Smart farming, an increasingly common part of food production, refers broadly to the innovative use of sensors, robotics, and artificial intelligence (AI) to streamline agriculture. Field crop examples of smart farming include monitoring soil health using small sensors, spotting signs of disease in plants via drones, and facilitating connections between smaller-scale farmers through consumer electronic devices. These smart innovations offer potential time savings and crop optimization benefits for farmers and may contribute to the wiser use of resources in food production.
Smart farming is also used in animal agriculture in forms such as Precision Livestock Farming (PLF) and Digital Livestock Farming (DLF). PLF uses sensors and small electronics to measure key indicators related to animals’ physiology and behavior. DLF then enhances these data with predictive capabilities. In modern industrial agriculture, many farmed animals lead lives characterized by extreme confinement and suffering in which their experiences are often unknown or dismissed. Do smart farming approaches like DLF offer a solution? Or do DLF and other forms of smart farming in animal agriculture benefit the animal industry more than animals?
Assessing the potential value of improvements to farmed animal welfare depends on personal perspectives and ethical frameworks. For some, the alleviation of any amount of suffering for farmed animals, when potentially multiplied across the tens of billions of farmed animals in the world today, represents a tangible way to decrease suffering. For others, improving the welfare of farmed animals who remain commodified for their bodily products may seem less motivating than freeing animals from exploitation as food.
Regardless of values or approach, smart farming in animal agriculture is an innovation deeply connected to industrial animal farming and the profit motives of agribusiness. As such, it presents a strategic leverage point for fighting animal exploitation worldwide in a variety of ways. The sections that follow explore the welfare improvements that smart farming can offer and its linkages to industrial animal agriculture.
The Rise of Digital Livestock Farming
Modern industrial technology has been impacting food and agricultural production since the Industrial Revolution. Now, however, with the advent of smart farming approaches such as DLF, “motorized devices are being replaced with AI,” says Ben Williamson, US Executive Director of Compassion in World Farming.
Precision Agriculture is a smart farming approach that rose to prominence in the 1990s by using electronic innovations to precisely control the delivery of water and nutrients to crops. By the early 2000s, the principles of Precision Agriculture had begun to be applied in animal agriculture under the name Precision Livestock Farming (PLF), which used digital tools to monitor the health, behavior, and living conditions of farmed animals.[1] DLF enhances PLF with predictive capabilities and AI, aiming to both increase animal well-being and improve productivity.
“Conventional PLF collects health or welfare-indicating-parameters-based data after the occurrence,” says Suresh Neethirajan, PhD, professor at the Farmworx Consultancy in Wageningen, The Netherlands, who studies how to improve the livelihood of small- to medium-scale farmers while enhancing the welfare of farmed animals.
For instance, with conventional PLF, wearable sensors and video monitoring technologies collect animal-based measurements related to behavior and health and send this data to a computer without requiring invasive procedures for animals or hands-on contact from human caretakers, potentially enhancing animal well-being while reducing the potential for pathogen transmission to humans.[2] Algorithms then process the collected data and create insights that enable farmers to make decisions. For instance, if there are variations or fluctuating hormone levels in the body of a cow, pig, or chicken, a farmer using those data could then decide how to care for the animal.
With the addition of DLF, farmers can move from reactive to predictive measurements, which may avert animal suffering as well as streamline farm operations. For example, data analysis could predict with high accuracy when a disease outbreak is likely to occur, arming the farmer with the information to perhaps prevent the outbreak, saving both the expense of treatment and the pain and suffering that the animals would endure. This predictive technology has also shown promise for accurately pinpointing illness, reducing the need for blanket treatment via the routine sub-therapeutic antibiotics typically fed to animals living in CAFOs.[3] “Digitalization is about creating actionable intelligence via data and creating meaningful value to all the stakeholders in livestock farming while ensuring the well-being of the animals,” Neethirajan says.
Implementation of smart farming technology can take many forms. At dairy farms across the United States, robots are managing the feeding of calves, and some farmers are attaching wearable technology to animals to track how much an individual animal is eating, how much that animal is moving, and if the animal is suffering any lameness. Some farmers in Russia and Turkey are even experimenting with placing virtual reality (VR) headsets on cows to show them images of green pastures. Their intention is to improve the cows’ mental well-being by letting them feel the peace associated with being outside.
Farmers have expressed concerns about narrow profit margins and questioned who owns the data collected and analyzed by electronic tools.[4] However, surveys show that most are still willing and happy to embrace smart farming technologies. “It helps them cut down the time in farm management and provides convenience such as remote management and predictive possibilities,” Neethirajan says.
Digital Livestock Farming’s Connection to Industrial Animal Agriculture
Despite the potential for avoiding suffering or enhancing farmed animals’ mental well-being, the close connection between smart farming innovations such as DLF and industrial animal agriculture is problematic. Smart farming technologies and DLF’s predictive abilities are most applicable in large-scale animal agriculture where devoting frequent personal attention to monitoring animals’ well-being is prohibitive. This scale incentive, plus the added expense of smart farming technologies themselves, means that smart farming is a tool most suited to industrial agribusiness, where the overriding goal for any innovation is the maximization of profit.
While admitting the potential positive welfare benefits of continuous health monitoring, delivery of individual care, and optimization of environmental conditions, Professor Marian Stamp Dawkins of the University of Oxford sums up the problematic nature of smart farming in animal agriculture, writing that “some people see the efficiency gains offered by the new technology as a direct threat to the animals themselves, allowing producers to get ‘more for less’ in the interests of profit.”[5]
Despite offering certain potential benefits to animals and farmers, smart farming allows animal agribusiness to increase the exploitation of farmed animals. While many researchers and promoters maintain that these new digital tools enhance animal welfare by providing real-time information about animal well-being,[6] that explanation is not the true driver of the use of this technology. The ultimate benefit of smart farming in animal production is to increase profits for the animal agriculture industry. “Smart farming is designed to increase the yields and productivity from farmed animals,” Williamson says. And there is proof it is working.
One farmer in Turkey, for example, reported that fitting cows with VR headsets showing green pastures raised daily milk production from 22 liters to 27 liters per cow per day, presumably by lowering cows’ stress. Research reported by the European Commission found that when cows were outfitted with smart ear tags and feeders to precisely deliver mineral supplements, milk yield increased by 1% while the need for veterinary assistance fell by 25%, and the cost of mineral supplementation fell by 10%. Although optimizing health through precision nutrition may benefit the well-being of dairy cows, increases in milk quantity and the overall cost efficiency of milk production provide powerful incentives for implementing DLF and other smart farming technologies for reasons of profit rather than animal well-being.
VR headset technology is still limited, but smart farming in other forms is increasingly being used worldwide, even in lower-income countries. The precision agriculture market, with its focus on sensors and small-scale technological tools, is projected to reach $19.24 billion by 2030 at an annual growth rate of 14.95%, according to data from Straits Research.[7]
Challenges in the Assessment of Farmed Animal Welfare
Companies producing smart farming technology mention improved animal health and well-being as a key goal. But the impact of improved health on overall farmed animal welfare depends on how animals’ physical condition is assessed and translated to a measurement of welfare. Factors that will shape the impact of smart farming on animal welfare include how welfare is defined, how well technology can measure it, and whether it can be proven that smart farming technology increases welfare metrics.[8]
There are many ways animal welfare can be defined, but “it essentially comes down to the animals’ happiness and level of suffering,” says Jacy Reese Anthis, social scientist, Co-Founder of the Sentience Institute, and author of The End of Animal Farming. “Animals want to live unconfined, be in good health, and have healthy social relationships.” Unfortunately, conventional US industrial animal agriculture often falls far short of ensuring animals’ well-being.
According to estimates by the Sentience Institute, 99% of US farmed animals live in confined animal feeding operations (CAFOs).[9] Meanwhile, US per capita meat consumption rose from 252 pounds per person on average in 1999 to 264 pounds per person on average in 2020.[10] To meet consumer demand for more animal protein, CAFOs house thousands of animals in crowded, confined quarters. Effectively monitoring the welfare of so many animals can prove challenging, if not impossible, and welfare improvements in an industrial animal farming context likely have limits. It is doubtful that even the most advanced and ubiquitous smart farming technology could fully ensure the welfare of animals commodified as food within industrial agriculture.
Neethirajan’s research has identified three major challenges to effectively tracking animal welfare with smart farming’s digital tools, including the cost of implementation, the validity of measurements, and the timing of insights.[11] Yet he notes that the existing alternative methods for collecting animal health metrics are also limited by being time-consuming, labor-intensive, and, therefore, costly. These barriers lead to negative impacts on animal welfare. As an example, animal farmers often rely on stock people to spot health and welfare issues in the animals, but in some commercial pig farms, there may be only one stock person for every 300 animals, which means that some animals who are suffering will undoubtedly be missed.
Digital tools could theoretically allow for more effective monitoring, which could deliver welfare benefits. However, the ultimate outcomes for animals depend on additional understanding of animals’ physical and emotional states. Welfare monitoring will be impossible without a deeper understanding of animals’ experiences and accurate measurement of welfare indicators.[12]
Although Neethirajan says that DLF technology is created with an animal needs-centered focus, it is still unclear whether smart farming in a commercial animal agriculture context can truly prioritize animal welfare. “If the goal is to make these animals happy on these farms, it would require unimaginable logistics and financial costs to these companies and consumers, not to mention an army of veterinarians, better feed and more space for animals,” Anthis says. “I don’t think society is willing to pay for that, which is why if we care about animals, we should not exploit them in the first place.”
Centering the Intrinsic Value of Animal Well-Being
Although animal welfare improvement is a stated goal of DLF, and many smart farming interventions related to nutrition and health offer clear comfort benefits to animals, it is still worth considering whether the implementation of smart farming tools is likely to be guided by compassionate concern for animals. The eventual impact of DLF and its tools will depend on the values that underlie its implementation. Will animal welfare be improved out of empathy for farmed animals and concern for the intrinsic value of their lived experiences as sentient beings? Or, will welfare improvements be sought as a means to optimize the economic value of their ongoing commodification?
Even modest improvements to animals’ health and comfort could incrementally increase animal well-being compared to the abysmally low baseline of standard CAFO conditions in the industrial animal industry. While any improvement in welfare reduces suffering, and some animal welfare improvements also decrease the cost efficiency of industrial animal exploitation, welfare improvement does not necessarily alter the overarching ethical framework of animal commodification.
It is possible to imagine that applications of DLF might look very different if guided primarily by a concern for animals’ well-being as sentient beings. Some technological tools might remain, such as tools for optimizing nutrition and disease detection or tools for measuring animals’ agitation. However, if implemented with a guiding value of protecting animals from harm, such data would likely invite responses that prioritize animal well-being over the cost-efficiency of animal farming. Other digital tools might not make sense, with animal protection as a guiding value. For example, smart farming interventions like VR headsets for dairy cows could provide a cost incentive for the dairy industry while merely masking the grim reality of farmed animal confinement.
Animal Advocacy Approaches Related to Digital Livestock Farming
Smart farming approaches like DLF present a range of entry points for different styles of animal protection advocacy. Advocates focused on incremental welfare improvement for farmed animals may find value in some applications of DLF that increase animals’ health and comfort while perhaps remaining opposed to others that offer clearer benefits for the animal industry.
Some research related to DLF has sought to specifically assess and reveal farmed animals’ emotional states. Researchers note that assessment of animals’ emotions via direct human observation is inherently variable and subjective, and DLF can help to standardize assessments.[13] Such tools and data could provide a potentially useful way for animal advocates to document animal suffering as part of expanding humanity’s moral circle to include non-human animals.
Conversely, advocates approaching animal protection through the lens of inclusive food system transformation may find DLF concerning both for its potential to further entrench animal exploitation and because it only offers benefits to a limited number of food system interest groups. By contrast, other interventions that reduce or eliminate animals from food production in favor of increasing equitable access to environmentally sustainable plant-based foods can offer overlapping benefits to all members of the food system.
Conclusion
By increasing productivity and lowering costs for agribusiness, smart farming approaches such as DLF could intensify animal suffering even while delivering useful data for improving farmed animal welfare. Technology use may increase animal productivity and decrease the need for human observation of farmed animal well-being, which could reduce production costs for animal agribusiness. If smart farming thus strengthens the value proposition of animal commodification, there will be a greater incentive to exploit farmed animals, which could lead to more animals being farmed.
The value proposition of animal farming still hinges on consumer demand and societal acceptance. A strong financial incentive plus high consumer demand for animal protein could lead to further growth of the farmed animal industry, exactly when reduction is critical, for various reasons ranging from animal protection to climate change mitigation.[14]
The need for digital tools to improve welfare for industrially farmed animals would become moot if societies shifted to a largely plant-based diet, a change the Intergovernmental Panel on Climate Change has called for in its 2022 report as a key strategy for confronting global climate risk.[15] However, the dominance of animal products in the current US food system and persistent disparities in the accessibility of plant-based eating present barriers to dietary transitions that would better support farmed animal well-being, human health, and environmental sustainability.
When considered in the context of the necessity of diet change, smart farming in animal agriculture stands as a potential solution to only one problem among many. Technological improvements such as DLF cannot fully solve animal health and welfare concerns in industrial farming, nor can they deliver the same benefits for animals or for overall food system sustainability as reducing the production and consumption of animal-based foods. Moreover, DLF offers benefits to only a few food system interest groups and concerns, failing to address the well-being of food production workers, the economic health of rural communities, the financial and dietary interests of food consumers, and many measures of environmental damage associated with industrial animal production.
Yet while it is most impactful to minimize or remove animals from the smart farming equation, that does not mean smart farming technologies are inherently problematic. Using smart farming in crop agriculture may streamline plant-based protein production, aligning more closely with the intrinsic value of animals’ well-being and the inclusive transformation of food systems for the common good.
A version of this post appeared previously at Sentient Media on April 27, 2022.
[1] Juliette Schillings, Richard Bennett, and David Christian Rose, “Exploring the Potential of Precision Livestock Farming Technologies to Help Address Farm Animal Welfare,” Frontiers in Animal Science 2 (May 13, 2021): 639678, https://doi.org/10.3389/fanim.2021.639678.
[2] Suresh Neethirajan and Bas Kemp, “Digital Livestock Farming,” Sensing and Bio-Sensing Research 32 (June 1, 2021): 100408, https://doi.org/10.1016/j.sbsr.2021.100408.
[3] Juliette Schillings, Richard Bennett, and David Christian Rose, “Exploring the Potential of Precision Livestock Farming Technologies to Help Address Farm Animal Welfare,” Frontiers in Animal Science 2 (May 13, 2021): 639678, https://doi.org/10.3389/fanim.2021.639678.
[4] See endnote 2.
[5] Marian Stamp Dawkins, “Does Smart Farming Improve or Damage Animal Welfare? Technology and What Animals Want,” Frontiers in Animal Science 2 (2021), https://www.frontiersin.org/articles/10.3389/fanim.2021.736536.
[6] Sigfredo Fuentes et al., “The Livestock Farming Digital Transformation: Implementation of New and Emerging Technologies Using Artificial Intelligence,” Animal Health Research Reviews 23, no. 1 (June 2022): 59–71, https://doi.org/10.1017/S1466252321000177.
[7] Straits Research, “Precision Agriculture Market Size Is Projected to Reach USD 19.24 Billion by 2030, Growing at a CAGR of 14.95%: Straits Research,” GlobeNewswire News Room, August 1, 2022, https://www.globenewswire.com/en/news-release/2022/08/01/2489650/0/en/Precision-Agriculture-Market-Size-is-projected-to-reach-USD-19-24-Billion-by-2030-growing-at-a-CAGR-of-14-95-Straits-Research.html.
[8] See endnote 5.
[9] Jacy Reese Anthis, “US Factory Farming Estimates” (Sentience Institute, April 11, 2019), https://sentienceinstitute.org/us-factory-farming-estimates.
[10] Gretchen Kuck and Gary Schnitkey, “An Overview of Meat Consumption in the United States,” Farmdoc Daily 11, no. 76 (May 12, 2021), https://farmdocdaily.illinois.edu/2021/05/an-overview-of-meat-consumption-in-the-united-states.html.
[11] See endnote 2.
[12] Suresh Neethirajan, “Affective State Recognition in Livestock—Artificial Intelligence Approaches,” Animals 12, no. 6 (January 2022): 759, https://doi.org/10.3390/ani12060759.
[13] Suresh Neethirajan, “Happy Cow or Thinking Pig? WUR Wolf—Facial Coding Platform for Measuring Emotions in Farm Animals,” AI 2, no. 3 (September 2021): 342–54, https://doi.org/10.3390/ai2030021.
[14] Xiaoming Xu et al., “Global Greenhouse Gas Emissions from Animal-Based Foods Are Twice Those of Plant-Based Foods,” Nature Food 2, no. 9 (September 2021): 724–32, https://doi.org/10.1038/s43016-021-00358-x.
[15] IPCC, “Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change” (Cambridge, UK and New York, NY, USA: Cambridge University Press, April 2022), https://report.ipcc.ch/ar6wg3/pdf/IPCC_AR6_WGIII_FinalDraft_FullReport.pdf.
