You can lead a cow to water…but can you make her drink?

Water. The simplest of compounds, two hydrogen atoms bound to an oxygen atom, and yet it is the critical solvent for most of terrestrial life’s sustaining reactions. Water is the transport medium for delivery of nutrients to our cells and removal of waste from our organs. It is the solvent that carries messages within tissues locally and to distant organs.

It is the principal mediator of body temperature for warm-blooded animals. Water is so crucial to mammalian life, that a loss of 20% of the body’s water is fatal to most organisms.

Despite the central role that water plays in mammalian physiology, we tend to take it for granted. It is even more crucial to animals that produce copious quantities of milk, a product that is roughly 87% water. All too often, we fail to provide this crucial nutrient in adequate quantities or in acceptable quality for our cows.

Deficiencies in either quantity or quality can have effects ranging from undetectable to profound. This discussion will touch briefly on water quantity and availability, the water cycle and its impact on water chemistry, integrate some concepts of water chemistry into our understanding of rumen function, and lastly consider some of the primary detrimental constituents found in water.


Water quantity and availability

It is no surprise that the modern lactating dairy cow needs a lot of water. As a very general rule of thumb, free water intake in liters per day, exclusive of water contained in feeds, averages 2.0 times milk production in kg.

For example, and for those occasions when there is no calculator handy, a cow producing 35 kg of milk per day will require about 70 L of water per day. Obviously, this is a gross simplification and only useful for average production levels, intakes and ambient temperatures. When more precision is needed, Holter and Urban (1992) have provided a free water intake prediction equation from a study that included more cows, over a wider range of diets, intakes and conditions than had previously been reported (Little and Shaw, 1978; Murphy et al., 1983).

FWI = -32.39 + 2.47 × DMI + 0.6007 × milk + 0.6205 × DM % + 0.0911 × JD - 0.000257 × JD²

(R2=0.69, P<0.0001, n=329)
FWI – Free water intake, exclusive of water consumed in feed, L/day
DMI – Dry matter intake, kg/day
Milk – milk production, kg/day
DM % - dietary dry matter %
JD – Julian calendar day, where January 1 = 1

However, the National Research Council’s 2001 Dairy NRC recommends use of the equation of Murphy and co-workers (1983) because the milk coefficient is closest to representing the water content of milk and the other factors are all known to influence free water intake. That equation is:

FWI = 15.99 + 1.58 × DMI + 0.9 × Milk + 0.05 × Na + 1.2 × minimum temperature o C

(R²=0.59, P<0.05, n=18)
FWI – Free water intake, exclusive of water consumed in feed, L/day
DMI – Dry matter intake, kg/d
Milk – Milk production, kg/d
Na – Sodium intake, g/d

Assessing water consumption on the farm is not difficult as long as the water supply line to the cows can be isolated. Commercial water meters, up to 650 L/min flow rates, are readily available for a few hundred dollars and can be installed to monitor water intakes for groups of cows. In many instances, we have been able to make corrections to water supplies for cows and measure consumption responses of over 50%.

Leaving aside the issues of water quality, optimal water intake assumes that water is available in sufficient locations, free of obstructions and at adequate rates of flow to meet the needs of groups of cows at peak drinking times. For cows housed in freestalls, water troughs should be spaced about every 30-35 meters down the length of the barn.

At that spacing a cow will never be more than 15 meters from a water station. Drinking behavior of cows is such that, given the opportunity, they will alternate feeding and drinking episodes. Peak water consumption will coincide with peak dry matter intake (Nocek and Braund, 1985). There should also be enough linear space at watering stations to accommodate 7–10% of the group simultaneously (McFarland, 2003).

Cows housed in tie-stall facilities presumably have access to water 24 hrs per day.

However, flow rate of water into water bowls will have a significant effect on consumption.

Anderson et al. (1984) reported that as flow rate into water bowls increased from 2 L/ min to 12 L/min, water consumption increased from 77.4 L/day to 87.4 L/day. In grazing systems, particularly in arid or hot conditions, the preference is to have water available within the paddock. However, in research from both Scotland and Australia, for cows grazing lush forage with <25% DM, milk yield was not affected by restricting water access to twice daily while the cows were being milked (Castle and Watson, 1973; King and Stockdale, 1981).


Where water comes from

The source of water has a large impact on its quality. Regardless of the source of water for cows, whether it is a well, pond, river or municipal system, it has picked up solutes along the way. What those solutes are and in what quantities they occur depends on the atmospheric and geologic conditions of the regional water cycle (Figure 1).




Figure 1. The water cycle.



As water evaporates from open water bodies, or transpires from vegetative cover, then condenses and falls back to earth as rain, sleet, hail or snow, it will have picked up constituents from the atmosphere. Carbon dioxide, nitrogen oxide and sulfur dioxide are the three most common, forming weakly acidic solutions ranging from pH 5.6 down to pH 4.3.

Upon its return to earth, this acidic solution will either run off the surface to be collected in lakes, ponds and rivers or infiltrate the ground and work its way down to the water table. In either case, the water dissolves some minerals and takes on the characteristics of the soils and bedrock through which it travels. Hence, an appreciation of the geology which influences your water sources is warranted.

A few generalizations will suffice for this discussion of groundwater sources:

• The ability of water to dissolve minerals depends on the type of rock through which water flows, such that: sedimentary > metamorphic > igneous, or in terms of common rocks: limestone or dolomite > shale or sandstone > granite or basalt

• Concentration of dissolved solids increases with depth of the water source: deep well (277 mg/L) > shallow well (243 mg/L) > spring (150 mg/L), depth being implicitly linked to residence time in the groundwater system (Fleeger, 1999).

Acquiring a geologic map of your region is an easy and instructive first step. A geologic map provides clear direction to the types of minerals likely to be found in a given groundwater source. For instance, a water sample from a dairy near Cornwall, PA, which has underlying high iron Triassic diabase bedrock, has an iron level of 14.5 ppm, one of the highest observed by this author.


Water chemistry, mineral speciation, pH and the strong ion difference

As a result of its passage through the water cycle, water acquires a chemical composition other than simply H₂O. Within water, each mineral or element can exist as a hydrated ion, as a molecule, or be complexed with another ion or molecule. The forms that an ion takes are referred to as its speciation. The dissolved and suspended mineral constituents of water will dissociate and re-associate with other ions to achieve the lowest possible energy state. The solutes thus formed represent a wide array of ionic species that vary depending on the concentrations of all the other constituent solutes (Stumm and Morgan, 1996).

Using the hydrogeochemical model PHREEQC (Parkhurst and Appelo, 1999) one can calculate the speciation of minerals in a water sample. As an example, Figure 2 demonstrates that as the level of sulfate increases from 20 to 1200 ppm, everything else being constant, the concentration of free iron (Fe⁺²) increases, with a concomitant decrease in the other iron species. Similarly, Figure 3 depicts the effect of increasing sulfate concentration, from 120 ppm to 1200 ppm, on the concentration of a few selected ion species in water.

To add one more level of complexity, we must consider also the solubility of each of these solute species. The relative solubility or insolubility will largely govern the availability in, and the extent of absorption from, the rumen. Table 1 has speciation and solubility parameters for calcium and iron in a representative water sample. Solubility is expressed as K_sp, the solubility product, and is the molar concentration at which that mineral specie reaches saturation and becomes insoluble (Stumm and Morgan, 1996).

The larger the K_sp, the higher the degree of solubility. The conclusion to be drawn from this discussion is that the ultimate form, concentration and availability of ingested inorganic minerals within the rumen and/or intestine is highly dependent on these interactions, whether the minerals come from water or trace mineral supplementation.

Therefore, simply knowing the concentration of a mineral in water or feed really tells us nothing of its speciation or solubility and hence its availability or potential for toxicity (NRC, 2005).

One reason for this speciation behavior is the requirement that aqueous solutions maintain electrical neutrality. The means by which water achieves electrical neutrality is the instantaneous dissociation and association of the water molecule itself such that the H+ and OH- ions are available to associate with unpaired ions of the opposite charge.

The ions that exert the largest effect are those which dissociate completely in water, with dissociation constants > 10^-4 Eq/L, and are known as the strong ions (Stewart, 1983). These strong ions are generally familiar to us in the dairy nutrition field because they are also the ones we include when calculating the Dietary Cation Anion Difference or DCAD, notably Na⁺, K⁺, Mg⁺⁺, Ca⁺⁺, Cl^- , PO₄ -, and SO₄ -.

In a seminal 1983 publication, How to Understand Acid-Base, Peter Stewart outlines the principles of the ‘strong ion difference’(SID), a version of which we now know as DCAD (Goff, 2006). A study of this principle leads one to the understanding that pH is not, as most of us understand it, an independent variable, but rather a dependent variable, driven entirely by the strong ion difference of a solution (Constable, 2003). Consequently, to consider or attempt to mitigate aberrant pH levels is to focus on the effect rather than the cause.




Figure 2. Effect of varying concentrations of sulfate on free iron concentration of water.





Figure 3. Concentrations of select ion species in water of high or low sulfate concetration.



Table 1. Ion specie of calcium and iron, concentration and solubility in a representative water sample.





DCAD and SID, water quality effects on cows

Most research on the effects of individual water constituents, namely minerals, on dairy cows has yielded unremarkable results. This has led many to conclude that water-borne minerals have little, if any, effect even at elevated levels.

Given the foregoing discussion of the complex interactions of ion species and their relative insolubility in an aqueous solution, i.e., rumen fluid, it is not surprising that naturally occurring minerals in drinking water sources rarely lead to acute toxicoses. Rather, elevated mineral levels tend to result in more subtle, chronic conditions of poor performance or increased health problems (NRC, 2005). These effects may be mediated through alteration of rumen function (Durand and Komisarczuk, 1988), antagonisms amongst the minerals (Spears, 2003), or oxidative stress (Miller et al., 1993).

Recent research in France and New Zealand has demonstrated significant effects of increasing DCAD in lactating dairy cows. Apper-Bossard and co-workers (2006) provided diets with DCAD values of 0, 150 or 300 mEq/kg of dry matter to cows consuming either high or low starch diets. Linear effects of DCAD on dry matter intake, milk production and fat-corrected milk were observed only in the high starch group.

Roche and co-workers (2005) drenched grazing cows twice daily to achieve DCAD levels of 230 to 880 mEq/kg of dry matter. There was a linear effect of DCAD on fat-corrected milk yield. On the basis of indigestible markers, they observed no change in dry matter intake, although they surmised that observed increases in milk fatty acids were consistent with increased dry matter intake. Interestingly, both studies measured increases in cis-9, trans-11 conjugated linoleic acid with the high DCAD diets, a result suggestive of altered rumen fermentation patterns (Lock et al., 2006).

Conversely, providing a water source high in chloride and sulfate (e.g., a negative CAD (cation-anion difference)) to lactating dairy cows significantly reduced milk yield (Sanchez et al., 1994). Water supplies with negative CAD of -1.9 mEq/L (Solomon et al., 1995) and -4.4 mEq/L (Challis et al., 1987) demonstrated similar reductions in milk production. When water with a negative CAD or SID is introduced to the rumen of a modern dairy cow consuming high quantities of rapidly fermentable carbohydrate and fiber with the concomitant production of anionic lactate, volatile fatty acids and non-esterified fatty acids, it may be enough to exacerbate sub-acute rumen acidosis.

In fact, recent work (Constable, personal communication) suggests that the beneficial effects of higher DCAD diets are due to an improved rumen fermentation. In summary, the rumen does not care whether the source of ions is from water or feed. It responds to an excess of anions, a negative CAD or SID, with perturbations in fermentation and hence its efficiency.

I recently summarized results from 617 water samples submitted to the DairyOne Water Lab in Ithaca, NY (Tables 2 and 3). Of note, is the wide range in SID values, from -46 mEq/L up to 339 mEq/L. If water with an SID of -1.9 mEq/L can cause significant decreases in production as in Sanchez (1994), imagine what a water supply with an SID of -46 mEq/L might do to the rumen.

In a 1994 farm case in which I was involved, we were able to change dry matter intakes by 3 lb/cow/day by switching between two water sources. One well supplied water with 300 ppm sulfate, the other had 1200 ppm sulfate. At the time, we attributed the depression in dry matter intake to the high sulfate concentration, per se. Re-examining those water samples in terms of SID, the high sulfate water supply had an SID of -8.07 mEq/L whereas the lower sulfate water had an SID of 1.14 mEq/L.

Considering the extent of variation in SID of water supplies, and the potential of its impact on the efficiency of fermentation in the rumen, I believe it is warranted to evaluate water samples not only for the individual mineral constituents, but also for SID.


Table 2. Characteristics of 617 water samples analyzed by the DairyOne Water Lab: strong ion contents.





Table 3. Characteristics of 617 water samples analyzed by the DairyOne Water Lab: pH, SID and weak ions.





Individual compounds of potential concern in water

There are many minerals which, if consumed in excess, can have deleterious effects on animals. These have recently been reviewed in an exhaustive publication by the NRC, Mineral Tolerance of Animals (2005). Likewise, the potentially harmful minerals in water have been written about in numerous publications (NRC, 2001; Beede, 1991). A brief discussion of a few factors which may be of concern in water follows.


NITRATE

While nitrate can be used in the rumen as a source of nitrogen for microbial protein synthesis, its reduction to nitrite can be a cause of concern when levels are elevated.

Nitrite is rapidly absorbed into the bloodstream where it can reduce the oxygen carrying capacity of hemoglobin. Water nitrate levels in excess of 133 ppm, or nitrate-N levels in excess of 20 ppm may lead to poor growth, infertility, abortions and general ill thrift (NRC, 1974). Of the 617 samples analyzed by the DairyOne Water Lab, only six had nitrate levels in excess of 133 ppm.


SULFUR

The level at which sulfur becomes a problem is not well defined, except in the case of hydrogen sulfide (H₂S). Hydrogen sulfide is the form that causes the distinctive odor of rotten eggs. Levels as low as 0.1 ppm can reduce water intake, though this is probably more an aesthetic effect of taste or odor. Because of similar physical and chemical properties, sulfur and selenium are antagonistic.

In lactating cows, increasing levels of sulfur result in a linear decline in plasma selenium (Ivancic and Weiss, 2001). The three-way interaction between sulfur, molybdenum and copper in the formation of thiomolybdates results in a reduction in the net availability of copper (Spears, 2003). As sulfur levels in water increase above about 1000 ppm, a mild and transient laxative effect may be observed, though cattle will adapt to this level (NRC, 2001).


BACTERIA

Sanitary standards for dairies place strict limitations on the presence of bacteria in water supplies. Generally, <1 colony forming unit/100 mL is recommended as a safe level.

More specifically, an analysis of coliforms is recommended if the sample is bacteriologically positive. If fecal coliform count > fecal streptococci, the source of contamination is likely human. When the reverse is true, the source of contamination is likely of animal origin (NRC, 2001).

There may also exist an interaction between mineral levels and bacterial populations in water supplies. Sulfur-reducing bacteria can flourish in low oxygen environments, such as wells and water heaters, and will reduce sulfur and sulfate to hydrogen sulfide, with resulting odor and palatability issues.

Similarly, sulfur-oxidizing bacteria utilize hydrogen sulfide and convert it to sulfate. This becomes an issue only in that, like iron and manganese-reducing bacteria, they form a black, smelly, filamentous slime in water troughs and tanks that can reduce palatability and consumption. The potential effects of these bacteria, as toxic principles, is at this time unknown, but worthy of investigation.

References

Anderson, M., J. Schaar and H. Wiktorsson. 1984. Effects of drinking water flow rates and social rank on performance and drinking behavior of tied-up dairy cows. Livest. Prod. Sci. 11:599.

Apper-Bossard, E., J.L. Peyraud, P. Faverdin and F. Meschy. 2006. Changing dietary cation-anion difference for dairy cows fed with two contrasting levels of concentrate in diets. J. Dairy Sci. 89:749.

Beede, D.K. 1991. Mineral and water nutrition. Vet Clin. North Am. Food Anim. Pract. 7:373.

Castle, M.E. and J.N. Watson. 1973. The intake of drinking water by grazing dairy cows: the effect of water availability. J. Br. Grassland Soc. 28:203.

Challis, D.J., J.S. Zeinstra and M.J. Anderson. 1987. Some effects of water quality on the performance of high yielding cows in an arid climate. Vet. Rec. 120:12.

Constable, P.D. 2003. Hyperchloremic acidosis: The classic example of strong ion acidosis. Anesth. Analg. 96:919.

Durand, M. and S. Komisarczuk. 1988. Influence of major minerals on rumen microbiota. J. Nutr. 118:249.

Fleeger, G.M. 1999. The geology of Pennsylvania’s groundwater. Pennsylvania Geological Survey (3rd ed.), 4th ser., Educational Series 3, p. 34.

Goff, J.P. 2006. Mineral disorders of the transition period: origin and control. Proceedings of the World Buiatrics Congress. Nice, France.

Holter, J.B. and W.E. Urban, Jr. 1992. Water partitioning and intake in dry and lactating Holstein cows. J. Dairy Sci. 75:1472.

Ivancic, J. and W.P. Weiss. 2001. Effect of dietary sulfur and selenium concentrations on selenium balance of lactating Holstein cows. J. Dairy Sci. 84:225.

King, K.R. and C.R. Stockdale. 1981. Milk yield of dairy cows given restricted access to water in a Mediterranean-type climate. Aus. J. Exp Ag. Anim. Husb. 21:167.

Little, W. and S.R. Shaw. 1978. A note on the individuality of the intake of drinking water by dairy cows. Anim. Proc. 26:225.

Lock, A.L., T.R. Overton, K.J. Harvatine, J. Giesey and D.E. Bauman. 2006. Milk fat depression: impact of dietary components and their interaction during rumen fermentation. Proceedings of the Cornell Nutrition Conference for Feed Manufacturers, Syracuse, NY, USA.

McFarland, D.F. 2003. Nutritional interactions related to dairy shelter design and management. Adv. Dairy Tech. 15:69.

Miller, J.K., E. Brezezinska-Slebodzinska and F.C. Madsen. 1993. Oxidative stress, antioxidants and animal function. J. Dairy Sci. 76:2812.

Murphy, M.R., C.L. Davis and G.C. McCoy. 1983. Factors affecting water consumption by Holstein cows in early lactation. J. Dairy Sci. 66:35.

National Research Council. 1974. Nutrients and Toxic Substances in Water for Livestock and Poultry. Natl. Acad. Sci., Washington, DC.

National Research Council. 2001. Nutrient Requirements of Dairy Cattle. Natl. Acad. Sci., Washington, DC.

National Research Council. 2005. Mineral Tolerance of Animals. 2nd rev. ed. Natl. Acad. Sci., Washington, DC.

Nocek, J.E. and D.G. Braun. 1985. Effect of feeding frequency on diurnal dry matter and water consumption, liquid dilution rate and milk yield in first lactation. J. Dairy Sci. 68:2238.

Parkhurst, D.L. and C.A.J. Appelo. 1999. User’s guide to PHREEQC. U.S. Geological Survey Water - Resources Investigations Report 99-4259, p. 312.

Roche, J.R., S. Petch and J.K. Kay. 2005. Manipulating the dietary cation-anion difference via drenching to early-lactation dairy cows grazing pasture. J. Dairy Sci. 88:264.

Sanchez, W.K., D.K. Beede and M.A. Delorenzo. 1994. Macromineral element relationships and lactational performance: Empirical models from a large data set. J. Dairy Sci. 77:3096.

Solomon, R., J. Miron, D. Ben-Ghedialia and Z. Zomberg. 1995. Performance of high producing dairy cows offered drinking water of high and low salinity in the Arava desert. J. Dairy Sci. 78:620.

Spears, J.W. 2003. Trace mineral bioavailability. J. Nutr. 133:1506S.

Stewart, P.A. 1983. How to understand acid-base: a quantitative acid-base primer for biology and medicine. Elsevier North Holland, Inc., New York, NY, USA, p. 86.

Stumm, W. and J.J. Morgan. 1996. Aquatic chemistry: chemical equilibria and rates in natural waters. 3rd ed. John Wiley & Sons, Inc., New York, NY, USA, p. 1022.