Heat stress reduces milk yield (Silanikove et al., Reference Silanikove, Shapiro and Shinder2009) and milk synthesis, decreasing its protein, fat and lactose contents (Bernabucci et al., Reference Bernabucci, Basiricò, Morera, Dipasquale, Vitali, Piccioli Cappelli and Calamari2015). Raw milk suitability to industrial processing involving heating treatments is evaluated with stability tests, such as the ethanol test, which is still used in several countries such as Brazil, Russia, Taiwan, China, Argentina, Uruguay, South Africa (Chavez et al., Reference Chavez, Negri, Taverna and Cuatrín2004; Horne, Reference Horne, McSweeney and O'Mahony2015; Machado et al., Reference Machado, Fischer, Stumpf and Stivanin2017). Certain physiological and pathological conditions of cows may induce changes in milk's chemical composition (Fagnani et al., Reference Fagnani, Beloti and Battaglini2014), resulting in abnormal alterations in the concentrations of protein, sodium, chloride and lactic acid that might impair milk's stability. There are few studies evaluating the effects of heat stress on functional characteristics of raw milk such as stability and acidity. In the cheese making process, milk coagulation properties, such as clotting time and curd firmness were negatively affected when the temperature–humidity index (THI) was higher than 75 (Bernabucci et al., Reference Bernabucci, Basiricò, Morera, Dipasquale, Vitali, Piccioli Cappelli and Calamari2015). To our knowledge, there are no studies on the provision of shade to mitigate the adverse effects of heat stress on milk stability.
Our objective was to test the hypothesis that during the summer in the subtropics, access to shade helps to maintain functional properties of raw milk acceptable to industrial processing, this study aimed to investigate the effects of tree shade provision for dairy cows on stability and acidity of raw milk.
Materials and methods
Local description, animals and management
The trial was conducted on a commercial dairy farm in Brazil (latitude of −27°05′05″, longitude of −53°35′38″; altitude of 206 m above sea level) during the summer, from January to March and lasted 38 d. Handling of animals was performed following Brazilian National guidelines and was approved by the Ethics Committee of the Federal University of Rio Grande do Sul, under the number 21901.
Sixteen lactating Holstein cows were selected from the dairy herd. At the beginning of the experiment animals had 520 ± 74 kg body weight (BW), body condition score (BCS) of 3.0 ± 0.3, yielded 21.5 ± 4.2 kg/d milk (MY) and were in milk for 120 ± 61.2 d (days in milk – DIM). Cows were divided into two groups of eight cows each considering initial milk production, age and lactation period.
Each group was placed in its own specific but adjacent rectangular-shaped paddock (40 × 700 m each) with Tifton 85 (Cynodon dactylon) grass, and this separation lasted until the end of the trial. Cows within groups grazed a new strip of pasture of approximately 0.8 ha every day, moving the front and back fence lines inside the paddocks. The shaded area of each paddock, approximately 200 m2, was located at the end of the paddock (shorter side of the rectangle) and was not covered with pasture. Cows had permanent access to water in drinkers installed in the paddocks.
During the first 14 d (day 1 to 14 – pre-stress period) all cows had free access to the shaded area of its specific paddock, with at least 10 m2 of shade per animal. On days 15 to 24 (stress period), one group of cows was deprived of accessing the shaded area by the use of a fence line behind the projection of the tree's shadow. On days 25 to 38 (post-stress period), fence line that prevented access to shade from one group was removed and all the two groups of cows had free access to their respective shaded areas. Since the shaded area was not covered with pasture, cows with access to shade in the stress period did not have increased pasture allowance when compared with cows without access to shade.
Production and milk composition traits
Milk yield (kg) was recorded at the morning and evening milking using Westfalia milk meter. Individual milk samples were taken daily at the morning and evening milking. Proportional volumes of milk from morning and evening milking from each cow were mixed and composed 200-ml samples. Concentrations of fat, total protein and lactose in milk were determined by an infrared analyzer (Bentley 2000® equipment). Acidity and density were determined by titration with 0.1 N NaOH solution and using a thermolactodensimeter, respectively (Tronco, Reference Tronco1997). Ethanol stability was determined according to Machado et al. (Reference Machado, Fischer, Stumpf and Stivanin2017); results were expressed as the minimal ethanol concentration in the alcoholic solution that induced milk coagulation. Heat stability was estimated using the coagulation time test (CTT) described by Negri et al. (Reference Negri, Chavez, Taverna, Roberts and Speranza2001) and Machado et al. (Reference Machado, Fischer, Stumpf and Stivanin2017). Somatic cell count (SCC) was determined by flow cytometry with Somacount 300®. Detailed materials and methods are provided in the online Supplementary File.
Statistical analysis
Variables with continuous values were submitted to normality test using Shapiro–Wilk test. Somatic cell count was transformed using log 10 (log10 SCC). The power analysis of the sample size was calculated using the POWER procedure
Data were analyzed by univariate variance analysis with cows as the experimental units according to a completely randomized design with a factorial arrangement (n = 2: treatments – shade and no shade) and period (n = 2, Stress and Post-stress) and repeated measurements over time (days). PROC MIXED (SAS®) was used to evaluate a model: Yij = m + Ti + Pj + TPij + β(Xij − Xm) + εij, where Yij is the observed response, m = constant, Ti is the effect access to shade (n = 2, with or without shade), β is the linear regression coefficient between the covariate and the response, Xij is the observed value of the characteristics measured on pre-stress period (used as covariate), Xm is the mean of the covariate, Pj is the effect of period (n = 2, stress and post-stress periods), TPij is the effect of the interaction between treatment and period and εij is the error term; using the following parameters: method = REML, covariance matrix = AR(1), repeated = day. Mean separation was performed by using PDIFF option. The significance criterion was taken as P < 0.05. Data measured on pre-stress period, when all cows had access to shade were used as a covariate. Additionally, long term effect of shade on milk ethanol stability was analyzed with PROC MIXED on all days of measurement, covariance matrix = AR(1).
Results
The sample sufficiency was confirmed by a power analysis with the result of 0.99 for all variables tested.
During the stress period, overall means (±se) of air temperature (AT – °C) were 25.9 ± 2.7 and 28.4 ± 3.1, relative humidity means (RH – %) were 72.9 ± 8.3 and 60.0 ± 12.7 and Temperature-Humidity Index means (THI) were 75.4 ± 3.7 and 77.4 ± 3.5 at the shaded and unshaded areas of the paddocks, respectively (online Supplementary Fig. S1).
There were significant interactions between treatment and period (P < 0.05) for milk yield, milk crude protein, density and acidity. During the stress period, cows without access to shade showed lower values (P < 0.05) for milk protein concentration (2.8 vs. 3.0 g/100 g, se = 0.04) and density (1028.3 vs. 1029.6, se = 0.38) but with higher acidity (18.7 vs. 15.4 °D, se = 1.0) compared with cows with access to shade. The stressed cows also exhibited a numerical but non-significant decrease in milk yield (17.7 vs. 19.4 kg/d, se = 1.0). The largest effects of shade unavailability in the stress period were increased milk acidity and reduced stability both in the ethanol and in the coagulation time test (Table 1). In unshaded cows, milk ethanol stability dropped below the minimal values required by the dairy industries (e.g. 70 to 72 °GL, depending on the country's legislation) on days 22 and 24 (stress period), while ethanol stability in shaded cows showed much less variation, remaining above the minimal values required. Recovery to stability values similar to those registered during the pre-stress period took approximately 14 d (online Supplementary Fig. S2). Values for coagulation time were lowest for the cows without access to shade both in stress and post-stress periods. Values for milk urea during the pre-stress were lowest, increasing for both groups, but the highest values were noticed for the cows with access to shade during the stress period. During the post-stress period, there were no significant differences (P > 0.05) for milk yield, milk crude protein, acidity and SCC between groups of cows. There were no effects (P > 0.05) of treatment (access to shade) or interaction between treatment and period for fat and lactose concentrations in milk (Table 1).
LSmean values for milk yield and composition of cows under alert THI conditions with or without access to shade at Stress and Post-stress periods
a−bLSmeans within a row and within periods (pre-stress, stress and post-stress) with different superscripts are statistically different (P < 0.05).
1 Log10 SCC = Somatic cell count logarithmically transformed (log10).
2 Dornic degrees (°D), 1 Dornic degree (1°D) is equal to 0.1 g of lactic acid per liter of milk.
3 The volume of ethanol (ml) in 100 ml of an alcoholic solution, in Gay-Lussac (GL) degrees.
Discussion
The main contribution of the present study is to highlight the positive effects of providing natural tree shade to grazing cows on milk stability and acidity during the summer in the subtropics. Functional characteristics of milk such as stability are important during milk thermal processing in the dairy industry. Ethanol stability is a cheap and fast test run at farm and at industry reception to check raw milk suitability for industrial purposes (Horne, Reference Horne, McSweeney and O'Mahony2015), and usually the minimal acceptable value varies from 70 °GL ethanol, as in Argentina (Chavez et al., Reference Chavez, Negri, Taverna and Cuatrín2004) to 72 °GL ethanol, as in Brazil (Machado et al., Reference Machado, Fischer, Stumpf and Stivanin2017). However, a minimal value of 74 °GL or even higher is recommended for UHT milk (Shew, Reference Shew1981). Milk stability may also be measured by the time needed for milk to coagulate at high temperature, e.g. 144°C (Chavez et al., Reference Chavez, Negri, Taverna and Cuatrín2004). It should be taken into consideration that one of the limitations of this last test is the lack of a clear definition of the value for the low threshold for milk stability, although it might be assumed that, the longer milk resists treatment without the formation of clots, the greater is its thermal stability (Negri et al., Reference Negri, Chavez, Taverna, Roberts and Speranza2001; Machado et al., Reference Machado, Fischer, Stumpf and Stivanin2017).
Cows with access to shade produced milk with ethanol stability (76.5 °GL) suitable for UHT process. In cows with shade access throughout the study, milk stability presented small and no significant variation between periods, which was expected. When cows without shade access regained shade in post-stress, partial recovery of ethanol stability to pre-stress levels took approximately 14 d, showing long term effects (online Supplementary Fig. S2). Stability decrease in cows without shade might be partially related to the high acidity in milk, which could have increased the ionic calcium concentration in milk, acknowledged as important factor down-regulating milk stability (Chavez et al., Reference Chavez, Negri, Taverna and Cuatrín2004; Horne, Reference Horne, McSweeney and O'Mahony2015). It is important to emphasize that the lower ethanol concentration used in the study was 68 °GL, so it is possible that milk stability in unshaded cows was even lower, but out methodology did not enable us to detect it.
Shade provision during stress period had only a small (non-significant) but positive effect on milk production but did significantly increase milk crude protein concentration. The negative effects of severe heat stress on milk production (Silanikove et al., Reference Silanikove, Shapiro and Shinder2009) can be greater than those provoked by lower feed intake and are related to metabolism disturbances (Gao et al., Reference Gao, Guo, Quan, Nan, Fernandez, Baumgard and Bu2017). Chronic heat stress is known to reduce milk component concentration such as protein and lipids due to lower synthesis in the mammary gland (Cowley et al., Reference Cowley, Barber, Houlihan and Poppi2015).
The increase in SCC observed for the cows with access to shade in stress period might be related to higher humidity at the ground with accumulated mud in the shaded area, conditions usually associated with high SCC (Sandrucci et al., Reference Sandrucci, Bava, Zucali and Tamburini2014). Also, a behavioral study (Vizzotto et al., Reference Vizzotto, Fischer, Thaler Neto, Abreu, Stumpf, Werncke, Schmidt and McManus2015) showed that cows with access to shade spent more time lying, increasing the udder's exposure to environmental pathogens, especially after milking. However SCC is not usually related to milk stability (Machado et al., Reference Machado, Fischer, Stumpf and Stivanin2017) as increases in pH reduce ionic calcium (Horne, Reference Horne, McSweeney and O'Mahony2015), probably counteracting the changes in minerals (e.g. increases in chlorides and sodium) (Chavez et al., Reference Chavez, Negri, Taverna and Cuatrín2004) and serum proteins (Horne, Reference Horne, McSweeney and O'Mahony2015) that usually lower milk stability.
In conclusion, under the prevailing environmental conditions in the study when cows faced THI that could be classified as in the alert range, access to shade maintained milk acidity and stability within an acceptable range, increasing its suitability for dairy processing. Shade provision should be considered in dairy production in the tropics.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022029920000916.
Acknowledgements
This work was supported by National Council for Scientific and Technological Development (CNPq – grant number 476556/2008-2) and was performed with the collaboration of the Eidt Family.
