A review of megatrends in the global dairy sector: what are the socioecological implications?

The global dairy industry is undergoing a period of expansion and consolidation, alongside heightened critique and competition from non-dairy alternatives. This review identifies four key megatrends within the global dairy sector, focusing in on the socioecological challenges associated with each. The megatrends were identified through a literature review of recent publications within the dairy science and social science fields, as well as a review of grey literature from intergovernmental and institutional reports. Key findings include geographical range shifts in production and consumption of dairy milk from the Global North to the Global South; intensification of production agendas that strive for mechanisation, standardisation, and corporatisation of the sector; increasing awareness of the ecological impacts of intensive dairying; and finally, disruptions to the sector driven by plant-based milks and, potentially, synthetic milks. We identify under-researched socioecological challenges associated with each of these trends. Although dairy milk may be homogenous in its final form, the sector remains heterogenous in its impacts across spaces, places, and scales, as increasingly intensive dairying systems fundamentally reshape human–cattle relations. The combined impacts of these trends bring into question the mythologies of milk and the assumed desirability of ever-expanding dairy industries. Our review finds that the future of dairy is not clear nor uncontroversial and that more attention needs to be directed to maximising and broadening the social benefits of the dairy and dairy alternatives, minimising the human and non-human costs, and limiting contributions to global climate change.


Introduction
Dairy 1 milk exists as a fresh product (as milk or by-products of milk), in powdered forms, and as a synthesized set of proteins and fats in industrial foodstuffs. The United Nations Food and Agriculture Organisation (FAO 2019) estimates that over 80% of the world's population consume dairy products on a regular basis. Innumerable human and nonhuman lives and associated ecologies are entangled in milk production and consumption, creating social and ecological (socioecological) vulnerabilities that are casting clouds over the future of the sector. Based on the analysis below, we believe the global dairy sector is likely to undergo a period of significant transition over the next few decades.
There have been increasing calls to move beyond animalbased food systems to more sustainable forms of food production in a bid to address the true costs of animal agriculture and build more resilient food systems (for example, see Poore and Nemecek 2018;Springmann et al. 2018;Willet et al. 2018;Pieper et al. 2020). Morris et al. (2021, p. 2) characterise this shift as a "societal grand challenge". In this paper we engage with this challenge by synthesising current trends in dairy and identifying emerging challenges and opportunities. The aim is to generate knowledge that can contribute to research agendas and policy dialogues about how to live (and eat) sustainably in the current epoch of global environmental change.
Our approach differs from other reviews of the sector conducted via institutional sustainability reports, dairy sciences, 1 3 or agricultural economics that tend to focus upon expansion and growth within the sector. Instead, we analyse the socioecological dimensions of change. In doing so we are interested in how people, places and environments are being affected by trends within the dairy sector in order to highlight unintentional or hidden impacts of change.
The first section of the paper explains the methodological framing for this review. Following this, we identify four megatrends, where megatrends are approached as trajectories of social, economic, environmental, or technological change occurring over the coming decades that "continually reshape the world around us" (Hajkowicz 2015, p. 3). The four trends we identify are: (1) Shifting geographies and scales of production and consumption, (2) Intensification of capital, land and animals, (3) Growing awareness of the ecological impacts of dairy, (4) Alternative milk distruptions.
The discussion section of the paper synthesises the four trends to reveal how they come together to inform the scale and diversity of challenges facing the sector and their associated socioecological implications. Our analysis raises questions about the mythologies of milk, particularly the often-unspoken assumption that an ever-expanding industry is good for people, cattle or the planet.

Methodology
In this section, we explain our methodological approach. We outline our data collection process, how we are approaching socioecological impacts, and how we define and identify megatrends. We also briefly reflect on the limitations of our review through what was included and excluded through this process. Before continuing, it is important to note our focus is on planetary trends, however we are also interested in differences between the countries commonly grouped as the Global North and Global South. Countries in the Global North include European, North American, Australian and New Zealand dairy industries, whereas the Global South incorporates predominantly Asian, African and Latin American markets. We draw on these imagined geographies to discuss trends, though we acknowledge that the reality of these categories is much more nuanced and a detailed synthesis of each region's dairy industry is likely to reveal much greater diversity and difference.

Data selection process
To begin the process for this review, we read through recent grey literature produced by influential multilateral institutions that reported on the current state and future modelling of the global dairy sector. This included reports from the United Nations FAO (see FAO 2019FAO et al. 2020), the OECD-FAO coalition (see OECD-FAO 2010, as well as information from the Intergovernmental Panel on Climate Change (IPCC) special report on land and climate change (see Mbow et al. 2019). We also analysed high profile papers that have examined the role of livestock within future food systems in the context of climate change (see Springmann et al. 2019;Willet et al. 2018). After reading these reports, key search terms were chosen based upon reoccurring themes in the literature. These were used in Scopus and Google Scholar database searches and included terms like "dairy trends", "dairy review", "dairy" AND "sustainability", "climate change", "plant-based milks", "synthetic dairy", "synthetic milks" and similar searches.
In this scoping study, we prioritised articles from the period 2017 to 2021. 2 We followed a similar review process to Aschemann-Witzel et al. (2020) work on plant-based trends to inform our selection of academic literature, focussing on peer-reviewed journal articles published in the past 5 years that highlighted the most current trends and issues in the sector. Our search highlighted academic literature that came predominantly from the fields of dairy science, food innovation and social science scholarship. We initially identified academic reviews of the dairy sector, such as those published within the Journal of Dairy Science (see Beaver et al. 2020;Schuster et al. 2020;Britt et al. 2018;Pulina et al. 2018;McCarthy et al. 2017) and other key review papers across different environmental, social and nutritional journals about dairy production and consumption (see Gauly et al. 2013;Aschemann-Witzel et al. 2020;Fiel et al. 2020;Roy et al. 2020;Stephens 2020;Hjalsted et al. 2020;Lonkila and Kaljonen 2021;Cogato et al. 2021;Wankar et al. 2021). We narrowed in on the reviews that we felt had particular relevance to our socioecological framing. We also purposively selected 40 case study papers referred to in the reviews that concentrated upon a socioecological dimension of a megatrend we were interested in, such as plant-based milks or intensification. Our data selection process is shown in Fig. 1. A list of all reviewed articles is available in "Appendix".

Socioecological analysis
Our review centres upon the megatrends that are contributing to socioecological change within dairy industries. Rather than see ecological or social impacts bracketed off from one another as may be the case in an economic or environment review, we attempt to bring the multiple human and nonhuman actors that comprise dairying, including the economies, environments and bodies that enable and are shaped by dairying, into view and consideration. Our approach has been influenced by political ecology, more-than-human geography and the broader environmental humanities, which highlight that the relations between things as an important factor in shaping the world (see Whatmore 2002;Latour 1999;Haraway 2003). Concepts like "co-becoming" recognise how humans, cattle and environments never evolve on their own but are caught up in complex relations that change over time. Here we are interested in identifying the megatrends that are reshaping these relations within dairy industries and the ramifications for the human and non-human actors within it. This requires recognising cattle as "lively" more-than-human actors (Collard and Dempsey 2013;Holloway and Bear 2017;Bear and Holloway 2019;Gillespie and Collard 2015;Gillespie 2021) who shape and are shaped by these trends and highlighting the intended and unintended outcomes of dairy industries at different scales (for example, see McGregor et al. 2021). It also requires a level of abstraction, as we cannot do justice to the in-depth ethnographic work on dairying, and instead draw upon selected case study research to help think through some of the impacts of megatrends for socioecological communities.
In doing so, we build upon and extend the work of Clay and Yurco (2020) on the political ecology of milk, Clay et al.'s (2020b) examination of the drivers of dairy intensification and possibilities of dairy alternatives, Mylan et al.'s (2019) exploration of the possibilities of plant-based milk and Zafrilla et al.'s (2020) review of the sustainability challenges to dairy livestock systems. These researchers have begun challenging the implicit assumption in much of the literature that dairy should be encouraged to grow and expand. Instead, our socioecological analysis problematises this thinking by tracing how megatrends in dairy are related to changing relations and socioecological wellbeing of humans, non-humans, and environments. In doing so we follow Sexton et al. (2019, p. 48) who argue that such approaches can create more openings for narratives that evoke and invent "more compassionate human-animal relationships".

Data analysis: identifying megatrends
The idea of megatrends was first introduced by Naisbitt (1982) as a means identifying new ideas, interpretative communities, ways of knowing and worldviews for subsequent analysis (Slaughter 1993). The mega refers to the macro-scale of the processes and the likely impacts of the trend, with Mittelstaedt et al. (2014, p. 254) arguing that megatrends are complex "social science constructs" which are "seismic in their effect, both in time and space". They are "seismic" because as Visconti et al. (2014, p. 363) assert, a megatrend "is the beginning of a new trajectory that interrogates the established direction a system has followed before". As such megatrends destabilise the status quo creating new opportunities and challenges for those reliant upon existing structures. They are social because, as Rohner (2018, p. 30) contends, megatrends "impacts the lens with which society views the world, thus influencing values and thinking". We approach megatrends within dairy as sectorwide processes that are transforming how and where the dairy sector operates and how socioecologies are affected by and interact with dairying. Megatrends shape and are shaped by existing socioecological structures including institutional systems, organisations, operations, environments and human and non-human actors (Turner 2005).
We identified megatrends in the global dairy sector through an iterative process. We first identified a broad range of reoccurring themes that appeared in the grey and academic literature that related to socioecological challenges and opportunities. Each of these themes are relevant to the dairy sector, are discussed in dairy industry reviews, and tend to attract case study research. We then sought to group 1 3 these themes to identify the broader megatrend of which they were a part. For example, robotics and automated milking is an important theme that is transforming elements of the sector however the disruptive influence of automation is much more significant when considered alongside the related themes of corporatisation and standardisation, both of which are aligned with automation, collectively contributing to a megatrend around dairy intensification. Similarly, while dairy emissions are a source of much analysis, when combined with various other issues in the literature, such as the climate change induced heat stress or resource depletion, they combine to form a sector wide megatrend about the growing concern of various aspects of dairying's relationship with the environment. Themes shifted to megatrends when they could be grouped in a way that suggested macroscale, sector-wide changes. The themes and megatrends are evident in Fig. 2. In focusing on socioecological dimensions of megatrends and identifying them in this iterative way we do not attempt to identify all the trends happening across the dairy sector. There are two main gaps in our analysis. First, we do not focus in on trends within dairying subfields such as food science, genetics, nutrition, or veterinary science. While important advancements are occurring in each of these fields it is only at the point in which trends in these fields impact the broader socioecological relations of the dairy industry that they would become incorporated into this analysis. Second, and relatedly, our focus is upon large scale trends that are occurring now in the dairy sector rather than a focus on emerging research priorities or small-scale changes. As such we acknowledge, for example, that there is a great deal of promising discussion about climate smart and regenerative agriculture amongst key multilateral bodies (for example, see FAO 2021), however, based on our review of the literature, this has yet to have widespread impacts that are causing a seismic shift within the sector. Instead, we see these sorts of initiatives as themes that emerging as part of a broader megatrend concerned about the environmental impacts of dairying. Indeed, if such initiatives are to gain widespread uptake, they will need to engage with other more established megatrends, such as intensification, that seems to push in counter directions. We now discuss each megatrend in turn.

Shifting geographies and scales of production and consumption: from North to South
Consumption of dairy milk 3 was historically contingent on regional availabilities of resources, arable land, population growth and access to cattle. Cattle originated in the Middle East but became domesticated in many parts of the world and Mind map used to identify themes and megatrends process embedded in local dairy systems. Cattle ranges expanded considerably through colonialism when these "creatures of empire" played a key role in territorial expansion (Anderson 2004). By the twentieth century, narratives about milk had become mythologised, associated with generalised goodness and story lines of "modernisation, progress and nation building" within European and North American markets (Clay and Yurco 2020, p. 3). High intake of dairy products has been widely promoted in Northern countries, often for bone health due to its high calcium content (Willet et al. 2019). The dairy industry is now well established in the Global North with consumption of fresh dairy projected to reach 25.2 kg per capita of milk solids by 2030, compared with 12.6 kg per capita 4 (OECD-FAO 2021, p. 179) in the South. However, as we discuss below, demand for dairy milk has flattened out or is declining in some countries in the North where plant-based milks are capturing part of the market.
Dairy milk has since proven globally attractive with demands for animal-based proteins (meat and dairy) increasing in poorer countries as incomes rise (Stephens 2020). Increasing dairy consumption is part of the so-called "nutrition transition" (Popkin 2003) that necessitates an accompanying "livestock revolution" (Delgado 2003) to produce more meat and dairy. Advertising campaigns supporting the consumption of milk (see Zafrilla et al. 2018) are increasing demands for fresh dairy products in growing markets across the Global South, particularly in India, Pakistan and parts of Africa (OECD-FAO 2020, p. 175). For example, OECD-FAO (2020) find that more "away-from-home" eating in Southeast Asian countries is driving up demand for dairy consumption, such as cheese, as consumers eat more processed fast-foods.
Part of the success of dairy expansion has come from the adaptability of milk products, which can be transported in powdered forms. At present, fresh dairy products tend to be more expensive in the Global South (such as Sub-Saharan Africa, Eastern and South-eastern Asia) than in the Global North as fresh milk is more expensive to trade due to its highly perishable nature . To ensure a constant and cheaper supply of longer-life milk, fresh milk is converted to powdered forms, for which an integrated network of export markets have been developed. Most dairy imports and exports are in the form of whole milk powder (WMP) or skim milk powder (SMP) (OECD-FAO 2020). The major dairy exporters are New Zealand, the European Union and the United States (OECD-FAO 2020).
The OECD-FAO Agricultural Outlook (2021) projections for 2021-2030 have modelled increased production across developed and developing countries, according to kilo tonnes per week (kt pw) shown in Table 1. This is in line with OECD-FAO (2010) findings from a decade earlier, which projected more milk will be produced outside of the OECD and that growth is expected for all dairy products with WMP, butter and cheese having the strongest growth. Production in the Global South now outstrips that of the North, much of it associated with rapid growth in India and China. However given approximately 80% of the world's population is in the South, it is still lower on a per capita basis. Nevertheless, the present outlook clearly identifies a period of rapid growth for the dairy sector in the Global South as it is predicted to expand by approximately a third in the next 10 years as shown in Table 1.
According to Rabobank's annual listing of top global dairy companies by turnover for 2020 (see Table 2), most large dairy organisations are from the Global North (Ledman and van Buttum 2020). Each of these multibillion-dollar companies owns multiple brands of both dairy products (milks, yoghurts, cheeses) as well as by-products that contain milk (confectionary, pet food, cereals) (Nestle 2021). However, dairy companies from China have moved up in ranking in recent years due to expanding total net worth and production capacity (through more mechanised forms of milk production). The Yili Group, which is known as Asia's largest dairy firm (Dairy Industries International 2020), has expanded its global reach, with partnerships involving Thailand, New Zealand, Italy, the Netherlands and Uruguay (Oceania Dairy 2020).
The recent and accelerating growth in dairy production in the South is occurring when the sector looks very different to what it did when it originally expanded in the North. As will be discussed, dairy is more corporatised, mechanised and mobile, and must compete in national and international markets, not just local ones. Current conditions favour capital-intensive rather than labour-intensive operations, favouring larger businesses over more traditional smaller family run farms. Dairy has become a key site of investment and capital accumulation, influencing the form and shape of the industry as it expands in the Global South. This presents land use and livelihoods challenges for small scale farms and farming communities and raises questions about who benefits from the expansion of dairy in the Global South across spaces and scales. The rapid growth of dairy production and consumption in the Global South means that the socioecological challenges of dairy expansion will be concentrated there, as we discuss below.

Intensification of capital, land, and animals
Multiple and inter-related processes of dairy intensification represent a second megatrend in the dairy sector. The growth of large dairy companies (see Table 2) has resulted in an intensification of capital amongst a smaller group of influential actors, who have, in turn, sought to maximise profits by intensifying production, making the most out of dairying land and dairying bodies, both human and non-human. While milk production has historically been undertaken by pasture-based smallholder farms, economies of scale favour larger corporate entities over these smaller traditional businesses. This can result in smaller farms being bought out or replaced through a variety of means and consolidated into large dairy production systems who can invest in mechanisation and meet industry standards for greater profitability. Improved national and international transportation and processing systems enables this transition as more distant players can compete in what used to be mainly local markets. Further consolidation of dairy farms through processes of mergers, acquisitions and strategic alliances in the dairy industry are expected to continue (Knips 2005) as less competitive smaller dairy farms exit the market (for example, see Sewell 2021).
One of the drivers of capital intensification are the sanitation and quality standards demanded in contemporary food systems. Quality assurance for food is now an essential element within the global food industry and it is widely acknowledged that as standards become stricter, food manufacturers are required to invest in infrastructure and quality controls to remain competitive (Kotsanopoulos and Arvanitoyannis 2017). Agri-food chains are governed by strong interdependencies among retailers and controlling bodies that impose a variety of regulations (Gianni et al. 2017). The dairy sector is no exception to this, with increased public awareness regarding food safety pushing the dairy sector to improve the safety and image of dairy products (Ding et al. 2019). Britt et al. (2018) argue that structural consolidation of dairy farming will continue as the sector becomes more vertically integrated and that this requires more inputs to ensure milk products are meeting product quality demands, resulting in more resource and energy intensive production lines. From the collection of milk to the process of pasteurisation, homogenisation, storage, packaging, and distribution, each of these components comes with its own social, political, economic and environmental costs and expectations. Dairy processing, for example, is recognised as one of the most energy intensive sectors within the food industry (Briam et al. 2015;Challis et al. 2017;Ladha-Sabur et al. 2019), having particular safety requirements due to the perishable nature and limited shelf life of milk (Douphrate et al. 2013).
Meeting these standards can be difficult for small dairy producers and "traditional" household producers due to a lack of access to modern processing technologies and the time and expense meeting such standards may incur (Britt et al. 2018). Cattle bodies and outputs like milk vary dramatically across space and environments. While the push for quality standards advantages larger players as a form of risk management, it is having significant impacts upon those who cannot meet those standards . In a case study on family farms in the Global South, Bosc et al. (2018, p. 313) found that the functioning of family farms is often far removed from the "standards of agricultural specialisation" required by public policies. Regional differences in terms of development, regulations and exporting opportunities (Kotsanopoulos and Arvanitoyannis 2017) risks further marginalising smaller and poorer farmers who may not be able to meet and adapt to changing expectations. For example, Minten et al. (2020) found that although dairy milk production, quality and supply is increasing due to the prevalence of larger scale farms and their capacity to adopt new technologies and resources, smaller farms are increasingly excluded "because of relatively higher coordination costs that downstream firms in the value chain incur in their commercial engagements with these small farms" (Minten et al. 2020, p. 2). Reardon et al. (2009), similarly found that there was a trend of shifting from local sourcing in countries in the Global South to national, regional, and global networks to reduce cost and increase efficiencies within a competitive market context. China estimates that at least 100,000 small-scale dairy farmers have stopped farming altogether since 2010, not helped by a ban on the collection of milk from many small-scale family farms due to concerns about milk quality (Bai et al. 2018).
A second driver and outcome of a more consolidated and intensive dairy sector has been investment in new technologies such as robotics to reduce labour costs. Automatic milking systems (AMS) represent the most advanced shift towards mechanisation in the sector. Introduced in the 1990s (Jacobs and Siegford 2012), AMS essentially involves robots automatically milking cows, minimising the need for conventional human intervention/labour. The world's largest "milking carousel" was recently acquired by one of China's leading producers of raw milk, Lvyuan Animal Husbandry (GEA Group 2020). This machine is capable of milking 10,000 cows, 3 times per day. While technical apparatuses such as milking machines facilitate a more efficient production process, thereby reducing costs and providing a competitive advantage over smaller non-automated producers, it also generates a disembodied and increased moral distance between humans and the cattle as both become progressively alienated from one another (Clarke and Knights 2021;Holloway and Bear 2017). AMS encourages a shift away from animal husbandry and multispecies relations to a much more mechanical relationship where the metrics for animal health and wellbeing are quantified in artificial intelligence systems.
Mechanisation in milking, along with the increased prominence of feedlot rather than pasture-based production systems, are intensifying dairy production in increasingly corporatized environments (Clay et al. 2020b). Increased intensification of dairy agriculture reinforces objectifying constructions of cows as commodities whose primary value comes from their ability to produce milk that can be sold for profit. Alternative ways of engaging with them as feeling complex beings (Schuster et al. 2020) becomes more and more difficult as human-cow encounters become rarer, routinised and limited. Farmers are now producing more for less, however there are also less farmers and farm workers needed to produce this milk, and cows have less opportunity to build more felt and caring relationships with humans. These intensive systems of production face the accusations of animal cruelty, a lack of transparency, and "placelessness", that have long been levelled at intensive animal agricultural systems (Sexton et al. 2019, p. 64;Gillespie 2018). The effort to make milk quicker, cheaper and more homogenous is central to efforts to expand dairy industries, favouring consumers and large dairying companies at the expense of small farms and farm animals. Much more work is needed to track the socioecological impacts of the transition in the Global North and South from distributed small scale farming systems to larger intensified production systems where the livelihoods and wellbeing of farmers, farm workers and farm animals are given appropriate consideration (Steenveld et al. 2012).

Growing awareness of the ecological impacts of dairy
It is increasingly recognised that the global dairy sector has an important role to play in the reduction of global emissions, being a significant contributor of greenhouse gasses, mostly through methane released through eructation, but also through land use change, refrigeration, transport and various other sources ( Bar-On et al. 2018). The ecological impacts of animal agriculture are well documented with livestock industries also associated with land degradation due to overgrazing, soil erosion and salinisation, deforestation, biodiversity loss, and the contamination of surface and groundwater (Saari et al. 2020). The findings of the UN Food and Agriculture Organisation's (2006) controversial and ground breaking report Livestock's Long Shadow first raised the alarm at a global scale and have been bolstered by studies ever since (e.g. Willet et al. 2019). Poore and Nemeck (2018, p. 4), for example, found that even the lowest-impact animal products exceed the environmental impacts of vegetablebased proteins, to the extent that meat, aquaculture, eggs, and dairy use ~ 83% of the world's farmland, contribute 56-58% of food-related emissions, but provide only 37% of protein intake. These and similar findings are placing pressure on dairy industries to lower emissions and develop more sustainable systems. The most recent Sixth Assessment Report by the IPCC (MassonDelmotte et al. 2021) notes substantive increases in methane emissions, with animal agricultural being one of the largest sources.
Research into enteric fermentation, the metabolic process that creates methane in cattle rumen, has targeted the digestive system of cattle to limit the production of methane. Despite thirty years of experiments however, improvements have been minimal and inconsequential when compared with the increasing size of the global herd (McGregor et al. 2021). Lively cattle bodies have resisted technological control. Even if much touted but somehow eternally immanent methane mitigating "breakthroughs" are achievable at scale, the ever-increasing demand, production and wastage of food that is part and parcel of the industrial food system (Ormond 2020) means that the raft of other environmental problems associated with dairy would remain. The ongoing expansion of the dairy industry will exasperate pollution issues from acidification to eutrophication on land and freshwater systems (Poore and Nemecek 2019;Rotz 2018;Fiel et al. 2020), as well as the displacement of natural carbon sinks (Harwatt et al. 2019) if current production and consumption trajectories persist. While there are efforts to price the environmental and social externalities of dairy to provide a fairer price and contribute to climate change mitigation (Pieper et al. 2020) there are also clear limits to this approach (see McAfee 1999). Zafrilla et al. (2020) reflected on recent reports by the IPCC to argue that livestock production is being positioned as the subject of new forms of climate governance, or as Ormond (2020, p. 163) suggests "corporate carbon commitments" are being redirected to farm level. For example, climate friendly(er) dairy will rely on improvements in cattle breeds and feed (Wankar et al. 2021) and improvements in animal health to minimise "wastage" (Mylan et al. 2019). However, each of these changes come with their own costs and challenges. For example, some forms of seaweed have been identified as potential feed additives that can reduce methane emissions (Kinley et al. 2020). In addition to reducing biogenic methane, the cultivation of seaweed could have net benefits for local labour forces and supporting regional economies (Kinley et al. 2020). However, like any intensive production system, this raises questions about the sustainability of large-scale aquaculture, as the cultivation of seaweed for the global meat and dairy cattle herd of over a billion animals will invariably have negative impacts on marine ecosystems (Hasselström et al. 2018). The logistics appear challenging at the very least.
Further case studies and empirical research that examines the impacts of that climate mitigation strategies may have upon human and non-human stakeholders is needed to envision what climate friendly cattle may look like in the future (see Ormond 2020). For example, if feed additives are a requirement for low emissions dairy production this is likely to increase the costs of production, favouring larger players over smaller ones and further threaten the viability of small farms. In contrast there is some promising work on agro-ecology and regenerative farming being promoted by institutions like the FAO that recasts well managed cattle as assets in the fight against climate change and ecological degradation (for example, see Teague and Kreuter 2020). Agroecology principles as defined by the FAO aim to include social values in conjunction with environmental practices that reflect "dignity, equity, inclusion and justice, associated with gender and intergenerational equality and access to decent jobs" (Barrios et al. 2020, p. 236) and tend to favour the more traditional small scale family farms still prevalent in the Global South. While interest in this area is growing to counter the "monocultures of industrial agriculture" and help small scale farms to prosper (Altieri et al. 2015, p. 874), it has yet to attract the mass investment required at sufficient scale to counter opposing processes favouring intensification of production.
For some, promissory narratives (Sexton et al. 2019) of low emissions cattle through technological or agricultural fixes are a bridge too far, heightening the need for systemic shifts towards plant-based proteins as a replacement for animal-based protein. Rather than being a radical idea the "societal grand challenge" of moving away from or limiting animal-based food systems (see Mylan et al. 2021) is being echoed in major intergovernmental reports, including the IPCC Special Report on Climate Change and Land (Mbow et al. 2019) and the EAT-LANCET Commission (Willett et al. 2019), as well as the earlier UN Environment Programme report (Hertwich et al. 2010, p. 82) which argued for a substantial reduction of animal products in human diets in order to lessen the negative environmental impacts from agriculture more broadly. Dairy industries need to develop much more sustainable production processes if they are to mitigate growing consumer concerns about the environmental impacts of the industry.

Alternative milk disruptions
Alternative proteins, in the form of plant-based milks and the gradual development of synthetic milks (grown from fermentation techniques and/or animal cells) are emerging as a fundamental disruption to conventional animal milk production and consumption. Although milks such as soy and nutmilks have had a role throughout human history (Valenze 2011), the popularity of these products has increased in volume and expanded in range in recent times driven by consumers who are motivated by both ecological and health concerns (Lonnie et al. 2018). This change in consumer preferences is particularly evident in the Global North (see Stokel-Walker 2018; Franklin-Wallis 2019; Mintel 2019). Lonkila and Kaljonen (2021) argue that the future viability of the global dairy industry is contingent on the attention paid to the "entangled changes" of food production, technological development and consumer demands. Demographics play a role in driving these changes with Stewart et al. (2020) finding that the consumption of dairy milk in the US drops off steadily with age (Moshfegh et al. 2019). Stokel-Walker (2018) refers to this process as older consumers "aging out of the market", while younger consumers are also shifting towards dairy alternatives. McCarthy et al.'s (2017) US based study found that animal welfare and environmental concerns tend to be primary drivers as to why people might purchase alternative milk over cow's milk. In their research, they found that the main reasons consumers drink milk is out of habit or "for the flavour" rather than its nutritional quality. Debates around the nutritional properties of milk are out of scope for this review but are more questioned than at any time in the past (for example, see Michaëlsson et al. 2014;Aune 2015;Jakobsen et. al 2021). We discuss each form of alternative milk below.

Plant-based milk
Plant-based foods and proteins have become a focal point in the pursuit of sustainability goals (Aschemann-Witzel et al. 2020). Sales of plant-based milks are on the rise in the Global North, contributing to a decline in per capita consumption of dairy milk in some countries, including the US (Stewart et al. 2020) where dairy milk sales in the US fell to $12 billion in 2019 from $15 billion in 2015, resulting in some large American dairy producers filing for bankruptcy (Garabito 2020). Neo and Lim (2021) found that interest in alternative protein production and consumption were creating opportunities for more localised production in Asia, driven in part by a push for more nutritious "immunity-boosting" products in countries such as China. The OECD-FAO (2020) report on global dairy and agriculture argues that plant-based dairy substitutes are growing in popularity due to consumer awareness around lactose intolerance, and the health and environmental impact of dairy products (see also Mylan et al. 2019). Business opportunities for plant-based products are viewed as positive (Acschmann-Witzel et al. 2020) and are now also being pursued by some existing dairy companies around the world. For example, Danone acquired the plant-based foods company WhiteWave in 2017, which they note was a "strategic move towards meeting consumer expectations for healthier and sustainable choices… enriching our portfolio and complimenting dairy fermented products" (Danone 2021). Early adopters of this trend are set to increase sales and production outputs as the increase in demand for plant-based alternatives continues to reconfigure the market.
The interest in plant-based milks derives from a number of negative associations with dairy relating to health (Clay et al. 2020a), environmental impacts (see Willet et al. 2018;Springmann et al. 2018) and animal welfare, particularly in intensive systems where there is a "perceived disconnect from naturalness" (Beaver et al. 2020, p. 5749). Land use change from pasture to crops to produce plant-based milks is a growing possibility on the horizon (Philippidis et al. 2021) if novel dietary patterns such as veganism within Western food culture becomes more mainstream (Chiorando 2018). However, a widespread transition to plant-based milk production remains distant and faces many socio-cultural, economic and political hurdles which beckons research from both the physical and social sciences in order to fully understand the socioecological ramifications of such a shift (Philippidis et al. 2021). For example, dairy industries across North America, Europe and Australia are engaging in contests over labelling animal-free alternatives such as milk, butter and cheese (Sexton et al. 2019). In the US, calls for the reintroduction of the "Dairy Pride Act" have been made to protect dairy industries from plant-based alternatives by claiming milk can only be made by animals (Keller and Heckman LLP 2019). Alternative milk brands in the European Union face a ban on using terms or imagery on packaging which refer to or evoke dairy products (Bonadio and Borghini 2021). This is positioned as a move to avoid confusion for consumers at the supermarket, however such strategies clearly reflect the interests of dairy industries keen to protect their markets. More research is needed on the potential growth of plant-based alternatives and what contribution they may make to food transitions (Mylan et al. 2019;Lonkila and Kaljonen 2021).

Synthetic milk
Synthetic milks grown through cellular agriculture (milk that shares the same biochemical make up on animal milk but produced without animal bodies) are emerging as an additional competitor to animal-based milks. Cellular agriculture is an emergent field in which agricultural productsmost typically animal-derived agricultural products-are produced through processes operating at the cellular level, as opposed to (typically farm-based) processes operating at the whole organism level (Stephens 2020). Cellular agriculture can be defined as "attempts to sustainably supplant animal products with biomass cultured from cells" in-turn creating openings for new food production and novel economic geographies (Jönsson 2020, p. 922). This new form of food has attracted significant investment from venture-capitalists, particularly concentrated in the Silicon Valley region of San Francisco (Sexton 2020) but is also being developed in many richer parts of the world including Israel, Australia and Europe. The US-based company Perfect Day has recently partnered with the company Brave Robot to create the world's first animal-free dairy ice creams (Starostinetskaya 2021). They market their product by claiming cow's milk ice cream to be "tasty, but unsustainable", plant-based ice cream as "more sustainable, but not tasty" and their product as "both sustainable AND tasty" (Brave Robot 2021). Unlike synthetic meat which faces difficulties in matching the complexity and textures of different meat products, synthetic milk is likely to be indistinguishable from dairy milk due to its liquid form. If it can be produced more cheaply than dairy milk the potential for a relatively rapid change in the dairy industry brought on by cellular agriculture is high. However, at present, the technology is still evolving and the scalability of cellular milk is a challenge (Pandya 2016). Reality has not matched the promises envisioned by innovators in the alternative protein sector thus far.
If synthetic milk can replace dairy as an ingredient in the industrial food processing sector (von Massow and Gingerich 2019) this could present significant challenges for large-scale producers who depend on the exportability of powdered milk products. In their analysis of the possibilities offered by synthetic milks, Jönsson et al. (2019) characterise these novelties as "post-animal products" which "carry new realities, new ontologies with them" (Mol as cited in Jönsson et al. 2019, p. 72). These new forms of food appear better for animal welfare on the basis that there are no livestock animals harmed in the making (synthetic milk can be produced from yeast). Thus, the creation of cellular agriculture and synthetic milks offer the possibility of unfettered production of animal products without conventional spatial constraints and welfare concerns (Jönsson 2020), as no animal bodies are concealed or abstracted in the production of milk (Leroy et al. 2020). These disruptions in the dairy sector may steer humanity towards radically different food systems (Leroy et al. 2020) where synthetic milks compete with traditional dairy. A recent report into the future of dairy by think tank Rethinkx (Tubb and Seba 2019) argued that by 2030, new fermentation industries could create up to 700,000 jobs in the US alone. However, Lonkila and Kaljonen (2021, p. 10) warn that synthetic meat and milk alternatives are entangled in capital-intensive agriculture and may further the unequal distribution of protein-rich diets across spaces. As such, alternative proteins do not necessarily challenge corporatisation, homogenisation or consider questions of justice within global food production and consumption (for humans, non-humans and environments), instead they may represent a new stage of consolidation that further marginalises low tech dairy systems. Capital intensive synthetic milk production could eventually further displace many people from the global dairy sector, albeit with potential environmental and animal welfare gains. However, with these new technologies, it may become a possibility that small scale, independently owned breweries could be rolled out in ways that fracture current consolidation processes and empower animal-less local producers in new and interesting ways.

Other animal milks
In addition to plant-based and synthetic milks, other animal milks are, in some contexts, being developed to compete with dairy milk production. On smaller scales, Numpaque et al. (2020) find that for different regions of the world, other animal species such as mare, donkey, yak, reindeer, camel and llama have an important share in milk consumption. At present, world milk production is estimated as 81% cow milk, 15% buffalo milk, and 4% total for other animal milks (OECD-FAO 2020). In their review of trends in the dairy sector, Pulina et al. (2020) found that within the last 50 years, sheep and goat milk production has more than doubled and is expected to increase up to 26% for sheep milk and 53% for goat milk by 2030. Miller and Lu (2019) find that rising consumer demand, strong prices, and climate change are influential factors for the uptake in new goat milk industries in countries such as New Zealand, China and the US. Goat milk has been found to be less allergenic, contain more nutritional benefits than plant-based alternatives (Park 2021) and softer gastric digestion than other non-cattle milks (Roy et al. 2020). Additionally, the production of small ruminant milks has been suggested as more environmentally friendly and socially appropriate for some rural communities (Pulina et al. 2020). At present, the upscaling of these industries faces barriers as Vouraki et al. (2020) find that sector lacks the professionalisation, management training and supply chain integration that cow dairy milk has, which results in varied levels of productivity. It is also questionable if the intensification such scaling up would require is desirable for people, animals or the planet.

Discussion
In this section we look across the four megatrends to reflect on emerging socioecological challenges for dairy industries. One of the most striking trends is the shift in the volume of global dairy production and consumption from North to South. While Northern markets are saturated with milk, there is room for immense growth in parts of the Global South. This creates opportunities for new jobs in the dairy sector and promises nutritional benefits for under-nourished human communities, as long as they can access affordable milk products. The geographic expansion of dairy is being driven by large multinational companies who are investing in more intensive forms of farming. This includes automated milking systems and more confined animal production systems that make milk faster and cheaper, by slashing labour costs and maximising bodily production. Dairying is becoming more homogenous and placeless as it adapts to increasingly stringent industry standards oriented around consumer safety and risk management. Hence the form of dairy expansion that is likely to take place in new areas of the Global South, is more mechanised, corporate and standardised than what has occurred in the past.
While the global growth of dairy is often positioned as a positive outcome for food security and economies, replicating older mythologies about progress and modernisation, our analysis suggests such narratives should be viewed with caution. The literature tends to emphasise the productivity and economic gains for economies and consumers but there is a dearth of work analysing the social costs of dairy farming that focuses on the welfare of humans and non-humans. It is evident, for example, that the corporatisation and intensification of dairy production favours larger companies over small dairy producers. Further mechanisation of dairy production raises questions about the size and longevity of the dairy labour force and is likely to exacerbate the already established trend of pushing smaller dairy milk producers out of competitive markets. In addition, if dairy production is predicted to grow significantly in the Global South over the next decade (OECD-FAO 2021), this raises concerns about where the land and resources to enable such an increase is to come from. It is likely that land use pressures will increase the risk of land use conflicts and deforestation, as even intensive dairy farming still requires extensive feed production systems. Dairy cattle require a significant amount of land and resources for the volume of food produced. For example, 1 l of cow's milk can use up to 1020 l of water throughout the whole process (Rotz 2018). It is likely many of the environmental pressures associated with dairy in the Global North will be replicated or intensified. particularly where environmental regulations are lax, in the South, including river and groundwater pollution, greenhouse gas emissions and soil degradation.
As the environmental costs of dairy production become more well known the pressure on smallholders is likely to increase as societies seek to minimise impacts. Expectations for low emissions food products are increasing and may eventually result in regulations requiring feed additives or vaccinations to reduce methane emissions. While larger businesses are likely to be able to absorb these costs, environmental regulations may add to the challenges faced by smallholder farmers. There is also the possibility for bespoke markets to evolve that support small holder farms employing regenerative or agro-ecological principles. However, the general trend is towards intensification or production and more research is needed to fully understand the implications of intensification, standardisation and regulations on human and non-human lives (Holloway and Bear 2017). Currently, it seems that many of the benefits of expansion are likely to be captured by large multinational companies, but the immediate social and environmental costs are to be felt locally. As such, many rural communities risk being disadvantaged by such a transition without sufficient government protections. Alternatively, as Weis (2013) reminds us, there is nothing inevitable or even necessarily desirable about ever-expanding animal-based diets and production systems, particularly for regions such as Asia, where a large proportion of the population are historically lactose intolerant (Valenze 2011). At a global scale the climate impacts of dairy production are significant and any increase in dairy cattle in the Global South should be at least matched by decreases of dairy cattle in the Global North. Efforts to expand dairy as the world struggles with global heating is irresponsible and goes against calls for richer countries to "declare a timeframe for peak livestock" after which countries steadily decrease livestock numbers in order to pursue global climate goals (Harwatt et al. 2019, p. 9).
The world seems a long way from this sort of governmental intervention and so, in the absence of industry-led limits on the global herd, consumers and food tech investors are taking their own action through alternative milk products. Plant and synthetic milks mimic dairy milk in terms of colour, texture, consistency, and taste while at the same time, distancing the product from the unfavourable aspects of dairy, from animal welfare through to health and environmental concerns. Plant-based milks have successfully captured parts of the market and synthetic milks show as yet unrealised promise to make a large impact. However, by mimicking milk, both approaches subtly reinforce dairy milk (and animal products in general) as the norm for modern diets . A critical question for alternative milks is whether replacing one form of industrial production (animal-based dairy) with another (plant and cellular-based dairy) will be socially and environmentally beneficial. Mylan et al. (2019) argue that alternative milks and their functioning within the food system in relation to the organisation of markets and consumption remains largely unchanged. Additionally, Sexton et al. (2019) suggest that upscaling alternative proteins (and other animal protein analogues) takes society further away from cultivating more localised food systems based on care and trust. Addressing issues of sustainability through alternative dairy may have some environmental and animal welfare benefits but may accelerate trends of corporatisation and centralised control of food systems, with potentially devastating impacts for small farmers. Such innovations risk merely changing the components of dairy systems (inputs from plants rather than animals) while extending corporate control over land and labour. The social costs of transitioning from dairy milk to plant-based or synthetic milks would be immense for rural communities and as such requires much more research to explore the socioecological dimensions of transitions away from animal-based dairy systems and how, and if, such transitions can be done in just and fair ways.

Conclusion
We have identified four megatrends affecting the global dairy sector and highlighted key challenges for the future. The impacts of shifting geographies, intensification, ecological pressures and the possibilities offered by milk alternatives are creating openings for interdisciplinary scholarship to engage with the under-researched impacts of these trends. Such research should recognise the vitality and agency of cattle and other actors within dairy systems, and that the expansion of dairy industries is not just an economic process, but one that is embedded with more-than-human relations (Collard and Dempsey 2013). Cattle not only require food, water, land and labour to produce milk, but they increasingly require robots, quality standards, methane mitigation technologies, intensive farming, investment and international markets if they are to survive. It is the liveliness of cattle that has shaped these relations, a liveliness that dairy industries will never fully control or contain and is resulting in the megatrends we have identified here. While some megatrends reflect an enthusiastic embrace and expansion of these increasingly corporatized assemblages, others reflect concerns about their destructive socioecological impacts, resulting in a growing market for dairy alternatives. There is a need for more critical case study research that focuses on the complex impacts dairying, and dairy alternatives, are having on place, people and cattle.
Our review suggests that the future of dairy is not clear nor uncontroversial. The mythology that milk is unquestionably good and universally beneficial is not borne out by this review and the almost teleological assumption that dairy will simply roll out across the planet to benefit all deserves careful scrutiny to assess its true costs. If dairy is to live up to its promises, much more attention needs to be directed to maximising and broadening the socioecological benefits of the industry, minimising the human and non-human costs, including those inflicted on cattle, and limiting its contributions to global environmental change. With some trends working against these goals, the social license of dairy is likely to erode in mature markets, creating further openings for milk alternatives, each of which comes with its own set of challenges. While dairy is likely to continue expanding in the immediate future, the megatrends and socioecological challenges we have identified here are substantial and suggest its medium-term future trajectories are much less certain. Funding Open Access funding enabled and organized by CAUL and its Member Institutions.

Appendix
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.