How does reproduction account for dairy farm sustainability?

Abstract Sustainability - the new hype of the 21st century has brought discomfort for the government and society. Sustainable agriculture is essential to face our most concerning challenges: climate change, food security, and the environmental footprint, all of which add to consumers' opinions and choices. Improvements in reproductive indexes can enhance animal production and efficiency, guaranteeing profit and sustainability. Estrus detection, artificial insemination (AI), embryo transfer (ET), estrus synchronization (ES), and multiple ovulations are some strategies used to improve animal reproduction. This review highlights how reproductive strategies and genetic selection can contribute to sustainable ruminant production. Improved reproductive indices can reduce the number of nonproductive cows in the herd, reducing methane emissions and land use for production while preserving natural resources.


Introduction
Sustainability -the new hype of the 21 st century has brought discomfort for the government and society. But is this topic a novelty in research and politics areas?
The concept of sustainability was first addressed in forestry near the 17 th and 18 th centuries -with the idea the never harvest more than the forest could yield in new cycles (Wiersum, 1995). However, it was only in 1987, when the United Nations World Commission on Environment and Development (WCED) published the Brundtland Report, that the term 'sustainable development' became popular and was defined as "development that meets the needs of the present without compromising the ability of future generations to meet their own needs". After that, agendas and declarations were built to guide 'sustainable development', but all of society did not accept the idea. Later, in the mid-1990s, the concept was brought into evidence again, gathering researchers' and politicians' attention (Purvis et al., 2019).
Sustainable agriculture is essential to face our most concerning challenges: climate change, food security, and the environmental footprint, all of which are added to consumers' opinions and choices. According to the United States Department of Agriculture (USDA, 2023), greenhouse gas (GHG) emissions from agriculture accounted for 11.2% of total United States of America (USA) emissions in 2020, where 5.6% is due to direct nitrous oxide, 4.2% to direct methane, 0.8% to direct carbon dioxide, and 0.6% to electricity-related. However, in 2016, Brazilian agriculture contributed 33.2% to total GHG emissions in Brazil (Brasil, 2023), evidencing the distinction on GHG emissions between countries in respect to the proportion of agriculture-based economics.
In addition, food derived from animal products (i.e.: dairy and beef) provides essential nutrients for the human diet. Thus, over the years, animal production has increased and adapted to feed the world population; however, ruminant production has contributed to GHG emissions, mainly due to enteric methane (CH 4 ).
Methane is an abundant non-CO 2 GHG with a shorter atmospheric lifespan, around nine years, and its reduction allows more rapid benefits for climate change (Ripple et al., 2014). The total GHG emissions from global livestock are 7.1 Gigatonnes (Gt) of carbon dioxide equivalent (CO 2 -eq) per year, representing 14.5% of all anthropogenic GHG emissions. From the 7.1 Gt CO 2 -eq, 44% of emissions are methane (CH 4 ), 29% as nitrous oxide (N 2 O), and 27% as CO 2 (FAO, 2023). There are distinct anthropogenic sources of CH 4 (ruminants, fossil fuel industry, landfills, biomass burning, and rice production); however, ruminants are the largest source ( Figure 1; Ripple et al., 2014). Moreover, CH 4 emission intensities vary from one commodity to another. The highest levels of CO 2 -eq in livestock are produced by beef (around 300 kg CO 2 -eq/kg of protein produced), followed by small ruminants (beef and milk; 165 and 112 kg CO 2 -eq/kg of protein produced, respectively) and cow milk, chicken and pork product, which are at the bottom of emission intensity list (below 100 CO 2 -eq/kg of protein produced) (FAO, 2023). To better address the importance of ruminants for climate change, first, we need to understand how they participate in GHG emissions. Ruminant digestion is a process of enteric fermentation in a multichambered stomach (Ripple et al., 2014), where ruminant microbes can convert plant carbohydrates to energy to benefit them and the animal (Knapp et al., 2014). In the reticulorumen and hindgut, carbohydrates are hydrolyzed by microbial enzyme activity -sugars are fermented to volatile fatty acids producing reducing equivalents (i.e., metabolic hydrogen). This metabolic hydrogen is then converted to H 2 by hydrogenase-expressing bacterial species and H 2 is converted to CH 4 by methanogenic archaea. This is an essential mechanism since H 2 can negatively impact carbohydrate degradation, microbial growth, and microbial protein synthesis (Knapp et al., 2014). Thus, it is imperative to focus on mechanisms to mitigate CH 4 production by ruminants, such as feeding management and nutrition, rumen modifiers, and an increase in animal production through genetics and reproductive approaches (Knapp et al., 2014). Improvements in reproductive indices can enhance animal production and efficiency, guaranteeing profit and sustainability (Hufana- Duran and Duran, 2020). Estrus detection, artificial insemination (AI), embryo transfer (ET), estrus synchronization (ES), and multiple ovulations are some strategies used to improve animal reproduction. Efficient reproduction is vital for dairy cows due to their high milk yields since low reproductive indices can increase days open, implying a more extended period in an unproductive phase (Pinedo et al., 2020). Furthermore, genetic selection associated with improved reproductive characteristics can promote sustainable livestock and decrease CH 4 emissions by 10 to 15% (Garnsworthy, 2004). Therefore, this review highlights how reproductive strategies and genetic selection can contribute to sustainable ruminant production.

Effect of calving intervals on greenhouse gases emissions
Dairy production comprises gestation cycles, calving, lactation, and a dry period preceding the next calving (Lehmann et al., 2016). Traditional dairy systems have managed cows to calve once a year (e.g., 12-month calving interval). This reproductive strategy is based on the idea that early conception benefits the production economy, which arose from 1960s studies showing that annual milk production was maximized by calving intervals between 12 and 13 months (Speicher and Meadows, 1967;Louca and Legates, 1968).
To achieve the 12-month of calving interval, the first insemination will occur when production levels are still high and a positive energy balance is yet to be re-established, increasing the risk of metabolic disorders and failed conception (Browne et al., 2015). Such conditions have made current dairy systems question the annual calving interval as an ideal practice. Moreover, because calving intervals are closely related to the number of calves and replacement heifers in the herd and the efficiency of milk production , recent research has focused on the role of calving intervals on GHG emissions. Mitigation strategies for GHG emissions from livestock have been pointed out as a critical part of climate obligations (Wall et al., 2012). Wall et al. (2012) examined the effects of three lactation length scenarios (305, 370, and 440 days) on GHG emissions using United Kingdom dairy herd data. The tested lactation lengths were equivalent to the conventional annual calving target, the UK's average calving interval (12.3 months), and an 18-month calving interval. The authors estimated that longer calving intervals required fewer milking cows and replacements to maintain milk yield levels; nonetheless, CO 2 equivalent (CE)/farm per year increased by 157 t when calving intervals were extended from 12 to 18 months. In this study, the annual herd milk yield remained constant, and the numbers of cows and replacements were allowed to vary to maintain yields for each lactation-length scenario.
When the number of cows in the herd was kept constant and calving intervals were manipulated through different timings of first insemination, Lehmann et al. (2019) reported decreases in carbon footprint (by up to 8.2% per annual cow) by extending calving intervals from 13 to 18 months due to less feed production and enteric fermentation. Similarly, Browne et al. (2015) reported lower total emissions and emissions intensity (t CO 2 e/t milk fat plus protein) for 18-month calving intervals compared to annual calving.
Several authors have advocated the extension of calving intervals and lactation in dairy cows (Lehmann et al., 2014(Lehmann et al., , 2016Sehested et al., 2019;Burgers et al., 2021). The possibility of reducing GHG emissions through longer calving intervals is mainly attributed to more lactation days and fewer dry days per cow per year (if the dry period length remains unchanged), and fewer calves and replacement heifers (reducing replacement rate per year; Lehmann et al., 2016). The GHG related to feed use by youngstock are accounted for in the milking herd; therefore, by reducing the number of youngstock, longer calving intervals could possibly aid in mitigating GHG emissions by reducing herd feed use per kilogram of milk produced and GHG emissions from animals not contributing to production Sakatani, 2022).
Although the efficacy of extending calving intervals for mitigation of GHG emissions is still under debate, Kok et al. (2019) observed a 1.0% and 1.7% increase in GHG (CO 2eq /t of milk fat plus protein) from heifers and cows when lactation was extended in two months and four months, respectively, but emissions were similar to baseline calving interval (mean of 390 days for primiparous and multiparous cows) or even reduced when lactation persistency or the lifespan of cows was increased. These results suggest that lactation persistency and production level (e.g., primiparous, or multiparous cows) may play a role in GHG emitted from cows managed under longer calving intervals.

Estrus detection and GHG as a tool for sustainability
More attention to cows' reproduction and technological strategies adopting can result in efficient performance, guaranteeing profitability and sustainability (Hufana- Duran and Duran, 2020). In addition, estrus detection is an essential factor affecting reproductive performance, and failure to detect it or misdiagnosis can result in significant economic losses (Senger, 1994).
The traditional and most used estrus detection method is the farm staff's direct observation (Palmer et al., 2010), resulting in efficiency below 50% up to 90% (Roelofs et al., 2010). However, estrus detection is a usual problem of dairy farms, mainly due to the labor required (Mayo et al., 2019) for cows' observation and the occurrence of short periods of estrus in high-producing dairy cows (Wiltbank et al., 2006), resulting in economic losses by $360 per missed estrus (De Vries, 2006).
Several devices for the automation of estrus detection have been developed to face the low rate of estrus detection (Firk et al., 2002). The use of pedometers, chin-ball markers, heat-mount detectors, devices that measure vaginal or milk temperature, and devices that measure the electrical impedance of the genitalia or vaginal mucus and radiotelemetry (Brehme et al., 2008;Duran et al., 2015) are examples. Results from studies indicate a considerable potential to detect estrus with more precision to improve detection rates and reduce error rates. In addition, estrus detection can reduce the environmental impact by reducing the number of nonproductive animals in the farms (Sakatani, 2022).
The efficiency of estrus detection and the time to the beginning of breeding after calving influenced the cost of production and methane emissions (Archer et al., 2015). For an average UK herd (126 cows and 7.353 annual milk yield per cow), this saved at least £50 per cow and a 3.6% reduction in methane emissions per liter of milk when the estrus synchronization of first insemination was used and compared with breeding based on observed estrus. So, estrus synchronization can contribute to reducing GHG emission.

Artificial insemination and GHG
Artificial insemination (AI) is essential to improve herds' genetic efficiency (Hufana-Duran and Duran, 2020). The genetic advance achieved with artificial insemination can increase milk production without expanding the number of animals in dairy herds (Gifford and Gifford, 2013); thus, indirectly, AI can enhance the system's sustainability. The adoption of AI, mainly in Brazil, is related to using other production systems as farm-housed cows (Santos et al., 2021), reducing production areas while preserving the natural resources.
According to Hristov et al. (2013a), assisted reproductive technologies, such as AI, have a high relative effectiveness in mitigating non-CO 2 GHG emissions ( Table 1). The improvement in fertility can reduce the number of unproductive animals kept on farms and the number of replacement heifers needed. Moreover, reducing culling rates from 35 to 30% may reduce whole-herd enteric CH 4 emissions by 3.1% when the age at first calving is around 26 months (Knapp et al., 2014).  Garnsworthy (2004) also observed that fertility scenarios guided by AI would result in different CH 4 outputs for cows and replacement heifers (ton/yr; Figure 2). Therefore, enhancement of fertility levels was likely to reduce CH 4 emissions by 24% and ammonia emissions by 17% (Table 2).

Figure 2.
Annual methane output per 100 cows in dairy herds with no milk quota and a mean annual milk yield of 6000 kg per cow, and with current levels of fertility (A) with 78 days to first insemination, 50% of estrus detection rate, 38% of conception rate to first AI and 37% conception rate to subsequent AI; 1995 levels (B) with 72 days to first insemination, 55% of estrus detection rate, 47% of conception rate to first AI and 46% conception rate to subsequent AI or ideal levels (C) with 70 days to first insemination, 70% of estrus detection rate, 65% of conception rate to first AI and 60% conception rate to subsequent AI. Adapted from Garnsworthy (2004).

Embryo transfer and farm sustainability
The ET started to be developed in farm dairy cows in the 1940s and 1950s (Rowson, 1951), consisting of the transfer of a viable embryo produced in vivo from a donor cow or produced in vitro after follicular aspiration to the uterine horn of a receiving cow. From this technique, it is possible to produce several embryos of superior cows, and the introduction of in vitro fertilization allowed to multiply the number of embryos produced, enhancing the positive effects of embryo transfer on genetic gain, and resulting in greater milk production (Lohuis, 1995).
As discussed earlier, enhancing the number of high-producing dairy cows enables the reduction or elimination of low-producing and non-producing cows in the dairy farm; it can mean a reduction of CH 4 intensity, mainly by the increase of conception rate and herd's genetic gain when ET is used (Hristov et al., 2013b). Furthermore, ET is a prime strategy to improve the fertility of heat-stressed high-producing dairy cows, increasing the pregnancy rate by 80.8% compared to the prostaglandin plus estrus technique (Baruselli et al., 2020).

Genetic improvement and GHG
For many years, livestock was blamed for the rise in GHG emissions. Over time, strategies such as genetic selection (Sypniewski et al., 2021) were implemented to reduce CH 4 production (Króliczewska et al., 2023).
The heritability for CH 4 traits is moderate, ranging from 0.12 to 0.45 (Breider et al., 2019;López-Paredes et al., 2020;Króliczewska et al., 2023). Furthermore, a high heritability (rg = 0.94) between daily CH 4 production and CH 4 intensity (de Haas et al., 2011) suggests that selecting for CH 4 will result in lower CH 4 units per milk produced (Kamalanathan et al., 2023) as described in temperate conditions studies (Table 3). Genetic selection is a powerful strategy for reducing CH 4 emissions. CH 4 intensity can be reduced by 1.25% per year by genetic selection (de Haas et al., 2021). These metrics have been incorporated as a goal in breeding programs, allowing for a reduction of 0.021 mg/L in five generations (Calderón-Chagoya et al., 2021).
Although CH 4 production increases as milk yield increases due to genetic selection (Hossein-Zadeh, 2022), the main should be on CH 4 intensity (g of CH 4 per unit of milk yield).
Reducing CH 4 at the expense of milk yield, DMI, or sacrificing economic gains should be avoided (Richardson et al., 2022;Króliczewska et al., 2023).
High-producing dairy cows can reduce GHG intensity. Lahart et al. (2021) compared the top 5% cows to a group representative of the national average genetic merit and showed that elite cows reduced GHG intensity and enhanced N efficiency. Interestingly, this study also evaluated three feeding systems (low grass allowance; high grass allowance; and high concentrate) and found that a high concentrate diet had greater GHG due to growing, manufacturing, and transportation of the additional concentrate used, indicating that other factors, other than animal model, must be considered.
As reported by Cairo (2023 forthcoming), the improvement in milk yield each year was 0.383 kg, despite Girolando and Jersey's cows increasing milk yield by over 0.5 kg per year. Similarly to other worldwide breeding program (Zhang et al., 2019), CH 4 production increased by 16.7% in Brazil. However, the CH 4 intensity was reduced by 0.82, 1.95, 1.70, 1.21, and 1.74% per year for Holstein, Girolando (Holstein x Gyr), Gyr, Jersey, and Guzera, respectively (Cairo, 2023 forthcoming).
CH 4 intensity has recently emerged as a viable measure for genetic selection (Kandel et al., 2018). As a result, it is better to have fewer cows producing more milk, diluting the CH 4 in the final product, rather than having more cows producing less CH 4 , but also less milk (de Haas et al., 2021). Furthermore, milk yield is positively correlated with CH 4 production, indicating that caution is required when the goal of genetic selection is lower CH 4 production (Breider et al., 2019). So, genetic selection appears to be a strategy to reducing GHG emissions and improving sustainability (Hossein-Zadeh, 2022;González-Recio et al., 2020).

Conclusions
Improved reproductive indices can reduce the number of nonproductive cows in the herd, reducing CH 4 emissions and land use for production while preserving natural resources. Only genetic selection as an approach for dairy farm sustainability may reduce CH 4 emissions by more than 1% per year.