Drinking behavior of dairy cows under commercial farm conditions differs depending on water trough design and cleanliness

In dairy cows, an adequate water supply is necessary for optimal feed consumption, productivity, health, and animal welfare. In the present study, we tested the hypothesis that dairy cows' drinking behavior is altered by the design and cleanliness of their drinking troughs. The study was conducted at a commercial dairy farm with a herd of 135 lactating cows held in a free-range barn. Drinking behavior at two identical tank troughs (length: 2.00 m; width, 0.43 m; depth, 0.15 m; volume, 70 L) and two identical double-valve troughs (length, 0.73 m; width, 0.32 m; depth, 0.10 m; variable volume, 5-15 L), each made of stainless steel, was video-recorded daily in the first 2 h after feeding for 15 d in each of two study periods (December 2019 and February 2020). To determine the effects of trough cleanliness on drinking behavior, one trough of each design was cleaned daily, and the other trough of each design was not cleaned (2 x 2 Latin square study design), allowing the cows to choose between clean and unclean troughs of each design. Drinking episodes were analyzed and characterized using Behavioral Observa-tion Research Interactive Software. Drinking water quality was analyzed at the start and end of the study periods and monitored daily using rapid tests. At tank troughs relative to double-valve troughs, the following were observed: a shorter total duration of drinking episodes (P < 0.001), higher odds for smelling behavior while tasting (P = 0.01), lower odds for drinking episodes consisting only of tasting behavior (P = 0.002), shorter (P = 0.03) and fewer drinking breaks (P < 0.001), lower odds for swallowing difficulties (P = 0.001), and higher odds for interruptions due to agonistic behavior (P = 0.001). The water quality at the start and end of the study periods did not differ significantly. Nevertheless, the cows' drinking behavior changed according to trough cleanliness. At unclean troughs relative to clean troughs, the following were observed: more (P = 0.02) and longer (P = 0.03) drinking breaks, a higher number of sips per drinking episode (P < 0.0001), tendentially higher odds for drinking episodes consisting only of tasting behavior (P = 0.08), and lower odds for swallowing diffi-culties (P = 0.001). In total, daily cleaning of the troughs altered 7 and trough design 8 out of 13 drinking behavior variables, giving additional insights in dairy cows drinking behavior.