The distribution of drinking water-to-cattle ratios in the summer across four feedlots in the Texas High Plains

doi:10.4236/as.2013.46040

Paper Menu >>

Journal Menu >>

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: [email protected], [email protected]

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

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)

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