Vol.4, No.6, 282-286 (2013) Agricultural Sciences
http://dx.doi.org/10.4236/as.2013.46040
The distribution of drinking water-to-cattle ratios
in the summer across four feedlots in the
Texas High Plains
Raju Gautam1*, Pablo J. Pinedo2, Sangshin Park1, Renata Ivanek1
1Department of Veterinary Integrative Biosciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M Univer-
sity, College Station, USA; *Corresponding Author:
2Texas A&M AgriLife Research & Extension Center, College of Veterinary Medicine and Biomedical Sciences, Texas A&M Uni-
versity, Amarillo, USA
Received 16 April 2013; revised 17 May 2013; accepted 1 June 2013
Copyright © 2013 Raju Gautam et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
In this short communication, we report the
findings of a cross-sectional pilot study of the
amount of water available per head of cattle
(water-to-cattle ratio) and the associated feedlot
and environmental factors across 26 pens in
four Texas feedlots. The water-to-cattle ratio
varied greatly among pens within and between
feedlots. Mixed-effect linear regression model-
ing with feedlot as a random effect indicated that
water in troughs with a higher water-to-cattle
ratio was generally warmer when compared with
water in troughs with a lower water-to-cattle ra-
tio. This may have implications in the transmis-
sion and persistence of pathogens in feedlot
cattle, such as Shiga toxin-producing Esche-
richia coli and Salmonella, because warmer wa-
ter has been reported to favor the growth of
these pathogens. Therefore, future field studies
in feedlot cattle are warranted to assess whether
the water-to-cattle ratio a ffects the prevalence of
these pathogens in the water itself or in feces
shed by the animals.
Keywords: Water-Trough; Feedlot; Water-to-Cattle
Ratio
1. INTRODUCTION
Drinking water is a critical component of everyday
nutrition of feedlot cattle. Commercial feed lot operations
manage drinking water in water troughs with an auto-
matic refill system. The water refill system is designed to
turn on and off automatically depending on the preset
holding capacity of the water trough. Thus, water starts
to flow into the trough when the water level goes below
the preset holding capacity level, and the flow stops
when the level is reached. Such a system for regulating
the volume of water in the water trough minimizes water
loss due to overflow, while ensuring the supply of an
adequate amount of drinking water to the cattle at all
times.
The daily water requirements of a feedlot cattle vary
anywhere from 4 - 12 gallons per head of cattle depend-
ing on animal weight, diet, and ambient temperature [1].
Maintaining an adequate amount of clean drinking water
for feedlot cattle is therefore essential to the welfare and
productivity of animals. However, excess amounts of
standing drinking water, particularly during warmer
months, may also encourage pathogen growth (e.g.,
Shiga toxin-producing Escherichia coli (STEC)), in-
creasing their concentration in water, and thereby facili-
tating the infection transmission among animals [2]. This
undesirable effect is expected to be amplified in periods
and regions characterized with warm weather because
pathogen growth is strongly modulated by temperature [3,
4]. However, there is a lack of information about the wa-
ter-to-cattle ratio and its variability in feedlot o perations.
Likewise, there is a knowledge gap in our understanding
of how ambient temperature may affect the temperature
of cattle drinking water in the feedlot. We report a cross-
sectional pilot study of 26 pens across four feedlots in the
Texas Panhandle. The objectives of the study were: 1) to
obtain information about variation in the ratio of the
amount of drinking water over the number of cattle per
pen under the existing feedlot operating conditions and 2)
to determine the relationship between the ambient tem-
perature and the drinking water temperature in feedlot
pens.

R. Gautam et al. / Agricultural Sciences 4 (2013) 282-286 283
2. MATERIALS AND METHODS
We conducted a questionnaire-based survey of 13
feedlots in the Texas Panhandle to obtain preliminary
information about the distribution of pen level water-
to-cattle ratios and to elicit the feedlot’s willingness to
participate in the subsequent cross-sectional study. A
total of 124 feedlot pens were covered during the survey.
The feedlots were identified through one of the investi-
gators’ (PP) extension contacts. Of these 13 feedlots, 11
expressed their willingness to participate in the cross-
sectional study. Due to financial restrictions, only four of
these feedlots were enrolled. The choice was based on
the convenience related to the proximity among feedlots
and the feedlot managers’ cooperativeness, both of which
were necessary to allow intensive sampling to be per-
formed over two days, covering two feedlots per day.
Six pens were selected for enrollment in three out of four
enrolled feedlots, and eight pens were selected in the
fourth feedlot. Purposive samplin g was used in the selec-
tion of the feedlot pens, such that in a specific feedlot
one-half of the enro lled p ens co nstituted a h igh er number
of animals per water trough and the other half consisted
of a lower number of animals per water trough. This
sampling strategy was implemented with the aim to cap-
ture the full extent of variation in the water-to-cattle ra-
tios. A structured questionnaire was administered to the
feedlot managers by personal interview to obtain infor-
mation about the number of animals in the pen and the
time in feeding. The number of water troughs in the pen,
and whether the water troughs were exposed to the sun
or managed under a shade were recorded by personal
observations, while ambient and drinking water tem-
peratures were measured using a graduated thermometer
(Testo 110, maker: Testo). Measurement of the drinking
water pH was made using a digital pH meter with tem-
perature compensation (OYSTER-10, maker: Extech).
Electrical conductivity of water was measured using
portable probes (Digital Conductivity meter 09-327,
maker: Fisher Scientific Traceable). The time of day the
measurements of ambient and water temperatures were
taken for a given pen was the same, but the time of day
varied between pens. Information on water trough capac-
ity was obtained by physical measurements of the length,
width, and height of the water column in the wa-
ter-trough using a measuring tape. To minimize measur-
ing and recoding bias, all measurements were performed
by one investigator (RG). Finally, water-to-cattle ratio
was calculated by dividing the volume of water in the
trough(s) with the number of cattle in the pen.
Summary statistics were calculated for all recorded
variables. A statistical model was developed to assess the
association of different pen-level factors (i.e., drinking
water temperature, water pH, water conductivity, length
of time in feeding, and the water trough exposure to the
sun) to the water-to-cattle ratio (the outcome variable).
Screening of the individual variables for association with
the outcome variable was performed by fitting a univari-
ate regression model using a liberal cutoff (α = 0.25) for
the significance of association. In this analysis, the asso-
ciation between the outcome (water-to-cattle ratio) and
the temperature of drinking water was assessed by using
a variable that represented the difference between the
ambient and water temperatures for a given pen (hereaf-
ter referred to as the “temperature difference”) to account
for the difference in measurement times during the day
between pens. Variables that had a potential association
with the outcome (i.e., a P-value < 0.25) based on the
univariate analysis were considered in a multivariable
mixed-effect linear regression model, which included the
feedlot as the random effect to account for clustering of
pens within feedlots. The final multivariable model was
selected using a forward selection procedure based on
the Akaike’s information criterion (AIC) of all nested
models and the significance of the variables in the model
was determined at α = 0.05. Model assumptions for nor-
mality and equal variance were assessed by generating
normal probability and residual plots using the model
residuals.
Statistical analysis was also conducted to assess the
potential relationship between the temperature of the
drinking water (the dependent variable) and ambient
temperature of the pen (the explanatory variable). The
relationship was assessed based on the graph ical plotting
of all data points and a linear regression model devel-
oped for the subset of data on water troughs (from 20
pens) that were exposed to the sun. All statistical analy-
ses were performed in R 2.13.1 version (R Development
Core Team, 2011) of the software package for statistical
computing.
3. RESULTS AND DISCUSSIONS
The overall distribution of the water-to-cattle ratio in
the 13 surveyed feedlots is shown in Figure 1(a). The
median water-to-cattle ratio was 0.25 gallons/head with a
median absolute deviation (MAD) of 0.11. While the
middle 50% of the water-to-cattle ratios were in a rela-
tively narrow range between 0.2 and 0.34 gallons/head,
the bottom and particularly the top quartiles indicated
considerable variation (Figure 1(a)). Figure 1(b) illus-
trates the distribution and variation in the water-to-cattle
ratios for each of the four feedlots enrolled in the
cross-sectional study. For these four feedlots, the median
water-to-cattle ratio was 0.30 gallons/head with MAD of
0.24. The purposive selection of pens with lower and
higher values of water-to-cattle ratio captured the tails of
the distribution of water-to-cattle ratios with an over

R. Gautam et al. / Agricultural Sciences 4 (2013) 282-286
284
(a) (b)
(c) (d)
Figure 1. (a) A box plot of the distribution of the overall water-to-cattle ratio in 13 surveyed feedlots; (b) The pen-specific water-to-
cattle ratios for each of the four feedlots enrolled in the cross-sectional study. In the lower panel, the relationship between the
ambient temperature and drinking water temperature managed (c) under the shade and (d) exposed to the sun.
representation of the higher values (Figure 1(b)). How-
ever, notably, some pens had the water-to-cattle ratio of
1.5 gallon/head or above while the highest ratios reported
in the survey were at around 1 .2 gallons/head suggesting
that the su rvey-based water-to-cattle ratios were underes-
timated due to the recall bias. Alternatively, the differ-
ence may be reflective of the dynamic nature of the wa-
ter-to-cattle ratio, which may have changed if the number
of animals moving in and out of the pens changed be-
tween the two time points of the study. Figures 1(c) and
(d) show the relationships between the temperature of
drinking water and the ambient temperature of the pen
for drinking water managed under shade and exposed
directly to the sun, respectively. Based on the visual in-
spection of the plots, there was no apparent increase in
the drinking water temperature with an increase in the
ambient temperature of the pen for water maintained
under the shaded area. However, when water was di-
rectly exposed to the sun, the water temperature in-
creased linearly with the increase in the ambient tem-
perature of the pen. For every 1˚C increase in the ambi-
ent temperature, the temperature of drinking water di-
rectly exposed to the sun increased on an average by
0.53˚C, SE = 0.08 (P-value < 0.0001, adjusted R2 =
0.71).
For the model exploring the relationship between the
pen-level factors and water-to-cattle ratio, univariate
analysis indicated two fixed effects (the “temperature

R. Gautam et al. / Agricultural Sciences 4 (2013) 282-286 285
difference” and water conductivity) to be linearly related
with the water-to-cattle ratio. The variance component of
the random effect was 0.18, which constituted approxi-
mately 50% of the total variation and the remainder (0.2)
was attributed to the residual error. The large amount of
variation attributed to the random component suggests
that there is a wide variation in the water-to-cattle ratio
from one feedlot to another. For every 1˚C increase in the
“temperature difference” between the ambient and water
temperatures there was a corresponding decrease of
0.085 (SE = 0.044) gallons (300 ml)/cattle in the wa-
ter-to-cattle ratio (P-value = 0.03) when controlling for
water conductivity and the feedlot. This means that pens
with a lower water-to-cattle ratio tend to have cooler
drinking water. This can be explained by a faster refilling
of the water troughs with fresh water, which implies that
the standing drinking water in the troughs is exposed to
high ambient temperature for a shorter period of time.
On the contrary, a high water-to-cattle ratio is related to a
slower rate of trough refilling and consequently the water
is exposed to high ambient temperature for a longer pe-
riod of time. In other words, water in troughs with high
water-to-cattle ratio is generally warmer. Because warmer
water may promote faster growth of bacteria important to
food safety [3,4], it would appear that having a lower
water-to-cattle ratio, which does not adversely affect the
adequate supply of drinking water, may actually be a tool
to control transmission of the infections mediated th rough
contaminated water. Alternatively, a system to allow con-
tinuous flow of water in the water trough may be used to
prevent steep rise o f drinking water temperatur e with the
increase in ambient temperature during a summer day.
However, lowering the water-to-cattle ratio is likely to be
a more attractive choice to the feedlot owners because it
would benefit the beef industry through a better use of an
already limited water resource in feedlot operations in
the Texas High Plains. One of the important factors that
could affect the refill rate of water in the pen is the ag e of
animals because heavier (adult) cattle in the feedlot are
expected to be drinking more water and thus increase the
rate of refill even if the water-to-cattle ratio is the same.
While we did not have information on age, we used ani-
mal time in feeding as a proxy to assess if it significan tly
affected the water-to-cattle ratio. The univariate analysis
did not suggest that time in feeding significantly affected
the observed water-to-cattle ratio in this study and th ere-
fore, we believe that the age effect could be ignored.
Previously, it had been reported that the prevalence of
E. coli O157:H7 in weaned calves could be reduced by
replacing large-volume water tanks with small volume
water troughs that facilitated high turnov er rate of drink-
ing water [5]. Several other studies tried to relate con-
tamination of drinking water with shedding prevalence of
E. coli O157:H7 and the findings were inconclusive [6-8].
However, these studies did not consider the potential
effect of increased water temperature on pathogen repli-
cation and the consequence of higher bacterial concen-
trations on E. coli O157:H7 shedding prev alence in cattle.
While the variation in the water-to-cattle ratio observed
in this study, between 0.2 and 1.2 gallons/head, may not
at the first glance appear meaningfully wide, the varia-
tion is however, sufficient to allow for a significantly
increased temperature of drinking water for higher values
of the water-to-cattle ratios. Thus, the results of this
study suggest that the assessment of the relationship be-
tween water-to-cattle ratio and prevalence of pathogens
in feedlots, such as E. coli O157:H7, is worth pursuing
further. Similarly, the information on replication of E.
coli O157:H7 in drinking water in feed lots is lacking and
whether the replicatio n reported by Vital et al. [3 ] in dis-
tilled water applies to the drinking water in feedlots
needs to be verified under experimental conditions.
The association between the water-to-cattle ratio and
water conductivity was marginally significant (P-value =
0.07). For every one unit increase in water conductivity,
there was an associated decrease in water-to-cattle ratio
of 0.15 gallons (0.568 ml)/head of cattle (SE = 0.1) after
controlling for the temperature difference and the feedlot.
In other words, water that refills faster has higher con-
ductivity. The validity of and the mechanism behind this
association is unclear. Generally, conductivity of water is
influenced by the concentration of inorganic dissolved
solids such as chloride, nitrate, sulphate, sodium, and
magnesium, as well as water temperature [9].
The pilot study described here has important limita-
tions due to the small number of enrolled feedlot opera-
tions, the narrow geographic study area covered, and the
observational study design used. However, the findings
may be of importance to the beef industry and food
safety. Moreover, while no prior information was avail-
able to allow estimation of sample size in the current
study, this report could support sample size estimation in
the future.
4. CONCLUSION
There is a wide variation in the water-to-cattle ratio
among feedlots in the Texas Panhandle. The water-to-
cattle ratio was associated with temperature, which was
an important factor that affected pathogen (e.g., STEC
and Salmonella) growth in drinking water for animals.
Therefore, future field studies should be conducted to
assess the association between the water-to-cattle ratio
and the prevalence of the corresponding infections in
feedlots.
5. ACKNOWLEDGEMENTS
This work was supported by the Texas A&M University’s College of
Veterinary Medicine Graduate Student Research Proposal Award to RG
and the National Science Foundation grant NSF-EF-0913367 to RI

R. Gautam et al. / Agricultural Sciences 4 (2013) 282-286
286
funded under the American Recovery and Reinvestment Act of 2009.
Any opinions, findings, and conclusions or recommendations expressed
in this material are those of the authors. The authors gratefully ac-
knowledge the assistance of the Texas Cattle Feeders Association in
contacting the participating feedlots.
REFERENCES
[1] Boyles, S., Loerch, S., Fluharty, F., Shulaw, W. and
Stanfield, H. Feedlot management primer. In: Beef In-
formation, Ohio State University Extension.
http://beef.osu.edu/library/feedlot/feedlot.pdf
[2] Gautam, R., Bani-Yaghoub, M., Neill, W.H., Döpfer, D.,
Kaspar, C. and Ivanek, R. (2011) Modeling the effect of
seasonal variation in ambient temperature on the trans-
mission dynamics of a pathogen with a free-living stage:
Example of Escherichia coli O157:H7 in a dairy herd.
Preventive Veterinary Medicine, 102, 10-21.
doi:10.1016/j.prevetmed.2011.06.008
[3] Vital, M., Hammes, F. and Egli, T. (2008) Escherichia
coli O157 can grow in natural freshwater at low carbon
concentrations. Environmental Microbiology, 10, 2387-
2396. doi:10.1111/j.1462-2920.2008.01664.x
[4] Wang, G., Zhao, T. and Doyle, M.P. (1996) Fate of en-
terohemorrhagic Escherichia coli O157:H7 in bovine fe-
ces. Applied Environmental Microbiology, 62, 2567 -2570.
[5] Kaspar, C.W. (2009) Waterborne dissemination of Es-
cherichia coli O157:H7. Unite States Food and Drug
Administration.
http://www.fda.gov/AnimalVeterinary/NewsEvents/CVM
Updates/ucm138286.htm
[6] LeJeune, J.T., Besser, T.E., Merrill, N.L., Rice, D.H. and
Hancock, D.D. (2001) Livestock drinking water microbi-
ology and the factors influencing the quality of drinking
water offered to cattle. Journal of Dairy Science, 84,
1856-1862. doi:10.3168/jds.S0022-0302(01)74626-7
[7] LeJeune, J.T., Besser, T.E., Rice, D.H., Berg, J.L., Stil-
born, R.P. and Hancock, D.D. (2004) Longitudinal study
of fecal shedding of Escherichia coli O157:H7 in feedlot
cattle: Predominance and persistence of specific clonal
types despite massive cattle population turnover. Applied
Environmental Microbiology, 70, 377-384.
doi:10.1128/AEM.70.1.377-384.2004
[8] LeJeune, J.T., Besser, T.E. and Hancock, D.D. (2001)
Cattle water troughs as reservoirs of Escherichia coli
O157. Applied Environmental Microbiology, 67, 3053-
3057. doi:10.1128/AEM.67.7.3053-3057.2001
[9] Environmental Protection Agency (2012) Conductivity. In:
Water: Monitoring and Assessment, Washington DC.
http://water.epa.gov/type/rsl/monitoring/vms59.cfm