Water and forage intake, diet digestibility, and blood parameters of beef cows and heifers consuming water with varying concentrations of total dissolved salts

Abstract

The objective of this study was to investigate the effects of water quality on water intake (WI), forage intake, diet digestibility, and blood constituents in beef cows and growing beef heifers. This was a replicated 5 × 5 Latin square with five drinking water treatments within each square: 1) fresh water (Control); 2) brackish water (100 BRW treatment) with approximately 6,000 mg/kg total dissolved solids (TDS); 3) same TDS level as 100 BRW achieved by addition of NaCl to fresh water (100 SLW); 4) 50% brackish water and 50% fresh water to achieve approximately 3,000 mg/kg TDS (50 BRW); and 5) same TDS level as 50 BRW achieved by addition of NaCl to fresh water (50 SLW). Each of the five 21-d periods consisted of 14 d of adaptation and 5 d of data collection. Animals were housed individually and fed mixed alfalfa (Medicago sativa) grass hay cubes. Feed and WI were recorded daily. Data were analyzed with animal as the experimental unit. Age, treatment, and age × treatment were fixed effects, and animal ID within age was the random variable for intake, digestibility, and blood parameter data. Water and feed intake were greater than expected, regardless of age or water treatment. No treatment × age interactions were identified for WI (P = 0.71), WI expressed as g/kg body weight (BW; P = 0.70), or dry matter intake (DMI; P = 0.21). However, there was an age × treatment tendency for DMI when scaled to BW (P = 0.09) in cows consuming 100 BRW compared with fresh water. No differences were found for the other three treatments. Heifers provided 50 SLW water consumed less (P < 0.05) feed (g/kg BW) compared with heifers provided fresh water and 100 BRW. No differences (P > 0.05) in water, DMI, feed intake, or diet digestibility were found due to water quality treatment. In conclusion, under these conditions, neither absolute WI, absolute DMI, nor diet digestibility was influenced by the natural brackish or saline water used in this experiment. These results suggest that further research is necessary to determine thresholds for TDS or salinity concentration resulting in reduced water and/or feed intake and diet digestibility.

Introduction

In many parts of the United States, rates of groundwater withdrawal are greater than those of recharge, resulting in decreased groundwater supplies, lower stream and lake levels, and/or land subsidence, with an increased reliance on water other than fresh water (Masters et al., 2005; El Shaer, 2010). Low rainfall can cause an increase in total dissolved solids (TDS) in the soil and groundwater (Yousfi et al., 2016), potentially affecting the groundwater used for livestock. Understanding how the quality of water available for drinking can impact water consumption and efficiency of livestock production is critical for current and future livestock production enterprises as well as the safety and security of the food supply. One characteristic of water quality is the concentration of total dissolved or soluble salts, often termed as the level of salinity. Drinking water with TDS above 1,000 mg/kg (i.e., mg/L) is termed “saline,” and brackish water usually refers to water with TDS between 1,000 and 10,000 mg/kg but can be much higher (USGS, 2013; Stanton et al., 2017; Stanton and Dennehy, 2017).

Cattle frequently consume water moderate to high in TDS globally. Brackish and saline ground water sources are widespread, including in countries such as Egypt (Assad and El-Sherif, 2002), Australia (McGregor, 2004), India (Sharma et al., 2017), Tunisia (Yousfi et al., 2016), Brazil (Castro et al., 2017), Asia, the eastern Mediterranean, Africa (Masters et al., 2005), and most of the central United States (Androwski et al., 2010; USGS, 2013).

Because of the increasing importance of drinking water with high TDS in the foreseeable future, it is important to gain an understanding of factors affecting the utilization of brackish/saline water by ruminant livestock. If adverse effects of consumption of drinking water moderate to high in TDS are documented, management practices could be altered to lessen the effects. Moreover, differences among species exist, such as generally greater salt tolerance by goats than sheep and older than young small ruminants (McGregor, 2004) and perhaps lower tolerance of cattle vs. goats and sheep (Squires, 1988; SCA, 1990). However, little research has been conducted in beef cattle.

Apart from the TDS level in water consumed by ruminants, estimates of the amount of fresh water required and consumed by ruminant livestock and methods of prediction vary considerably. While more recent data are available for growing (Brew et al., 2011) and finishing (Arias and Mader, 2011; Sexson et al., 2012; Ahlberg et al., 2018; Zanetti et al., 2019) cattle, the latest data available for estimating fresh water intake (WI) for mature beef cows were published in 1956 (Winchester and Morris, 1956). The NASEM (2016) recommendation of the water requirement of beef cows was based on these data with 400 kg cows despite today’s mature beef cows averaging between 488 (Walker et al., 2015) and 630 kg (McMurry, 2008). Therefore, the objective of this study was to determine the effects of age and water quality on WI, dry matter intake (DMI), digestion, and blood constituent concentrations in beef cows and heifers.

Materials and Methods

All procedures for animal use were approved by the Oklahoma State University Institutional Animal Care and Use Committee (AR-18-RS-270). This experiment was conducted at the Nutrition and Physiology Research Center (NPRC) of Oklahoma State University (Stillwater, OK, USA).

Animal and diet management

The experimental design was two simultaneous 5 × 5 Latin squares using cows and growing heifer calves similar to the design employed by Yirga et al. (2018) as described by Snedecor and Cochran (1967). The five drinking water treatments within each square included: 1) fresh water (Control); 2) brackish water (100 BRW treatment) with approximately 6,000 mg/kg TDS; 3) same TDS level as 100 BRW achieved by addition of NaCl to fresh water (100 SLW); 4) 50% brackish water and 50% fresh water to achieve approximately 3,000 mg/kg TDS (50 BRW); and 5) same TDS level as 50 BRW achieved by addition of NaCl to fresh water (50 SLW). Each of the five periods consisted of 14 d of treatment adaptation and 5 d of data collection (Harris, 1970; Kaufmann et al., 1980; Van Soest, 1982; Merchen, 1988). Five Angus cows (1,106 +/− 31 d, 547 +/− 46 kg) and five Angus heifers (362 +/− 10 d; 265 +/− 30 kg) were alternatingly placed and housed in individual 2.44 × 4.73 m partially enclosed concrete stalls with rubber mats covering the slats over a manure pit to ensure total fecal collection and improve animal comfort (Horn et al., 1954). Each pen was equipped with a covered 2.4 m concrete feed bunk and a 56-L rubber container used to provide drinking water. Ambient temperature (°C) and relative humidity (%) during the experimental period (September to December) are presented in Table 1.

Table 1.

Monthly rainfall, ambient temperature, and relative humidity¹

		Temperature, °C			Relative humidity, %
Month/year	Rainfall, cm	Mean	Min.	Max.	Mean	Min.	Max.
September 2019	2.56	26.2	19	35	72.6	33	100
October 2019	1.53	13.5	−5	33	69.1	17	99
November 2019	0.82	8.22	−10	25	67.2	14	100
December 2019	0.23	6.56	−7	22	65.5	16	100

¹Source: Oklahoma Climatological Survey and the Oklahoma Mesonet, Stillwater, OK.

Mixed grass/alfalfa (Medicago sativa) cubes (Table 2) and water (Table 3) were offered ad libitum. Throughout adaptation and experimental periods, animals were offered 120% of the preceding day’s DMI. Feed and water were delivered once daily at 0700 hours with additional feed and water delivered in the afternoon as needed to ensure ad libitum intake. Brackish water was transported from a well located at the American Institute for Goat Research, Langston University (Langston, OK, USA), as needed and stored in a large water tank at the research facility. Water treatment mixtures were made every other week or as needed and stored individually in clearly labeled intermediate bulk containers, including fresh water, so that conditions were similar among water treatments. Conductivity, pH, TDS, and salinity were measured with a Pocket Pro+ (Hach, Loveland, CO) during mixing to ensure consistency. Before delivery to animals, each water storage container was stirred for 2 min to minimize solid separation and accumulation at the bottom of the tank. The paddle used to stir tanks was rinsed with fresh water between each use. During the collection period, 19-L buckets of water for each treatment were set near the stalls and weighed at the beginning and end of the collection period to account for water vaporization. A mineral block was offered on an ad libitum basis to each animal (American Stockman, Stillwater Milling Co., Stillwater, OK, USA). The mineral block contained between 96% and 99% sodium chloride (NaCl), 2,400 ppm Mn, 2,400 ppm Fe, 260 to 380 ppm Cu, 320 ppm Zn, 70 ppm I, and 40 ppm Co. Animals were weighed 4 h post feeding at 3-wk intervals.

Table 2.

Chemical composition of alfalfa cubes¹

Item2	Load 1 (period 1 to 3)3	Load 2 (period 4 to 5)3	SEM	P-value
Nutrient
DM, %	95.4	95.2	0.05	0.06
OM, %	88.7	87.8	0.13	<0.001
CP, %	17.5	16.6	0.23	0.03
NDF, %	49.5	47.3	0.43	0.002
ADF, %	31.7	31.5	0.25	0.35

¹McCracken Hay, Elgin, OK, USA. Ad libitum access was provided throughout each of the five 21-d periods.

²ADF, acid detergent fiber; CP, crude protein; DM, dry matter; NDF, neutral detergent fiber; OM, organic matter.

³Feed was delivered in two loads. Load 1 was fed during periods 1 through 3, and load 2 was fed during periods 4 and 5.

Table 3.

Composition of water consumed by growing Angus heifers and Angus cows

	Treatment1
Item	Control	50 BRW	50 SLW	100 BRW	100 SLW	SEM
Total dissolved solids, mg/kg	276.3e	2,852d	3,309c	5,263b	5,878a	130.5
Sodium, mg/kg	38.9e	755.1d	1,001c	1,394b	1,860a	42.4
Sodium, %	4.07d	7.47c	9.48b	7.58c	9.71a	0.76
Calcium, mg/kg	34.5c	162b	34.3c	285.3a	35.8c	3.87
Magnesium, mg/kg	8.25c	36.9b	7.79c	62.6a	6.57c	0.62
Potassium, mg/kg	4.77c	4.69c	6.54b	5.69b,c	9.54a	0.59
Nitrate-N, mg/kg	0.91a	0.42b	0.76a	0.63a,b	0.80a	0.11
Chloride, mg/kg	33.2e	628d	1,585b	1,243c	2,782a	58.6
Sulfate, mg/kg	36.7c	1,038b	37.1c	1,985a	41.2c	17.2
Boron, mg/kg	0.10c	4.34b	0.09c	8.27a	0.12c	0.07
Bicarbonate, mg/kg	137.6c	184.2b	144.2c	210.4a	143.7c	4.46
Alkalinity2, mg/kg	112.8c	151.0b	118.2c	172.5a	117.8c	3.65
Hardness3, mg/L	119.9c	556.0b	117.1c	969.5a	116.5c	10.9
SAR4	1.54e	13.9d	40.2b	19.5c	75.5a	1.96
PAR5	0.10c	0.06c	0.15b	0.02d	0.23a	0.02
EC6, dS/m	405e	3,722d	4,846c	7,055b	8,700a	217.9
pH	8.00	8.70	7.96	8.01	7.98	0.29

¹Treatment descriptions: Control, fresh water; 50 SLW, 50% NaCl and 50% fresh water; 50 BRW, 50% brackish water and 50% fresh water; 100 BRW, 100% brackish water; 100 SLW, 100% NaCl water.

²Alkalinity as CaCO_3.

³Hardness guidelines = Soft: 0 to 60 mg/L; Moderately hard: 61 to 120 mg/L; Hard: 121 to 180 mg/L; and Very hard: ≥181 mg/L.

⁴SAR, sodium adsorption ratio.

⁵PAR, potassium adsorption ratio.

⁶EC, electrical conductivity.

^a–eWithin a row, water treatment means without a common superscript letter differ (P < 0.05).

Feed and fecal collection

About 125 g of grass/alfalfa cubes was collected each day during each collection period. Orts were collected daily prior to feeding and immediately weighed. Feed and ort samples were dried at 55 °C in a forced-air oven for 72 h, or until no additional weight loss was detected, and dried sample weights were determined to calculate dry matter (DM) content. Dried ort samples were pooled within period for each animal, thoroughly mixed, and a 100 g subsample was collected for grinding and chemical analysis. Feed and orts were ground through a 1-mm screen using a Fritsch Pulverisette 19 Cutting Mill (Markt Einersheim, Germany).

Total fecal excretion was determined for five consecutive days during each collection period. Fecal material was collected at 0700 and 1900 hours each day, stored in a sealed plastic trash bag, and weighed immediately after collection. Each day, 5% of the a.m. collection and 5% of the p.m. collection were combined in a foil pan and thoroughly mixed. From this daily pooled sample, a 23-cm disposable aluminum foil pan was filled to level, weighed, and immediately placed in a forced-air drying oven. Samples were stirred twice daily and dried at 55 °C for 72 h or until no further weight loss was detected. All samples were ground to pass through a 1-mm screen using a Fritsch Pulverisette 19 Cutting Mill (Markt Einersheim, Germany). After grinding, 50 g from each of the five daily fecal samples within a period were combined and thoroughly mixed. From this pooled sample, 100 g of dried, ground fecal sample was retained for chemical analyses. Feed and fecal samples were stored in re-sealable plastic bags to minimize moisture accumulation before and after grinding.

Blood collection

Blood samples were collected via jugular venipuncture in vacutainer tubes 4 h post feeding on day 14 and 4 h post feeding on day 19 (the last day of each period). A portion of blood was collected in 5-mL tubes coated with sodium fluoride and potassium oxalate, another portion of blood was collected in 9-mL tubes coated with sodium heparin, and a third portion was collected in 10-mL dry tubes for serum. Blood was permitted to clot on ice, and serum was separated by centrifugation at 2,000 × g for 10 min at 4 °C.

Immediately after sampling, whole blood in green-top tubes with sodium heparin was used to determine packed cell volume (PCV) with a Micro-hematocrit centrifuge (PSS Select; Model DSC-030MH) and hemoglobin, oxygen saturation, and Met- and CO-hemoglobin (MetHb, HbCO) with an OMS3 Hemoximeter (Radiometer America, Westlake, OH, USA). Whole blood oxygen concentration was calculated from total hemoglobin (tHb) and oxygen (O₂) saturation of hemoglobin (HbO₂), according to Eisemann and Nienaber (1990):

$${\displaystyle \begin{array}{l}{\displaystyle {\mathrm{O}}_2\left(\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{l}/\mathrm{L}\right)=(\left(\mathrm{H}\mathrm{b}{\mathrm{O}}_2/100\right)\ast \mathrm{t}\mathrm{H}\mathrm{b}\ast 10\ast 1.34)/22.4}\end{array}}$$

where O₂ = whole blood oxygen concentration, mmol/L, HbO₂ = hemoglobin oxygen saturation, %, and tHb = total hemoglobin, g/dL.

Plasma was collected from sodium heparin-coated tubes after centrifuging for 20 min at approximately 3,000 × g at 10 °C and was used to determine osmolality with a model 2020 Osmometer (Advanced Instruments, Inc., Norwood, MA, USA). Serum and plasma were frozen at −20 °C, and plasma from tubes with sodium fluoride and potassium oxalate was thawed and used to determine glucose and lactate concentrations with a YSI 2300 Plus Glucose & Lactate Analyzer (YSI Inc., Yellow Spring, OH). Thawed serum was analyzed for albumin (ALB), alanine aminotransferase (ALT), alkaline phosphatase (ALP), aspartate aminotransferase (AST), creatine kinase (CK), gamma-glutamyl transferase (GGT), Ca, total cholesterol (CHOL), chloride (CL⁻), creatinine (CREAT), glucose, lactate, Mg, K, Na, total triglycerides (TRIG), and blood urea nitrogen (BUN) with a Vet Axcel Chemistry Analyzer (Alfa Wassermann Diagnostic Technologies, West Caldwell, NJ) according to manufacturer’s instructions.

Feed and fecal analysis

All ground samples were analyzed for analytical DM by drying at 105 °C for 24 h. Following DM analysis, feed and fecal samples were combusted at 500 °C for at least 5 h to determine the ash content. Organic matter (OM) content was calculated by subtracting ash % from 100. Samples were also analyzed for acid detergent fiber (ADF) and neutral detergent fiber (NDF), with amylase and sodium sulfite used during NDF determination (ANKOM, 2020; method 15). Acid detergent insoluble ash was measured as the residue from ADF after ashing at 500 °C for at least 5 h (Van Soest et al., 1991). Crude fat (CF) was determined using ANKOM (2020; solvent extraction method for XT¹⁵) with ether extraction (EE, ANKOM^XT15 Extractor). Crude protein (CP) was determined by multiplying N (Leco TruMac CN, St. Joseph, MI, USA) concentration for each sample by 6.25. Both feed and fecal samples were analyzed simultaneously to avoid potential run error. Apparent total tract digestibility of DM (DMD) was based on feed DM intake and fecal DM output. Organic matter digestibility (OMD) was based on feed OM intake and fecal OM output. Intake of digestible OM (DOMI) was calculated by multiplying OM intake (OMI) by OM digestibility and is reported as g/kg BW.

Statistical analyses

Water chemical composition was analyzed as a repeated measure over time using the MIXED procedure in SAS (SAS 9.4; SAS Institute Inc., Cary, NC). The model statement included water treatment and sampling week. The repeated measure was sampling week. Data collected for WI, DMI, BW, blood constituents, total fecal output, digestibility, and water components were analyzed using the MIXED procedure in SAS (SAS 9.4; SAS Institute Inc., Cary, NC) as a side by side 5 × 5 Latin square where animal was the experimental unit. Animal within age class was used as a random variable for intake, digestibility, and blood parameters data. Period was used as a repeated measure for all analysis, and animal within age class was the subject for intake, digestibility, and blood variable analysis. Means separation was by least significant difference with a protected F-test at α ≤ 0.05. Significance was declared if P ≤ 0.05, and tendencies were declared at P > 0.05 and P ≤ 0.1.

Results and Discussion

Water composition

Least square means for the main effect of treatment on water chemical composition are presented in Table 3. Criteria used to classify water quality include organoleptic properties (odor and taste), physiochemical properties (TDS, pH, hardness, and O₂), toxic compounds, excess mineral compounds, and bacteria (NASEM, 2016). The United States has set National Primary and Secondary Drinking Water Standards for humans to ensure safe, palatable drinking water. Water should have a pH between 6.5 and 8.5, less than 500 mg/L TDS, and less than 1 mg/L of nitrites and 10 mg/L of nitrates, among other components. The fresh water (control) used in this experiment fell within both primary and secondary water standards with chlorine falling between the primary and secondary standards (4 mg/L maximum residual disinfectant level goal; 250 mg/L maximum contaminant level).

Brackish water is defined as water that contains 1,000 to 10,000 mg/L of TDS and can be classified as one of four groups (USGS, 2013; Stanton et al., 2017). Water group definitions of Stanton et al. (2017) are group 1: sodium bicarbonate-dominant water type in which sulfate contributes about one-third of the total anion equivalents; group 2: calcium sulfate-dominant water type in which sodium and magnesium each contribute about one-quarter of the total cation equivalents; group 3: sodium chloride-dominant water type and has a high mean concentration of TDS (8,440 mg/L); and group 4: mixture of cations and anions with low TDS (1,360 mg/L) and a high percentage of silica (1.7% of the total moles of cations and anions). In this experiment, the 100 BRW treatment would be classified in group 3. Additionally, treatment silica concentrations were much lower in this experiment than suggested for group 4.

Water with TDS < 3,000 mg/kg has historically been deemed safe for livestock consumption, and no negative side effects should be expected other than potential initial diarrhea (NRC, 2001). Stanton et al. (2017) define fresh water or safe water to be <1,000 mg/L. From 3,000 to 5,000 mg/kg, diarrhea should be expected and intake may be suppressed, inhibiting maximum performance (NASEM, 2016). From 5,000 mg/kg and greater, water should be avoided (NRC, 2001) for pregnant and lactating animals. Anything beyond 7,000 mg/kg should be avoided entirely (NRC, 2001). Therefore, the 100 BRW and 100 SLW treatments would be considered “unsafe” for livestock according to these guidelines (NRC, 2001). Similarly, according to these guidelines, the 50 BRW and 50 SLW would be considered safe for livestock.

Hardness is defined as the sum of calcium and magnesium reported in addition to other cations such as zinc, iron, and selenium (NASEM, 2016). According to water hardness guidelines in NASEM (2016), hardness (mg/L) from 0 to 60 is considered soft, from 61 to 120 mg/L is considered moderately hard, from 121 to 180 mg/L is considered hard, and ≥181 mg/L is considered “very hard.” Therefore, Control, 50 SLW, and 100 SLW treatments would be classified as “moderately hard.” Both brackish water treatments employed in this experiment would be classified as “very hard.”

All water treatments were below nitrate-nitrogen (NO_3-N) standards recognized as safe for beef cattle (less than 10 mg/kg; NASEM, 2016). Sulfate can be detrimental to water and DMI, and general recommendations are less than 500 mg/L for calves and less than 1,000 mg/L for adult cattle. Water sulfate concentration means for 50 BRW and 100 BRW treatments were greater (P < 0.01) than Control, 50 SLW, and 100 SLW and well beyond the recommended maximum concentrations. Sulfate concentrations in Control, 50 SLW, and 100 SLW were similar (P > 0.85) and well below these recommendations. Concentrations of Ca, Mg, B, and bicarbonate were greater (P < 0.01) in 100 BRW than in 50 BRW, and these minerals were more concentrated (P < 0.01) in 50 BRW than Control, 50 SLW, and 100 SLW. Sodium and Cl concentrations were 33% and 123% greater (P < 0.01) in 100 SLW than in 100 BRW, respectively. Electrical conductivity was greatest (P < 0.01) in 100 SLW and declined with declining TDS. There was no difference among water treatments in pH (P = 0.33).

Boron concentration in the 100 BRW treatment was considered toxic according to the 5 mg/L threshold concentration suggested by National Academy of Sciences (1972) for livestock drinking water. However, Green and Weeth (1977) reported no acute toxicity effects when drinking water for growing beef heifers contained up to 300 ppm Boron. These researchers noted reduced DMI and weight loss when water contained ≥150 ppm Boron. Green and Weeth (1977) suggested safe tolerance for Boron in drinking water between 40 and 150 ppm.

WI, feed intake, and digestion

Least square means for WI, DMI, and feed component digestibility are presented in Table 4. There were no significant age × treatment interactions (P ≥ 0.10), and, therefore, only means for main effects are reported. At 120 g/kg BW, water consumption for cows in this study was high compared with current predicted WI (Spencer et al., 2017) for mature cows and previous observed WI (Winchester and Morris, 1956) for mature cows. However, Spencer et al. (2017) estimated WI requirements for nonlactating cows consuming 22 g/kg BW in DMI compared with this study’s 36 g/kg BW in DMI. Winchester and Morris (1956) suggested a linear relationship between DMI and WI, which could account for a portion of the increased daily water consumption observed in this study.

Table 4.

Effects of age and water treatment on water intake, feed intake, and apparent total tract diet digestibility

	Age		Water treatment1						P-value
Item2	Heifer	Cow	Control	50 BRW	50 SLW	100 BRW	100 SLW	SEM3	Age	Treatment	Age × Treatment
Intake
BW, kg	310.4	604.2	457.5	458.1	460.1	454.4	456.5	14.0	<0.001	0.81	0.83
Water, kg/d	44.2	72.6	60.9	55.4	59.0	55.5	61.2	2.29	<0.001	0.10	0.71
Water, g/kg BW	144.9	120.7	139.2	126.9	132.6	126.8	138.6	5.72	0.006	0.28	0.70
DM, kg/d	12.6	22.1	18.0	17.5	17.3	16.8	17.0	0.59	<0.001	0.37	0.21
DM, g/kg BW	40.5	36.4	40.1	38.5	37.4	38.0	38.2	1.08	0.02	0.31	0.09
OM, kg/d	11.1	19.5	15.9	15.4	15.2	14.8	15.0	0.52	<0.001	0.37	0.20
DOM, kg/d	6.83	12.1	9.76	9.50	9.45	9.3	9.3	0.40	<0.001	0.83	0.69
Digestibility
DM, %	60.9	61.5	61.3	60.9	61.1	61.4	61.4	0.32	0.11	0.77	0.44
OM, %	61.0	61.7	61.1	60.9	61.6	61.9	61.4	0.80	0.51	0.87	0.83
NDF, %	55.9	57.6	57.9	55.2	55.0	58.7	57.0	2.48	0.50	0.77	0.40
ADF, %	52.2	53.9	54.3	51.1	51.3	55.2	53.3	2.71	0.56	0.75	0.43
CP, %	91.5	86.1	88.9	88.4	88.4	89.3	89.1	0.48	<0.001	0.61	0.28
EE, %	41.9	46.1	49.1	38.9	37.0	46.5	48.4	4.34	0.30	0.18	0.16

¹Treatment: Control, fresh water; 50 SLW, 50% NaCl and 50% fresh water; 50 BRW, 50% brackish water and 50% fresh water; 100 BRW, 100% brackish water; 100 SLW, 100% NaCl water.

²ADF, acid detergent fiber; BW, body weight; CP, crude protein; DM, dry matter; DOM, digestible organic matter; EE, ether extract; NDF, neutral detergent fiber; OM, organic matter.

³SEM, pooled standard error.

Similarly, WI in this study was high compared with previous reports for growing cattle. Sexson et al. (2012) reported growing steers consumed 21 g/kg BW in DMI of a total mixed ration and 79 g/kg BW in WI with an average ambient temperature of 21.96 °C. Arias and Mader (2011) found heifers consuming 21 g/kg BW in DMI consumed 72 g/kg BW in WI with average ambient temperature of 21.4 °C. With heifers consuming twice as much DMI g/kg BW in this experiment at an average temperature of 13.6 °C, it is not surprising to see WI, reported as g/kg BW, to double if WI and DMI have a constant relationship suggested by Winchester and Morris (1956). Winchester and Morris (1956) also reported WI to be constant up to 4.4 °C ambient temperature. Other researchers have also reported a constant relationship between DMI and WI (Murphy et al., 1983; Hicks et al., 1988; Loneragan et al., 2001). Additionally, animals fed a similar high-quality diet of alfalfa hay (Williams et al., 2018) consumed an average DMI of 34.5 g/kg BW, which is close to the DMI consumption observed in this study. Overall, the exceptionally high level of weight-adjusted WI observed in this experiment can at least partially be explained by the high level of DMI.

As expected, heifers consumed less water on an absolute basis (kg/d) than did cows. However, when expressed relative to BW, heifers consumed 24.2 g/kg BW more water (P = 0.006) compared with cows. This finding agrees with previous research of Sexson et al. (2012). Cattle weighing less than 500 kg had increased water consumption as BW increased. Conversely, cattle that weighed greater than 500 kg had decreased WI as BW increased (Sexson et al., 2012). The decline in WI associated with greater BW could be explained by the change in composition of gain, with an increasing proportion of fat and decreasing proportion of protein and water (NRC, 2000).

There was a tendency (P = 0.10) for a treatment effect when WI was expressed as kg/d. While daily consumption (kg/d) of 50 BRW and 100 BRW did not differ (P = 0.97), the tendency was due to less consumption of the two brackish treatments (50 BRW and 100 BRW) compared with fresh water and 100 SLW (P < 0.05). Patterson et al. (2004) also described lower consumption of brackish water containing ~3,000 mg/kg TDS up to ~7,000 mg/kg TDS compared with water with ~1,000 mg/kg TDS. When expressed relative to BW, water treatment did not influence water consumption (P = 0.28).

Like WI, consumption of alfalfa cubes was copious in this experiment with an average of 38 g/kg BW (Table 4). For example, in a recent experiment using Angus cows from the same herd, nonlactating cows consumed 28 g/kg BW of a grass hay and molasses-based liquid supplement diet (55.7% total digestible nutrients; Andresen et al., 2020). Williams et al. (2018) also reported copious DMI when lactating Angus beef cows consumed alfalfa cubes (34 to 35 g/kg BW, DM basis). As expected, cows consumed more (kg/d) DM, OM, and digestible OM (P < 0.001) than heifers. However, heifers consumed more feed DM than cows when expressed per unit of BW (P = 0.02).

There was a tendency (P = 0.09) for an age × treatment interaction for DMI when expressed as g/kg BW (Table 4). This tendency was due to cows consuming 100 BRW having lower (P = 0.05) DMI compared with cows consuming fresh water (34.4 and 38.3 g/kg BW, respectively), whereas there was no difference in DMI for heifers consuming 100 BRW compared with fresh water (42.0 and 41.6 g/kg BW, respectively; P = 0.82). Further work is needed to determine if brackish drinking water influences DMI differentially depending on age.

The NRC (2005) estimates that 1 g of NaCl/kg of BW can be consumed by ruminants with no effect on DMI. In this experiment, while assigned to the 100 SLW treatment, heifers and cows consumed 0.66 and 0.56 g/kg BW NaCL from the water source, respectively. Therefore, with no reduction in DMI found in the current experiment due to added NaCl in drinking water, these results support the previously published guidelines (NRC, 2005). The highest sulfate level in this experiment was 1,985 mg/kg in the 100 BRW treatment which is below the threshold (2,500 mg/kg) reported to cause a reduction in DMI in several experiments (Weeth and Hunter, 1971; Weeth and Capps, 1972; Digesti and Weeth, 1976; Loneragan et al., 2001; Grout et al., 2006; López et al., 2014, 2016). There was no difference among water treatments for OMI.

Neither age nor water treatment affected DOMI, DMD, OMD, NDF, ADF, or EE (P ≥ 0.11). Similar results were reported by López et al. (2016). Alves et al. (2017) recorded a decrease in NDF digestibility when cattle consumed water containing ~3,000 mg/kg TDS to 8,000 mg/kg TDS. Nevertheless, under these conditions, up to 5,878 mg/kg TDS did not influence diet or component digestibility. Heifers had greater CP digestibility (P < 0.001) than cows. Blaxter et al. (1966) reported greater N retention in 15-wk-old steers compared with 81-wk-old steers, primarily due to greater protein tissue energy retention in the younger animals.

Serum mineral concentrations, blood metabolites, and hematological indicators

Least square means for blood constituent concentrations are presented in Table 5. Concentrations of Cl, Ca, K, Mg, and Na were all within normal range for cattle (Fielder, 2015), although, according to Zelal et al. (2017), Mg concentrations approached hypomagnesemia in heifers at 1.96 mg/dL. However, Harvey and Bruss (2008) consider 1 to 1.8 mg/dL to be low and generally resulting in no clinical deficiency signs. An age effect was found for Ca (P = 0.007) with cows having lower serum Ca than heifers.

Table 5.

Effects of age and water treatment on blood constituent concentrations

	Age		Water treatment1						P-value
Item2	Heifer	Cow	Control	50 BRW	50 SLW	100 BRW	100 SLW	SEM3	Age	Treatment	Age × Treatment
Cl−, mmol/L	103.6	104.3	103.5	104.3	104.0	103.4	104.6	0.49	0.17	0.29	0.88
Ca, mg/dL	10.4	10.03	10.1	10.2	10.2	10.3	10.3	0.12	0.007	0.65	0.28
K, mmol/L	4.59	4.47	4.40	4.57	4.49	4.54	4.65	0.09	0.19	0.43	0.18
Mg, mg/dL	1.96	2.13	2.11a	2.10a	2.09a	2.05a,b	1.90b	0.06	0.02	0.08	0.03
Na, mmol/L	142.2	143.2	142.3	143.3	142.6	142.6	143.0	0.42	0.07	0.40	0.35
ALB, U/L	3.07	3.21	3.17	3.23	3.08	3.14	3.10	0.06	0.03	0.47	0.88
ALT, U/L	19.4	19.1	19.3	20.3	19.2	17.7	19.6	1.63	0.89	0.71	0.09
AST, U/L	62.2	57.4	61.5	58.8	61.5	54.5	62.7	4.21	0.44	0.46	0.65
ALP, U/L	92.1	42.3	71.6	61.3	64.3	66.1	72.7	6.14	0.0003	0.49	0.81
GGT, U/L	16.9	21.3	18.7	19.2	18.9	19.8	19.1	1.09	0.06	0.61	0.15
CK, U/L	185.3	144.4	179.9	166.6	158.9	162.4	156.6	9.54	0.0016	0.45	0.82
CREAT, mg/dL	0.95	0.94	0.93b	0.99a	0.92b	0.96a,b	0.96a,b	0.02	0.62	0.05	0.27
TRIG, mg/dL	30.8	29.7	29.8	31.0	29.6	31.4	29.6	1.39	0.59	0.67	0.82
CHOL, mg/dL	106.3	133.3	125.0	124.4	119.3	114.1	116.2	6.17	0.02	0.36	0.62
BUN, mg/dL	20.1	23.9	23.2a	22.4a,b	22.2a,b,c	21.0d	21.3c,d	0.76	0.02	0.0018	0.04
Glucose, g/L	0.46	0.46	0.46	0.47	0.46	0.45	0.46	0.02	0.79	0.98	0.19
Lactate, g/L	0.17	0.15	0.14	0.16	0.15	0.13	0.20	0.02	0.49	0.10	0.78
Osmolality	286.8	284.5	282.1	287.1	290.0	285.9	283.3	2.29	0.36	0.12	0.72
PCV, %	34.4	35.7	34.7	36.4	35.0	35.5	33.7	0.81	0.12	0.23	0.52
tHb, g/dL	12.9	13.5	13.2	13.5	13.4	13.2	12.7	0.35	0.16	0.44	0.66
HbO2, %	66.9	72.1	71.4	69.8	68.9	68.2	69.3	3.39	0.13	0.97	0.73
O2, mmol/L	12.1	13.7	13.3	13.6	13.0	12.3	12.2	0.75	0.07	0.64	0.87
HbCO, %	1.94	2.13	2.11	2.02	1.96	1.97	2.13	0.19	0.29	0.95	0.69
MetHb, g/dL	0.45	0.58	0.50b,c	0.43b,c	0.66a	0.41c	0.57a,b	0.05	0.08	0.008	0.59

¹Treatment: Control, fresh water; 50 SLW, 50% NaCl and 50% fresh water; 50 BRW, 50% brackish water and 50% fresh water; 100 BRW, 100% brackish water; 100 SLW, 100% NaCl water.

²ALB, albumin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CHOL, total cholesterol; CK, creatine kinase; CREAT, creatinine; GGT, gamma-glutamyl transferase; HbCO, carboxyhemoglobin; HbO2, hemoglobin oxygen saturation; MetHb, methemoglobin; PCV, packed cell volume; tHb, total hemoglobin; TRIG, total triglycerides.

³SEM, pooled standard error.

^a–dMain effect water treatment means without a common superscript letter differ (P < 0.05).

An age × treatment interaction (P = 0.03) was found for blood Mg concentration. Least square means were 2.17, 1.98, 1.87, 2.0, and 1.8 mg/dL for heifers and 2.0, 2.2, 2.3, 2.1, and 2.0 for cows consuming Control, 50 BRW, 50 SLW, 100 BRW, and 100 SLW, respectively (data not shown). Heifers consuming 50 SLW and 100 SLW had lower serum Mg compared with Control (P ≤ 0.015), whereas cows consuming 50 SLW had greater serum Mg concentration compared with Control (P = 0.03). An explanation for this interaction is unclear. Perhaps the slightly elevated K concentration in the SLW treatments resulted in reduced Mg absorption in heifers (Zelal et al., 2017). Neither age nor treatment effects nor age × treatment interactions were found for serum Na concentration.

Heifers had lower (P = 0.03) serum ALB concentration compared with cows. However, according to Harvey and Bruss (2008), ALB concentrations were within normal range for cattle (3.0 to 3.55 g/L). The difference of 0.14 U/L ALB, or about 4.5% greater (P = 0.03) concentration in cows, may not be biologically significant given both age classes were considered to be within the normal range.

A trend (P = 0.09) for age × treatment interaction was found for hepatic ALT although ALT concentration was within normal range for cattle (11 to 40 U/L; Ingvartsen, 2006; Fielder, 2015), and thus, the interaction was not deemed to be biologically significant. The main and interaction effects were not different (P ≥ 0.44) for hepatic AST. Nevertheless, AST concentrations were lower than the normal range reported for cattle (78 to 132 U/L; Ingvartsen, 2006; Fielder, 2015). High activity for hepatic AST and ALT is most often indicative of acute or chronic liver disease. Increased serum AST activity is considered a sensitive marker for identifying liver damage, even if the damage is subclinical (Kauppinen, 1984; Meyer and Harvey, 1998). In contrast, ruminant liver cells do not show high ALT enzyme activity, and increased serum activity from liver damage, even in necrosis, is insignificant (Forenbacher, 1993). The most sensitive marker to diagnose acute liver damage is ALT, while AST is more sensitive in reflecting the degree of damage (Kew, 2000).

An age effect for ALP (P < 0.01) was found with heifers having over twice the concentration (92.1 U/L) of cows with the normal range reported between 7 and 43 U/L (Putnam et al., 1986). Often, there is a linear relationship between activity of serum ALP and GGT in cholestatic liver injury (Meyer, 1983). Although GGT concentrations in this study are not increased as dramatically as ALP, they are still elevated overall. This relationship is presented in Table 5 with GGT overall being slightly above the normal range for cattle (Fielder, 2015). Therefore, further work is necessary to determine if long-term exposure to high-TDS water may increase risk of liver damage or malfunction. Greater serum ALP concentration is often observed in normal growing or adult animals with increased osteoblastic activity (Sun et al., 2015). Higher serum ALP concentration was also observed in young beef calves compared with their dams (Hidiroglou and Thompson, 1980). The skeletal growth of calves is the primary site of ALP activity and likely contributes most of the difference documented in this experiment (Moog, 1946). Hidiroglou and Thompson (1980) also suggested increased concentrations of serum ALP in young animals were the result of bone growth and development.

An age effect (P = 0.0016) was found for CK with heifers having higher serum concentrations than cows, but CK levels were within the normal range for both heifers and cows (0 to 350 U/L). A tendency for a treatment effect from the 50 SLW treatment was noted for CREAT (P = 0.052), although concentrations were within the normal range for both cows and heifers (0.5 to 2.2 mg/dL).

BUN was within the normal range (20 to 30 mg/dL) for both cows and heifers (Harvey and Bruss, 2008), although an age × treatment interaction (P = 0.04) was found for BUN concentration. Least square means were 20.5, 20.9, 21.0, 18.5, and 19.5 mg/dL in heifers and 25.8, 23.9, 23.4, 23.5, and 23.0 in cows consuming Control, 50 BRW, 50 SLW, 100 BRW, and 100 SLW, respectively (data not shown). Heifers consuming 100 BRW had lower (P ≤ 0.01) BUN than heifers consuming Control, 50 BRW, and 50 SLW. However, BUN concentration was similar among cows consuming 50 BRW, 50 SLW, 100 BRW, and 100 SLW (P ≥ 0.5). Moreover, when cows consumed the Control water source, BUN was elevated (P ≤ 0.02) compared with BUN concentration when cows consumed the other four water sources. Taken together, these data may indicate slightly lower N utilization when cattle consume diets with higher levels of TDS. Overall, BUN concentration was greater (P = 0.02) in cows than in heifers and this was unexpected because heifers consumed more feed per unit of BW and had greater CP digestibility. These results could be interpreted as either improved N utilization in the growing heifers or increased protein catabolism in the cows (Blaxter et al., 1966).

No effects or interactions were identified for total serum TRIG, but concentrations for total TRIG in both cows and heifers were twice the concentration considered normal (0 to 14 mg/dL; Harvey and Bruss, 2008). Cows had higher serum CHOL concentration than heifers (P = 0.02) although there was no treatment effect. Concentrations of CHOL for both ages were elevated compared with standard levels (58 to 88 mg/dL; Harvey and Bruss, 2008).

No effects of age, treatment, or age × treatment interaction were found for serum glucose concentrations with both age classes within the normal range reported for cattle (0.45 to 0.75 g/L; Fielder, 2015). There was a tendency for a treatment effect (P = 0.1) for serum lactate concentration, although no effect of age or age × treatment interaction. Lactate concentrations were within the normal range for cattle (0.05 to 0.2 g/L) according to Harvey and Bruss (2008).

There were no age or treatment effects for osmolality, PCV, O₂, tHb, HbO₂, and HbCO. Osmolality (270 to 300 mOsm/kg; Harvey and Bruss, 2008), PCV (24 to 46%; Fielder, 2015), and tHb (8 to 15 g/dL; Fielder, 2015) were all within normal range for cattle. A treatment effect occurred for MetHb (P = 0.008) due to both BRW treatments being depressed and both SLW treatments being elevated, but no other effects or interactions were found. There was a tendency (P ≤ 0.08) for MetHb and O₂ concentrations to be greater in cows than in heifers and may be explained by greater blood volume and circulation due to greater body and organ size.

Conclusions

Based on the results of this experiment, brackish or saline water up to 6,000 mg/kg TDS had little to no effect on WI, DMI, or diet digestibility in beef cows and heifers. Most blood mineral and metabolite concentrations were within normal ranges reported in the literature, suggesting little metabolic stress derived from consuming water containing greater concentrations of TDS from a natural brackish source or addition of sodium chloride. Further research is necessary to determine thresholds for TDS or salinity concentration in beef cattle water sources as well as the potential effects of long-term exposure to high-TDS water.

Glossary

Abbreviations

ADF

acid detergent fiber

BUN

blood urea nitrogen

BW

body weight

CF

crude fat

CHOL

total cholesterol

CK

creatine kinase

CP

crude protein

DM

dry matter

DMD

dry matter digestibility

DMI

dry matter intake

DOMI

digestible organic matter intake

EC

electrical conductivity

EE

ether extract

IBC

intermediate bulk containers

NDF

neutral detergent fiber

OM

organic matter

OMD

organic matter digestibility

OMI

organic matter intake

TDS

total dissolved solids

WI

water intake

Conflict of interest statement

The authors declare no conflicts of interest.