Please note our website will be undergoing maintenance on Tuesday, May 28, 2024. e-Commerce transactions and new registrations will be temporarily unavailable during this time. We apologize for any inconvenience this may cause.
Free access

Projected economic losses from milk performance detriments under heat stress in Quebec dairy herds

Publication: Canadian Journal of Animal Science
21 August 2020

Abstract

The objective of this study was to estimate economic losses associated with milk performance detriments under different climate scenarios. A dataset containing milk records of Holstein and daily temperature–humidity indexes compiled over 6 yr in two contrasting climatic dairy regions of Quebec [eastern (EQ) and southwestern Quebec (SWQ)] was used to develop equations for modeling milk performance. Milk performance, including milk, fat, protein, and lactose yields of dairy herds (kg·d−1), were then projected considering six plausible climate scenarios during a climatic reference period (REF: 1971–2000) and two future periods (FUT1: 2020–2049; FUT2: 2050–2079). Economic losses were assessed by comparing future to reference milk prices based on components. Results indicated that fat and protein yields could decline in the future, thus resulting in economic losses ranging from $5.34 to $7.07 CAD·hL−1 in EQ and from $5.03 to $6.99 CAD·hL−1 in SWQ, depending on the amplitude of future temperature and humidity changes and on whether it is milk quota or cow number which is adjusted to avoid under-quota production. The projected increase in occurrence and duration of heat stress episodes under climate change could result in substantial financial harm for producers, thereby reinforcing the necessity of implementing heat-abatement strategies on dairy farms.

Résumé

L’objectif de cette étude était d’estimer les pertes économiques associées au détriment de performance laitière sous différents scénarios climatiques. Une banque de données contenant les registres de lait des Holsteins et les indices quotidiens température-humidité compilés pendant 6 ans dans deux régions laitières climatiques contrastantes de la province de Québec [est du Québec (EQ) et sud-ouest du Québec (SWQ)] a été utilisée afin de développer les équations pour la modélisation de la performance laitière. La performance laitière, incluant les rendements en lait, en gras, en protéines, et en lactose des troupeaux laitiers (kg·j−1), a ensuite été projetée en considérant six scénarios climatiques plausibles pendant une période climatique de référence (REF — « climatic reference period » : 1971–2000) et deux périodes futures (FUT1 : 2020–2049; FUT2 : 2050–2079). Les pertes économiques ont été évaluées en comparant les prix du lait du futur aux prix de la période de référence selon les composantes. Les résultant indiquent que les rendements en gras et en protéines pourraient diminuer dans le futur, entraînant ainsi des pertes économiques variant de 5,34 à 7,07 $ CAD·hL−1 dans l’est du Québec, et de 5,03 à 6,99 $ CAD·hL−1 dans le sud-ouest du Québec, selon l’amplitude des changements futurs de température et d’humidité, et si c’est le quota de lait ou le nombre de vaches qui sera ajusté afin d’éviter une production sous-quota. L’augmentation projetée de l’occurrence et de la durée de stress thermique après changement climatique pourrait se solder par d’importantes pertes financières pour les producteurs, renforçant ainsi la nécessité d’implémenter des stratégies de réduction de chaleur sur les fermes laitières. [Traduit par la Rédaction]

Introduction

In homeothermic animals, maintaining a body temperature within the range of normal activity is a prerequisite to achieving maximum performance (Bernabucci et al. 2010). To maintain this equilibrium, produced metabolic heat must be continously dissipated to the environment at rates dependent on body-air temperature and water vapor gradients. Concurrent elevations of ambient temperature and relative humidity were reported to skew heat balance by impeding heat dissipation and ultimately increasing thermal load, thus provoking heat stress.
In response to heat stress, dairy cows adapt their physiology, metabolism, and behavior at multiple levels of vertebrate organization, to minimize heat production and (or) promote heat loss (Wheelock et al. 2010; Collier and Gebremedhin 2015). These homerhetic processes are critical to survival but are detrimental for milk (Wheelock et al. 2010; Bernabucci et al. 2014; Cowley et al. 2015) as well as for reproductive performance (Schüller et al. 2014), health (West 2003), and welfare (Polsky and Von Keyserglingk 2017). Therefore, exposure to hyperthermia is associated with economic losses estimated over $1.2 billion, $800 million, and $595 million for the US dairy industry when heat stress occurs during lactation (Key and Sneeringer 2014), the dry period (Ferreira et al. 2016), and in utero (Laporta et al. 2020), respectively.
Most heat-stress-related milk performance and economic studies have been conducted for regions with tropical, aride, and temperate climates (Galán et al. 2018), arguably because heat stress episodes are more pronounced under these climates. As for cold climates (we use here the common Köppen–Geiger terminology as it is reported by Peel et al. 2007), a recent study conducted by our group indicates that declines in milk fat and protein yields are also observed, although heat stress episodes are of relative mild intensity and less frequent (Ouellet et al. 2019). Despite recent advances in cooling systems and better understanding of the mechanisms underlying the condition, heat stress remains nowadays a financial burden for dairy producers and one of the primary constraints to efficient production and food security (St-Pierre et al. 2003; Ferreira et al. 2016; Laporta et al. 2020).
Climate change, defined as a significant modification in long-term weather statistics for variables such as temperature and rainfall, is ongoing and inevitable for the next decades (IPCC 2014). In the agricultural territory of the Province of Quebec, climate change is expected to result, from 1971–2000 and from 2041–2070, in an increase in summer temperatures ranging from 1.6 to 4.5 °C, in hotter daily minimum and maximum temperatures, as well as in longer heat waves (Ouranos 2015). As another manifestation of climate change, duration, frequency, and intensity of heat stress episodes are expected to increase (Hatfield et al. 2008), which would exacerbate current impacts on the dairy sector. Despite such forecasts, research concerning the potential costs of heat stress is limited in regions characterized by a cold climate without dry season and with warm summer (defined as Dfb in Köppen–Geiger terminology; Peel et al. 2007) such as most of Quebec’s agricultural territory. Establishing such costs could serve as a basis for determining whether adaptation is needed regarding this issue.
The objective of this study was to estimate, relative to a climatic recent-past period (REF: 1971–2000), future economic losses associated with milk performance impairments under heat stress. The tested hypothesis was that, relative to the reference period, projected future increases in heat stress occurrence and duration could lead to declines in fat and protein yields, thus ultimately resulting in important financial harm for Quebec’s dairy sector.

Material and Methods

Dataset description

The data used in this study have been described previously (Ouellet et al. 2019). Briefly, the dataset was provided by Valacta (now Lactanet, Dairy Production Center of Expertise, Quebec and Atlantic, Sainte-Anne-de-Bellevue, QC, Canada). It includes 606 031 milk analysis records compiled from January 2010 to November 2015, for 34 361 Holstein cows in 279 herds. Herds were located in two climatically contrasting regions of the Province of Quebec: southwestern (SWQ) and eastern (EQ) Quebec. Herein, we define EQ as the region of the province characterized by the Atlantic Maritime ecozone, whereas SWQ is characterized by the Mixedwood Plains ecozone. Parity average [± standard deviation (SD)] is 2.7 ± 1.7 lactations. Daily production traits, including milk, fat, protein and lactose yields, as well as the associated milk fat, protein, and lactose concentrations, were recorded monthly. The dataset also contained days in milk (DIM), lactation start date, calving month, calving year, parity number, and estimated breeding value for milk yield, components yields, and percent.

Dataset edits

Test-day (TD) records with missing data (for milk, fat, protein, lactose, or genetic values) and with DIM > 350 were discarded from the dataset. Additionally, only cows with a minimum of 8 TD records per lactation and with at least one record before 50 d of lactation were kept for further analysis. After editing, a total of 261 382 TD records were kept for the analysis, corresponding to 23 445 Holstein dairy cows with parity of 2.3 ± 1.8 (average ± SD) and from 189 dairy herds. Among these cows, 8379 (36%) were at first parity, 8026 (34%) were at second parity, and 6741 (30%) were at third parity or beyond. First-parity cows produced 28.1 ± 5.8 kg milk·d−1 with 4.0% ± 0.6% fat and 3.3% ± 0.3% protein. Second-parity cows produced 30.9 ± 8.3 kg milk·d−1 with 4.0% ± 0.7% fat and 3.4% ± 0.4% protein. Production at third parity or more was 32.1 ± 9.1 kg milk·d−1 with 4.0% ± 0.5% fat and 3.3% ± 0.4% protein.

Heat stress duration calculations

Heat stress duration was then included in the dataset. To do so, daily present-time (2010–2015) temperature–humidity indexes (THI) were calculated, considering observed meteorological data retrieved from the closest weather station relative to the dairy herds of our dataset and following the equation developed by the NRC (1971):
where T is the temperature (°C) and RH is the relative humidity (%). For each TD, daily THI values of the eight previous days (excluding the TD itself) were considered for calculating the associated heat stress duration. A window of 8 d was chosen as it was reported that this interval of time has a significant relationship with milk performance in dairy cows raised in a Dfb climate (Ouellet et al. 2019). Heat stress duration was defined as the longest sequence of days with THI ≥ 65 during the 8 d window. A threshold of 65 was chosen as it was previously reported to impair productivity for cows raised in a Dfb climate (Ouellet et al. 2019). Heat stress duration was ultimately separated into five categories: “0” for no day with heat stress, “1–2” for 1 or 2 d, “3–4” for 3 or 4 d, “5–6” for 5 or 6 d, “7–8” for 7 or 8 d.

Milk performance equations

Equations to model milk, fat, protein, and lactose yields under different heat stress durations were then developed in SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). To do so, the edited dataset was randomly split into three subsets: a training set (including 50% of the herds), a validation set (including 25% of the herds), and a test set (including 25% of the herds).
The training set was used to choose the variables to include in the equations. All variables retrieved in the edited dataset were assessed for co-linearity using (linear) Pearson’s and (rank) Spearman’s correlation coefficients. Variables with the highest correlations with the predictive outcome and without co-linearity and all of their possible interactions were first included in the model. Manual backward selection was then performed, and only significant variables (P ≤ 0.05) were ultimately retained for inclusion in the model. Furthermore, continuous variables were assessed for linearity by including a quadratic term in the model. When the quadratic term was significant, it was included in the model; otherwise, the variable was modeled as linear.
Fat and protein yields under different heat stress episodes duration were modeled as follows:
where Yfijkl is a measurement of fat or protein yields; expDIMf is the fixed effect of e−0.05 × DIM; DIMi is the fixed effect of the days in the milk category; DwithHSj is the fixed effect of the heat-stress duration category; Lactk is the fixed effect of the parity category; GNTICl is the fixed effect of the estimated breeding value for milk or fat or protein yields or fat and protein concentrations; DwithHSj × Lactk and DIMi × Lactk are fixed-effect interactions; and Calving_monthl is the fixed effect of the calving month. Because DwithHSj was not associated (P = 0.23) with milk or lactose yields, DwithHSj and DwithHSj × Lactk were not included in the equation to model those outputs.
The validation set was next used to confirm the selection of variables, and the test set was finally used to assess the goodness-of-fit of the final equation, by calculating the root-mean-square error (RMSE), coefficient of determination (R2), Bayesian information criteria, and Akaike information criterion. In addition, the residuals (observed minus predicted) were regressed against the centered predicted values, and the intercept and the slope were used to assess the mean and the linear bias, respectively (St-Pierre 2003).
Ultimately, the equations used to predict fat and protein yields led to R2 = 0.78 and RMSE = 0.17 kg·d−1 for fat yield and to R2 = 0.85 and RMSE = 0.10 kg·d−1 for protein yield. The equation used to model lactose yield led to R2 = 0.42 and RMSE = 0.3 kg·d−1. Finally, the model used to predict daily milk yield led to R2 = 0.40 and RMSE = 7.1 kg·d−1.

Climate scenarios used to project future THI

The future evolution of climate was based on an ensemble of six plausible climate scenarios for daily T and RH. A climate scenario is here defined as a bias-adjusted numerical climate model simulation, which must be distinguished from a greenhouse gas emission scenario considered for generating a climate simulation. Table 1 contains all information necessary to identify the simulations, along with the codes (SIM-01, SIM-02, SIM-03, SIM-04, SIM-05, and SIM-06) that will be used to refer to them hereafter. In climate change impact studies, it is important to consider more than one climate scenario as there is uncertainty on the amplitude of future climate change, pertaining mostly to uncertainty on future greenhouse gas emissions evolution through time and to inter-model differences in physics formulations (natural variability is also an important source of uncertainty for the close period FUT1). Here, different climate models are covered: MIROC5 (Watanabe et al. 2010), MRI-CGCM3 (Yukimoto et al. 2012), and CRCM5 (Separovic et al. 2013), the latter one being a limited area model driven by global models CanESM2 (Arora et al. 2011) and MPI-ESM-LR (Stevens et al. 2013). All global simulations were developed during phase 5 of the coupled model intercomparison project (CMIP5; Taylor et al. 2012). To put it simply, the ensemble covers two greenhouse gas emission scenarios, one “optimistic” (RCP4.5) and another “pessimistic” (RCP8.5) (van Vuuren et al. 2011). The “optimistic“ emission scenario considers that main greenhouse gas emissions will peak around 2040, whereas the “pessimistic“ emission scenario is often used as the basis for worst-case climate change scenarios, where emissions continue to rise throughout the 21st century (van Vuuren et al. 2011). Bias-adjustment was made with the quantile mapping technique (Themeßl et al. 2011; Grenier 2018), using data from two meteorological stations as the reference, with “Mont-Joli A” representing EQ and “St-Hubert A” representing SWQ.
Table 1.
Table 1. Specifications for identifying climate simulations used in this study.

Note: CRCM5 is version 5 of the Canadian Regional Climate Model. Expansions for acronyms of global models and modeling institutes can be found at http://www.ametsoc.org/PubsAcronymList.

Temperature and RH were used to calculate daily THI during each involved period (REF over 1971–2000, FUT1 over 2020–2049, and FUT2 over 2050–2079), following the equation developed by the NRC (1971) and presented in a previous section. Daily THI were then used to assess the occurrence of heat stress episodes, by calculating the average annual number of days potentially causing heat stress, namely days with THI ≥ 65, for each time period (REF, FUT1, and FUT2). Additionally, the intensity of heat stress episodes was assessed for REF, FUT1, and FUT2, by calculating the average number of days per year falling in the following THI categories: (65–70), (70–75), (75–80), (80–85), and ≥85. Finally, the duration of heat stress was quantified by calculating the average annual number of consecutive days with heat stress. All calculations were performed with the FREQ procedure of SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

Herds selection

Given the large number of herds included in the dataset, a subsample from the edited dataset had to be selected to facilitate the analysis. Selection criteria are (1) a minimum number of 40 lactating cows and (2) a proportion of primiparous cows of 35% ± 5%, of second lactation cows of 25% ± 5%, and of third (or subsequent) lactation cows of 40% ± 5%. Those criteria were determined to select herds comparable to the average of Quebec dairy herds. Overall, 71 dairy herds (37 from EQ and 34 from SWQ) met the selection criteria and were then kept for further analysis. Eastern Quebec herds had an average (± SD) number of 55.1 ± 10.7 lactating cows (with eight or more TD), whereas SWQ herds had an average (± SD) of 57.2 ± 11.5 lactating cows (with eight or more TD). In comparison, the provincial average (± SD) of lactating cows per herd was 59.2 ± 1.6 for the years 2010–2015 (Valacta 2011, 2012, 2013, 2014, 2015, 2016). Finally, average (± SD) EQ herd composition was as follows: 39.8% ± 1.4% for first lactation cows, 28.2% ± 3.2% for second lactation cows, and 27.3% ± 2.8% for third lactation or more. Average (± SD) SWQ herd composition was as follows: 40.6% ± 0.8% for first lactation cows, 26.2% ± 1.2% for second lactation cows, and 28.1% ± 2.9% for third lactation or more.

Projected performance

Annual herd performances (milk, fat, protein, and lactose yields) were subsequently projected during each year of each period (REF, FUT1, and FUT2), using the predictive equations described above and considering the THI calculated with each of the six climate scenarios. To do so, the lactation start date (day and month) of each animal was changed from the original dateset (2010–2015) depending on the studied time period. For example, a cow that had a lactation start date of 6 Oct. 2010 had a lactation start date of 6 Oct. 1971 for the year 1971, a lactation start date of 6 Oct. 1972 for the year 1972, and so on for all the years included in the analysis. All lactations were assumed to last 305 d. Heat stress duration was then added. Temperature–humidity index calculated for the 8 d prior to TD were then matched to the corresponding date. Test days were then separated in the five aforementioned categories (“0” for no day with heat stress, “1–2” for 1 or 2 d, and so on). Ultimately, annual milk performance per herd were averaged for REF, FUT1, and FUT2. All milk performance projections were made with estimated breeding value and parity held constant over time (thus remaining the same in each of the studied periods). Consequently, only heat stress duration and occurrence varied between the time periods and the climate scenarios.

Economic values and calculations

Several economic parameters are needed to calculate Canadian milk income, namely milk quota price [CAD·kg−1 of butterfat (BF)], interest of quota sold (%), BF, protein, and lactose price (CAD·kg−1), deductions on milk price, variable costs per cow, and interest associated with cow purchase (when necessary) (see Table 2). All economic parameters used in this study were averaged for the years 2010–2015. Average herds’ annual milk income were first calculated for REF, using the recent past segment of each climate scenario (separately). Average herds’ milk income values were then subsequently calculated with performance projected for FUT1 and FUT2 periods, again separately for each climate scenario. Because milk production in Canada is limited for each producer by the milk quota owned, two economic scenarios were investigated for FUT1 and FUT2 relative to REF. In the first scenario, the number of cows was maintained constant (same as REF), and quota was sold, in the perspective of avoiding under-quota production. In the second scenario, it is the amount of owned quota that was held constant (same as REF), whereas the number of cows was adjusted according to the amount of BF produced during the considered period, again mimicking the avoidance of under-quota production.
Table 2.
Table 2. Mean (2010–2015) economic parameters used to assess the economic effect of milk components variations during summer time.

Note: All prices were averaged for the years 2010–2015. BF, butterfat.

a
CAD were converted in USD at an exchange rate of 0.8963 (average rate 2010–2015; Bank of Canada 2020).
b
Averaged for the years 2013–2015 from Agritel Web database (http://agritel.fgcaq.com/Agritel), Fédération des groupes conseils agricoles du Québec (Longueil, QC, Canada).
In both economic scenarios, milk and components prices as well as milk deductions (marketing and transport), milk quota price, and variable costs related to cows were held constant over time; thus, same prices were used to calculate milk income and expenses by herd in all three periods. Ultimately, the evolution of economic losses associated with milk components modifications due to climate change was assessed using a partial budget methodology, comparing variations in income, and expenses projected under FUT1 and FUT2 with corresponding quantities for REF.

Results and Discussion

Heat stress is known to have detrimental effects on cows performance. In this project, we developed milk performance equations with a milk records dataset to project milk, fat, protein, and lactose yields under different climate scenarios. Ultimately, our goal was to assess the plausible time evolution of heat-stress-related economic losses from lactating cows of dairy herds located in a cool climate. This approach was chosen over modeling animal responses to heat stress with data retrieved from the literature (St-Pierre et al. 2003) and over the use of a stochastic frontier production function model (Key and Sneringer 2014) as it allowed to consider climate change, heat tolerance of animals raised in a continental-humid (Dfb) climate, and production contexts of the Province of Quebec.

Projected climate conditions

In all six considered climate scenarios and in both regions, projected summer THI increased over time, with the highest increase projected from REF to FUT2 (Fig. 1). More specifically, projected THI in EQ increased on average by 2.2 units in FUT1 and by 4.2 units in FUT2 (in both cases, relative to REF). As for SWQ, THI increased on average by 2.3 units in FUT1 and by 4.2 units in FUT2. It should be noted that the six considered climate scenarios do not fully cover the uncertainty in future summer temperature changes as obtained by considering a larger ensemble of CMIP5 simulations, but roughly speaking, they are representative of the interquartile range of that uncertainty (verified but not shown). Hence, derived quantities like mean THI changes are here presented as indicative of the sign and order of magnitude of the expected changes.
Fig. 1.
Fig. 1. Evolution of summer (June, July, and August) average maximum temperature (AT, °C) (± SD), and temperature–humidity index (THI) in eastern Quebec (EQ: AT, ; THI, ) and southwestern Quebec (SWQ: AT, ; THI, ) through reference (1971–2000), near (2020–2049), and distant future (2050–2079) time periods.

Projected occurrence and intensity of heat stress causing days

Projections also suggest an increase in the annual occurrence of days causing heat stress (Fig. 2). In EQ, cows are projected to be exposed (mean ± SD) annually to 77.3 ± 7.8, 95.3 ± 5.9, 108.5 ± 11.0 heat stress causing days in REF, FUT1, and FUT2, respectively. Meanwhile, SWQ dairy cows are projected to be exposed on average (± SD) to 120.8 ± 7.1, 138.1 ± 6.9, and 145.3 ± 7.5 d, for these same periods, respectively. In both regions, the highest projected annual number of heat stress causing days was obtained for FUT2 in SIM-06 (EQ: 131.8 ± 17.4 d; SWQ: 176.3 ± 10.2 d). Furthermore, in both regions, the frequency of occurrence of days with a THI falling in the higher-intensity categories increases in the future relative to REF (Fig. 2). Once again, highest increases are projected from REF to FUT2. Overall, our results suggest plausible future increases in the occurrence and severity of heat stress episodes in the two investigated regions.
Fig. 2.
Fig. 2. Mean (± SD) annual number days with a temperature–humidity index (THI) higher than 65 and per THI class [□, (65–70); , (70–75); , (75–80); , (80–85); ▪, ≥85] under different climate scenarios in eastern (EQ) and southwestern (SWQ) Quebec through reference period (1971–2000), near future (2020–2049), and distant future (2050–2079).

Duration of projected heat stress episodes

For EQ, projections have 36.1% ± 0.6% of days associated to heat stress during REF (Fig. 3). This proportion is projected to increase to 41.1% ± 0.9% in FUT1 and to 43.9% ± 2.3% in FUT2, reflecting longer heat stress duration in the future. A change of same sign and similar magnitude was also projected for SWQ as 46.0% ± 1.1% of annual days were classified in the heat stress class (THI > 65) during the reference period, whereas projections have 53.5% ± 1.6% and 54.3% ± 1.6% of heat stress causing days during FUT1 and FUT2, respectively. In addition, in both regions, the number of annual days classified into the “7–8” consecutive days with heat stress category is projected to increase over time. In EQ projections, an average (± SD) of 7.9% ± 0.2% of TD are classified into the longest heat stress duration category during REF. The proportion is projected to increase over time, with 12.8% ± 1.6% and 17.2% ± 2.9% classified in this category for FUT1 and FUT2, respectively. Likewise, in SWQ, 21.0% ± 0.2% of TD were classified into the “7–8” d with heat stress category for REF, whereas 25.8% ± 1.6%, and 29.6% ± 3.0% were projected to be classified in this category for FUT1 and FUT2, respectively.
Fig. 3.
Fig. 3. Mean annual proportion (%) of days in each heat stress accumulation category [□, 0 d with heat stress (HS); , 1–2 d with HS; , 3–4 d with HS; ▪, 5–6 d with HS; ▪, 7–8 d with HS] in eastern (EQ) and southwestern Quebec (SWQ) through 1971–2000, 2020–2049, and 2050–2079.

Projected milk, fat, protein, and lactose yields

Over the past years, several studies have reported production impairments when cows are exposed to heat stress conditions (Wheelock et al. 2010). A previous study reported a decrease in milk fat of 20 g·d−1·cow−1 after an exposure to heat during 1 or 2 d, whereas decreases in milk protein of 20 g·d−1·cow−1 for first and third and greater lactation cows and of 30 g·d−1·cow−1 for second lactation cows were reported after a heat stress exposure of 3 to 4 d (Ouellet et al. 2019). Those results are in line with previous work conducted by Hammami et al. (2013) and by Cowley et al. (2015). Furthermore, no relationship (P > 0.05) between heat stress episodes duration and milk, and lactose yields was obtained (Ouellet et al. 2019). This result was counter-intuitive, considering that milk synthesis is known to be highly sensitive to thermal stress as decreased yields of 35%–40% are common in dairy cows exposed to heat stress of medium to high intensity (West et al. 2003). As an explanation, it was suggested by Ouellet et al. (2019) that the absence of relationship could be related to the intensity of heat stress experienced by dairy cows in a Dfb climate. However, this has not been validated and does warrant further investigation. For this reason, milk and lactose yields were held constant in the three time periods of the present study (REF, FUT1, and FUT2). Based on the developed predictive equations, herds produced on average (± SD) 477 724 ± 95 193 kg milk·yr−1 (32.9 ± 4.6 kg·cow−1·d−1), in EQ, whereas SWQ herds produced on average (± SD) 496 332 ± 114 529 kg milk·yr−1 (30.8 ± 5.8 kg·cow−1·d−1). In addition, EQ herds produce on average 21 917 ± 4376 kg lactose·yr−1 (1.09 ± 0.2 kg·cow−1·d−1), whereas SWQ herds produce 22 646 ± 5233 kg (1.08 ± 0.6 kg·cow−1·d−1).
In EQ herds, compared with REF, projected fat yield decreased on average by 2405 kg in FUT1 and by 2431 kg in FUT2 (Table 3). Relative to REF, differences in projected milk fat production range from 2392 (SIM-03) to 2416 kg·yr−1·herd−1 (SIM-06) in FUT1 and from 2395 (SIM-03) to 2478 kg·yr−1·herd−1 (SIM-06) in FUT2. Comparable decreases were calculated in SWQ as projected fat yield declined on average by 2361 kg in FUT1 and by 2452 kg in FUT2 relative to REF (Table 3). Differences in milk projected fat production varied on a herd basis from 2083 (SIM-05) to 2507 kg·yr−1 (SIM-06) in FUT1 and from 2108 (SIM-05) to 2569 kg·yr−1 (SIM-06) in FUT2.
Table 3.
Table 3. Average (± SD) predicted fat yield per year through reference period (1971–2000), near future (2020–2049), distant future (2050–2070) under six different climate scenarios in eastern Quebec and southwestern Quebec dairy herds.

Note: SD, standard deviation.

Projected protein yields followed the same trend as decreases were obtained for both regions (Table 4). In EQ, projected protein yield declined on average by 1990 kg·yr−1·herd−1 in FUT1 and by 2013 kg·yr−1·herd−1 in FUT2 (relative to REF). Changes in milk protein production (on a herd basis) varied from 1981 (SIM-03) to 1999 kg·yr−1 (SIM-05; SIM-06) for FUT1, and from 1987 (SIM-03) to 2051 kg·yr−1 (SIM-06) for FUT2. Comparable corresponding decreases were calculated in SWQ, with declines of 2057 kg·yr−1 for FUT1 and of 2092 kg·yr−1 for FUT2 (Table 4). Differences in milk projected protein production also varied (on a herd basis) from 2022 (SIM-03) to 2080 kg·yr−1 (SIM-06) in FUT1 and from 2050 (SIM-03) to 2131 kg·yr−1 (SIM-06) in FUT2.
Table 4.
Table 4. Average (± SD) projected protein yield per year through reference period (1971–2000), near future (2020–2049), distant future (2050–2070) under six different climate scenarios in eastern Quebec and southwestern Quebec dairy herds.

Note: SD, standard deviation.

Key and Sneeringer (2014) incorporated projected annual average THI centered on the year 2030 in a production frontier model to estimate how the local thermal environment affects the technical efficiency of dairies across the United States. By 2030, the authors concluded that relative to their baseline year (2010), heat stress would lower milk production of the average dairy from 0.60% to 1.35%, depending on the climate scenario. The authors explained that the smallness of the estimated production effects is most likely related to the fact that only 20 yr separated the warming scenarios and the baseline year. In the current study, the future periods lag the reference period by ∼50 and ∼80 yr, respectively, which increases the chance for THI changes to emerge from inter-annual natural variability. Component yields were not reported in the Key and Sneeringer (2014) study. Herein, considering the Canadian production context that is milk payment based on raw milk components, projections of milk components with climate scenarios and through time were appropriate. Overall, the ensemble of six plausible climate scenarios led to projected fat and protein yields declining over time, with declines reaching a maximum in FUT2 period, with everything but THI duration and occurrence considered constant.

Economic losses

For both regions, evolution of future milk price (CAD·hL−1) was calculated for each climate scenario, to estimate plausible future heat-stress-related economic losses. Estimations assumed permanence of the Canadian regime of milk quota (in kilogram of BF produced per day and per herd), which dairy producers could meet either by selling quota or by buying more cows, in a context where fat yield was to decrease. Two corresponding economic scenarios were, therefore, considered, one with cow numbers kept constant (quota sold) and another with quota kept constant (purchase of additional cows). Milk quota and herd cow numbers for REF were of 50.8 kg BF·d−1 in all climate scenarios and an average of 55.1 lactating cows per herd for EQ; and 60.8 kg BF·d−1 in all climates scenarios (at the exception of SIM-05, with 59.9 kg BF·d−1 due to higher THI) and 57.2 lactating cows per herd for SWQ.

Scenario 1: Number of cows kept constant (milk quota sold)

Assuming that the number of cows is held constant through time periods (EQ: 55.1; SWQ: 57.2) implies that milk quota must be sold to avoid under-quota milk production in a situation when milk fat is decreasing. In this economic scenario, net income decreases from REF to FUT1 and to FUT2 in both regions, due to projected declines in milk components yields, thus resulting in less fat and protein incomes. Furthermore, this scenario included reductions in milk marketing deductions and reductions in costs related to quota interests.
In EQ, overall fat and protein projected declines would lead to an average (± SD) decrease in milk price of $8.84 CAD·hL−1 for FUT1 and of $8.94 CAD·hL−1 for FUT2 (relative to REF; see Table 5). In addition, future projected declines in fat and protein yields would imply that EQ herd owners must sell on average (± SD) 6.6 ± 0.03 kg BF·d−1 from REF to FUT1, which represents average (± SD) decreases in costs of $1.85 ± $0.01 CAD·hL−1 when interest of 5.2% are considered (Table 6). From REF to FUT2, EQ herd owners must sell on average 6.7 ± 0.1 kg BF·d−1, which represents decreases in cost of $1.87 ±$0.02 CAD·hL−1. Moreover, decreases in components income and decreases in costs related to quota interests and milk marketing deductions would result in average net income decreases of $6.86 CAD·hL−1 ($31 777.38 CAD·herd−1·yr−1 or $576.72 CAD·cow−1·yr−1) and $6.93 CAD·hL−1 ($32 101.64 CAD·herd−1·yr−1 or $582.61 CAD·cow−1·yr−1) in FUT1 and FUT2, respectively, relative to REF. When expressed in US dollars (average exchange rate 2010–2015), this would represent losses of $516.91 and $522.19 per cow in FUT1 and FUT2, respectively.
Table 5.
Table 5. Variations in income (CAD·hL−1), expenses (CAD·hL−1) by climate scenario through near (FUT1: 2020–2049) and distant future (FUT2: 2049–2050) relative to the climatic reference period (1971–2000) in eastern and southwestern Quebec when number of cows is held constant.
a
Considering average herd of 55.1 lactating cows producing 4632 hL yr−1 initially owning a quota of 58.8 kg butterfat (BF)·d−1.
b
Considering fat price of $9.67 CAD·kg−1, protein price of $8.89 CAD·kg−1, and lactose price of $1.75 CAD·kg−1.
c
Considering marketing deductions price of $0.14 CAD·kg−1 of solids and transport deductions price of $2.54 hL−1.
d
Assuming that quota is bought at $25 000 CAD·kg−1 of BF and financed at 5.2%.
e
Considering an average herd of 57.2 lactating cows producing 4819 hL·yr−1 initially owning a quota of 60.8 kg BF·d−1 (SIM-01; SIM-02; SIM-03; SIM-04, and SIM-06) or 59.7 kg BF·d−1 (SIM-05).
Table 6.
Table 6. Variations in income (CAD·hL−1), costs (CAD·hL−1) and net income (CAD·hL−1) by climate scenario through near (FUT1: 2020–2049) and distant future (FUT2: 2049–2050) relative to the climatic reference period (1971–2000) in eastern and southwestern Quebec when quota is held constant.
a
Considering average herd owning a quota of 58.8 kg butterfat (BF)·d−1 and initially owning 55.1 lactating cows.
b
Considering fat price of $9.67 CAD·kg−1, protein price of $8.89 CAD·kg−1, and lactose price of $1.75 CAD·kg−1.
c
Considering marketing deductions price of $0.14 CAD·kg−1 of solids and transport deductions price of $2.54 hL−1.
d
Considering variables costs related to cows of $4128.67 CAD·cow−1 from Agritel Web database (http://agritel.fgcaq.com/Agritel) and a value of $3000 CAD·cow−1 at an interest rate of 8%.
e
Considering an average herd owning 57.2 cows and a quota of 60.8 kg BF·d−1 (SIM-01, SIM-02, SIM-03, SIM-04, and SIM-06), and or 59.7 kg BF·d−1 (SIM-05).
In SWQ herds, milk components declines projected from REF to FUT1 and to FUT2 would cause average decrease in milk price of $8.53 and $8.78 CAD·hL−1 (Table 5), respectively. Moreover, future projected fat decreases would imply that SWQ dairy herd owners must sell on average (± SD) 6.5 ± 0.4 kg BF·d−1 from REF to FUT1, which represents average (± SD) decrease in costs of $1.76 ± $0.12 CAD·hL−1 when interest of 5.2% are considered. From REF to FUT2, SWQ herds would have to sell on average 6.7 ± 0.5 kg BF·d−1, which represents a decrease in costs related to quota interests of $1.83 ± $0.09 CAD·hL−1. At last, projected decreases in components income and decreases of costs related to quota interests and milk marketing deductions would result in average net income decreases of $6.64 CAD·hL−1 ($31 995.37 CAD·herd−1·yr−1 or $559.36 CAD·cow−1·yr−1) and $6.82 CAD·hL−1 ($32 854.88 CAD·herd−1·yr−1 or $574.39 CAD·cow−1·yr−1) from REF to FUT1 and FUT2, respectively. When expressed in US dollars (average exchange rate 2010–2015), this would represent losses of $501.35 and $514.83 per cow in FUT1 and FUT2, respectively.

Scenario 2: Milk quota held constant (additional cows required)

Assuming that the amount of owned milk quota is held constant through time entails that cows must be bought to avoid under-quota milk production in a situation when milk fat is decreasing. Relative to REF, the addition of dairy cows in herds would negatively affect protein income and negatively affect lactose income in both regions. Moreover, this scenario entailed increases in milk deductions (transport and marketing) and variable costs related to the additional cows.
In EQ, future projected fat decreases would imply that dairy herd owners must buy on average (± SD) 6.3 ± 0.9 cows in FUT1 and in FUT2 to meet milk quota owned during REF. This represents an average (± SD) increase in costs of $5.37 ± $0.02 CAD·hL−1 in FUT1 and of $5.43 ± $0.04·hL−1 in FUT2 when variable costs of $4128.67 CAD·cow−1·yr−1, purchase cost of cows of $3000 CAD and interest rate of 8% are considered (Table 6). On average (± SD), the higher number of cows during FUT1 and FUT2 would increase annual herd milk production by 525.08 ± 1.98 hL and by 529.09 ± 3.94 hL, respectively. Ultimately, relative to REF, projected variations in incomes from protein and lactose yields, variations in expenses related to milk transport and marketing, and in variable costs related to additional cows would result in net income decreases of $5.37 CAD·hL−1 ($27 693.16 CAD·herd−1·yr−1 or $451.40 CAD·cow−1) in FUT1 and of $5.43 CAD·hL−1 ($28 044.04 CAD·herd−1·yr−1or $456.72 CAD·cow−1) in FUT2. When expressed in US dollars (average exchange rate 2010–2015), this would represent losses of $404.50 and $409.36 per cow in FUT1 and FUT2, respectively.
In SWQ, projected fat decreases would imply an average (± SD) acquisition of 6.4 ± 0.4 lactating cows in FUT1 and of 6.6 ± 0.3 lactating cows in FUT2 to meet milk quota owned during REF. This represents an average (± SD) increase in costs of $5.76 ± $0.3 CAD·hL−1 in FUT1 and of $5.96 ± $0.28 CAD·hL−1 in FUT2 when variable costs of $4128.67 CAD·cow−1·yr−1, purchase cost of cows of $3000 CAD and interest rate of 8% are considered (Table 6). On average (± SD), the purchase of cows during future periods would lead to an annual herd milk production increase of 535.40 ± 31.85 hL (FUT1) and of 553.71 ±25.90 hL (FUT2) relative to REF. Ultimately, projected variations in milk components income and costs related to milk transport and marketing and additional cows, would result in net income decrease of $5.16 CAD·hL−1 ($27 604.69 CAD·herd−1·yr−1 or $434.04 CAD·cow−1) in FUT1 and of $5.28 CAD·hL−1 ($28 353.25 CAD·herd−1·yr−1 or $444.41 CAD·yr−1) in FUT2 relative to REF. When expressed in US dollars (with average exchange rate 2010–2015), this would represent losses of $389.03 and $398.32 per cow in FUT1 and FUT2, respectively.
Only a limited number of studies have evaluated the economic losses associated with heat stress, probably because its effects on performance are often difficult to quantify. When discussing heat stress economic losses in lactating cows, most authors refer to the work conducted by St-Pierre et al. (2003), who estimated economic losses in the US dairy sector from heat stress in lactating animals and young stock, considering different heat abatement scenarios. To do so, the authors used several previous studies conducted in the US to develop biological response functions to heat stress in dairy cattle. When considering all changes relative to dry matter intake, milk production, number of days open, pregnancy rate, culling rate and mortality associated with heat stress during lactation and on young stock, the authors reported average losses of $167 USD·cow−1·yr−1 without any abatement strategy. Higher losses per cow were reported in the present study even if less related heat stress impacts were taken into account. This is mainly explained by the differences in the methodology used to assess heat-stress-related impact on lactating cows, as well as to the consideration of different time periods between both studies, and to the Canadian milk payment being based on components instead of on milk volume as is the case in the US. Furthermore, Canadian milk price is maintained higher than milk prices offered to the dairy producers of the United States.

Limitations

The economic analysis presented in the current study relies on a series of assumptions that may have impacted our results in different ways. The current approach did not consider the ability of dairy producers to adapt to climate change by investing in proper abatement strategies and selecting heat-tolerant cows. Hence, calculations most likely represent an upper bound of economic losses from milk components variations associated with heat stress. Heat stress experienced by cows was based upon THI, which is the most common environmental indicator of heat stress. However, THI does not take into account air speed at which the animal are exposed and differences between individuals. Further, THI was calculated with data retrieved from weather stations as a surrogate to on-farm data, which could have led to an underestimation of the heat stress level actually experienced by cows. The milk performance equations developed had as primary objective to project milk performance under different environmental conditions. Other performance indicators such as dry matter intake, health, and reproductive performance are also affected by heat stress when the condition occurs during lactation. In addition, heat stress occurring in utero and during the dry period may also independently affect milk performance at adulthood and during the subsequent lactation (Ferreira et al. 2016; Laporta et al. 2020). Further studies will be needed to increase the scope of the analysis. In addition, the predictive equations developed to model milk, fat, protein, and lactose yields only considered the duration and occurrence of heat stress episodes, and it did not account for the intensity of the episodes, due to the limited number of milk analysis under high THI in the Province of Quebec. In addition, the use of the climatic reference period of 1971–2000, although necessary to represent climatic evolution, may have led to an overestimation of heat stress impacts compared with if we had taken a more recent reference period. In addition, present-time genetic value of cows was used in all three time periods (REF, FUT1, FUT2). Given the recent efforts put towards genetic selection of highly producing cows, this probably led to an overestimation of the cows’ performance during the climatic reference period and to an underestimation during the future periods. Furthermore, all economic variables were assumed to be constant during each period, whereas they are known to vary in time. Finally, aggregated impacts on the regional and national aggregated demand was not accounted for in our calculations.

Conclusion

An ensemble of six climate scenarios was used to explore plausible outcomes in terms of impacts on milk performance and income from Quebec dairy herds. These scenarios are representative of the interquartile range of the uncertainty in future summer temperature evolution in a larger ensemble of available climate simulations (CMIP5). Results suggest that anthropogenic climate change could increase cow heat stress and consequently cause declines in major milk components, which ultimately would have profound impacts on milk income of Canadian dairy producers if no heat-abatement or other nutritional or genetic strategies are put in place. Estimating future financial losses related to milk components variations in different climate scenarios highlighted the necessity to implement proper management strategies to alleviate heat stress impacts on the Canadian dairy sector, to maximize its profitability in the future, as well as cow welfare.

References

Arora V.K., Scinocca J.F., Boer G.J., Christian J.R., Denman K.L., Flato G.M., et al. 2011. Carbon emission limits required to satisfy future representative concentration pathways of greenhouse gases. Geophys. Res. Lett. 38: L05805.
Bank of Canada. 2020. Annual exchange rates. [Online]. Available from: https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/ [12 July 2020].
Bernabucci U., Lacetera N., Baumgard L.H., Rhoads R.P., Ronchi B., and Nardone A. 2010. Metabolic and hormonal acclimation to heat stress in domesticated ruminants. Animal, 4: 1167–1183.
Bernabucci U., Biffani S., Buggiotti L., Vitali A., Lacetera N., and Nardone A. 2014. The effects of heat stress in Italian holstein dairy cattle. J. Dairy Sci. 97: 471–486.
Collier R.J. and Gebremedhin K.G. 2015. Thermal biology of domestic animals. Annu. Rev. Anim. Biosci. 3: 513–532.
Cowley F.C., Barber D.G., Houlihan A.V., and Poppi D.P. 2015. Immediate and residual effects of heat stress and restricted intake on milk protein and casein composition and energy metabolism. J. Dairy Sci. 98: 2356–2368.
Ferreira F.C., Gennari R.S., Dahl G.E., and de Vries A. 2016. Economic feasibility of cooling dry cows across the United States. J. Dairy Sci. 99: 9931–9941.
Galán E., Llonch P., Villagrá A., Levit H., Pinto S., and del Prado A. 2018. A systematic rewiew of non-productivity-related animal-based indicators oh heat stress resilience in dairy cattle. PLoS ONE, 13: e0206520.
Grenier P. 2018. Two types of physical inconsistency to avoid with univariate quantile mapping: a case study over North America concerning relative humidity and its parent variables. J. Appl. Meteor. Climatol. 57: 347–364.
Hatfield, J., Boote, K., Fay, P., Hahn, L., Izaurralde, C., Kimball, B.A., and Mader, T. 2008. Agriculture. Pages 21–47 in The effects of climate change on agriculture, land resources, water resources, and biodiversity in the United States. A Report by the U.S. Climate Change Science Program and the Subcommittee on Global Change Research, 362. Washington, DC, USA.
Hammami H., Bormann J., M’hamdi N., Montaldo H., and Gendler N. 2013. Evaluation of heat stress effects on production traits and somatic cell scores of Holsteins in a temperate environment. J. Dairy Sci. 96: 1844–1855.
IPCC., 2014. Climate change 2014: synthesis report. In Core Writing Team, R.K. Pachauri, and L.A. Meyer, eds. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. [Online]. Available from https://www.ipcc.ch/report/ar5/ [10 Apr. 2015].
Key N. and Sneeringer S. 2014. Potential effects of climate change on U.S. dairies. Am. J. Agric. Econ. 96: 1136–1156.
Laporta J., Fereira F.C., Ouellet V., Dado-Senn B., Almerida A.K., De Vries A., and Dahl G.E. 2020. Late-gestation heat stress impairs daughters and granddaughters’ lifetime performance. J. Dairy Sci. 103: 7555–7568.
NRC. 1971. A guide to environmental research on animals. National Acadamy Science, Washington, DC, USA.
Ouellet V., Cabrera V.E., Pacheco L.F., and Charbonneau É. 2019. The relationship between the number of consecutive days with heat stress and milk production of Holstein dairy cows raised in a humid continental climate. J. Dairy Sci. 102: 8537–8545.
Ouranos. 2015. Vers l’adaptation. Synthèse des connaissances sur les changements climatiques au Québec, Montréal, QC, Canada. 417 p.
Peel M.C., Finlayson B.L., and McMahon T.A. 2007. Updated world map of the Köppen-Geiger climate classification. Hydrol. Earth Syst. Sci. 11: 1633–1644.
Polsky L. and Von keyserglingk M.A.G. 2017. Invited review: effects of heat stress on dairy cattle welfare. J. Dairy Sci. 100: 8645–8657.
Schüller L., Burfeind O., and Heuwieser W. 2014. Impacts of heat stress on conception rate of dairy cows in the moderate climate considering different temperature-humidity index thresholds, periods relative to breeding, and heat load indices. Theriogenelogy, 81: 1050–1057.
Separovic L., Alexandru A., Laprise R., Martynov A., Sushama L., Winger K., et al. 2013. Present climate and climate change over North America as simulated by the fifth-generation Canadian regional climate model. Clim. Dyn. 41: 3167–3201.
Stevens B., Giorgetta M., Esch M., Mauritsen T., Crueger T., Salzmann S.R.M., et al. 2013. Atmospheric component of the MPI-M Earth System Model: ECHAM6. J. Adv. Model. Earth Syst. 5: 146–172.
St-Pierre N.R., Cobanov B., and Schnitkey G. 2003. Economic losses from heat stress by US livestock industries. J. Dairy Sci. 86(E. Suppl.): E52–E77.
St-Pierre N.R. 2003. Reassessment of biases in predicted nitrogen flows to the duodenum by NRC 2001. J. Dairy Sci. 86: 344–350.
Taylor K.E., Stouffer R.J., and Meehl G.A. 2012. An overview of CMIP5 and the experiment design. Bull. Am. Meteorol. Soc. 93: 485–498.
Themeßl M.J., Gobiet A., and Heinrich G. 2011. Empirical-statistical downscaling and error correction of regional climate models and its impact on the climate change signal. Clim. Change. 112: 449–468.
Valacta. 2011. Évolution de la production laitière Québécoise 2010. Page 42 in Le Producteur de Lait Québécois. Numéro spécial.
Valacta. 2012. Évolution de la production laitière Québécoise 2011. Page 48 in Le Producteur de Lait Québécois. Numéro spécial.
Valacta. 2013. Évolution de la production laitière Québécoise 2012. Page 54 in Le Producteur de Lait Québécois. Numéro spécial.
Valacta. 2014. Évolution de la production laitière Québécoise 2013. Page 56 in Le Producteur de Lait Québécois. Numéro spécial.
Valacta. 2015. Évolution de la production laitière Québécoise 2014. Page 40 in Le Producteur de Lait Québécois. Numéro spécial.
Valacta. 2016. Évolution de la production laitière Québécoise 2015. Page 38 in Le Producteur de Lait Québécois. Numéro spécial.
van Vuuren D.P., Edmonds J., Kainuma M., Riahi K., Thomson A., Hibbard K., et al. 2011. The representative concentration pathways: an overview. Clim. Change, 109: 5–31.
Watanabe M., Suzuki T., O’ishi R., Komuro Y., Watanabe S., Emori S., et al. 2010. Improved climate simulation by MIROC5: mean states, variability, and climate sensitivity. J. Climate, 23: 6312–6335.
West J.W. 2003. Effects of heat-stress on production in dairy cattle. J. Dairy Sci. 86: 2131–2144.
Wheelock J.B., Rhoads R.P., VanBaale M.J., Sanders S.R., and Baumgard L.H. 2010. Effects of heat stress on energetic metabolism in lactating Holstein cows. J. Dairy Sci. 93: 644–655.
Yukimoto S., Adachi Y., Hosaka M., Sakami T., Yoshimura H., Hirabara M., et al. 2012. A new global climate model of the Meteorological Research Institute: MRI-CGCM3 — model description and basic performance. J. Meteorol. Soc. Jpn. 90: 23–64.

Information & Authors

Information

Published In

cover image Canadian Journal of Animal Science
Canadian Journal of Animal Science
Volume 101Number 2June 2021
Pages: 242 - 256
Editor: Filippo Miglior

History

Received: 8 May 2020
Accepted: 3 August 2020
Accepted manuscript online: 21 August 2020
Version of record online: 21 August 2020

Permissions

Request permissions for this article.

Key Words

  1. heat stress
  2. dairy (cows)
  3. climate change
  4. dairy economics

Mots-clés

  1. stress thermique
  2. laitière (vaches)
  3. changements climatiques
  4. économie laitière

Authors

Affiliations

Université Laval, 2425, rue de l’Agriculture, Québec, QC G1V 0A6, Canada.
P. Grenier
Groupe Science du climat et services climatiques, Ouranos, Montréal, QC H3A 1B9, Canada.
D.E. Santschi
Lactanet, Ste-Anne-de-Bellevue, QC H9X 3R4, Canada.
V.E. Cabrera
Department of Dairy Science, University of Wisconsin-Madison, Madison, WI 53706, USA.
L. Fadul-Pacheco
Department of Dairy Science, University of Wisconsin-Madison, Madison, WI 53706, USA.
É. Charbonneau
Université Laval, 2425, rue de l’Agriculture, Québec, QC G1V 0A6, Canada.

Notes

Copyright remains with the author(s) or their institution(s). Permission for reuse (free in most cases) can be obtained from copyright.com.

Metrics & Citations

Metrics

Other Metrics

Citations

Cite As

Export Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by

1. Trends in the risk of heat stress to Canadian dairy cattle in a changing climate
2. Trends in the risk of heat stress to Canadian dairy cattle in a changing climate
3. Comparing thermal conditions inside and outside lactating dairy cattle barns in Canada
4. Thermal stress influence on the productive and economic effectiveness of Holstein-Friesian dairy cows in temperate climate
5. How heat stress conditions affect milk yield, composition, and price in Italian Holstein herds
6. Critical Temperature-Humidity Index Thresholds for Dry Cows in a Subtropical Climate

View Options

View options

PDF

View PDF

Get Access

Login options

Check if you access through your login credentials or your institution to get full access on this article.

Subscribe

Click on the button below to subscribe to Canadian Journal of Animal Science

Purchase options

Purchase this article to get full access to it.

Restore your content access

Enter your email address to restore your content access:

Note: This functionality works only for purchases done as a guest. If you already have an account, log in to access the content to which you are entitled.

Media

Media

Other

Tables

Share Options

Share

Share the article link

Share on social media

Cookies Notification

We use cookies to improve your website experience. To learn about our use of cookies and how you can manage your cookie settings, please see our Cookie Policy.
×