Colwell. 1987. Survival Strategies of Bacteria in the Natural
Environment. Micobiological Reviews, Vol.51, No.3, pp.365-379.

Singh, K.K., and W.S. Vincent. 1987. Clumping Characteristics and Hydrophobic
Behaviour of an Isolated Bacterial Strain from Sewage Sludge. Applied Microbiology
and Biotechnology, Vol.25, No.4, pp.396-398.

Singh, S. 2007. Investigation of Bacterial Fecal Indicators and Coliphage Virus in
Sediment and Surface Water of Parks and Beaches along the Grand River (MI) and Lake
Michigan (MI). Master’s Thesis. Michigan State University.

Soby, S., and K. Bergman. 1983. Motility and Chemotaxis of Rhizobium meliloti in Soil.
Applied and Environmental Microbiology, Vol.46, No.5, pp.995-998.

Solo-Gabrielle, H.M, Wolfert, M. A., Desmarais, TA. and C.J. Palmer. 1999. Sources of
Escherichia coli in a Coastal Subtropical Environment. Applied and Environmental
Microbiology, Vol.66, No.1 pp.230-237.

T. A. Stenstrom. 1989. Bacterial Hydrophobicity, an Overall Parameter for the
Measurement of Adhesion Potential to Soil Particles. Applied and Environmental
Microbiology, Vol.55, No.1, pp.142-147.

Stevik, T.K., Aa, K., Ausland, K., and J .F. Hanssen. 2004. Retention and Removal of
Pathogenic Bacteria in Waste Water Percolating through Porous Media: A Review. Water
Research, Vol. 38, No.6, pp.1355-1367.

Stoddard, C.S., Coyne, M.S., and J .H. Grove. 1998. Fecal Bacteria Survival and
Infiltration through a Shallow Agricultural Soil: Timing and Tillage Effects. Journal of
Environmental Quality. Vol.27, No.6, pp. 1516-1523.

101

Tisdall, J .M, and J .M. Oades. 1982. Organic Matter and Water-stable Aggregates in
Soils. Journal of Soil Sciences, Vol. 33, No. 2, pp.141-163.

Unc, A., and M.J. Goss. 2003. Movement of Faecal Bacteria Through the Vadose Zone.
Water, Air, and Soil Pollution, Vol. 149, No. 1-4, pp. 327-337.

Unc, A., and M.J. Goss. 2004. Transport of Bacteria from Manure and Protection of
Water Sources. Applied Soil Ecology, Vol. 25, No.1, pp.1-18.

Wang, W., Kravchenko, A. N., Ananyeva, K. A., Smucker, A. J. M., Chun, H., Mazher,
M. A., Rose, J. B., and M. L. Rivers. 2010. How to best utilize synchrotron X-ray
computed microtomography imaging for 3D soil aggregates characterization? GeoX:
Third International Workshop on X-ray CT for Geomaterials, February 28-March 3.

Wildy, P., and TE Anderson. 1964. Clumping of Susceptible Bacteria by Bacteriophage
Tail Fibres. Journal of General Microbiology, Vol.34, No.2, pp.273-283.

J .F. Wilkinson. 1958. The Extracellular Polysaccharides of Bacteria. Bacteriological
Reviews, Vol.22, No.1, pp.46-73.

Zimmerman, A.M., Rebarchik, D.M., Flowers, A.R., Williams, J .L., and DJ. Grimes.
Escherichia coli Detection using mTec Agar and Flourescent Antibody Direct Viable
Counting on Coastal Recreational Water Samples. Letters in Applied Microbiology, Vol.
49, No.4 , pp.478-483.

102

Appendix Raw Data for Analysis

Table A. 1. Data for heterotrophic plate counts for bacterial extraction from whole aggregate experiments.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

10'2 10'3 104 Aggregate Aggregate
Dilution Dilution Dilution weight (g) treatment

Sample/Dilution (CFU) (CFU) (CFU)

0.1801 T1
V4 49, 37, 38 5, 3, 9 0, 1, 0

0.1397 T2
V5 17, 13, 13 1,1, 0 0, 0,1

0.1501 T7
V6 69, 58, 65 7, 7, 10 0, 0, 1

0.15 T1
V10 46,31,44 4,2,1 -

0.1802 T2
V11 47,41,40 0,3,2 -

0.1821 T7
V12 43, 49,31 0,4,2 -

0.1013 T1
V13 48, 35,51 4, 5,- -

0.1606 T2
V14 57, 58, 52 10, 5, 6 -

0.1004 T7
V15 56, 50, 52 4, 8, 9 - .

0.1602 T1
V16 43, 34, 41 4, 0, 7 -

0.1787 T2
V17 66, 79,82 5, 3, 6 -

0.1637 T7
V18 24,36, 32 3,1,3 -

0.1989 T1
V19 57, 60, 77 5, 6, 2 -

0.1095 T2
V20 40, 38, 39 8, 1, 3 -

0.1098 T7
V21 56, 63, 51 10, 12, 4 -

 

*Note: CFU values for dilution plates are in triplicates for each sample.

103

 

Table A2. Data for whole aggregate desiccation and E. coli recovery experiments.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

, Aggregate
Air-drying 10'2 10'3 Aggregate E.coli weight
Date of time Dilution Dilution dry solution after air-
Experiment (minutes) (CFU) (CFU) weight (g) weight(g) mg (g)
3/12/2009 None 53,53, 57 4,7,5 0.2531 0.0485 -
120 0, 0, 0 - 0.3117 0.0487 0.3160
3/13/2009 None 28, 42, 24 6,2,2 0.2238 0.0481 -
120 0, 0, 0 - 0.2657 0.0498 0.2656
3/14/2009 None 39, 43, 49 - 0.2160 0.0495 -
120 0, 0, 0 - 0.2461 0.0462 0.2412
None 79,61,57 7,5,4 0.2585 0.0487 -
4/15/2009 30 74, 70 - 0.2450 0.0491 0.2771
60 2, 2 - 0.2825 0.0500 0.2936
None 45, 44, 49 1,4,2 0.2062 0.0468 -
4/16/2009 30 57, 49, 57 2, 4, 4 0.1997 0.0473 0.2286
60 7, 13,7 - 0.2510 0.0489 0.2631
90 0, 0, 0 - 0.2548 0.0483 0.2562
None 68,62, 56 5,7,8 0.2358 0.0445 -
15 67, 74, 72 4, 11,7 0.1884 0.0492 0.2266
5/11/2009 30 65, 83, 76 6, 9, 7 0.1630 0.0495 0.1924
45 64, 61, 70 8, 7, 7 0.2035 0.0493 0.2248
60 41, 41,45 — 0.1562 0.0480 0.1717
None 51,33,45 0,4,3 0.2070 0.0496 -
1 1/17/2009 30 39,33,36 3,3,6 0.2055 0.0492 0.2324
60 14,28,13 4,2,2 0.2324 0.0508 0.2458

 

104

 

 

Table A3. Data for heterotrophic plate counts for bacterial extraction from aggregate interior and exterior

 

 

 

 

 

 

 

 

 

 

 

 

 

 

layers.

10‘2 10‘3

Dilution Dilution Aggregate
Samgle CFU) (CFU) Weight (g) Treatment Diluent
T7 107 A 41, 36, 33 3, 2, 2 0.0375 T1 10ml
T7 107 C 35, 36, 40 3, 3, 5 0.0387 T2 10ml
T1 105 A 63, 53, 51 4, 0, 5 0.0741 T7 10ml
T1 105 C 26, 16, 30 O, 0, 1 0.0686 T1 15ml
T2 154 A 36, 41, 37 3, 4, 1 0.083 T2 10ml
T2 154 C 18, 38, 37 3, 2, 5 0.0851 T7 10ml
T7 111 A 27, 19, 17 2, 1, 1 0.0296 T1 10ml
T7 111 C 28, 28, 31 4, 6, 1 0.031 T2 10ml
T2 144 A 13, 1.5, 19 2, 0, 0 0.0527 T7 10ml
T2 144 C 12, 19, 10 5, 2, 1 0.0475 T1 10ml
T1 136 A 64, 64, 63 16, 8, 9 0.0704 T2 10ml
T1 136 C 94, 94, 103 11, 12, 8 0.064 T7 10ml

 

 

 

 

 

 

*Note: The letter “A” after the sample ID indicates the sample is the outer aggregate layer. The letter “C”
after the sample ID indicates the sample is the inner aggregate layer.

105

 

Table AA. Data for flow chamber experiments using saturated aggregates.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

R . Aggregate Effluent 532:; Soil Bacterial
eplrcate concentration concentration . .
(CFU) (CFU) concentratron treatment Specres
(CFU)
1 1.70E+03 4.98E+02 2.75E+03 T1 E. coli
2 9.00E+02 6.30E+02 6.50E+03 T1 E. coli
3 2.10E+03 2.34E+02 4.08E+03 T1 E. coli
4 3.27E+03 7.28E+01 4.75E+03 T1 E. coli
5 3.44E+03 3.44E+03 3.25E+03 T1 E. coli
1 3.70E+03 4.17E+00 2.75E+03 T2 E. coli
2 3.53E+03 1.25E+01 6.50E+03 T2 E. coli
3 2.83E+03 l.04E+02 4.75E+03 T2 E. coli
4 2.10E+03 3.92E+02 3.67E+03 T2 E. coli
5 2.10E+03 3.92E+02 3.67E+03 T2 E. coli
1 3.27E+03 N.D. 2.75E+03 T7 E. coli
2 4.77E+03 N.D. 6.50E+03 T7 E. coli
3 3.33E+03 4.53E+00 4.08E+03 T7 E. coli
4 3.57E+03 6.23E+01 4.75E+03 T7 E. coli
5 3.40E+03 1.22E+03 3.25E+03 T7 E. coli
1 1.30E+03 1.20E+01 1.17E+03 Tl Ent. Faecium
2 4.33E+02 3.78E+02 1.15E+03 T1 Ent. Faecium
3 1.40E+03 1.85E+01 1.38E+03 T1 Ent. Faecium
4 1.70E+03 0.00E+00 1.3 8E+03 T1 Ent. F aecium
5 1.27E+03 8.74E+02 1.20E+03 Tl Ent. Faecium
l 1.27E+03 N.D. 1.17E+03 T2 Ent. Faecium
2 1.77E+03 N.D. 1.15E+03 T2 Ent. Faecium
3 1.57E+03 N.D. 1.38E+03 T2 Ent. Faecium
4 1.53E+03 N.D. 1.38E+03 T2 Ent. Faecium
5 5.00E+02 1.36E+02 8.67E+02 T2 Ent. Faecium
1 1.27E+03 N.D. 1.17E+03 T7 Ent. Faecium
2 l.77E+03 N.D. 1.15E+03 T7 Ent. Faecium
3 1.57E+O3 N.D. 1.38E+03 T7 Ent. Faecium
4 1.53E+03 N.D. 1.38E+03 T7 Ent. Faecium
L 5 5.00E+02 4.36E+02 1.2015403 T7 Ent. Faecium

 

 

 

 

*Note: N.D. indicates that there were no bacteria detected.

106

 

 

 

Table A.5.Data for flow chamber experiments using air-dry aggregates.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A e ate Effluent Spiked . .
Replicate Congciiitfation concentration bacteria 8011 Bacterial
(CFU) (CFU) concentration treatment Specres
(CFU)
1 6.10E+03 N.D. 4.42E+03 T1 E. coli
2 6.97E+03 N.D. 3.83E+03 T1 E. coli
3 6.10E+03 N.D. 4.42E+03 T1 E. coli
4 8.97E+03 N.D. 4.83E+03 Tl E. coli
5 1.02E+04 N.D. 5.50E+03 T1 E. coli
1 6.13E+03 4.30E+00 4.42E+03 T2 E. coli
2 6.57E+03 N.D. 3.83E+03 T2 E. coli
3 6.13E+03 4.30E+00 4.42E+03 T2 E. coli
4 9.17E+03 1.95E+01 4.83E+03 T2 E. coli
5 9.10E+03 1.05E+02 5.50E+03 T2 E. coli
1 5.67E+03 1.31E+01 4.42E+03 T7 E. coli
2 6.97E+03 N.D. 3.83E+03 T7 E. coli
3 5.67E+03 1.31E+01 4.42E+03 T7 E. coli
4 8.83E+03 7.42E+01 4.83E+03 T7 E. coli
5 1.00E+04 N.D. 5.50E+03 T7 E. coli
1 2.47E+03 N.D. 1.22E+03 Tl Ent. Faecium
2 2.93E+03 N.D. l.l6E+03 T1 Ent. Faecium
3 2.47E+03 N.D. 1.22E+03 T1 Ent. Faecium
4 2.73E+03 N.D. 1.25E+03 Tl Ent. Faecium
5 3.03E+03 N.D. 1.28E+03 T1 Ent. Faecium
1 2.17E+03 N.D. 1.22E+03 T2 Ent. F aecium
2 3.20E+03 N.D. 1.16E+03 T2 Ent. Faecium
3 2.17E+03 N.D. 1.22E+03 T2 Ent. Faecium
4 3.17E+03 9.73E+00 1.25E+03 T2 Ent. Faecium
5 2.80E+03 1.58E+01 1.28E+03 T2 Ent. Faecium
1 2.63E+03 1.75E+01 1.22E+03 T7 Ent. Faecium
2 3 .47E+03 N.D. 1.16E+03 T7 Ent. F aecium
3 2.63E+03 1.75E+01 1.22E+03 T7 Ent. Faecium
4 2.63E+03 5.30E+00 1.25E+03 T7 Ent. Faecium
5 3.33E+03 N.D. 1.28E+03 T7 Ent. Faecium

 

107

 

3

Table A6. Data for slicing experiments using T1 aggregates.

 

Moisture content

0%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Weight Replicate Weight Replicate Weight

Subsection Replicate 1 (g) 2 g) 3 (g)
1 2.00E+02‘ 0.042 0.00E+00 0.019 8.67E+02 0.0289
2 4.00E+02 0.073 0.00E+00 0.023 1 .03E+03 0.0465
3 1.33E+02 0.043 0.00E+00 0.015 8.67E+02 0.121
4 7.33E+02 0.031 6.67E+01 0.024 7.33E+02 0.0141
5 2.67E+02 0.012 2.67E+02 0.047 8.67E+02 0.0939
6 0.00E+00 0.024 0.00E+00 0.074 9.00E+02 0.2109
7 4.67E+02 0.04 0.00E+00 0.02 6.00E+02 0.1617
Influent 7.50E+03 - 7.50E+03 - 7.50E+03 -
Moisture content 1 5%

Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
l 6.67E+01 0.0072 6.33E+02 0.0181 5.00E+02 0.0282
2 5.67E+02 0.0997 4.33E+02 0.0155 1.33E+02 0.0038
3 6.67E+01 0.0607 2.67E+02 0.0303 3 .00E+02 0.0073
4 2.33E+02 0.0261 7.00E+02 0.0354 3 .00E+02 0.0505
5 7.00E+02 0.0277 1.00E+02 0.0191 3 .00E+02 0.0258
6 1.00E+02 0.2442 2.67E+02 0.0214 1.67E+02 0.0267
7 4.00E+02 0.1505 1.33E+02 0.0344 3.67E+02 0.0216
Influent 2.3 5E+03 - 4.48E+03 - 5 .92E+03 -
Moisture content 30%

Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
1 3.50E+02 0.0548 1.00E+02 0.0124 4.67E+02 0.0134
2 2.50E+02 0.0304 2.33E+02 0.0071 1.33E+02 0.0072
3 2.50E+02 0.0251 3.33E+01 0.0075 3.33E+02 0.001
4 6.00E+02 0.026 2.67E+02 0.014 2.67E+02 0.005
5 8.50E+02 0.061 3.33E+02 0.0277 2.67E+02 0.0012
6 4.50E+02 0.0037 3.33E+02 0.0323 3 .00E+02 0.003
7 4.00E+02 0.0246 3 .00E+02 0.0023 5 .67E+02 0.01 3
Influent 4.51E+03 - 4.51E+03 - 8.47E+03 -

 

 

 

 

 

 

 

*Note: these E. coli concentrations are in CFU/ aggregate.

108

 

Table A.7. Data for slicing experiments using T2 aggregates.

 

Moisture content

0%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
1 8.00E+02 0.022 0.00E+00 0.015 6.00E+02 0.1412
2 2.00E+02 0.025 0.00E+00 0.046 0.00E+00 0.0045
3 2.67E+02 0.036 2.00E+02 0.016 3.00E+02 0.1694
4 1.53E+03 0.037 4.00E+02 0.034 6.33E+02 0.1527
5 9.33E+02 0.015 3.33E+02 0.033 1.20E+03 0.0651
6 6.67E+02 0.0001 2.00E+02 0.03 1.03E+03 0.0564
7 6.67E+01 0.004 6.67 E+01 0.026 6.67E+02 0.1274
Influent 7.50E+03 7.50E+03 - 7.50E+03 -
Moisture content 15%

Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
1 2.67 E+02 0.1707 3.33E+02 0.0265 3.33E+02 0.0354
2 3.33E+02 0.0407 6.67E+01 0.0385 2.00E+02 0.0178
3 0.00E+00 0. 1093 1 .00E+02 0.0405 0.00E+00 0.0002
4 6.67E+01 0.1665 7.67E+02 0.0588 5.00E+02 0.0145
5 1.67E+02 0.0987 4.00E+02 0.0319 5.00E+02 0.0457
6 1.33E+02 0.1006 8.33E+02 0.02 4.00E+02 0.0195
7 2.33E+02 0.0836 1.67E+02 0.0151 4.33E+02 0.0424
Influent 2.35E+03 4.48E+03 - 5.92E+03 -
Moisture content 30%

Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
1 6.67E+01 0.0118 2.67E+02 0.0183 3.00E+02 0.013
2 6.67E+01 0.003 6.67E+01 0.0159 1.33E+02 0.011
3 1.33E+02 0.0076 5.00E+02 0.0336 1.67E+02 0.00013
4 3.00E+02 0.0234 1.33E+02 0.0035 3.33E+02 0.0105
5 3.00E+02 0.0236 4.00E+02 0.0418 2.33E+02 0.0023
6 4.33E+02 0.0218 6.33E+02 0.0219 3.33E+02 0.0128
7 1.00E+02 0.0176 3.00E+02 0.0217 1.67E+02 0.0019
Influent 4.51E+03 4.51E+03 - 8.47E+03 -

 

 

 

 

 

 

109

 

Table A8. Data for slicing experiments using T7 aggregates.

 

Moisture content

0%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
1 1.40E+03 0.087 8.00E+02 0.028 9.67E+02 0.1287
2 7.33E+02 0.026 9.33E+02 0.0001 6.00E+02 0.1899
3 1.40E+03 0.046 2.13E+O3 0.059 6.67E+02 0.1909
4 4.67E+02 0.03 3.33E+02 0.01 1.97E+03 0.637
5 7.33E+02 0.009 1.47E+03 0.059 9.33E+02 0.1241
6 6.67 E+02 0.034 1.07E+03 0.03 7.00E+02 0.0016
7 8.00E+02 0.035 2.67E+02 0.02 1.03E+03 0.131 1
Influent 7.50E+03 7.50E+03 - 7.50E+03 -
Moisture content 15%
Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
1 3.33E+02 0.0985 5.67E+02 0.0938 6.67E+02 0.1302
2 1.33E+02 0.2204 5.67E+02 0.0805 7.67E+02 0.0754
3 6.67E+01 0.2298 1.33E+03 0.0365 8.00E+02 0.0284
4 4.67E+02 0.2345 5.33E+02 0.1154 9.67E+02 0.0583
5 6.67E+02 0.3126 1.23E+03 0.0017 8.00E+02 0.1596
6 3.67E+02 0.0391 7.00E+02 0.015 1.07E+03 0.0234
7 2.67E+02 0.0516 7.33E+02 0.0197 3.33E+02 0.0311
Influent 2.35E+03 - 4.48E+03 - 5.92E+03 -
Moisture content 30%
Replicate Weight Replicate Weight Replicate Weight
Subsection 1 (g) 2 (g) 3 (g)
1 4.33E+02 0.0153 3.33E+01 0.0053 6.67E+02 0.0211
2 4.67E+02 0.0396 3.67E+02 0.01 14 6.00E+02 0.0297
3 5.00E+02 0.0435 1.33E+02 0.013 3.67E+02 0.0028
4 5.33E+02 0.0176 3.00E+02 0.0021 5.67E+02 0.0382
5 1.43E+03 0.0853 4.33E+02 0.0033 8.67E+02 0.0098
6 6.33E+02 0.0317 6.67E+01 0.005 1.03E+03 0.0319
7 3.67E+02 0.0109 3.00E+02 0.0043 5.33E+02 0.0086
Influent 4.51E+03 - 4.51E+03 - 8.47E+03 -

 

 

 

 

 

110