1 Introduction

One of the key factors for equine production is water, its availability in sufficient quantity and adequate quality [1, 2]. Water is essential for horses since it takes part in vital and essential functions, such as tissues development, disposal of metabolic products through urine and sweating, milk production in mares, also contributing to the transport of food substances.

If the quality of drinking water for horses is not adequate, the animals may become sick and eventually die. For example, excessive chlorides and magnesium can lead to profuse diarrhea that leads to a painful symptomatology typical of the equine known as colic equine syndrome that could lead to the death of the animal [3].

The daily consumption of water may be between 20 and 60 L [4, 5] depending on each animal and factors such as physical activity, work and exercise done, the atmospheric conditions as temperature, humidity, whether the animal is in lactating period, as well as the nature of the ration, since some food provides a certain amount of water. On the other hand, the type of activity the horse does also has an influence in water consumption (Table 1). For example, water consumption of gestating horses is 32–41 L/animal/day and full lactating 41–50 L/animal/day [4, 5].

Table 1 Water consumption according to the activity done by the horse and levels of chlorides in samples

Studies about water quality in equine production worldwide are limited [6,7,8] and even more limited in Argentina (South America) [9, 10]. Hooda et al. [7] reviewed a water quality in livestock farming areas of the UK determining that the rational use of manure and mineral fertilizers can help reduce pollution problems arising from livestock farming practices and avoid the contamination of the water sources. Moreover, in Belgian equine farms, Burgess et al. [6] attributed for the first time horse deaths to extremely poor water quality (high salinity and sulfate toxicity).

Despite the amount of horses in Argentina declared by National Service of Agri-Food Health and Quality [11www.senasa.gov.ar] is of 2,442,130 heads, most of them are in Buenos Aires Province. Buenos Aires Province is the area of Argentina which has largest number of equine farms.

In this province are located the Federal Capital of Argentina (the city of Buenos Aires) and the great Buenos Aires, where more than 17 million people live. This is one of the most populous cities in the world with the higher industrial development of the country. This megacity and its periphery do not have an adequate water sanitation network. Domestic and industrial effluents are partially treated or untreated, and they are discharged into water bodies, which is a source of contamination for groundwater. So in many cases, agricultural production and small towns near the city of Buenos Aires find it difficult to have access to water quality sources because aquifers have different degree of pollution [12, 13]. In this sense, the equine production closest to the city of Buenos Aires could have a lower water quality than those located further. The aim of this paper is to characterize water quality in equine production farms in Buenos Aires Province, Argentina.

2 Materials and methods

The 26 equine production farms studied are located in different areas of Buenos Aires Province (Pilar, Open Door, Exaltación de la Cruz, General Las Heras, Malvinas Argentinas, Moreno, Hurlingham, Cañuelas, Chivilcoy, Lujan, San Andrés de Giles and San Isidro) (Fig. 1). The selection of these farms was based on its importance in the production of jumping horses of the province of Buenos Aires. These equine farms have from 11 to 410 horses each. Some of the sampling locations are in the outskirts of city of Buenos Aires mainly in the northern and western areas. In this area, the groundwater comes from the Puelches and Pampeano aquifers. The Pampeano aquifer acts as a way to recharge and discharge the Puelches aquifer with the resulting transfer of polluting substances. These aquifers have in their high and low zones sodium bicarbonate waters and sodium sulphated chloride, respectively [12, 14]. The pH of the same zones fluctuates between 7.4 and 8.03, the total dissolved solids are between 800 and 2200 mg/L, and the conductivity between 800 and 5000 µs/cm.

Fig. 1
figure 1

Sampling Sites of Farms. 1-11-12- 14-15-22- General Las Heras, 2-San Isidro, 3-Malvinas Argentinas, 4-Moreno, 5-6-8-10-11-13-21-23 Pilar, 7- Cañuelas, 9-18 San Andrés de Giles, 12- Open Door, 16-17 Moreno, 19-20-26 Hurlingham, 24- Chivilcoy, 25- Exaltación de la Cruz

Groundwater samples were taken from a faucet directly connected to the water well in each equine production farm.

The depth of the water source was registered, and water samples were taken following the protocol [15] in sterilized recipients and were kept refrigerated until they were taken to the laboratory.

The following physicochemical parameters were registered, pH, conductivity (µS/cm) through Hanna Hi 255, total hardness (mg/L) determined by EDTA titration method [15] chlorides (mg/L) by the Argentometric method, nitrates (mg/L) by the cadmium reduction method with HACH DR 890 and the total dissolved solids (TDS, mg/L) using the drying method at 103–105 °C (Method 2540C, [15]). TDS is a measure of the total ionic concentration of dissolved minerals in water. TDS is composed of the following principal cations (or positively charged ions): sodium (Na+), calcium (Ca2+), potassium (K+), magnesium (Mg+2) and anions (or negatively charged ions): chloride (Cl), sulfate (SO42−), carbonate (CO32−), bicarbonate (HCO3) and to a lesser extent by nitrate (NO3), iron (Fe3+), manganese (Mn2+) and fluoride (F). Chloride is a major component of dissolved solids. The total dissolved solids were determined in a sub-sample of ten farms that had three different levels of chlorides (high concentrations: > 100 mg/L; moderate concentrations: 40–50 mg/L and low concentrations < 20 mg/L). Parameter values were compared to the recommended guide levels for drinking water for livestock of the CCME [16] and USEPA [17]. These guide values are general values for animal production and not specific to horses.

The values of physicochemical parameters were correlated with the distance to the city of Buenos Aires and the depth of the wells using the Spearman correlation (p < 0.05).

A principal components analysis (PCA) was performed to identify the main physicochemical parameters that influence water quality in relation to the distance to the city of Buenos Aires. The selection of PCs (axes) for interpretation was performed using a screen plot [18]. Data analyses were made by using the Infostat software [19].

3 Results and discussion

Groundwater sources from the farms studied are generally drillings of Puelches and Pampeano aquifers. The wells depth was between 14 and 80 m, the wells of shallow depths the ones associated with the Pampeano aquifer, whereas the ones between 70 and 80 m come from Puelches aquifer.

Results show that pH values of the water samples from the water table and semi-artesian wells are on the range of 6.50–7.73 and did not exceed the optimal range for animal drink water suggested by Canadian Environmental Quality Guidelines CEQFs [20], Equine Facilities Assistance program [21] and USEPA (2007) (Table 2). The pH was not correlated with the analyzed parameters (Table 3).

Table 2 Physicochemical parameters of water samples, depth of underground water sources and distance of farms to urban centers
Table 3 Values of r of Spearman’s correlation

The conductivity of the water samples from the farms studied does not go beyond the values recommended by the literature (Table 2) in most of the equine farms. In relation to the majority of ions (water table and semi-shallow groundwater), the calcium and magnesium concentrations were found within the acceptable limits for animal drink water (Ca2+ < 500 mg/L, Mg2+ < 250 mg/L and Cl from 2 to 23) (Tables 1, 2).

Chlorides incorporated into the daily intake of water from farms with high and moderate concentrations of chloride (Table 1) are higher than the recommended levels (3000 mg/L, [22]). This should be considered because high levels of chloride intake can cause various diseases in horses as electrolyte imbalances and colic. For this reason, the water quality, particularly chloride concentration, should be monitored every 48–72 h, before symptoms of osmotic diarrhea appear since it can lead to equine colic. To avoid the disease, the equine farms can use better quality alternative water sources (surface or network), or groundwater from different wells and mix it to improve water quality to ensure animal health.

Hardness does not correlate with the parameters analyzed, while conductivity has a significant positive correlation with chlorides (r = 0.905, p = 0.0001), with TDS (r = 0.996, p = 0.0001) and depth (r = 0.634, p = 0.0049), and a negative correlation with the distance to the city of Buenos Aires (r = − 0.839, p = 0.002) (Table 3). This suggests that the more remote localities from the city have water with lower conductivity. High values of conductivity from the farms near the city of Buenos Aires may be due to domestic and industrial effluents (partially treated or untreated).

These results are also coincident with the PCA, where the first and second components explained 69% of the total variability. The first component (PC1) explained 43% of total variance and was mainly determined by total dissolved solids (eigenvector: 0.53), conductivity (0.51) and chlorides (0.49). Farms with the best water quality are the ones at the longest distance (> 80 km) from the city of Buenos Aires (Fig. 2), meanwhile those close to the city have the greatest total dissolved solids, conductivity and chlorides (Table 2). PC2 accounts for 26% of the total variance and has strong positive weight for hardness (0.59) and nitrates (0.59), and negative score with depth (− 0.46). The second axis showed a clear farms separation according to wells depth (negative end of the axis) and water nitrates and hardness (positive end of the axis).

Fig. 2
figure 2

Principal component analysis plot based on water quality parameters measured in 26 equine farms of the Buenos Aires Province, Argentina

The values of total dissolved solids of all the samples were below the values suggested by Lewis [22]. However, in farm 19 (Hurlingham) this value is higher, although this is not considered to be within the range of dangerous values at which animals reject water consumption, even though productive losses have not shown yet (4000–3000 mg/L) [5, 22]. The TDS presents a significant positive correlation with chlorides (r = 0.913, p = 0.0001) and water well depth (r = 0.666, p = 0.035), and a significant negative correlation with the distance to the city of Buenos Aires (r = − 0.852, p = 0.002). Chloride also is a concern in groundwater, and concentrations are increasing in many aquifers in the world, especially in urban areas. Elevated concentrations of chloride in streams can be toxic to some aquatic life and increase the potential corrosively of the water [23, 24].

Nitrates’ concentration in 79% of the farms exceeded the limit considered optimal by CCME [16] and USEPA [17]. However, these values were below the levels suggested by Lewis [22] as triggers of conditions in equines (Table 2).

Nitrates have a significant negative correlation (r = − 0.65, p < 0.005) in the depth of the water source (Table 3). The samples’ number 9, 14, 18 and 22 shows a higher concentration of nitrates (Table 2). This is possible since the water coming from the water table is liable to be affected with high concentrations of this ion for being in direct contact with the non-saturated zone; this can be seen in the most superficial samples.

Human activities can affect concentrations of dissolved solids in groundwater. Groundwater pumping can pull deep saline water upward to shallow depths, or in from the coast into freshwater aquifers. Human activities can add dissolved solids to recharging groundwater. Detergents, water softeners, fertilizers, road salt, urban runoff, and animal and human waste all contain elevated concentrations of dissolved solids that are delivered to groundwater by wastewater disposal, septic systems or direct application to the land surface [25, 26]. The TDS results correspond with the results showed for the conductivity and to characteristics of the aquifers.

The results of the work show that there is a relative association between the water quality of drink used in the farms and the distance from these locations to the most important urban area (city of Buenos Aires). This suggests that sources of groundwater near to the city of Buenos Aires present a lower water quality for equine production [13, 27,28,29].

4 Conclusion

The rise in the demand of water resource and the scarcity and decline in its quality has exposed the necessity to evaluate the availability of the resource and its quality in order to guarantee the development of equine production [30] In conclusion, the water equine production farms in Buenos Aires Province was negatively related to the distance from Buenos Aires city. Equine farms closest to the city (< 80 km) may present different problems to be considered: the excessive salinity and the presence of nitrates from anthropic origin.