Using bacteria to monitor the influences of cattle wastes on water quality

Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices.

h

L

5 71/67

>3.

Using Bacteria To Monitor
the Influences of Cattle Wastes

on Water Quality

<3 -o

3

■» >-A

V(1

r°

u>

%C\
cl "3-

iP

cP

n tP

\f i'o

<3-

<3%

U.S. Department of Agriculture
Science and Education Administration
Agricultural Research Results ARR-NE-3
September 1979

Trade names are used in this publication solely for the
purpose of providing specific information. Mention of
a commercial or a proprietary product does not consti-
tute recommendation, guarantee, or warranty of the
product by the U.S. Department of Agriculture or an
endorsement by the Department over other products not
mentioned .

Science and Education Administration, Agricultural Research Results,
Northeastern Series, No. 3, September 1979

Published by Agricultural Research (Northeastern Region), Science and Education
Administration, U.S. Department of Agriculture, Beltsville, Md . 20705

ii

USING BACTERIA TO MONITOR THE INFLUENCES OF
CATTLE WASTES ON WATER QUALITY^/

By S. H. Kunkle^/

ABSTRACT

Four bacterial indices of water pollution — total
coliforms, fecal coliforms, fecal streptococci, and
enterococci — were sampled in the surface runoff from a
0.1 ha sprinkle-irrigated field plot in Vermont. The
field study was made on mowed-grass pasture during
spring to late summer. The runoff was sampled during
three irrigations before and ten irrigations after
treating the field plot with a single application of
recently excreted dairy cattle manure.

Results showed that the fecal coliform group,
especially if used together with the total coliform
group, was a good index for discriminating between
contamination from new as compared with old manure.

KEYWORDS: Monitoring, overland flow, non-point source

pollution, water quality index, FC/FS ratio,
coliforms, bacterial indicators, animal
wastes, surface runoff.

INTRODUCTION

Surface waters in rural areas must be reasonably free of contamination by
pathogenic organisms to protect water supplies and to safeguard public health
in recreation areas. One source of biological contamination of water is
field-applied or stored animal wastes that may wash into streams during storms.
From a health viewpoint, the flushing of recently excreted animal wastes into
streams is of greatest concern because older wastes lose many of their
pathogens through natural die-off processes. Although several bacterial
groups are good indices of fecal contamination as determined in laboratory
studies, these groups are not always so good for detecting contamination under

1/ Cooperative study between the Northeast Watershed Research Center,
Science and Education Administration-Agricultural Research, U.S. Department
of Agriculture and the Pennsylvania Agricultural Experiment Station, The
Pennsylvania State University.

2/ S. H. Kunkle served as hydrologist for the USDA, Agricultural Research
Service (now SEA-AR) at the time this study was carried out. He is presently
with the International Forestry Staff, U.S. Forest Service, U.S. Department of
Agriculture, P.0. Box 2417, Washington, D. C. 20013.

1

field conditions. For example, the "total coliforms" include many nonenteric
organisms that are mainly soil organisms. Thus, surface runoff from almost
any farm field contains a high total coliform count, even from experimental
fields devoid of animal wastes, as shown in studies by this author (Kunkle
1970) and others. Consequently, an index — bacterial or other — is needed that
clearly establishes the presence of recently excreted animal wastes in water.
Many laboratory studies have shown that the fecal coliform bacterial group can
be used as such an index for human or animal contamination and can serve as a
good indicator for the potential presence of pathogens spread by human or
animal wastes. Most of these were laboratory based or municipal pollution
studies .

The general objective of this study was to determine which of four common
bacterial groups is the best index to monitor contamination by recently excre-
ted animal wastes in surface runoff. There has been little field investigation
of the fate of bacterial indicators in farm surface runoff despite the routine
use of these same indicators for monitoring the water quality of streams and
water supplies.

The specific objectives were (1) to evaluate the relative concentrations
of four common bacterial indicators in runoff from a field which was free of
animal wastes as compared with those found in runoff from the same field after
cattle manure was applied, and (2) to observe the relative die-off (reduction
in numbers) of these four bacterial groups in surface runoff for several weeks
after manure application. The runoff was generated approximately once a week
by sprinkle irrigation and, occasionally, by natural storms.

The study was part of a series of stream and plot studies on runoff and
water quality made on the Sleepers River Research Watershed in Danville, Vt.
This plot was used in earlier related studies (Kunkle 1970, 1971).

MATERIALS AND METHODS

A surface runoff plot of about 30 x 35 m (^0.1 hectare) was constructed
in a hilly area consisting of a deep and permeable soil, the moderately well-
drained Cabot silt loam. The plot had a 14 percent slope and faced southwest.
Vegetative cover was a mixture of timothy, bluegrass, red clover, and red top
and was maintained at a 10-20 cm height by hand cutting. The plot elevation
was 300 m above mean sea level. During the period of this study (spring and
summer 1970) daily maximum air temperatures were typically 25-30 °C (only
rarely above 33 °C) , and night air temperatures dropped to the 10-20 °C range.

"Rainbird" type sprinklers were set on supports in a grid fashion 1.2 m
above ground, with the sprinkler heads about 6 m apart. The plot was sprinkle
irrigated 13 times during June-September at a rate of about 25 mm/h. Each
irrigation lasted for 1-1/2 to 3 hours. Figure 1 gives the total irrigation
and runoff amounts for each date. All natural storms that produced runoff
(June 27 and 30) were sampled. The amount of runoff on June 30 was very small
(<100 liters) and is suspected to be unrepresentative. The analysis of the
June 27 runoff (also minor) fits between those observed for the June 25 and
July 1 irrigation induced runoff events.

2

Plastic runoff gutters, inserted just under the top 5-10 cm of soil on the
downhill edge, served to catch the surface runoff. Runoff was measured with
H-type flumes. Rain or irrigation depths were determined at 32 collectors
scattered over the plot. For each irrigation event, up to 20 sample bottles
of runoff were collected, plus several control samples of the irrigation water.

For each sample bottle at least duplicate analyses at various dilutions
were made for total coliform bacteria (M-Endo broth) , enterococci (M- Enterococ-
cus agar) , fecal streptococci (KF-Streptococcus agar) , and fecal coliforms
(M-FC broth) using membrane filter procedures. For each of the four bacterial
groups there were at least 364 individual readings; 13 irrigations x ^14 sample
bottles (observations) /irrigation x 2 or more analyses (plates) /bottle . Four
to six control samples of the irrigation water for each irrigation were ana-
lyzed in the same way. All samples were iced immediately upon collection and
then filtered within a few hours. Sediment concentrations in the runoff water
were very minute, so no prefiltration was needed.

The fecal coliforms were incubated in a water bath at 44.5+0.5 °C.

Other bacteria groups were incubated at 35+0.5 °C, according to "Standard
Methods" (American Public Health Association 1971) . Colorimetric analyses
of phosphate were made (Bausch and Lomb Spectronic-20) using a stannous
chloride method.

Manure Application

The runoff plot was ideal for observing background levels before treatment
because it had not been grazed, fertilized, or manured for about 10 years be-
fore the study. The bacterial contamination in runoff before manuring was
monitored in 1968, 1969, and 1970 to establish background levels for the four
bacterial groups. Three irrigations were made before applying manure. On
June 22, 800 kg of manure was spread over the plot 1-2 hours before the fourth
irrigation. There were nine more irrigations over a 3-month period (fig. 1).

3

About 1 m (0.8 ton) of wet dairy manure was applied to the 0.1 ha plot;
i.e., an 8 ton/ha application rate. The manure was 1-5 days old and contained
10 percent sawdust and 56 percent water. It was spread evenly over the plot
but never closer than 4 m from the gutters on the downhill edge.

The manure was analyzed (one sample, five aliquots) for fecal coliform/
fecal streptococci (FC/FS) and fecal colif orm/total coliform (FC/TC) ratios.

It was hauled, applied to the plot, sampled, and analyzed on the same day.

Each sample was iced immediately and then prepared for analysis within a few
hours. Based on a serial dilution, FC/FS averaged about 0.5. In three
parallel dilutions per aliquot, FC/FS ratios ranged from 0.05 to 1.37,
averaging 0.53. The same five aliquots and dilutions also showed the total
coliform bacteria to be comprised of about 71 percent fecal coliforms with a
FC/TC ratio ranging from 0.46 to 1.00, averaging 0.71.

Since the manure had about 300,000 fecal coliforms/g wet manure, according
to analysis, the 1 m^ application contained 240,000 x 10^ fecal coliforms.
According to Geldreich (1966), this would be about 44 cow days of fecal

3

coliform bacteria applied. (Cow days are defined as the estimated per capita
contribution of indicator microorganisms from a single cow, which averages
5,400 x 10^ fecal colif orms/day) . Assuming a cow produces 23 kg/day of manure,
the 800 kg of manure that had been spread on the plot would equal about 35 cow
days. In summary, according to the manure volume or bacterial numbers analyzed
in the manure, between 35 and 44 cow days of manure were applied to the plot.

RESULTS AND DISCUSSION

The results of the study are presented in figures 1 and 2. Figure 1 shows
concentrations for the bacterial groups by irrigation event during the study
period. Each point represents an average concentration of approximately 14
observations (bottles) for that event. The individual observation was the
best quality plate of at least two isolated plates. Figure 2 shows individual
observations for fecal coliform and phosphate (PO^-P) for some of the events.
Phosphate analyses for the samples provided an index for comparing the bac-
terial concentrations to the nutrient levels in the runoff water before and
after the manuring. The time of manuring in both figures is designated by the
arrow. The phosphate levels were almost zero before manuring, rose sharply to
about 2.5 mg/1 in the first irrigation event after manuring, then fell to
about 0.5 mg/1 or below for all following events.

The irrigation water remained very clear throughout the study, with
typically zero or near-zero fecal colif orms/sample , Concentrations for the
other bacterial groups were so low as to be insignificant by any comparison
to the runoff water. For example, on the first irrigation event after
manuring (June 22) , the concentrations in irrigation water averaged total
coliform-40, fecal coliform-15, fecal streptococci-35, and enterococci-20/100
ml, compared to values in the 10^ and 10^ range in the runoff water as shown
in figure 1.

Pretreatment Concentrations

The pretreatment (background) bacterial concentrations in runoff before
June 22 were very similar for three of the four bacterial groups — total
coliforms, fecal streptococci, and enterococci — ranging mainly from 10,000 to
100,000/100 ml. On the other hand, fecal coliform concentrations were mainly
at the less-than-100/100 ml level (fig. 1). The pretreatment bacterial con-
centrations (June 11, 16, and 18) resembled those found in 200 runoff samples
taken from the same plot during the previous summer.

Since soil and vegetation are the source of large numbers of nonfecal
coliforms (Geldreich 1966, Kunkle 1970 and 1971, van Donsel et al . 1967), the
large numbers of total coliforms observed in the runoff are not surprising.
Analysis of a few surface soil samples on the study plot showed nonfecal
coliform concentrations up to 30,000/g of soil.

Precip and Runoff

Totals in 1000 I B a ct er i a / 1 00 m I

Date of Event

Figure 1. — Average concentrations for the four bacterial groups for the
13 irrigation events . Each point is the average of about 14
samples. The large arrow and "T" show the time of manure
application.

5

30

2.5

T)

2.0 &

“0

15 5 —

!.oi -
\

(9.5 <D —
5

XI

150 5
o

ioo r

Figure 2. — Concentrations of fecal coliform bacteria and phosphate (PO^-P)
for selected events showing individual observations.

Background concentrations of fecal coliforms were quite low but always
present in surface runoff from the plot. Small mammals, birds, and other
wildlife transmit some of these bacteria to soil. Their subsequent survival
depends on climate and other factors (Geldreich 1966, van Donsel et al. 1967).

Immediate Effects of Manuring

Concentrations of all four bacterial groups increased after the manure
treatment on June 22, but only the fecal coliform group increased by several
orders of magnitude (fig. 1). The fecal coliform counts approached those of
the total coliforms, indicating that many of the coliforms after manure
application were of fecal origin, as indicated by analysis of the manure.
Concentrations of all four groups similarly increased immediately after
treatment, but the relative increase in fecal coliforms was the greatest
between pretreatment and posttreatment. The trends in the fecal coliform
concentrations resembled those observed for phosphate levels in the runoff
water (fig. 2).

6

Long Term Effects - Die-Off

Which bacterial group or combination of groups appeared to be the best
indicator of recent animal excrement? This question was answered by analyzing
the relative decline in bacterial concentrations in the runoff during the weeks
after manuring. As the manure aged, the irrigation was repeated and,
presumably, organisms died off during exposure to the elements and predators.
Among the four bacterial groups, the fecal coliform counts declined much more
rapidly than did those of the other three groups. After about 70 days, the
fecal coliform concentrations returned to pretreatment levels. The total
coliform concentrations also decreased and returned to pretreatment levels
although the recession was less rapid. On the other hand, fecal streptococci
and enterococci remained at essentially the same concentration for many weeks
following the manure application. Therefore, these latter two groups couldn't
be used to distinguish the recent animal excrement in the plot's runoff from
that found after several weeks of ageing.

Other Comparisons

Geldreich (1966) compared fecal coliforms to fecal streptococci using an
FC/FS ratio and found, in controlled laboratory studies, that generally a ratio
under 1 indicates animal fecal influences, whereas a ratio of 4.0 or more
indicates contamination by human waste. In this study, mean ratios were calcu-
lated for the individual runoff samples after manuring. Ratios in the applied
waste averaged 0.53. As shown in table 1, the individual ratios were nearly
all under 1.0, which substantiated Geldreich 's findings.

Table 1. — Averages for the fecal colif orm/f ecal streptococcus (FC/FS) ratios
for samples taken during 13 irrigation events

Date

Mean FC/FS-^

Date

Mean FC/FS

Pretreatment

June

11

0.015 (.002-. 051)

July

8

0.005

(.002-. 010)

June

16

0.132 (.000-. 300)

July

15

0.007

(.00 3-. 014)

June

18

0.002 (.001-. 004)

July

22

0.006

(.000-. 021)

July

29

0.003

(.000-. 006)

Posttreatment

Aug.

12

0.001

(.000-. 002)

Aug.

19

0.017

(.002-. 036)

June

22

0.748 (.071-1.579)

Sept .

16

0.001

(.000-. 004)

June

25

0.040 (.015-. 106)

July

2

0.029 (.011-. 051)

1V Minimum and maximum values of VL4 samples in parenthesis.

7

SUMMARY

The main study objective was to evaluate some common bacterial indices for
discriminating between recently excreted vs. old animal wastes under field
conditions. In addition, I wanted to further substantiate and define previous-
ly published studies which showed that fecal coliform concentrations in streams
could be used to distinguish grazed from ungrazed catchments (Kunkle 1971) .

The results showed that fecal coliform populations in runoff waters
increased substantially more than did three other bacterial groups after
newly excreted cattle manure was added to the plot. Fecal coliforms were an
insignificant fraction of total coliforms in runoff before manuring, but their
concentrations approached those of the total coliforms immediately after
manuring. Concentrations then decreased quickly during the weeks after the
manure had been applied. In about 70 days, their numbers returned to pre-
treatment levels. Conversely, the other bacterial groups did not show a
similar decrease in concentrations. The fecal streptococci and enterococci
remained at about the same concentrations in runoff for many weeks after
manuring.

In conclusion, the fecal coliform group, especially when used in conjunc-
tion with the total coliforms, was a very useful index for discriminating
between runoff contamination from new as compared with old or no manure from a
field. The other bacterial groups tested did not discriminate as well.

LITERATURE CITED

American Public Health Association. 1971. Standard methods for the examina-
tion of water and wastewater. 13th ed., 874 pp.

Buckhouse, J. C., and Gifford, G. F. 1976. Water quality implications of
cattle grazing on a semi-arid watershed in southeastern Utah. J. Range Mgt.
29(2): 109-113.

Geldreich, E. E. 1966. Sanitary significance of fecal coliforms in the

environment. U.S. Dept. Int., Environ. Protect. Agency Pub. WP-20-3, 122 pp . ,
Cincinnati, Ohio.

Kunkle, S. H. 1970. Concentrations and cycles of bacterial indicators in
farm surface runoff. Cornell Univ. Conf . Agr . Waste Mgt. Proc . , Jan.

19-21, 1970.

Kunkle, S. H. 1971. Sources and transport of bacterial indicators in rural
streams. Amer. Soc. Civ. Eng. Symp. Interdisciplinary Aspects of Watershed
Mgt. Proc., Montana State Univ., Aug. 3-6, 1970.

van Donsel, D. J., Geldreich, E. E., and Clarke, N. A. 1967. Seasonal

variations in survival of indicator bacteria in soil and their contribution
to storm-water pollution. Appl. Microbiol. 15(6): 1362-1370.

8

GP 0 860 426

U. S. DEPARTMENT OE AGRICULTURE
SCIENCE AND EDUCATION ADMINISTRATION
NORTHEASTERN REGION
AGRICULTURAL RESEARCH CENTER WEST
0ELTSVILLE, MARYLAND 2070B

OFFICIAL BUSINESS

PENALTY FOR PRIVATE USE, S300

POSTAGE AND FEES PAID
U. S. DEPARTMENT OF
AGRICULTURE

AGR 101