Water Quality and Dairy Cattle Production: Saskatchewan Surveys

Abstract

The quality of water in Saskatchewan is less than ideal and poor water quality can affect livestock adversely. The sulfate content of the water can affect ruminants by interfering with Cu metabolism. other factors such as Zn, Fe and Mo content of the feed can also have an effect on cu metabolism. For these reasons two surveys were undertaken to investigate the source and quality of water on Saskatchewan dairy farms and the effect of this water on dairy cattle.

Survey One found that 92.8% of the 656 farms used the Holstein breed of dairy cow with an average milking herd size of 42 cows. Average milk production for Holsteins was 20.5 kg d-1. Other breeds used included Brown Swiss (0.2% of farms) producing 19.2 kg cow-1d-1, mixed breed herds (4.3% of farms) producing 18.8 kg cow-1d-1, Ayrshires (1.7% of farms) producing 18.4 kg cow-1d-1, and Jerseys (1.0% of farms) producing 13.9 kg cow-1d-1. 39% of the farms surveyed used either DRAS or ROP milk recording programs.

Water quality and water sources varied throughout the province. The average well depth was 148.2 meters with 84.4% of farms using water from wells. Other sources of water included dugouts, springs, reservoirs and treated city water. The average water on Saskatchewan dairy farms contained 750.7 mg 1-1 hardness, 595.9 mg 1-1 sulfate, 22.8 mg 1-1 nitrates and 1971.4 ����S cm-1 conductivity. 76.7% of farms had water hardness levels less than 1000 mg 1-1. 81.8% of farms had water sulfate levels

of less than 1000 mg 1-1• 83.3% of farms had water nitrate levels of less than the maximum recommended level for livestock of 22 mg 1-1, however, 28 farms (4.2%) had nitrate levels of greater than 100 mg 1-1• 60.4% of farms had water conductivity levels of between 1000 and 3000 ����S cm-1. The levels of constituents in the water varied significantly with well depth in accordance with accepted theory on the chemical development of aquifer water. While the average water appears to be of reasonable quality there was a wide range of constituent levels that varied greatly between farms.

stepwise regression found a small but significant negative effect of hardness on milk production (r2 = 0.0158, P < 0.01). In the second survey 12 farms were selected on the basis of water sulfate level, record-keeping to provide a wide range of water sulfate levels and allow accurate measurement of cow milk production and reproduction.

The average water on farms in Survey Two contained 727.6 mg 1-1 hardness, 816.4 mg 1-1 sulfate, 5.9 mg 1-1 nitrates, 2313.3 ����S cm-1 conductivity, 396.3 mg 1-1 alkalinity, 148.2 mg 1-1 Ca, 31.6 mg 1-1 Mg, 320 mg 1-1 Na and had a pH of 7.3. Farms were chosen to provide a wide range of water sulfate levels and thus water sulfate ranged from 84 mg 1-1 to 2220 mg 1-1.

Average 4% FCM production in Survey Two was 31.7 kg cow- 1day-1. Stepwise regression found that 4% FCM production was significantly affected by DMI and Days in Milk (r2 = 0.28, P < 0.01) but not significantly affected by the intakes of protein,

nitrates, sulfur, Cu, Zn and Fe. Correlation analysis indicated significant correlations between 4% FCM production and hardness, sulfates, nitrates and Ca levels in the water. Stepwise regression using 4% FCM production and water quality parameters found that nitrates had a negative effects on production (r2 = 0.14, P < 0.05).

This survey indicated that cows could receive from 1.4 to 43% of their Ca requirements from Ca in the water. Cows can also receive from 5.8 to 62.8% of their daily S intake from S in the water. Na in the water can account for 6.6 to 288% of the cow's Na requirement.

CU status, as measured by plasma CU content, was found to be correlated to Mo intake, Days in Milk and Services per conception. There was not significant correlation between plasma Cu and Total S intake however, there was significant positive correlation between S intake and serum K, glucose, CPK, and significant negative correlation between S intake and serum urea. Total S intake was not significantly correlated to other serum parameters.