Continuing challenges of elephant captivity: the captive environment, health issues, and welfare implications

Enclosures

Although quantity of space is not explicitly mentioned in Vicino & Miller’s (2015) opportunities to thrive, it would seem to be a necessary component insofar as large, complex areas can sustain larger social groups of elephants, thus affording more opportunities for positive welfare by allowing choice, autonomy, and increased behavioral diversity. Alternatively, smaller elephant groups could also benefit from the positive welfare provided by larger spaces. Significant public attention has been directed at the housing and care of elephants (Meehan et al., 2016b), with that attention often focused on space. The scientific literature addressing free-living elephants indicates that, based on their large size and natural history, space is a key component of physical, behavioral, social, and mental well-being (Poole & Granli, 2009). Meehan et al. (2016b) note that space is a limited resource for zoos, and state that adequate space is required to support larger social groupings as well as robust feeding and enrichment programs; however, an “adequate” amount of space is not specified. An increase in space available to an elephant may also benefit metabolic health, but only if the increase in space is “substantial” (Morfeld & Brown, 2017, p. 11). Space limitations may be particularly problematic for male elephants, who range more widely in the wild than do family groups (Hartley, Wood & Yon, 2019). During musth, a period of heightened aggression and sexual drive (Ghosal et al., 2013), free-ranging males increase their range sizes and intermingle with multiple family units as they search for females in estrus (Fernando et al., 2008). Captive musth males also increase their movement and overall activity even though they remain constrained by their enclosures (LaDue et al., 2022a). Unlike free-ranging male elephants, captive males appear to have an elevated stress response during musth, although LaDue et al. (2022a) urge caution when interpreting wild-zoo differences due to sampling issues. Whereas providing access to females for mating may improve welfare, this may not be an option in all situations. It is worth noting that enclosure size is considered to be more important by scientists than by zoo personnel (Gurusamy, Tribe & Phillips, 2023). Just as the size of a species can affect its perception of enclosure complexity (de Azevedo et al., 2023), the size assessments that captive elephants make of their enclosures may potentially suffer from a floor effect insofar as all captive spaces may simply be too small from the perspective of elephants, whose home ranges are several orders of magnitude larger than even the largest zoo enclosure (Atkinson & Lindsay, 2022).

In zoos, researchers have found that factors other than space have an effect on elephant welfare (Meehan et al., 2016b). In their epidemiological study of African and Asian elephant welfare in 68 accredited North American zoos, Meehan et al. (2016b) determined that measures of space were not identified as risk factors for certain welfare outcomes (e.g., stereotypies, obesity, female reproductive dysfunction). However, in accordance with a possible floor effect, they acknowledged that their investigation was “limited to the range of exhibit sizes at participating North American zoos [i.e., 118 to 15,765 m²], and future studies incorporating larger areas could potentially find associations between space and welfare outcomes” (p. 7). With relatively little variation in zoos, and with no zoos approximating even the smallest range space available to elephants in the wild, measuring the effect of zoo enclosure size on welfare is challenging (Mason & Veasey, 2009). In their natural habitat, elephants have expansive home ranges, extending from tens to 10,000 km² (Fernando et al., 2008; Ngene et al., 2017; Bahar, Kasimi & Hambali, 2018). Enclosure sizes for captive elephants vary. Some zoo accrediting agencies specify minimum space standards for outdoor spaces ranging from 500 m² per elephant (AZA, 2021b) to a minimum shared space of 3,000 m² (BIAZA, 2019). Minimum indoor stall size varies from 47 m² for one female elephant (AZA, 2021b) to 300 m² for four females (BIAZA, 2019), even though an elephant can spend a significant amount of time indoors (Greco et al., 2016a). Although some larger exhibits exist (e.g., Disney’s Animal Kingdom: 28,300 m²; San Diego Zoo Safari Park: 13,000 m²), the actual size of most outdoor enclosures in a sampling of 20 Western zoos (and three Eastern elephant centers) appears to be in the range of 17–6,937 m² per animal (Taylor & Poole, 1998), which is orders of magnitude smaller than their natural habitat—we note that more recent data on this issue are necessary. More recently, Meehan et al. (2016a) evaluated not only total exhibit size, but the individual elephant’s experience based on the time they spent in different subdivisions of the exhibit. Calculated Total Space Experience (measurements of area as a function of time) values ranged from 117 to 15,742 m² (mean Overall Total Space Experience for African elephants was 3,642 and 1,781 m² for Asian elephants). For elephants in captive environments, given that an elephant can typically cover every section of a 10,000 m² enclosure in less than an hour, Atkinson & Lindsay (2022) suggest that an enclosure should be one or two orders of magnitude (i.e., ~1 km²) greater than traditional zoo exhibits. They note that such an enclosure then would allow for (1) a sufficient quantity and diversity of vegetation for normal foraging behaviors as well as visual screening, and (2) dozens of widely distributed focal points (e.g., pools, rubbing rocks, sand mounds) for socialization and possible avoidance behaviors. Moreover, they suggest that a more tropical climate, similar to African and Asian elephants’ natural habitat, would be preferable (as opposed to an arctic climate as in northern Canada). Atkinson & Lindsay (2022) conclude that range countries are the only places where elephants can truly flourish.

The issue of space in zoos is challenging, as zoos tend to be situated in urban centers, limiting possibilities for significant expansion. Schmidt & Kappelhof (2019) acknowledge that huge investments would be necessary to transform European elephant exhibits into complex fission-fusion housing systems and to create more capacity for male elephants. For zoos to maintain compatible herds, facilitate social behaviors, and manage male elephants, it would require “substantial financial investment, large physical spaces of many hectares [1 hectare = 10,000 m²] and considerable expertise in elephant management” (Hartley, Wood & Yon, 2019, p. 74). Some zoos in the U.S. have been investing in exhibit renovations at great cost, for relatively small expansions. The Fort Worth Zoo recently invested $32 million to improve their 12,141 m² elephant enclosure (Fort Worth Zoo, 2024); the Cincinnati Zoo will spend $50 million for a 20,234 m² area (Miller, 2023); and the Memphis Zoo recently announced a $250 million plan to renovate their exhibits for African elephants, rhinoceroses, and giraffes (Moore, 2024). Future research will be required to determine if these renovations have the effect of significantly improving elephant well-being. For now, improvement in zoo association elephant standards for space, social groupings, and male elephants may force zoos to invest in exhibit changes. In addition, research conducted in non-zoo facilities that provide much larger, more complex spaces (e.g., sanctuaries, conservation centers, semi-captive situations) could potentially improve our understanding of the effect of these spaces on welfare, with applications to zoos and other captive situations.

Space complexity and enrichment

The attributes of an enclosure would seem to be an essential element in Vicino & Miller’s (2015) requirement that animals have quality spaces to express species-specific behavior. In this regard, space complexity has indeed been shown to be important to the welfare of elephants in zoos. Scott & LaDue (2019) examined the behavior of two elephants in three different treatments: a small simple space (with a pool and shade structures), a small complex space, and a large complex space (the latter two with a pool, shade structures, elevated feeders, and varied substrate). Results indicated an increase in foraging and self-maintenance and a decrease in time spent stationary in both the small and large complex spaces, underscoring the importance of space complexity. Scott & LaDue (2019) determined that increased exhibit complexity was sufficient to significantly improve behavioral diversity, overall activity levels, and other behaviors indicative of positive welfare, leading them to conclude that complexity should be prioritized over space (see also DiVincenti, McDowell & Herrelko, 2023). However, the authors’ own results show that activity and behavioral diversity were highest in the large complex space. Increased investigatory behavior was significantly higher in only the large complex yard, with stereotypic behavior decreasing “drastically” in that condition. This indicates that the combination of space and complexity achieved even more impressive results than did complexity alone. Atkinson & Lindsay (2022) suggest that more expansive spaces, with varied topography, natural substrates, and natural vegetation for foraging, can better provide the level of environmental complexity necessary to stimulate natural behaviors and provide choice, autonomy, and diversity of experience. A similar conclusion was reached by Brown et al. (2020), who noted better elephant welfare in enriched, less constricted, and more stimulating environments.

Creating space complexity can include the addition of various forms of enrichment to an enclosure in order to promote the expression of important behaviors (e.g., foraging, self-maintenance) and provide opportunities to engage in exploration, play, and problem solving, as well as exercise choice (Hoy, Murray & Tribe, 2010; French, Mancini & Sharp, 2018; Greco et al., 2016b). Importantly, in their natural habitat, animals use their cognitive skills to overcome problems that may directly or indirectly affect their survival. Unfortunately, this type of problem solving has historically been absent in captivity (Meehan & Mench, 2007). Only in more recent years has greater focus been placed on more cognitive types of enrichment, including time-delay puzzle feeders, high tech enrichment devices, and insightful problem solving tasks (Foerder et al., 2011; Highfill et al., 2016; Krebs & Watters, 2017; French, Mancini & Sharp, 2018; Barrett & Benson-Amran, 2020). Such cognitive enrichment, especially when species-appropriate abilities are stimulated, appears to provide some positive behavioral benefits by increasing the animals’ interest in their environment (Krebs & Watters, 2017; Holland, 2018). Nevertheless, the long-term benefits of such enrichment remain unclear. Furthermore, cognitive enrichment may be provided infrequently in zoos, with keepers perceiving that semi-permanent and permanent exhibit features (e.g., pools, logs, scratching areas, puzzle feeders) are more important than cognitive enrichment (Hoy, Murray & Tribe, 2010; Greco et al., 2016b). Greco et al. (2016b) found that problem-solving opportunities were components in the enrichment programs of 97% of the 63 zoos they examined; however, these opportunities were provided infrequently. In fact, the frequency of problem-solving enrichment was so low that Greco et al. (2016b) were unable to test its effects in their welfare models. Such a low frequency may be attributed to the time required to prepare and distribute enrichment, as opposed to feeding enrichment, tactile enrichment, and human-animal interactions (Hoy, Murray & Tribe, 2010).

There are few studies on the effects of enrichment for elephants, with most of them focused on feeding enrichment (Greco et al., 2016b). As such, more research is necessary in this area. Just as importantly, zoos must focus on the suite of enrichment options available to them, especially in the area of problem-solving enrichment. Impediments to the provision of enrichment include conflicting priorities, lack of available time, and uncertainty about which enrichment practices are most effective (Hoy, Murray & Tribe, 2010; Tuite et al., 2022). This indicates the need for more support at all staff levels and for increased resources to allow for the effective implementation and evaluation of enrichment (Hoy, Murray & Tribe, 2010).

Exercise

Vicino & Miller’s (2015) third opportunity to thrive is that the animal should have an environment that promotes good health. Certainly, an important welfare issue for elephants is the amount of exercise they can experience, which is often measured by distance traveled (Holdgate et al., 2016). Estimates of walking distances in nature vary by age and sex (Slotow & van Dyk, 2004), season (Loarie, van Aarde & Pimm, 2009), and resource availability (Gadd, 2002). Nevertheless, ~8–12 km/day are normal, with much greater distances (up to ~50 km/day) being common (Wall et al., 2013; Miller, Hogan & Meehan, 2016). Zoo-based studies of walking in elephants do not use uniform outcome measures, making comparisons to natural habitat studies difficult. At the San Diego Zoo Safari Park, Miller, Andrews & Anderson (2012) and Miller, Hogan & Meehan (2016) estimated an average walking distance of 8.65–9.82 km/day for African elephants. Similarly, Brady, McMahon & Naulty (2021) estimated Asian elephants traveled 9.35 km/day at the Dublin Zoo (enclosure size: 8,500 m²). Rowell (2014) calculated a distance of 9.05 km over an 18-hour period at the Royal Melbourne Zoological Gardens (enclosure size: 5,143 m²). Holdgate et al. (2016), who gathered data on both African and Asian elephants across 30 North American zoos, concluded that elephants walked only a distance of 5.3 km/day, far shorter distances than in the wild. In general, the larger the enclosure, the greater the opportunity for purposeful walking. Importantly, Holdgate et al. (2016) also state that the range of exhibit sizes in the study population may not have been sufficient to show the effect that very large areas can have on walking distances, and that elephants might walk more if provided larger enclosures than are currently available in North American zoos (cf., Meehan et al., 2016b). Holdgate et al. (2016) further determined that, overall, the distances elephants walked in zoos were influenced most significantly by feed-related factors. However, Hacker, Miller & Schulte (2018) note that Holdgate et al. (2016) compared walking across institutions and not within the manipulated space of one facility. In their own study, Hacker, Miller & Schulte (2018) manipulated space and food in different conditions: access to two yards with food (Both); access to one yard with food (Half); access to both yards with food in only one (Both/Half). They then measured walking distances and behavior among elephants belonging to different dominance groups. Increased walking distances for middle-ranked African elephants and a significant correlation between the change in behavioral diversity across the Both and Half treatments, along with dominance ranking, led the authors to conclude that both food and space impact elephant walking and behavior at their institution. More clarity is needed to assess the importance of space relative to food and how it affects walking distances.

Although the functional need for walking appears to be reduced in zoos due to the provision of social and nutritional resources, ~60 million years of evolution has provided Proboscidea with a range of anatomical and physiological specializations for long distance living (Shoshani, 1998; Poole & Granli, 2009). For example, elephant foot morphology allows them to load the most lateral aspect of their feet when walking which, along with decreasing pressure below the foot’s fat pad, helps them maintain foot health (Panagiotopoulou et al., 2016). This locomotion mechanism does not work well for elephants housed in small enclosures with hard substrates (Panagiotopoulou et al., 2016). The exercise elephants obtain by walking is also a crucial component of an enriched environment because it supplies the brain with oxygenated blood, increases the effectiveness of the immune system, and enhances an animal’s cognitive abilities (Jacobs et al., 2021), while also providing metabolic benefits (Morfeld & Brown, 2017). Finally, Holdgate et al. (2016) contend that walking is important because it supports exploratory behavior and its information-gathering function may be rewarding outside of the acquisition of resources.

Clearly, more research is required to determine the relationships between walking, space, food, and social conditions for elephants in zoos. However, the lack of significantly larger enclosure sizes will continue to constrain such investigations. As previously stated, research in non-zoo facilities with larger, more complex spaces could enhance our knowledge, potentially improving captive elephant welfare or helping to determine whether elephants should be kept in captivity at all.

Sensory experience

Estimates of elephants’ walking distances in zoos often miss a critical element: how the captive elephant’s sensory experiences while walking differ from those of a free-ranging elephant whose sensory environment is composed of a broad and often-changing variety of sounds, smells, sights, and tactile and taste sensations. In nature, elephants exist in different types of large, dynamic ecosystems (e.g., forests, open savanna, wet marsh, and desert; Clubb & Mason, 2002), where they interact with diverse ecological features and collect information about the physical and social environment. In captivity, sensory experiences are largely predicted by the temporal and spatial features of their enclosures, which are relatively limited and unchanging, and the enrichment provided within these limited spaces (Lucas & Stanyon, 2017). Elephants in zoos typically are confined to one or more outdoor yards, smaller holding areas, and/or a barn with individual and/or group stalls (Lucas & Stanyon, 2017). Outdoor enclosures may contain a variety of natural and artificial features such as soil and hard substrates (e.g., concrete), water features, wet and dry wallows, and (concrete) rockwork or tree features to accommodate maintenance behaviors (Clubb & Mason, 2002; BIAZA, 2019; AZA, 2021b). Note that concrete substrate/rock/tree features challenge Vicino & Miller’s (2015) requirement that the opportunity to self-maintain includes substrate that is species-specific. If sensory stimulation via exploratory behavior and information-gathering in the captive environment is indeed rewarding in itself (Holdgate et al., 2016), more diverse and natural surroundings with the opportunity for more varied sensory experiences may be important to improving elephant welfare. Facilities providing this type of environment currently include three Global Federation of Animal Sanctuaries accredited elephant sanctuaries (two in the U.S. and one in Brazil) and the AZA accredited White Oak Conservation Foundation in Florida. For example, the elephant area at White Oak measures 68.8 km², encompassing forest, open grasslands, ponds, and wetlands, and a variety of plant (food) species (White Oak Conservation, 2022). The Elephant Sanctuary in Tennessee spans 12.4 km² and features spring-fed lakes, pastures and woodlands, and natural forage (The Elephant Sanctuary in Tennessee, 2024).

Insofar as human visitors are another prominent feature of the zoo environment, they are also part of an elephant’s sensory experience. Decades of research indicate that the effect of zoo visitors can be negative, neutral, or positive in terms of animal behavior and welfare (Hosey & Melfi, 2014; Sherwen & Hemsworth, 2019). Species-specific differences, the nature of the visitor interactions, enclosure design, and individual animal temperament may contribute to the variation in zoo animal responses (Sherwen & Hemsworth, 2019). Visitor effects on elephants are also mixed, as are the facilities and conditions under which they were studied. Based on research at the St. Louis Zoo, Krishnan & Braude (2014) concluded that large audiences could serve as a source of behavioral enrichment, as indicated by decreased stereotypies and the elephants’ use of space. However, their definition of a “high” crowd size was five or more people and “low” as fewer than five, raising the question of whether their methodology was sensitive enough to support their conclusion. Manning et al. (2023) studied elephants in a South African tourism park that offered human-elephant interactions (e.g., walks, feeding); observations were conducted before, during, and after COVID pandemic closures. They concluded that changes in behavior, including elevated self-directed behaviors (a novel indicator of anxiety in African elephants), demonstrated that returning tourist numbers, whether high or low, were stressful for the elephants. In examining public feeding of elephants in a zoo, Fernandez, Upchurch & Hawkes (2021) suggested that public feedings may function as enrichment for some elephants. Although the study was limited, the authors found that public feedings appeared to result in increased social activity and a decrease in stereotypies (only one elephant engaged in the behavior regularly) when compared with nonpublic feeding days for two of three elephants. All three elephants showed increased foraging and were more active in the period after a public feeding session. Williams et al. (2023) found that elephants had positive responses to visitors more frequently than would be expected by chance; however, they noted that enclosure design may reduce the negative impact of zoo visitors on elephants because of the necessity of keeping a safe distance between visitors and elephants. Although Quadros et al. (2014) did not specify elephants in their study of the effects of visitor noise on 12 mammal species, they found significant behavioral differences at the individual level. They concluded that the noise generated by zoo visitors negatively affects the welfare of individual animals, especially when noise amplitude exceeded 70 dB(A). Given the popularity of elephants in zoos, and the noise that can be generated by the public (especially in indoor spaces), further study is clearly required to determine whether and in what ways such human interaction affects elephant welfare across different exhibit designs.

Caretakers represent another human facet of an elephant’s environment. Because elephants are highly managed, keepers are essential contributors to elephant welfare (Carlstead, Paris & Brown, 2019), an important aspect of Vicino & Miller’s (2015) opportunity for optimal health. Elephants typically spend just over 50 percent of their daytime hours under behavioral control in managed activities (e.g., exercise sessions, foot and skin care, training; Greco et al., 2016a). There appears to be an inverse relationship between time spent with keepers and an elephant’s rate of stereotypy, even outside of time spent in keeper-directed activities. Greco et al. (2016a) attribute this to the social relationships that elephants form with their keepers. Keepers may also benefit from these relationships, as the stronger the bond felt by keepers, the less likely they were to report dissatisfaction with their jobs (Carlstead, Paris & Brown, 2019). However, keepers generally appear to develop stronger bonds with Asian rather than African elephants, which is reflected in job satisfaction measures (Carlstead, Paris & Brown, 2019). Management and housing differences may account for such measures (Carlstead, Paris & Brown, 2019) because Asian elephants are managed for a significantly greater proportion of their day than African elephants, who tend to be more independent (Greco et al., 2016b). On average, African elephants spend more time in unique environments, and experience greater total space and more outdoor space than Asian elephants (Meehan et al., 2016a). Bonds between keeper and elephant may be related to welfare benefits for both. Carlstead, Paris & Brown (2019) found that positive keeper attitudes were associated with lower mean serum cortisol concentrations in elephants, suggesting that good keeper-elephant relationships lower stress responsiveness in elephants. In the future, zoos could substantially expand exhibits for Asian and African elephants, potentially giving the animals greater autonomy and requiring less keeper-managed time. This may call for reassessing the nature of keeper-elephant relationships in order to maintain a beneficial relationship for both keepers and elephants. Another important area of reassessment has been the end of bullhook use in AZA zoos (AZA, 2022) and the phase-out of these devices in EAZA facilities (EAZA, 2019), as their use presents serious challenges to the safety and well-being of both elephants and keepers (Wilson et al., 2015).

Because the study of keeper-elephant relationships has received little attention in zoos (Carlstead, Paris & Brown, 2019), more research in this area is necessary. Research that focuses on the attitudes and beliefs of keepers relative to welfare measures in elephants could contribute to positive welfare for elephants and ensure that keeper-elephant relationships are mutually beneficial (Carlstead, Paris & Brown, 2019).

Sociality

Facilitating appropriate social structures for captive elephants is a particular challenge for Vicino & Miller’s (2015) provision that the opportunity to express species-specific behavior requires appropriate social groupings. Free-ranging elephants tend to live in matriarchal, multi-generational family groups of two to 10 adult females and their offspring (Sukumar, 2006; Vance, Archie & Moss, 2009; de Silva, Schmid & Wittemyer, 2017). Elephant family groups share a fission-fusion structure, separating and merging with larger groups of up to several hundred elephants (Poole & Moss, 2008; de Silva, Ranjeewa & Kryazhimskiy, 2011). In nature, female elephants remain with their natal herd, forming strong lifelong bonds with related females, although family fissions are also common (Garai, 1992; Sukumar, 2006; Poole & Moss, 2008). Males differ in that they remain with their family group until sexual maturity, when they disperse (Lee et al., 2011). They nevertheless maintain complex social ties with conspecifics of all ages and both genders (Hartley, Wood & Yon, 2019).

Historically, captive elephants have often been limited to small groups of mostly unrelated adult females with very few infants or juveniles (Clubb & Mason, 2002; Rees, 2009), making the inability to replicate natural social structures an area of concern for the welfare of elephants in zoos (Williams et al., 2019b). Current research suggests that captive elephants should be held in related, multi-generational groups for optimal welfare (Williams et al., 2019b; Finch et al., 2020; Harvey et al., 2018; Hartley & Stanley, 2016). In North America, only 20% of the 226 adult elephants in U.S. zoos studied by Meehan et al. (2016a) had the opportunity to interact with young elephants, albeit an absence of calves in a group does not necessarily lead to poor welfare (Williams et al., 2019b). Currently, zoo standards allow for as few as three or four elephants (AZA, 2021b; BIAZA, 2019) and some zoos continue to house females singly (AZA, 2021a). The AZA’s three-elephant minimum does not specify that all elephants must be females, which can contribute to isolating female elephants (e.g., two males housed separately from one female), or even the same species. Schmidt & Kappelhof (2019) report that most of the female reproductive elephants in the Asian Elephant European Association of Zoos and Aquaria Ex situ Programme (EEP) are being held in family groups, an improvement over previous years. However, in general, there has been uneven progress in increasing mean group size for both Asian and African elephants across North American and European zoos, with significant increases found only for African elephants in North America (Rees, 2021). In North American zoos, mean group sizes remain at 4.45 and 3.3 for African and Asian elephants, respectively (Rees, 2021). Although Williams et al. (2019b) did not find a statistically significant link between group size and prosocial behavior, they observed that the largest social group with the greatest number of calves was the most socially interconnected. Other recent studies have found more affiliative behaviors and fewer aggressive behaviors among related elephants (Harvey et al., 2018; Finch et al., 2020), supporting the recommendation for keeping elephants in related, multi-generational groups to optimize welfare. Using free-ranging elephants as a model for structuring elephant groups in zoos has several benefits, including promoting and facilitating social learning and natural social behaviors, reducing abnormal behaviors and stereotypies, and improving health, welfare, and reproduction (Hartley, Wood & Yon, 2019). However, to hold the greater number of elephants that comprise multi-generational groups, zoos would require far larger and more complex facilities. This presents a significant challenge due to the space and resources that would be required.

The social environment has a considerable impact on stereotypic behavior rates, an important indicator of compromised welfare (Greco et al., 2016a, 2017). Spending more time with larger numbers of conspecifics and the amount of time spent with juveniles is associated with reduced stereotypy rates, whereas being housed separately increases stereotypic behavior (Greco et al., 2016a). Inter-zoo transfers also increase the risk because they can be disruptive to social groups and break important social bonds (Clubb & Mason, 2002; Armstrong & Johnson, 2021). Transfers are usually for breeding purposes, coordinated through programs such as the Species Survival Plan that AZA accredited zoos have adopted to promote genetic diversity (Rees, 2011). However, elephants can also be moved between facilities as a result of space limitations, sex of offspring (Williams et al., 2019b), or exhibit closure. The transfer of group members can affect social stability and limit long-term success in zoo social groups (Williams et al., 2019b). As many as 80% of the elephants in North American zoos in 2012 had experienced at least one inter-zoo transfer (Prado-Oviedo et al., 2016). Clubb et al. (2008) found that inter-zoo transfers lessened Asian elephant survivorship, an effect that lasted four years post-transfer, and Asian calves removed from their mothers at a young age tended to have poorer outcomes. Considering the complexity and importance of elephant social ties, it would be important to study the effects of transfers on elephants based on age, sex, and species.

Male elephants pose a particular challenge in captive facilities due to their strength, social needs, aggressiveness, and strong sexual and competitive motivations (Lee & Moss, 2009; Hartley, Wood & Yon, 2019). The proportional increase in the captive male population in recent years presents an additional challenge. Going forward, the number of males born will exceed the number that can be maintained in breeding groups and housing will be needed to accommodate multiple groups of males (Readyhough et al., 2022). In 2017, males made up only 21% of the Asian elephant population in North America but will eventually approach 50% (LaDue et al., 2022b), requiring greater consideration of their physical and social needs and living space. Schmidt & Kappelhof (2019) suggest that Asian elephant husbandry guidelines in Europe may need to be redefined to, for example, encourage zoos to hold male offspring longer. Still, more space would be needed to hold males. They also suggest potentially requiring zoos to restrict breeding to keep the number of males manageable.

Increasing evidence indicates that male elephants display extensive social behaviors that rival the complexity of that of females (LaDue et al., 2022b). Despite their social nature, males are typically held separately from females and other males although some zoos have integrated adult males into social groups. In North America and Europe, the majority of males are kept at zoos with females but no other males, restricting social learning from older males and the development of appropriate social and reproductive behaviors (Hartley, Wood & Yon, 2019). Although the importance of sociality for male elephants is stressed by some (LaDue et al., 2022b; Readyhough et al., 2022), and although there are welfare benefits from social interactions (e.g., reduction in stereotypies, Readyhough et al., 2022), progress appears to be slow. Some U.S. and European zoos have created all-male “bachelor” groups in response to the evidence for male sociality and the need to provide space for an increasing number of captive males (Hartley, Wood & Yon, 2019; Readyhough et al., 2022). In North America, there are currently two zoos holding bachelor groups (AZA, 2019, 2021a); in Europe, there are nine (Schmidt & Kappelhof, 2019). However, Hartley, Wood & Yon (2019) caution that the focus of these groups should be on the social development of the elephants rather than on creating bachelor groups as a convenient solution to housing males. Subadult males of similar age and experience level who are held together in bachelor groups are at risk of developing abnormal behaviors or a limited behavioral repertoire due to the absence of social learning from mature males and family groups, including failure to learn reproductive behaviors (Hartley, Wood & Yon, 2019). Mixed-age bachelor groups that include a mature male may facilitate social learning (Readyhough et al., 2022). The social makeup of bachelor groups is subject to change, as males may be moved more often for breeding purposes (Schmidt & Kappelhof, 2019), potentially breaking social bonds. The effects of such moves on individual males and groups is an area for future study. Importantly, the creation of bachelor groups is a relatively new development in zoos; as such, it is too early to determine if it will be feasible over a long period of time (Schmidt & Kappelhof, 2019). Also, many zoos prefer to keep their male with a female group for his well-being (Schmidt & Kappelhof, 2019). Clearly, providing appropriate housing and social conditions for males will continue to be a challenge as the number of male elephants continues to increase.

Optimizing the social lives of captive elephants may involve a variety factors: increasing group size (Lasky et al., 2020), facilitating highly related multi-generational groups (Hartley & Stanley, 2016; Harvey et al., 2018; Williams et al., 2019b; Finch et al., 2020; Armstrong & Johnson, 2021), integrating males and females (Hartley, Wood & Yon, 2019; Lasky et al., 2020), developing fission-fusion housing (Schmidt & Kappelhof, 2019), and creating mixed-age bachelor groups (Hartley, Wood & Yon, 2019; Readyhough et al., 2022). These improvements all require significant resources, including larger, more complex environments that allow for the management of males and females (Hartley, Wood & Yon, 2019). At the same time, changes must be carefully approached and science-based. For example, while fission-fusion housing strategies are suggested as a way to improve welfare, uncontrollable social changes and social frustration could generate locomotor stereotypies (Greco et al., 2017). More research is necessary to determine a management strategy that would enable fission-fusion through free choice of social associations and indoor/outdoor locations on a round-the-clock basis. Although the opportunity for appropriate social contact is considered to be more important for welfare than space (Williams et al., 2020), Veasey (2020) cautions that zoos with limited space will face challenges in addressing such important behaviors as elephant sociality, feeding, and foraging. Certainly, greater capacity is required to hold large or growing groups of elephants and to accommodate males (Hartley, Wood & Yon, 2019; Schmidt & Kappelhof, 2019). In addition, there is a need to better understand captive elephants’ social networks and social behavior over longer periods of time and how social changes affect individuals and groups (Williams et al., 2020). The resource-intensive conditions (e.g., space, finances, personnel) required to facilitate appropriate social conditions for captive elephants could stand in the way of optimizing welfare and could determine whether elephants should be kept in captivity.

Feeding challenges

Vicino & Miller (2015) stress that animals should have the opportunity for a thoughtfully presented, well-balanced diet. For elephants, feeding involves much more than meeting dietary requirements. It often involves cognitive and physical challenges that are essential to an elephant’s behavioral ecology and well-being. Foraging involves movement, exploration, and choosing areas in a landscape that hold high foraging opportunities, then homing in on preferred foods (Das, Kshettry & Kumara, 2022). The handling of food may involve manipulation (e.g., prying bark from a tree, digging, kicking dirt off of grass roots) before consumption (Poole & Granli, 2009). Feeding is often done in concert with the family group, making it a social activity. Individual elephants vary their diets based on availability, preferences, and physiological needs (Gill et al., 2023). Free-ranging elephants are highly diverse feeders, consuming more than 100 seasonally and geographically varying food species (e.g., grasses, trees, bark, roots, fruits, and aquatic plants; Dierenfeld, 2006; Campos-Arceiz & Blake, 2011; Das, Kshettry & Kumara, 2022). As large-bodied herbivores, elephants naturally spend 60–80% of their waking hours foraging over long distances (Poole & Granli, 2009). In contrast, zoo diets tend to be more restricted in terms of the variety of foods, are seasonally invariant, and (often excessively) high in calories (Carneiro et al., 2015; Schiffmann et al., 2018). Diets for elephants in zoos generally consist of dried forage, usually hay, supplemented with commercial concentrate feed pellets, vitamins, fruits, and vegetables (Dierenfeld, 2006). Feeding schedules tend to be temporally predictable with high spatial predictability and only moderate variation in feeding methods (Greco et al., 2016b), although some zoos try to diversify the way animals are fed as a form of limited enrichment (e.g., puzzle/feeder balls, elevated nets with hay; Ramírez et al., 2023). Not unexpectedly, a more varied and unpredictable feeding regime enhances well-being in captive elephants by promoting walking (Holdgate et al., 2016). Nearly three-quarters of the elephants across 65 AZA accredited North American zoos were determined to be overweight or obese, putting them at risk of health conditions such as ovarian acyclicity (Morfeld et al., 2016), joint disease/osteoarthritis (West, 2006), foot ailments, and gait problems (Roocroft & Oosterhuis, 2001).

Timing, frequency, and methods of food provision are key components of elephant feeding programs (Greco et al., 2016b). These areas nevertheless require continuous attention to avoid temporal and spatial predictability. As with enrichment, feeding programs require considerable keeper time, resources, and biologically informed policies. Including browse in the diet appears to be beneficial for elephant welfare as it increases opportunities for foraging behaviors and the amount of time spent feeding, decreases inactivity, and allows elephants to engage in naturalistic feeding behaviors (e.g., manipulating branches, stripping leaves and bark; Stoinski, Daniel & Maple, 2000; Lasky et al., 2020), although it may not decrease stereotypic behaviors (Lasky et al., 2020). In highlighting the importance of foraging, AZA and BIAZA guidelines state that browse should be made available to elephants; however, neither organization specifies the amount of foraging time or type of foraging opportunities (Veasey, 2020). This is an area that must be addressed in order to standardize feeding practices and improve elephant welfare.

Brain and behavior

Physical health