Abstract
In order to achieve a better understanding of the factors that might have led our hominin ancestors to transition to a more terrestrial niche, including sleeping on the ground, we have conducted a study on the ground nesting behavior of chimpanzees (Pan troglodytes). Chimpanzees, like all other species of great apes, build nests in which to sleep each night, but little is known about regional differences in their nesting habits. Previously, nesting on the ground was considered typical of gorillas, but rare in most populations of chimpanzees. Using data acquired during our extensive chimpanzee nesting survey conducted between 2004 and 2013 across a > 50 000 km2 region in northern Democratic Republic of the Congo, we report a distinctive ground nesting behaviour of eastern chimpanzees (P. t. schweinfurthii). We have mapped the geographical distribution of ground nesting and compared its frequency at 20 survey areas on both sides of a large river, the Uele. We found that ground nests made up more than 1% of total nests at 15 of the 20 survey regions. For a subset of 16 of these regions, we utilized statistical models to investigate whether forest type and structure, as well as the abundance of carnivores and large herbivores, and the activities of humans impacted the frequency of ground nesting and nest height. We predicted that higher encounter rates of human and dangerous animal signs would be associated with lower rates of ground nesting as well as increased nest height. Overall, 10.4% of the Bili-Uéré chimpanzee nests were terrestrial, but the frequency of ground nesting varied extensively between the survey areas (0-29% of nests). The occurrence of ground nests was positively associated with denser forests (p = 0.004), herb patches (p < 0.001), and light gaps (p < 0.001). Light gaps (p < 0.001), herb patches (p = 0.044), and vine tangles (p = 0.016) also had a strong negative effect on nest height. Hunting by humans had a negative effect on the probability of the occurrence of ground nests (p = 0.001) and a positive one on nest height (p = 0.013), with a similar but likely marginal effect of large herbivores on nest height (p = 0.023). In addition, the chimpanzees nested at significantly lower heights with increasing distance from roads and settlements (p < 0.001). Carnivore encounter rates, however, had no significant impact on ground nest frequency or nest height. Our results indicate that ground nesting can no longer be considered a rare and patchily-occurring phenomenon in Pan troglodytes, but is instead a major component of the chimpanzee behavioural repertoire across a considerable fraction of the range of the Eastern subspecies. Our study highlights that neither the large body size of gorillas nor the taming of fire are necessary conditions for hominids to sleep overnight on the ground, even in areas inhabited by multiple species of large carnivore. Human hunting, however, appears to reduce the probability of ground nesting, or eliminate the behavior altogether.
Introduction
Following the evolutionary split of hominins from the genus Pan between approximately 6.5 and 9.5 million years ago (Almécija et al., 2021), our ancestors gradually transitioned to a more terrestrial lifestyle. The selective pressures leading to this shift, however, remain poorly understood. By four million years ago, the genus Australopithecus displayed clear anatomical adaptations for bipedal walking (White and Gen, 1987; Berger et al., 2010), although the similarity of its gait to that of modern humans is still debated (Stern, 2000; Ward, 2002), and it appeared to retain some morphological adaptations for climbing (Berge, 1994). By at least two million years ago, early members of the genus Homo had evolved the full suite of bipedal characters present in modern humans, including a longitudinally arched foot (Berger et al., 2010, Hatala et al., 2016). Eventually, the increased terrestriality of hominins would likely have involved sleeping on the ground, perhaps in ground nests, as we see today in the genus Gorilla and some populations of chimpanzees (Tagg et al., 2013; Hicks, 2010). It remains uncertain how small-bodied hominins would have managed to sleep on the ground given the large number of predators living in the Pliocene (Lewis and Werdelin, 2007), some of which are suspected of preying on hominins (Berger and McGraw, 2007; Treves and Palmqvist, 2007; Isbell et al., 2018). Wrangham (2009) hypothesized that the taming of fire approximately two million years ago may have been the necessary precondition for this innovation, although direct evidence remains lacking. To speculate, perhaps when hominins began to make use of resources in more open country habitats (Finlayson et al., 2011), they faced a shortage of suitable nesting trees and thus slept on the ground, possibly even protecting themselves with bomas (spikey defensive structures fashioned from thorny vegetation; Isbell et al., 2018). In addition, building nests on the ground may have decreased the hominins’ visibility to large carnivores and/or hostile conspecifics (in relation to the ‘landscape of fear’ (Bleicher, 2017)), as well as saved them energy. Given that all nonhuman great apes construct night nests (Fruth and Hohmann, 1996) and that non-lithic technology is found in several species of great apes and monkeys (van Lawick-Goodall, 1968; Mannu and Ottoni, 2009), it is likely that these behaviors were characteristic of the earliest hominins as well. Understanding behavioral variation in nest-making in our closest relatives, the African great apes, should be informative about the likely range of behaviors of our ancestors and the selective pressure that might have pushed them to spend more time on the ground. Thus we have conducted a multi-regional study of ground nesting behavior in eastern chimpanzees (Pan troglodytes schweinfurthii) in the northern Democratic Republic of the Congo (DRC).
In the domains of tools use, gestural communication, and self-maintenance, much evidence has been amassed over the past three decades for the existence of cultural variation between populations of chimpanzees and other nonhuman great apes (e.g., in orangutans: van Schaik (2003), bonobos: Hohmann and Fruth (2003), gorillas: Robbins et al. (2016)), to a degree likely approaching that of australopithecines and early Homo (McGrew, 2004, 2010). Nonhuman great apes show extensive variation in their behavioural repertoires, from food choice to tool use and gestural communication, in ways that resemble human culture (McGrew, 2004, 2010; Whiten et al., 2001). With some caution, we can thus use behavioural variation in African great apes to triangulate the behavior of our common ancestor.
Little is known about variation in nest-building, despite the fact that it is a fundamental aspect of great ape behaviour, distinguishing them from hylobatids and monkeys; it has even been considered to represent the ‘original tool’ of hominids (Fruth and Hohmann, 1996). The construction of a nest or bed in which to spend the night is a behaviour shared by all species of great apes (Fruth and Hohmann, 1996; McGrew, 1992). These nests are also used for other activities such as grooming, play, and sex (Fruth et al., 2018). Given our close evolutionary relationship with our fellow great apes, especially the genera Pan and Gorilla (Prüfer et al., 2012; Scally et al., 2012), the study of ground nesting in these extant species can inform our understanding of hominin sleeping patterns (Koops et al., 2012b). Most populations of nonhuman great ape species, including chimpanzees, build their nests primarily in the trees (bonobos (Pan paniscus): Serckx et al. (2016), orangutans (Pongo spp.): van Casteren et al. (2012), but see Ashbury et al. (2015) for a case of one orangutan nesting on the ground). Both species of gorilla (Gorilla gorilla and G. berengei), however, frequently build night nests in both the trees and on the ground (Brugiere and Sakom, 2001; Mehlman and Doran, 2002; Sanz et al., 2007). All gorilla populations studied to date include ground nesting in their repertoire, ranging from nearly 100% of nests at Virunga, Rwanda (Sarmiento et al., 1996), and La Belgique, Cameroon (Willie et al., 2014), to 14% in Ngotto, Central African Republic (Brugiere and Sakom, 2001) and only 7.4% at Petit Loango, Gabon (Furuichi et al., 1997). The ubiquity of ground nesting in this taxon may be related to comfort or as an adaptation to keep cool on hot nights (Mehlman and Doran, 2002), or by the large size of gorillas and their greater degree of terrestriality; silverback males in particular may have difficulties finding branches capable of supporting their weight (Yamagiwa, 2001).
At the majority of sites across their range, including at many of the long-term research sites, chimpanzees prefer to nest in trees. Ground nests appear in almost all populations of chimpanzees as a very small proportion of total nests, due to the fact that old, injured, or ill chimpanzees sometimes sleep on the ground; this falls generally below 1% of total nests, as is the case in Gombe, Tanzania (van Lawick-Goodall, 1968), Kanyawara-Kibale, Uganda (R. Wrangham, pers. comm. in Hicks, 2010), Budongo, Uganda (Reynolds and Reynolds, 1965), and Comoé, Côte d’Ivoire (J. Lapuente, pers. comm., 20 July 2021). Nevertheless, it is becoming apparent that nonhuman great apes show more flexibility in where they build their nests than was previously assumed (e.g., McLennan, 2018). Earl et al. (2020) demonstrated that, in captivity, both gorillas and chimpanzees are capable of building ground nests if conditions are appropriate.
As more chimpanzee populations have been studied in recent years, researchers have documented a significant occurrence of ground nesting in some regions (supplementary table S1). Tagg et al. (2013) showed that some populations of all four chimpanzee subspecies include frequent ground nesting in their behavioural repertoire. Koops et al. (2007) proposed mate-guarding as a possible explanation for ground nesting. In order to compare the occurrence of this behaviour at different sites, we first need to agree upon a baseline condition for which a population can be considered a ground nesting one. To account for the occasional use of ground nests by sick or old individuals that likely occurs in all chimpanzee populations, we propose that only those populations with a proportion of ⩾ 1% of total nests being terrestrial be considered ground nesting ones. Using this measure, ground nests are common at only 19% (3/16) of the habituated chimpanzee populations listed in supplementary table S1 (this table is limited to sites for which we have reliable information).
It is difficult, however, to assess whether the structures that researchers at different sites have described as ground nests are consistent with one another. For example, at Taï Forest, Boesch (1995) documented a number of what he termed ‘day nests’ made on the ground, but it is unclear what percentage of total nests these are, and if they might more properly be referred to as ‘leaf cushions’ (i.e., plants placed on the ground and sat on without any arrangement; Hirata et al., 1998). As another example, at Haut Niger, Guinea, Fleury-Brugiere (2002) did not mention ground nests, but they included a photograph of an old one. Consequently, it is unknown whether some of the 1074 nests they mention might have been terrestrial. These examples highlight the difficulty in interpreting the published data as to the presence or absence of ground nests at a particular site.
Although some studies have focused on the structural details of tree nests (e.g., Humle, 2003; Stewart et al., 2018), as well as the influence of habitat type in the context of hominin nesting sites (Hernandez-Aguilar, 2009), very little comparative work has been conducted on the construction styles of chimpanzee ground nests, nor of the raw materials used to build them. Some evidence suggests, however, inter-site variability in ground nest construction styles. For example, at two sites in Senegal, Fongoli (Stewart, 2011b) and Bagnomba Hill (Badji et al., 2018), researchers documented ground nests made of elephant grass (Andropogon sp.). Such nests were never found in the current study in northern DRC, despite the fact that savanna grass was abundant, including species of the genus Andropogon (Hicks, 2010). No ground nests were found at Ishasa (DRC), despite abundant elephant grass (J. Sept, pers. comm., 19 July 2021). Of course, such differences in vegetation choices may reflect differences in availability or abundance of construction materials.
An important aspect of any study of ground nests is to define what a ground nest is used for and how it is constructed, especially given that their definition varies between studies. With habituated chimpanzees, it is often possible to observe directly whether a ground nest is made during the day for resting or in the evening for sleeping at night (e.g., at Bulindi, Uganda, M. McLennan (pers. comm., 1 September 2022) observed a pregnant chimpanzee making a ground nest to rest in during the day). For non-habituated subjects, however, researchers must infer how the nest was used based on the presence of other nests, as well as dung, hairs, and urine (Hicks, 2010), or, if they are lucky, by encountering the chimpanzees at their nest groups in the early morning or observing them resting in day nests. To compound the confusion, at a number of sites, chimpanzees also sometimes sit on leaf cushions. This is the case at Kibale (Kanyawara), Uganda (R. Wrangham, pers. comm., 18 January 2010 in Hicks, 2010), and Kalinzu, Uganda (Furuichi and Yashimoto, 2000), Bossou, Guinea (Hirata et al., 1998), Taï, Côte d’Ivoire (Boesch, 1995), Goualougo, the Republic of Congo (Sanz and Morgan, 2007), as well as at Tenkere, Sierra Leone, in the latter case to protect themselves from thorns (Alp, 1997). At some research sites, it is not clear that these are categorized separately from ground nests, as we do here.
Another problem arises from a failure of researchers to observe ground nests, perhaps because the structures are not included in their ‘search image’. This can be the case especially during nesting surveys, where ground nests may be missed by surveyors who keep their eyes in the canopy scanning for tree nests.
Finally, another difficulty is that at some research sites where gorillas and chimpanzees share the same forest, distinguishing ground nests made by the two species is difficult (e.g., in Petit Loango, Gabon (Furuichi et al., 1997), Ngotto, Central African Republic (Hicks et al., 2009), and Goualougo, the Republic of Congo (Sanz et al., 2007)). This is, however, not an issue in the current study, as gorillas do not occur at Bili-Uéré.
Hypotheses proposed to explain ground nesting in chimpanzees
Ground nesting represents a divergence from what is considered the ‘standard’ and likely ancestral arboreal nesting behaviour in chimpanzees, and a number of hypotheses have been postulated to explain its occurrence (table 1). Here, we briefly review those tested in the current study.
Hypotheses postulated to explain the occurrence of ground nesting behaviour in chimpanzees.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Ecological factors (seasonality, climate, elevation above sea level, and vegetation availability) may play a role in the frequency of ground nesting or nest height. Pruetz et al. (2008), for example, found that chimpanzees built their nests significantly higher during the wet season at Fongoli. At their respective sites in northern DRC and Cameroon, Hicks (2010) and Tagg et al. (2013) reported that chimpanzees made more ground nests during the dry season. Differences in elevation (Matsuzawa and Yamakoshi, 1996), forest type, and forest structure (Tagg et al., 2013) may also influence whether chimpanzees build ground nests.
Human activities such as hunting, agriculture, and logging likely also affect the frequency of ground nesting, but this is complex and may be linked to differences in the relationships between chimpanzees and humans at different research sites (for examples of some of these differing patterns of interaction, see Hicks et al., 2012 and McLennan and Hill, 2010). In addition, van Dijk et al. (2021) observed that at Bulindi, western Uganda, chimpanzees nested at lower heights after a substantial portion of their forest had been cleared for agricultural fields. Tagg et al. (2013) found a higher frequency of ground nesting with increased human encounter rate, whereas Hicks (2010) and Last and Muh (2013) found the opposite. Chimpanzees in western Uganda preferred to build ground nests in swamps, likely in order to avoid detection by humans (McCarthy et al., 2017). At some research sites, ground nesting may be favored by a lack of large carnivores (e.g., Fongoli, Senegal: Pruetz et al. (2008), Seringbara, Côte d’Ivoire: Koops et al. (2012a)) and potentially large dangerous animals like elephants, but this is not always the case (Hicks, 2010). Absence of predators and large herbivores is, however, often a result of human hunting, and it is thus difficult to disentangle the effects of these factors.
The final factor that might explain ground nesting is culture, defined as ‘a behavior that is transmitted repeatedly through social or observational learning to become a population-level characteristic’ (McGrew, 1992). Ground nesting as a socially-transmitted behavior may explain the scattered distribution of the phenomenon in chimpanzees (Hicks, 2010). If we can rule out the potential impact of genetic, morphological, or ecological factors (see table 1) as well as of humans and other fauna on ground nesting, then the ‘method of exclusion’ (Whiten et al., 2001) would make the cultural explanation more likely. This can be a challenge, however, as most primate behavior is shaped by a complex mix of genetic, ecological, and social influences (McGrew, 1992; Whiten et al., 2001).
Ground nesting in the chimpanzees of northern Democratic Republic of the Congo
The frequent construction of complex ground nests by chimpanzees has now been documented at multiple areas in northern DRC (Hicks, 2010), as well as approximately 500 km to the southeast at Epulu (Hart, unpubl. data; Hicks, unpubl. data), and along the border with Uganda in the Relict Altitude Forest Fragment of the Albert Lake Escarpment (RAFALE) (A. Laudisoit, pers. comm., 20 September 2021). They also occur just across the border at Bugoma, Uganda (Hobaiter et al., 2022). Frequent ground nesting thus represents a potential large-scale behavioural variant (Hicks et al., 2019a).
In order to better understand the distribution of chimpanzee ground nesting across the Bili-Uéré landscape in northern DRC, we used data collected by Thurston Hicks (TH) and team over a 9-year period from 20 survey areas to map the proportion of ground nests out of total nests. We then used data collected at 16 of these survey areas to investigate the impact of the factors detailed below on the frequency of ground nesting behaviour across the region, as well as the height at which chimpanzees built their nests (see a map of the surveyed area in supplementary fig. S1, as well as Hicks et al. (2014, 2019a and 2020) for more details).
Habitat-type and seasonality
We investigated whether ecological factors such as seasonality and rainfall had an impact on the frequency of chimpanzee ground nesting and nest height in northern DRC. We also examined whether forest type, structure, and density influenced the likelihood of a nest being a ground nest, as well as nest height. We predicted that ground nests would make up a larger proportion of nests during the dry season, as was found at Dja, Cameroon (Tagg et al., 2013; see also Hicks, 2010). This we considered to be especially likely given that during the dry season in northern DRC, many swamps and streambeds dry up completely, thus leaving more ground available for nesting (Hicks, 2010). Due to the fact that chimpanzee ground nests have been documented in a variety of habitats and altitudes (e.g., swamps (McCarthy et al., 2013), and mountain slopes (Koops et al., 2012)), we had no specific predictions about the impact of forest structure and density on ground nesting and nesting height.
Carnivores and large herbivores
Chimpanzees are known across Africa to fall prey to large carnivores such as leopards (Panthera pardus) (Boesch and Boesch-Achermann, 2000) and lions (Panthera leo) (Tsukahara, 1993). Based on their known danger to humans, hyenas (Crocuta crocuta) represent a potential predator as well. All three carnivores are present in northern DRC (see Materials and methods). Large herbivores such as forest buffaloes (Syncerus caffer nanus) and African forest elephants (Loxodonta cyclotis) are potentially dangerous to chimpanzees, as they are to humans, and these species are present at multiple sites in northern DRC (Hicks, 2014). We hypothesized that increased abundance of these large mammal species would dissuade chimpanzees from nesting on the ground and encourage them to nest higher in the trees. We hence tested whether the encounter rates of large carnivores and herbivores affected the frequency of chimpanzee ground nesting as well as nest height at the 16 survey areas.
Human impact
Human disturbance is known to affect the behaviour of nonhuman great apes (e.g., orangutans (van Schaik, 2002), chimpanzees (Hicks et al., 2013; Kühl et al., 2019; Tédonzong et al., 2020). Hicks (2010) reported a negative impact of human hunting on ground nesting in northern DRC, and this was also observed at Lebialem-Mone, Cameroon (Last and Muh, 2013). We thus predicted that increased levels of human impact and close proximity to roads would reduce the occurrence of ground nesting as well as lead to an increase in nest height.
Materials and methods
The study region
Over a total non-consecutive period of 41 months between 2004 and 2013 (see supplementary table S2 for details), TH and coworkers conducted surveys for chimpanzees and other mid-to-large-sized mammals across an approximately 50 000 km2 region in northern DRC. These surveys took place in 20 areas between the Tele River in the south and the Mbomu River in the north with the towns of Bondo and Bambesa representing the western and eastern limits, respectively (supplementary fig. S1). This area was bisected by a major river, the Uele. Eight of the survey areas were to the north of the Uele River and 12 were to the south. Survey regions were relatively cohesive geographic areas, separated from other sites by roads, rivers, or settlements, and generally sharing the same major habitat type. In three cases (Akuma-Yoko, Nawege-Zaza and multiple locations at Bili South), because satellite maps revealed them to be connected as a unit and relatively homogeneous, we combined geographically separate forest regions to achieve an adequate sample size. Conversely, for analysis we separated two adjoining areas in the Bili-Gangu area because their predominant habitat types differed substantially: Camp Louis is mostly composed of savanna, savanna-woodland and regenerating forest, whereas Gangu is made up of mostly old growth moist tropical forest, drier to the north of the Gangu River and more seasonally flooded to the south. Given the complex farmland-forest mosaic nature of Bambesa, the region was counted as one unit (Hicks et al., 2019a).
We found chimpanzee nests in all of the surveyed regions (Hicks et al., 2014). The chimpanzee population appeared to be continuous across the area. The chimpanzees were unhabituated, and in most of the regions they were fearful of humans, likely due to heavy hunting pressure (Hicks et al., 2010, 2012). The exception was the Gangu Forest, where human pressure was very low (see supplementary table S3), and the chimpanzees behaved in a naïve way towards researchers. In the Camp Louis Forest, we estimated the density of weaned chimpanzees per km2 to be 0.46-1.43, and as high as 1.01-2.95 in the Gangu Forest. The population was stable between 2005 and 2012 (Hicks et al., 2014). Nest encounter rates were comparable across the entire region surveyed, and the chimpanzees shared elements of a single widespread material culture (termite mound and snail pounding, lack of tool-assisted predation on termites of the genus Macrotermes, frequent ground nesting and the use of tools to prey on epigaiec Dorylus and ponerine ants (Hicks et al., 2019a, 2020)).
Hicks (2010) and Hicks et al. (2019a) provide detailed descriptions of the climate across the Bili-Uéré region, using data their team collected at several sites as well as from the historical record. Annual rainfall is similar across the region (1652 mm at Hicks’ weather station at Aketi vs 1644 mm at Tukpwo further north (Gerard, 1960), with a wet season lasting from April to November. The seasonality of rainfall was more extreme to the north (Hicks, 2010): during two dry seasons at Bili, for instance, no rain fell and streams that had been over two meters deep during the wet season dried up completely.
The predominant habitat to the north of the Uele River was a mosaic of savanna woodland and moist tropical forest intersected by large patches of open grassland (Hicks, 2010). The forests and savannas were interlaced with streams; in the grasslands, these were surrounded by gallery forests. The moist tropical forest to the south of the Uele River lacked savannas. Mono-dominant stands of Gilbertiodendron dewevrei were present but rare to the north of the Uele and occurred more extensively along streams to the south (Hicks, 2010). Near settlements and alongside roads across the region, slash-and-burn agriculture had created large areas of farmland and regenerating forest, characterized by oil palms (Elaeius guineensis), introduced species of bamboo such as Bambusa vulgaris and, at a later stage of succession, stands of Musanga cecropioides.
We found leopard traces on both sides of the Uele River (Hicks, 2014). We encountered evidence (encounters, scats, and tracks) of hyenas and (more rarely) lions only to the north of the Uele River (supplementary table S3; Hicks, 2010, 2014). Forest elephants and buffaloes occurred in a patchy distribution across the survey region, with both species being more common to the north than the south (supplementary table S3), where they were absent at a number of sites, likely due to human hunting (Hicks, 2010; Hicks et al., 2019b). As reviewed in Hicks (2010), the human population density to the south of the Uele River was roughly triple that in the north. Artisanal mining for diamonds and gold was concentrated in the south, especially in proximity to the city of Buta. On our surveys, we found much higher encounter rates of human activities to the south than to the north of the Uele River (10.2 times higher for hunting evidence and 6.4 times higher for other human signs). Camp Louis, Gangu, and Leguga had the lowest human encounter rates, whereas Mbange West, Ngume, and Zongia had the highest (supplementary table S3). Gangu was the only region in which we surveyed distances >15 km from the nearest road or settlement, up to 41.8 km. Prior to 2007, Gangu had lower human encounter rates than Camp Louis (0.07 vs. 0.09 signs per km walked (Hicks et al., 2013), and for both of these areas, the majority of human signs were limited to a few old, abandoned mining pits. At Gangu, we found these pits only on the eastern edge of the forest; deeper inside, the forest was nearly pristine.
In 2007, there was a rapid incursion of thousands of gold miners into the formerly isolated Gangu area in the north, leading to a 4-year disruption of research in the area. During this time, elephant densities plummeted in the Gangu Forest, and in 2011, researchers encountered a bushmeat camp in Gangu with a chimpanzee carcass (Hicks, 2014). TH and team also documented such disturbances upon their return to the area in 2012, with recent hunting and fishing camps present in both Camp Louis and Gangu, and abundant signs of elephant poaching (Hicks, 2014). This led to an increased number of human signs observed in Gangu, compared to Camp Louis. Between 2005 and 2012, however, Hicks et al. (2014) found no significant change in chimpanzee nest encounter rates in the Gangu and Camp Louis regions, indicating that no population decline had occurred, and this continued to be the case during surveys in 2016 and 2019 ([Funk et al., 2022). South of the Uele River, Hicks et al. (2010) documented a thriving market for bushmeat, including chimpanzee meat and infants.
Surveys
During our Bili-Uéré surveys, we collected data on nesting behaviour (Hicks et al., 2019a), as well as encounter rates of humans and other mammalian species (Hicks et al., 2013; Hicks, 2014). Our surveys consisted of reconnaissance forest walks, generally following the path of least resistance and actively searching for chimpanzees (recces: 2627.1 km to the north of the Uele River and 476.7 km to the south). In 2005, to collect data on chimpanzee nest encounter rates, we conducted 160 km of line transects through the Camp Louis-Gangu regions (consisting of three parallel lines randomly placed on the map, separated from one another by 4 km), and repeated 26 km of these in 2012. In 2012, we conducted an additional 45 km of line transects in the Bambilo and Dume areas (supplementary fig. S1). In addition to nests, we recorded data (dung, prints, sightings) on carnivores and other mid-to-large herbivores as well as signs of human presence and impact, including agriculture, hunting, and mining (for more details on the methodology of these surveys, we refer the reader to supplementary tables S2 and S3, Hicks (2010, 2014), Hicks et al. (2013, 2019a)). The use of recces may have led to a bias in nest encounter rates, but this is of little concern given that the focus of this paper is on the proportion of ground nests to total nests, not chimpanzee density (see Discussion).
Chimpanzee nest surveys
We recorded GPS points for all chimpanzee nest groups encountered on both recces and line transects. We excluded from our encounter rates any nests to which we had been led by local guides who had encountered them prior to the surveys. We considered nests as belonging to the same group if they were of the same age (age categories adapted from Tutin and Fernandez (1984): fresh, recent, old, and rotten, with the addition of skeleton for old nests with only leafless branches remaining). We refer to nest aggregations of the same age and within 20 m of one another (Tutin and Fernandez, 1984; Furuichi et al., 1997; Blom et al., 2001; Dupain et al., 2004; Kouakou et al., 2009) as ‘nest groups’. This threshold distance has varied between studies: 30 m (Furuichi and Hashimoto, 2000; Humle, 2003; Dupain et al., 2004; Koops et al., 2012b; Tagg and Willie, 2013; Tagg et al., 2013), 50 m (White and Edwards, 2000; Morgan et al., 2006; Hicks et al., 2009), and 100 m (Stewart and Pruetz, 2013; Badji et al., 2018). Hicks and the other observers on his team estimated nest height by eye, following a period of training together. We recorded for each nest whether it was covered by canopy (fully or partially) or open to the sky, in addition to habitat type and whether or not feeding remains, chimpanzee dung, hair, or urine were present. We considered a nest to be a ground nest when its height was ⩽50 cm above the ground (for details on the definition of nests at different research sites, see additional data 1 of the online repository), including when a nest at this height had been built on a root or fallen tree with no open space between it and the ground. We consider 50 cm a reasonable cut-off point, as any nest constructed higher than this would likely have required the chimpanzee to climb up off the ground. In a few cases, although components of old ground nests had over time sprung up >50 cm, when they were reconnected as they had originally been, we could verify that the nests had touched the ground.
We categorized ground nests according to the degree of complexity of their construction as flimsy or complex. Complex ground nests involved the use of multiple components (branches, vines and/or herb samplings) woven together into a complicated structure, often in a circular bowl shape. ‘Flimsy’ ground nests involved minimal arrangement of multiple parts, lacking interweaving of branches and/or herbs, usually flat and with little discernable structure (although often roughly circular in shape). The latter possibly represent ‘day nests’ used for resting or feeding, as described for the Taï Forest by Boesch (1995). We also documented a previously undescribed nest type: half ground nests, a subcategory of complex ground nest constructed in a half-circle shape, often with the built side raised like a chair back (fig. 1). We also documented leaf cushions, which are not ground nests but consist of one or more leaves or branches simply placed on the ground with no arrangement or construction (fig. 1 and Hirata et al., 1998). Given the possibility that leaf cushions or flimsy ground nests may be used for different purposes than complex ground nests, and thus might show different patterns of usage and wear, we recorded data on dung, hair, and feeding remains associated with these artifacts.
(A) Fresh chimpanzee tree nest in the Gangu Forest, North Uele, January 2007 (accompanied by a Cercopithecus ascanius). (B) Local guide Chipula with a freshly-constructed complex ground nest made of herbs in the Zapay Forest, North Uele, December 2006. (C) Old complex ground nest made of herbs in the Mbange East Forest, January 2008. (D) Half ground nest, Gangu Forest, North Uele, September 2012. (E) Flimsy ground nest, Gangu Forest, January 2007. (F) Leaf cushion, Mbange East Forest, March 2008.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Fauna
Following Hicks et al. (2013), in order to assess the effect of the abundance of carnivores and large herbivores on the likelihood of chimpanzees building ground nests, on our forest walks we systematically recorded direct sightings as well as dung, tracks and feeding remains of all mid-to-large-sized mammal species. See supplementary table S3 for the number of kilometers walked per species.
Human activities
In order to compare the impact of human activities on chimpanzees in the different regions, we systematically recorded observations of human evidence along the transects and recces, following the methodology used by Hicks et al. (2013). ‘Hunting evidence’ includes bushmeat carcasses or skins, hunting camps, snares, and cartridges as well as gunshots heard or hunters encountered, while ‘Other human evidence’ refers to cultivated fields, footpaths, mining camps, fishing huts, graffiti, signs of fishing and honey collection, machete cut marks, encounters with unarmed people, and any other direct or indirect signs. Some of these categories, such as mining, were likely associated with hunting as well (Hicks et al., 2010).
Habitat type and forest density
Following Hicks et al. (2013), in order to assess the impact of forest density on ground nesting, we categorized forest into dense (visibility ⩽ 1 m), medium (visibility > 1 m, but ⩽ 5 m), and open (visibility > 5 m). In addition, based on Gerard (1960), Doumenge (1990), and Hicks (2010), we grouped habitat types into the categories listed in tables 2 and 3, and supplementary table S4.
Definitions of habitat types used for nesting by chimpanzees.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Definitions of the additional categories of habitat types which we included as fixed effects in Models 1 and 2.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Statistical analyses
We fitted two statistical models to test the impact of the environmental and faunal-related factors detailed below on the proportion of ground nests out of total nests. We then reran the same models with nest height as the response variable. We did not include leaf cushions in these analyses, since they were likely only used to be sat upon during the day (see online repository additional data 2 for a description of the leaf cushions found at Bili-Uéré). For the two models, we included only 16 of the 20 survey areas, omitting Difongo, Dume, Membulu, and Bambilo. Although the nest encounter rates were accurate for these areas, we lacked data for one or more of the predictors.
Model 1: factors impacting the likelihood of a nest group consisting of ground or tree nests
In our first model, we evaluated the impact of ecological and human impact factors on the likelihood that a nest group consisted of ground nests or tree nests. Originally, we aimed to fit this model on the level of the individual nests. In the majority of nest sites, however, either all or none of the nests were a ground nest. As a consequence, the corresponding model revealed an extreme estimate for the random intercepts effect of nest group, which led to various issues with regard to the estimation of the fixed effects. More precisely, for most of the estimates the model revealed identical and also very small standard errors. As a consequence, almost all p-values were unrealistically small. As it seemed likely to us that this was caused by the extremely large random intercept effect of the nest group, we omitted the 80 groups containing both ground and tree nests together (7% of the nest groups). We also omitted an additional 63 nest groups in which our four categorical predictors (presence of vine tangle, herb patch, light gap, and hill side) and forest density varied between nests at the same nest site. We then averaged the other predictors per nest group and considered as the response whether we found only tree nests or only ground nests at the nest site.
In order to estimate the effects of forest density, habitat type, forest structure (operationalized as ‘yes’ or ‘no’ for each of hillside, herb patch, and light gap; see table 4), distance from roads, encounter rates of human hunting and ‘other activities’ signs, and encounter rates of carnivore, buffalo and elephant signs (the latter two pooled; see below) on the likelihood of a nest group consisting of ground nests, we used a Generalized Linear Mixed Model (GLMM (Baayen, 2008), Model 1a) with binomial error structure and logit link function (McCullagh and Nelder, 1989). We included all of the above-mentioned predictors, with the exception of habitat type, as fixed effects (supplementary table S5). Given that ‘habitat type’ had multiple levels, we included this factor as a random intercept effect to control for its potential impact (for details, see table 2, supplementary table S5, and additional data 3 of the online repository). Given that we expected that predation threat from large carnivores might have a different effect on ground nesting than that of accidental encounters with elephants and buffaloes, we separated large herbivores from carnivores. We pooled elephants and buffaloes as ‘large, potentially dangerous herbivores’ because both can be dangerous to humans, and we reasoned that they would also likely represent a threat to chimpanzees they encountered sleeping on the ground.
Generally, the forests in northern DRC are seasonally-flooded, which means that the chimpanzees likely have more dry ground available for ground nesting during the dry season. This initially led us to consider seasonality as a predictor of the likelihood of a nest group consisting of ground nests. A visual inspection of the distribution of the residuals of the full model against date did not, however, reveal any pattern of temporal autocorrelation. Seasonal variation in the probability of ground nesting, on the other hand, would lead to temporal autocorrelation in these residuals. Thus, we did not include seasonality as a predictor (supplementary fig. S2).
To test the combined impacts of the nine fixed effects and to avoid cryptic multiple testing (Forstmeier and Schielzeth, 2011), we compared the full model with a null model lacking the fixed effects but being otherwise identical using a likelihood ratio test (Dobson, 2002). Prior to fitting the model, we checked the distribution of all of the predictors. As a consequence, we log-transformed ‘distance from road’ and ‘human signs’ in order to achieve a more symmetrical distribution and to avoid leverage issues. An additional reason for log-transforming these two predictors was that they otherwise could have a non-linear effect: as an example, one kilometer further away from the road is likely to have a larger effect at smaller distances from the road as compared to the same measure at larger distances. We then z-transformed forest density, distance from road and human and carnivore signs to a mean of zero and a standard deviation of one to achieve comparable model coefficients and to ease model convergence. We assessed model stability by comparing the estimates obtained from the model based on all of the data with those obtained from models based on subsets of the data with habitat types excluded one at a time (Nieuwenhuis et al., 2012). This revealed the models to be fairly stable (tables 5 and 6).
To rule out collinearity, we determined Variance Inflation Factors (VIF; Field, 2005) using the function vif of the R package car (version 3.0-10; Fox and Weisberg, 2011) and applied a standard linear model excluding the random effects, which revealed maximum VIFs well under 2.6 for all factors except elephant and buffalo signs (2.6), distance from road (2.7) and for other human signs encounter rates (3.1) (supplementary fig. S3). To check whether the correlation between hunting evidence and other human signs could distort inference, we fitted two additional models excluding each of these two predictors one at a time. For hunting evidence, the two models with and without other human signs revealed essentially the same results. For other human signs, however, the model without hunting evidence revealed a coefficient of the opposite sign as compared to the full model, and in this model, the effect of other human signs was also significant whereas it was not in the full model (see supplementary table S6 for details). Since there was little variation in the fixed effect predictors within the levels of the random effects, it was not possible to include random slope into the model. As we had no predictions regarding interactions between the predictors, we did not include interactions in the models. In addition, we fitted a second GLMM (Model 1b), pooling both hunting evidence and other human signs into the more general predictor of human activity and combining large carnivores and herbivores into the general predictor of ‘dangerous animals’ (supplementary table S6).
Model 2: factors impacting nest height
Given that there is an arbitrary dimension to our classification of ground nests as nests ⩽50 cm elevation, and given the substantial amount of variation in the heights of the tree nests (0.51-40 m), we fitted a second model in which we substituted the binomial response of ground nests (yes – no) with the Gaussian response of nest height. The dataset used for this model comprised one data point for each individual nest. We used a Linear Mixed Model (LMM (Baayen, 2008), Model 2a) to evaluate the effects of the same predictors as in Model 1 and included again habitat type, but this time also nest group random intercept effects (supplementary table S5). Degrees of collinearity were almost identical to that of Model 1, and we determined model stability and the significance of the full model as described for Model 1 (for assessing model stability, we also excluded nest sites one at a time). We also fitted a second LMM (Model 2b) to test the same predictors as in Model 2a, but lumped together all human signs and all large carnivore/herbivore signs, respectively, as described for Model 1b.
Model implementation
We fitted all models in R (version 4.1.0 or higher; R Core Team, 2020), using the function lmer of the package lme4 (version 1.1-26;0 Bates et al., 2015). We derived 95% confidence intervals of model estimates and fitted values by means of parametric bootstraps (function bootMer of the package lme4), using 1000 bootstraps and bootstrapping over the random effects. We used likelihood ratio tests (R function drop1; Barr et al., 2013) to derive the significance of the individual fixed effects. As suggested by Nakagawa et al. (2017), we used the function r.squaredGLMM from the package MuMIn (version 1.43.17; Barton, 2020) in order to determine effect sizes. The data analyzed using the two models comprised, respectively, 1021 nest groups and 16 habitat types (Model 1: supplementary table S5) and 2498 nests in 1091 nest groups and 20 habitat types (Model 2: supplementary table S5). The number of data-points per estimated term was 92.8 (Model 1a) and 178.4 (Model 2a).
Results
Ground and tree nesting across the region
Ratio of ground nests to tree nests
Across the region, we found 2991 chimpanzee nests in 1277 nest groups. Of these, 311 (10.4%) were ground nests in a total of 179 nest groups (14% of groups) (range: 0-29% of all nests; and 0-60% of all nest groups) (supplementary tables S2 and S7). Although ground nests occurred in the majority of the survey areas (15 of the 20), both to the north and south of the Uele River, we did not find them at the South Uele sites of Buta, Ngume, Lingo, and Mbange West, nor at Nawege-Zaza to the north of the Uele (table 4, fig. 2, supplementary tables S2 and S7). In 10 of the 20 survey regions, we found 5% or more ground nests, and in six we found 10% or more.
Number of ground nests encountered (by complexity) and number of leaf cushions per survey area. Ground nests include complex, flimsy and half ground nests, the latter being a subset of the complex ones.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Proportions of ground nests (in black) out of total nests per survey area. Circle area represents the number of nests found per survey area (range: 23 to 934). See supplementary table S7 for more details.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Characteristics of nest groups across the region
Overall, we found an average of 2.3 nests per nest group (median = 1; Q1 = 1; Q3 = 3; range: 1-26). Limiting ourselves to only the 179 nest groups with ground nests present, there was an average of 4.1 nests per group (median = 2; Q1 = 1; Q3 = 4.8; range: 1-26), averaging 2.4 tree nests (median = 0; Q1 = 0; Q3 = 3; range: 0-24) and 1.7 ground nests per group (median = 1; Q1 = 1; Q3 = 2; range: 1-13). In 52% (93/179) of the nest groups containing ground nests we found no associated tree nests, and in 36% (64/179) of the nest groups with ground nests present there were two or more ground nests; 5% (9/179) had 5 or more (supplementary table S8). We found two nest groups at Bambesa with large numbers of ground nests of the same age: one had a mix of 9 ground nests and 9 tree nests and another consisted entirely of 12 ground nests. One nest group in the Gangu Forest contained 13 ground and 11 tree nests of the same age. The average height of nests across the region was 9.0 m (median = 8.0; Q1 = 5.0; Q3 = 12.0; range: 0-40; n = 2649); excluding ground nests, it was 10.2 m (median = 8.0; Q1 = 6.4; Q3 = 13.0; range: 0.6-40; n = 2334) (supplementary table S9).
The majority of ground nests found both to the north (94% of 236 ground nests) and south (96% of 77 ground nests) of the Uele River were complex ground nests (table 4). To the north of the Uele River, 3% (7/236) of ground nests were half ground nests, vs. 22% (17/77) south of the Uele; all of these were of complex construction. Gangu, Bambesa, Mbange East and Lebo had the highest percentage of half ground nests; these were also found at Leguga, and Camp Louis, but not at the other survey areas. Eight of the 24 half ground nests (35%) were found on hillsides, compared to only 8% (25/311) of other ground nests and 7% (185/2639) of tree nests. In four cases, we found ground nests built on logs that were touching the ground. In 23 cases, the trees used to make ground nests had sprung up into the air with the passage of time, which made it difficult to estimate precisely their original height, although it was below 0.5 m.
We found that 19% (27/141) of fresh and recent complex ground nests were associated with hairs only, 6% (8/141) with dung only, 13% (18/141) with dung and hairs, and 10% (14/141) with feeding remains. No fresh and recent flimsy ground nests were associated with hairs, 17% (1/6) were associated with dung, and 50% (3/6) with feeding remains (supplementary table S10).
As can be seen in supplementary table S11, ground nests made up 13% (220/1712) of the nests which had been made recently enough to retain their original form (age categories 1-3), whereas for older, decayed nests (age categories 4 and 5), only 7% (87/1222) were terrestrial, a significant difference (Mann-Whitney U-test: W = 375 233, p = 0.008). This may indicate that ground nests at Bili-Uéré decay more quickly than tree nests, as has been documented at Bugoma, Uganda by Romani et al. (in press).
Concerning habitat types, to the north of the Uele River we found the majority of tree nests (52% (972/1857)) in mixed forest. To the south but not to the north of the Uele, the majority of tree nests were in Gilbertiodendron monodominant forest (51% (396/770) in the south vs. 6% in the north (115/1857)) (supplementary table S4), and only 38% (293/770) of tree nests were in mixed forest in the south. Of the 292 ground nests for which we recorded data on habitat type, the majority were also in mixed forest (71% (122/173) north vs. 52% (62/119) south). We also found ground nests in riparian forest (27% (46/173) north vs. 1% (1/119) south), swamp forest (28% (33/119), in the south only), and Gilbertiodendron dewevrei monodominant forest (1% (2/173) to the north vs. 19% (23/119) to the south). We found 29% (82/286) of the ground nests in dense forest and 53% (151/286) in medium dense forest, compared to 19% (451/2400), and 63% (1510/2400), respectively, for tree nests (see supplementary table S4 and additional data 4 of the online repository).
Leaf cushions
We found leaf cushions in 8 of the 20 survey areas (table 4); 50% (11/22) of these were associated with chimpanzee nests, 9% (2/22) with tree nests and 41% (9/22) with ground nests. Considering only the fresh leaf cushions, 44% (7/16) were associated with feeding remains, 31% (5/16) with hairs, and 31% (5/16) with dung (supplementary table S10).
Factors potentially impacting ground nesting and nest height in the chimpanzees of northern Democratic Republic of the Congo
Ground vs. tree nest groups (Model 1)
In Model 1a, we evaluated the impacts of forest structure, habitat type, human presence, distance from road, and dangerous animals on the probability of a nest group consisting of ground nests (table 5). Overall, the model showed that several fixed effects had a clear impact on the likelihood of a nest group consisting of ground nests (likelihood ratio test comparing full and null model:
Effects of ecological factors, human activities, carnivores and large herbivores on the probability of a nest group being composed only of ground nests (Model 1a).
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Effect of signs of hunting evidence per km (A), and other human signs per km (B) on the proportion of ground nest groups/total nest groups. The area of the dots represents the number of nests found at each area (min = 5; max = 735). The dashed lines depict the fitted model (Model 1a) and the dotted lines its upper and lower confidence limits for all other predictors centered to a mean of zero. GN, ground nest.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Habitat-type
Ground nest groups were significantly associated with herb patches and light gaps (table 5, supplementary fig. S4A, B), as well as with dense forest (supplementary fig. S5). The model revealed little variation in the random effect of habitat type in relation to the fixed effects (estimated standard deviation in link space: 0; see supplementary table S12 for more information).
Impact of large fauna
We found no significant impact of encounter rates of carnivores or large herbivores on the likelihood of a nest group consisting of ground nests, whether combined or divided into carnivores and herbivores (table 5, and see additional data 5 of the online repository), although there was a trend for fewer ground nest groups to be found in association with carnivores. We provide more information on the distribution of these carnivores, elephants, and buffaloes across the region in supplementary table S3.
Human impact on ground nesting
The model revealed a significant negative impact of human hunting on the probability of a nest group consisting of ground nests (table 5, and fig. 3A). In three of the areas most impacted by hunting (Mbange West, Buta, and Ngume; supplementary table S3), we found no ground nests at all (supplementary table S2). At Mbange East, which had a high impact of human hunting, only one of the 47 nests was terrestrial. Also, Gangu and Leguga, two of the areas with the lowest levels of human impact, hunting or otherwise, had some of the highest proportions of ground nests out of total nests (13% (119/934) and 29% (33/115), respectively; see supplementary table S2 for details). Encounter rates of non-hunting human signs, however, had no impact on the proportion of ground nest sites (table 5, and fig. 3B).
Factors impacting nest height (Model 2)
In addition, we modelled nest height using the same predictors included in Model 1 (table 6). We present the results for each group of predictors separately: habitat type, large fauna and human signs. Overall, the combined impact of the 10 fixed effects had a clear impact on nest height (full-null model comparison:
Habitat type
The model showed considerable variation in the random effect of habitat type in relation to the fixed effects (estimated standard deviation = 4.02; supplementary table S12). Furthermore, we found that light gaps, vine tangles, and herb patches had a negative effect on the height chosen by the chimpanzees to build their nests (table 6; see also additional data 6 of the online repository).
Impacts of large fauna and humans
We found a significant positive impact of large herbivores, but not carnivores, on nest height (table 6, and supplementary fig. S6). This impact of herbivores was, however, very small. Nest height significantly decreased with increasing distance from roads (table 6, fig. 4, and supplementary table S13), and significantly increased with higher encounter rates of hunting evidence, but not with other human signs (table 6, and supplementary table S3; see also additional data 7 of the online repository).
Results of the LMM model of the effects of environmental factors, humans and animals on nest height (Model 2).
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Nest height as a function of distance from road (Model 2a; significant: p-value < 0.001). Note that the model depicted lacks the random effect of nest group as its inclusion led to a very small intercept. We binned the predictor into 49 bins and the circle area is proportionate to the number of nests per bin (range: 7-320). The dashed line depicts the fitted model and the dotted lines its confidence intervals for all other predictors centered to a mean of zero.
Citation: Folia Primatologica 94, 4-6 (2023) ; 10.1163/14219980-bja10010
Discussion
We found ground nesting to be an important component of chimpanzee behaviour across the Bili-Uéré landscape: in 75% (15/20) of the surveyed areas, 1% or more of the nests were ground nests, and overall 14% (179/1277) of nest groups contained ground nests. We categorized 95% (296/313) of these ground nests as being of complex construction, and thus likely night nests (for more evidence that some of these nests were slept in at night, see supplementary text S1). At 10 of the 20 survey regions, 5% or more of the nests were ground nests, and at 6 regions ground nests made up 10% or more of total nests. These results indicate that ground nesting in northern DRC is a widespread and consistent behaviour spanning two sides of a major river, and is common in both the savanna woodland mosaic habitat to the north and the moist tropical forest habitat to the south. This is to be contrasted with the distribution of ground nesting in other chimpanzee populations, which is much more sporadic (supplementary table S1). Bili-Uéré appears to be the largest continuous area which has been surveyed over which chimpanzee ground nesting occurs (10.4% (311/2991) of all nests, with a fairly evenly distribution across the > 50 000 km2 area surveyed; fig. 2, table 4, supplementary table S2 and S7). This behaviour is found as well in the Epulu and RAFALE regions of DRC further east (Hart, unpubl. data; Hicks, unpubl. data; Laudisoit, unpubl, data), and ground nests are common at some localities in Uganda along the border with DRC (Maughan and Stanford, 2001; McCarthy et al., 2017; Hobaiter et al., 2022). This supports the inclusion of ground nesting into the widespread suite of behaviours described by Hicks et al. (2019a) as the Bili-Uéré Behavioural Realm. Given that forest density increased the likelihood that a nest group would contain one or more ground nests (table 5), and that on our recce walks we generally followed the path of least resistance (thus sometimes avoiding denser vegetation), it is possible that we undercounted ground nests across the region.
The relatively uncommon flimsy ground nests may have been used, like leaf cushions, to rest in or feed in during the day (Koops et al., 2007). In fact, when looking at fresh and recent nests, a substantial proportion of the flimsy ground nests and leaf cushions (50% (3/6) and 44% (7/16), respectively) were associated with feeding remains, whereas only 10% (14/141) of complex nests were. More work remains to be done investigating the styles of ground nest construction in these different regions, the plant types used, and the sex of the nesters, which will help us to determine whether this represents one continuously-distributed behavior or rather multiple independent evolutions of ground nesting, based on happenstance, similar ecology or genes. In addition, more nesting surveys should be conducted in the vast forests between these areas to ‘connect the dots’, if indeed they can be connected. It should be remembered, though, that surveys carried out conveniently close to roads or in areas with high levels of human hunting may only uncover subpopulations of chimpanzees that have recently lost ground nesting due to human threat.
Regarding the potential social role of terrestrial night nesting (e.g., mate guarding) we observed that more than half of the nest groups containing ground nests lacked any associated tree nests. In addition, 36% (65/179) of the ground nest sites had two or more ground nests, and 2% (3/179) of nest groups containing ground nests had 9 or more ground nests, including a group of 12 ground nests found with no associated tree nests (supplementary table S8). This makes it unlikely that mate guarding alone can explain the prominence of this behavior in northern DRC.
Models
We used statistical models to investigate the potential impact of several factors on the ratio of ground nests to tree nests and also nest height. These can be grouped into habitat type, dangerous fauna and human activities.
Habitat type
Our results revealed that in northern DRC, chimpanzees nested more often on the ground in mixed forest (63% (184/292) of ground nests vs 48% (1265/2627) of tree nests), but they also used riparian forests (16% (47/292) of ground nests vs 20% (517/2627) of tree nests) and swamps (11% (33/292) of ground nests vs 10% (266/2627) of tree nests) (supplementary table S4). South of the Uele River, but not to the north, chimpanzees often built both ground nests (19% of the total) and tree nests (51% of the total) in Gilbertiodendron dewevrei monodominant forest: given that this forest type is much more common to the south of the Uele (Hicks, 2010), this is perhaps not surprising. We cannot rule out that differences in vegetation availability, as we could expect to find in remote forests vs. more anthropogenic ones, might impact ground nesting. Although at Bili-Uéré trees and herbs used to construct tree and ground nests are common close to the road as well as in the deep forest, future studies should attempt to assess the impact of vegetation type on terrestrial nesting. Considering the major differences in the predominate habitat types on the two sides of the Uele River (forest-savanna mosaic to the north vs. semi-deciduous moist tropical forest to the south), it is unlikely that any particular environmental factor is sufficient to explain the unusually high frequency and wide distribution of ground nesting behaviour in northern DRC. Nevertheless, dense forests as well as light gaps and herb patches were strongly associated with ground nests across the region. Given that light gaps and herb patches were often associated with abundant herb presence, and that 44% (137/311) of ground nests used herbs in their construction (Hicks, unpublished data), perhaps the pattern is driven by availability of resources. Future studies should take into account more detailed information on the availability of preferred nest construction materials between survey areas, which may influence the frequency of ground nests.
In addition, several factors might affect nest decay rates (e.g., vegetation-type and/or nest complexity) (see Bessone et al. (2020) for such an analysis of bonobo nests), and these have not been investigated at Bili-Uéré. Morgan et al. (2016) analyzed the factors affecting the decay of arboreal and terrestrial nests of western lowland gorillas, finding that decay was influenced by forest structure and habitat type, as well as number of components. To address the lack of information on nest decay in chimpanzees, we have recently carried out a study on chimpanzee ground nest decay in the Bugoma Forest, Uganda (Romani et al., in press).
Dangerous animals
We predicted that the threat of large dangerous animals such as carnivores, buffaloes, and elephants would dissuade chimpanzees from sleeping on the ground, as suggested by Pruetz et al. (2008), Hicks (2010), Stewart and Pruetz (2013) and Badji et al. (2018). The literature on this relationship is contradictory: Tagg et al. (2013) found no influence of abundance of potential predators and large herbivores on ground nesting, whereas Pruetz et al. (2008), and Koops et al. (2012a) documented abundant ground nests in forests where large carnivores were rare or absent. In northern DRC, especially in the remote Gangu Forest, we often encountered ground nests in the same swamps in which we found the dung and footprints of large carnivores and herbivores. Comparing the different survey areas, we found that encounter rates of large herbivores had no impact on the frequency of ground nesting, but that there was a trend for carnivore encounter rates to be associated with a greater likelihood of a nest site being composed of ground nests (this trend disappeared when carnivores and large herbivores were combined) (table 5). In addition to this being only a trend, it was a very small effect. When considering the impact of these factors on nest height (Model 2), only large herbivores had a significant positive effect, although this small difference in heights was likely to have been of minimal biological importance: when the encounter rate of elephant and buffalo signs increased by one standard deviation of its log transformed predictor, nest height increased by only 0.62 m (SE = 0.3, table 6, and supplementary fig. S6). Ground nests were particularly common at Gangu (13% of total nests), which also had the highest encounter rates of these potentially dangerous animals. Other localities such as Buta, Ngume, Mbange West, Lebo, and Lingo had very few or no large mammals and also few or no chimpanzee ground nests. Given that large terrestrial mammals were also much less common near roads and in association with human signs, probably due to hunting pressure, this factor is likely linked to human activities. In addition, elephants, unlike the other large mammals under study, are capable of migrating en masse across the landscape, disappearing from one area and appearing in another over a short period of time (Hicks, pers. observ.). We observed this pattern in the forests of Camp Louis and inferred it in Mbange East (Hicks, 2010, 2014). To speculate, perhaps this unpredictability of the movement of elephants (and to a lesser degree buffaloes) might have led to the observed minor effect on chimpanzee nesting habits.
Human impact
Human disturbance is known to have an impact on the behavior of non-human great apes (orangutans: van Schaik (2002), chimpanzees: Hicks et al. (2013), Kühl et al. (2019)), often leading to impoverishment or loss of their traditions, but sometimes producing innovations (McLennan et al., 2019). Across Africa, the impact of human activities on ground nesting can be complex. At La Belgique, Cameroon, ground nests were positively associated with human evidence (Tagg et al., 2013), whereas the opposite was the case at Lebialem-Mone (Last and Muh, 2013). In northern DRC, we found, in accordance with Hicks (2010), but using an expanded data set, a significant negative effect of hunting signs, but not of non-hunting human signs or distance from roads, on the likelihood of a nest group consisting of ground nests (table 5, and fig. 3); similarly, for nest height we found a significant negative impact of hunting (table 6) and a positive one of distance from roads (table 6, and fig. 4). When the log-transformed distance from road increased by one standard deviation of its log-transformed predictor, nest height decreased by 1.2 m (SE = 0.3, table 6). In addition, when the log-transformed number of hunting signs increased by one standard deviation of its log-transformed predictor, nest height increased by 0.63 m (SE = 0.3, table 6); the latter is small and likely of minimal biological importance. Taken together with the absence or rarity of ground nests in the areas most impacted by human hunting (Buta, Ngume, Akuma-Yoko, and Mbange East and West) and their abundance in two of the least human-impacted areas (Gangu and Leguga), these results provide modest support for the hypothesis that the northern DRC chimpanzees build their nests slightly lower with a greater distance from human hunting activity, and appear to cease building ground nests altogether in areas where they experience extreme danger from hunting.
Ground nesting could also be a result of high rates of injury from snares or hunting in a particular forest region (chimpanzees missing hands or feet may have trouble climbing into the trees) or a larger number of old or sick individuals. This is also unlikely, however, given that ground nests were found at 15 out of our 20 survey areas, in multiple nest groups, and that a number of nest groups contained multiple ground nests. TH observed large groups of healthy-looking chimpanzees across the Bili-Uéré region, often actively climbing through or displaying in the treetops, and saw none that were afflicted with missing hands or feet or other obvious snare injuries (see supplementary video S1 for links to clips of uninjured Bili-Uéré chimpanzees). In addition, as already mentioned, snares and other hunting signs were rare or absent at the sites with the largest percentages of ground nests, such as Gangu and Leguga. The lack of human impact on chimpanzees in Gangu is actually more pronounced than our analysis revealed, because we found the overwhelming majority of human signs at the far eastern edge of what was otherwise a mostly pristine forest.
Potential social factors
At Nimba, Guinea, and Fongoli, Senegal, ground nesting has been proposed to be a sex-linked behaviour (Koops et al., 2007; but see Koops et al., 2012b). We have not included any genetic analysis in this paper. Nevertheless, given that we found more than half of the ground nests in nest groups lacked tree nests, and that we found multiple nest groups with large numbers of ground nests, we consider it unlikely that the behaviour was limited to one or the other sex, and thus it cannot be fully explained by mate-guarding. In the future, however, this should be tested with genetic analyses on dung and hair samples gathered from the nests.
Conclusions
The implications of our results are that neither the large body size of gorillas nor the taming of fire (Wrangham, 2009) are necessary for medium-sized hominids such as chimpanzees to regularly construct night nests on the ground, even in areas with high densities of carnivores. Chimpanzees have been documented to attack leopards (Hiraiwa-Hasegawa et al., 1986; Boesch, 2009), pirate prey from them (Nakamura et al., 2019; and possibly Hicks et al., 2020b), and consume a leopard carcass (Hicks et al., 2020b). They can, however, also fall prey to these carnivores (Boesch, 1991). In addition, chimpanzees at Loango, Gabon have been observed to attack gorillas (Southern et al., 2021). Thus, it is possible that in northern DRC some as-yet-unknown behavioural anti-predator adaptation may contribute to the apes’ increased comfort in sleeping on the ground, and it is notable that they frequently sleep together in multiple ground nests. The only dangerous predator that seems to limit chimpanzee ground nesting is their close cousin, Homo sapiens: in areas where human hunting was heaviest, ground nesting disappeared completely.
The Bili-Uéré region of northern DRC, with its complex mix of ecotones, provides us with a perfect natural laboratory in which to study the impact of factors that, millions of years ago, may have also pushed our own ancestors towards the gradual transition to a more terrestrial life. Hominins of the genera Ardipithecus, Australopithecus, and the earliest Homo were roughly similar in body size to modern chimpanzees and may have experienced some of the same selection pressures (McHenry, 1992), including niche competition with fellow hominids (Haile-Selassie et al., 2016) as occurs between chimpanzees and gorillas in many areas today (Oelze et al., 2014; Deblauwe, 2009; Southern et al., 2021). They likely faced more abundant large terrestrial predators than chimpanzees do today (Turner, 1990), but they did not have to reckon with the most dangerous hunter of all, Homo sapiens. These results add significantly to our knowledge of the behavioural diversity in the sleeping habits of one of our closest cousins. Given the enormous geographical area over which chimpanzee ground nesting occurs in northern DRC, and its distribution across a diverse range of habitats, P. troglodytes now joins G. gorilla and H. sapiens as a great ape species that, in at least some populations, frequently nests on the ground. Nonhuman great ape cultures are fragile and easily diminished by human encroachment (Kühl et al., 2019), and we now know that ground nesting disappears quickly in areas with increased human hunting pressure. To safeguard this fascinating aspect of chimpanzee behavioural diversity, it is essential that we provide protection for the Bili-Uéré chimpanzees to ensure that ground nesting does not disappear.
Corresponding author; e-mail: t.romani@uw.edu.pl
Acknowledgements
We thank the Ministère de l’Environnement of the DRC and the Institut Congolais pour la Conservation de la Nature (ICCN) for granting us permission to work in the DRC. We thank Chief Zelesi Yakisi for inviting us to set up a field site in his collectivity. Laura Darby-Singh and Adam Darby-Singh assisted with the South Uele surveys, and Karsten Dierks and Jeroen Swinkels assisted in the north. Field assistant Chipula verbally agreed that his photograph could be used in fig. 1. We thank the following supporters of our project: The Wasmoeth Wildlife Foundation, Lucie Burgers Foundation, International Primate Protection League, The African Wildlife Foundation, Le Centre de Rehabilitation des Primates de Lwiro, US Fish and Wildlife Service, the Panafrican Programme of the Max Planck Institute for Evolutionary Anthropology, Karl Ammann, Hans Wasmoeth, Jan Sevink, and Jan van Hooff. We also thank Terese and John Hart, Christophe Boesch, Hjalmar Kühl, Claudia Nebel, Andreas Walther, and Rainer Benz. At The Faculty of ‘Artes Liberales’, we thank Jerzy Axer, Robert Sucharski, Dorota Łagodzka, Joanna Pijanowska, Joanna Romanowicz, Ewa Jakubiec, Robert Przybysz, Tatyana Lewińska, Ilona Szewczyk, Jan Miernowski, and Monika Stobiecka. We are grateful to the hard work and knowledge of our field assistants, in particular Team Leaders Ephrem Mpaka, Henri Silegowa, Gilbert Pakulu, Bebe Bofenda, Ligada Faustin, Chief Mbolibie Cyprian, and Kisangola Polycarpe, as well as Olivier Esokeli, Seba Koya, Dido Makeima, Makassi Constant, Likambo, Likongo, Garavura Roger, and Kongonyesi. Anne Laudisoit, Alejandra Pascual-Garrido, Anna Nekaris, Vittoria Estienne, Lydia Luncz, Jan van Hoff and Carel van Schaik provided useful ideas in discussions which helped us interpret the data. Last but not least, we are thankful to the four anonymous reviewers for providing a number of invaluable remarks and comments.
Statement of ethics
Our research on chimpanzee nesting behaviour was non-invasive and based on artifacts left behind by the apes. We conducted all of our research in accordance with national and international laws regulating the protection of endangered species. All fieldwork took place with the permission from L’Institut Congolais pour la Conservation de la Nature (ICCN), the Ministère de l’Environnement of DRC and with the consent of local landowners and community members.
Conflicts of interest
The authors have no conflicts of interest to declare.
Funding sources
This work has been developed within the project “A Survey of Chimpanzee Material Culture in Eastern DR Congo” financed by the National Science Center of Poland on the basis of the Decision no. DEC-2017/25/B/NZ8/02730. Additional funding sources: The Institute for Biodiversity and Ecosystem Dynamics of the University of Amsterdam, Karl Ammann, The Wasmoeth Wildlife Foundation, The Lucie Burgers Foundation, African Wildlife Foundation, The US Fish and Wildlife Service, and The PanAfrican Program of the Max Planck Institute for Evolutionary Anthropology.
Author contributions
In addition to the tasks listed below, TR organized the data and co-wrote the manuscript. TH conceived the research, led the Bili-Uéré survey missions, organized the data and co-wrote the manuscript. Statistical analyses, maps and figures: RM and TR, with input from TH. Research protocol design: TH, RM and SM. Comments on the manuscript: ST, PR, AM, RM and MK.
Data availability
All data used in the analyses, including the basic script for the two models, are available at the following link: Romani T, Hicks TC (2023), Data Bili-Uere chimpanzee ground nesting, Mendeley Data, V1, doi: 10.17632/ssdftsv88y.2 (https://data.mendeley.com/datasets/ssdftsv88y/2).
Supplementary material
Supplementary material is available online at: https://doi.org/10.6084/m9.figshare.22093430
References
Almécija S, Hammond AS, Thompson NE, Pugh KD, Moyà-Solà S, Alba DM (2021). Fossil apes and human evolution. Science 372(6542). https://doi.org/10.1126/science.abb4363.
Alp R (1997). “Stepping-sticks” and “seat-sticks”: new types of tools used by wild chimpanzees (Pan troglodytes) in Sierra Leone. American Journal of Primatology 41(1): 45–52.
Ashbury AM, Posa MRC, Dunkel LP, Spillmann B, Atmoko SSU, van Schaik CP, van Noordwijk MA (2015). Why do orangutans leave the trees? Terrestrial behavior among wild Bornean orangutans (Pongo pygmaeus wurmbii) at Tuanan, Central Kalimantan. American Journal of Primatology 77(11): 1216–1229. https://doi.org/10.1002/ajp.22460.
Baayen RH (2008). Analyzing Linguistic Data. Cambridge, Cambridge University Press. https://doi.org/10.1017/CBO9780511801686.
Badji L, Ndiaye PI, Lindshield SM, Ba CT, Pruetz JD (2018). Savanna chimpanzee (Pan troglodytes verus) nesting ecology at Bagnomba (Kedougou, Senegal). Primates 59: 235–241. https://doi.org/10.1007/s10329-017-0647-2.
Barr DJ, Levy R, Scheepers C, Tily HJ (2013). Random effects structure for confirmatory hypothesis testing: keep it maximal. Journal of Memory and Language 68(3): 255–278. https://doi.org/10.1016/j.jml.2012.11.001.
Barton K (2020). MuMIn: Multi-Model Inference. R package version 1.43.17.
Bates B, Maechler M, Bolker B, Walker S (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67: 1–48. https://doi.org/10.18637/jss.v067.i01.
Berge C (1994). How did the australopithecines walk? A biomechanical study of the hip and thigh of Australopithecus afarensis. Journal of Human Evolution 26: 259–273.
Berger LR, McGraw WS (2007). Further evidence for eagle predation of, and feeding damage on, the Taung child. South African Journal of Science 103(11–12): 496–498.
Berger LR, de Ruiter DJ, Churchill SE, Schmid P, Carlson KJ, Dirks PHGM, Kibii JM (2010). Australopithecus sediba: a new species of Homo-like Australopith from South Africa. Science 328(5975): 195–204. https://doi.org/10.1126/science.1184944.
Bessone M, Booto L, Santos AR, Kühl HS, Fruth B (2021). No time to rest: how the effects of climate change on nest decay threaten the conservation of apes in the wild. PLoS One 16(6): e0252527.
Bleicher SS (2017). The landscape of fear conceptual framework: definition and review of current applications and misuses. PeerJ 5: e3772. https://doi.org/10.7717/peerj.3772.
Blom A, Almasi A, Heitkonig IMA, Kpanou JB, Prins HHT (2001). A survey of the apes in the Dzanga-Ndoki National Park, Central African Republic: a comparison between the census and survey methods of estimating the gorilla (Gorilla gorilla gorilla) and chimpanzee (Pan troglodytes) nest group density. African Journal of Ecology 39(1): 98–105.
Boesch C (1991). Teaching among wild chimpanzees. Animal Behaviour 41: 530–532.
Boesch C (1995). Innovation in wild chimpanzees (Pan troglodytes). International Journal of Primatology 16(1): 1–16. https://doi.org/10.1007/BF02700150.
Boesch C, Boesch-Achermann H (2000). The Chimpanzees of the Taï Forest Behavioural Ecology and Evolution. Oxford, Oxford University Press.
Boesch C (2009). The Real Chimpanzee: Sex Strategies in the Forest. Cambridge, Cambridge University Press. https://doi.org/10.1017/CBO9780511627132.
Brugiere D, Sakom D (2001). Population density and nesting behaviour of lowland gorillas (Gorilla gorilla gorilla) in the Ngotto forest, Central African Republic. Journal of Zoology 255(2): 251–259. https://doi.org/10.1017/S0952836901001315.
Deblauwe I (2009). Temporal variation in insect-eating by chimpanzees and gorillas in southeast Cameroon: extension of niche differentiation. International Journal of Primatology 30(2): 229–252.
Dobson AJ (2002). An Introduction to Generalized Linear Models. Boca Raton, Chapman & Hall/CRC.
Doumenge C (1990). La Conservation des Ecosystèmes Forestieres du Zaïre. UICN, L’Alliance Mondiale pour la Nature, La Commission des Communautés Européenes. 242 pp.
Dupain J, Guislain P, Nguengang GM, De Vleeschouwer K, Van Elsacker L (2004). High chimpanzee and gorilla densities in a non-protected area of the northern periphery of the Dja faunal reserve, Cameroon. Oryx 38: 1–8.
Earl S, Hopper LM, Ross SR (2020). Same space, different species: the influence of exhibit design on the expression of zoo housed apes’ species-typical retiring behaviours. Animals 10(5): 836. https://doi.org/10.3390/ani10050836.
Field A (2005). Discovering Statistics Using SPSS. London, Sage.
Finlayson C, Carrión J, Brown K, Finlayson G, Sánchez-Marco A, Fa D, Rodríguez-Vidal J, Fernández S, Fierro E, Bernal-Gómez M, Giles-Pacheco F (2011). The Homo habitat niche: using the avian fossil record to depict ecological characteristics of Palaeolithic Eurasian hominins. Quaternary Science Reviews 30(11–12): 1525–1532.
Fleury-Brugiere MC (2002). Les chimpanzee de la zone integralement Protegee Mafou, Parc de Haut Niger, Republique de Guinee. Rapport Parc National du Haut Niger, Programme AGIR.
Forstmeier W, Schielzeth H (2011). Cryptic multiple hypotheses testing in linear models: overestimated effect sizes and the winner’s curse. Behavioural Ecology Sociobiology 65: 47–55. https://doi.org/10.1007/s00265-010-1038-5.
Fox J, Weisberg S (2011). An R Companion to Applied Regression, 2nd ed. Thousand Oaks, CA, Sage.
Fruth B, Hohmann G (1996). Nest building behaviour in the great apes: the great leap forward? In Great Ape Societies (McGrew WC, Marchant LF, Nishida T, eds.), pp. 225–240. Cambridge, Cambridge University Press.
Fruth B, Tagg N, Stewart F (2018). Sleep and nesting behavior in primates: a review. American Journal of Physical Anthropology 166(3): 499–509. https://doi.org/10.1002/ajpa.23373.
Funk SM, Nkono J, Lushimba A, Fa JE, Williams D (2022). Determining baselines and trends of Eastern chimpanzees and forest elephants in a Central African protected area after civil strife. Journal for Nature Conservation 70: 126296. https://doi.org/10.1016/j.jnc.2022.126296.
Furuichi T, Inagaki H, Angoue-Ovono S (1997). Population density of chimpanzees and gorillas in the Petit Loango Reserve, Gabon: employing a new method to distinguish between nests of the two species. International Journal of Primatology 18: 1029–1046. https://doi.org/10.1023/A:1026356432486.
Furuichi T, Hashimoto C (2000). Ground beds of chimpanzees in the Kalinzu Forest, Uganda. Pan Africa News 7(2): 26–28.
Gerard P (1960). Étude écologique de la forêt dense à Gilbertiodendron dewevrei dans la région de l’Uele. Série Scientifique Number 87. Brussels, Publications de L’Institut National pour l’Étude Agronomique du Congo Belge (I.N.É.A.C.).
Haile-Selassie Y, Melillo SM, Su DF (2016). The Pliocene hominin diversity conundrum: do more fossils mean less clarity? Proceedings of the National Academy of Sciences 113(23): 6364–6371.
Hatala KG, Roach NT, Ostrofsky KR, Wunderlich RE, Dingwall HL, Villmoare BA, Green DJ, Harris JWK, Braun DR, Richmond BG (2016). Footprints reveal direct evidence of group behavior and locomotion in Homo erectus. Scientific Reports 6(1): 1–9.
Hicks TC (2010). A chimpanzee mega-culture? Exploring behavioural continuity in Pan troglodytes schweinfurthii across northern DR Congo. PhD dissertation, University of Amsterdam University.
Hicks TC, Darby L, Hart J, Swinkels J, January N, Menken SBJ (2010). Trade in orphans and bushmeat threatens one of the Democratic Republic of the Congo’s most important populations of Eastern Chimpanzees (Pan troglodytes schweinfurthii). African Primates 7: 1–18.
Hicks TC, Roessingh P, Menken SBJ (2012). Reactions of Bili-Uele chimpanzees to humans in relation to their distance from roads and villages. American Journal of Primatology 74: 721–733. https://doi.org/10.1002/ajp.22023.
Hicks TC, Roessingh P, Menken SBJ (2013). Impact of humans on long-distance communication behaviour of Eastern chimpanzees (Pan troglodytes schweinfurthii) in the northern Democratic Republic of the Congo. Folia Primatologica 84: 135–156. https://doi.org/10.1159/000350650.
Hicks TC (2014). Faunal diversity and human impact in two protected areas of northern DR Congo: Bili-Uéré and Rubi-Télé. Lukuru Foundation, USFWS Final Report 169.
Hicks TC, Tranquilli S, Kühl H, Campbell G, Swinkels J, Darby L, et al. (2014). Absence of evidence is not evidence of absence: discovery of a large, continuous population of Pan troglodytes schweinfurthii in the Central Uele region of northern DRC. Biological Conservation 171: 107–113. https://doi.org/10.1016/j.biocon.2014.01.002.
Hicks TC, Kühl HS, Boesch C, Dieguez P, Ayimisin AE, Fernandez RM, Roessingh P (2019a). Bili-Uéré: a chimpanzee behavioural realm in northern Democratic Republic of Congo. Folia Primatologica 90: 3–64. https://doi.org/10.1159/000492998.
Hicks TC, Menken SBJ, Laudisoit A, Hart J (2019b). Handling and consumption of vertebrate prey by chimpanzees (Pan troglodytes schweinfurthii) in the Northern Democratic Republic of the Congo. In Chimpanzee Behavior: Recent Understandings From Captivity and the Forest (Jensvold ML, ed.), pp. 1–34. Hauppauge, NY, Nova Science Publishers.
Hicks TC, Kühl H, Boesch C, Menken SBJ, Hart J, Roessingh P, et al. (2020). The relationship between tool use and prey availability in chimpanzees (Pan troglodytes schweinfurthii) of northern Democratic Republic of Congo. International Journal of Primatology 41: 936–959. https://doi.org/10.1007/s10764-020-00149-4.
Hiraiwa-Hasegawa M, Byrne RW, Takasaki H, Byrne JM (1986). Aggression toward large carnivores by wild chimpanzees of Mahale Mountains National Park, Tanzania. Folia Primatologica 47: 8–13.
Hirata S, Myowa M, Matsuzawa T (1998). Use of leaves as cushions to sit on wet ground by wild chimpanzees. American Journal of Primatology 44: 215–220.
Hobaiter C, Klein H, Gruber T (2022). Habitual ground nesting in the Bugoma Forest chimpanzees (Pan troglodytes sweinfurthii), Uganda. BioXiv. https://doi.org/10.1101/2022.12.14.520400.
Hohmann G, Fruth B (2003). Culture in bonobos? Between-species and within-species variation in behavior. Current Anthropology 44: 563–571.
Humle T (2003). Culture and variation in wild chimpanzee behavior: a study of three communities in West Africa. PhD dissertation, University of Stirling.
Isbell LA, Bidner LR, Van Cleave EK, Matsumoto-Oda A, Crofoot MC (2018). GPS-identified vulnerabilities of savannah-woodland primates to leopard predation and their implications for early hominins. Journal of Human Evolution 118: 1–13.
Koops K, Humle T, Sterck EHM, Matsuzawa T (2007). Ground-nesting by the chimpanzees of the Nimba Mountains, Guinea: environmentally or socially determined? American Journal of Primatology 69: 407–419. https://doi.org/10.1002/ajp.20358.
Koops K, McGrew WC, de Vries H, Matsuzawa T (2012a). Nest-building by chimpanzees (Pan troglodytes verus) at Seringbara, Nimba Mountains: antipredation, thermoregulation, and antivector hypotheses. International Journal of Primatology 33: 356–380. https://doi.org/10.1007/s10764-012-9585-4.
Koops K, McGrew WC, Matsuzawa T, Knapp LA (2012b). Terrestrial nest-building by wild chimpanzees (Pan troglodytes): implications for the tree-to-ground sleep transition in early hominins. American Journal of Physical Anthropology 148: 351–361. https://doi.org/10.1002/ajpa.22056.
Kouakou CY, Boesch C, Kühl HS (2009). Estimating chimpanzee population size with nest counts: validating methods in Taï National Park. American Journal of Primatology 71(6): 447–457. https://doi.org/10.1002/ajp.20673.
Kühl HS, Boesch C, Kulik L, Haas F, Arandjelovic M, Dieguez P, et al. (2019). Human impact erodes chimpanzee behavioural diversity. Science 363: 1453–1455. https://doi.org/10.1126/science.aau4532.
Last C, Muh B (2013). Effects of human presence on chimpanzee nest location in the Lebialem-Mone Forest landscape, southwest region, Cameroon. Folia Primatologica 84: 51–63. https://doi.org/10.1159/000346305.
Lewis EM, Werdelin L (2007). Patterns of change in the Plio-Pleistocene carnivorans of eastern Africa. In Hominin Environments in the East African Pliocene: an Assessment of the Faunal Evidence (Bobe R, Alemseged Z, Behrensmeyer AK, eds.), pp. 77–105. Springer. https://doi.org/10.1007/978-1-4020-3098-7_4.
Mannu M, Ottoni EB (2009). The enhanced tool-kit of two groups of wild bearded capuchin monkeys in the Caatinga: tool making, associative use, and secondary tools. American Journal of Primatology 71(3): 242–251. https://doi.org/10.1002/ajp.20642.
Matsuzawa T, Yamakoshi G (1996). Comparison of chimpanzee material culture between Bossou and Nimba, West Africa. In Reaching Into Thought: the Minds of the Great Apes (Russon AE, Bard KA, Parker S, eds.). Cambridge, Cambridge University Press.
Maughan JE, Stanford CB (2001). Terrestrial nesting by chimpanzees in Bwindi impenetrable National Park, Uganda. American Journal of Physical Anthropology 32: 104.
McCarthy MS, Lester JD, Stanford CB (2017). Chimpanzees (Pan troglodytes) flexibly use introduced species for nesting and bark feeding in a human-dominated habitat. International Journal of Primatology 38: 321–337. https://doi.org/10.1007/s10764-016-9916-y.
McCullagh P, Nelder JA (1989). Generalized Linear Models. London, Chapman & Hall.
McGrew WC (1992). Chimpanzee Material Culture: Implications for Human Evolution. Cambridge, Cambridge University Press.
McGrew WC (2004). The Cultured Chimpanzee. Cambridge, Cambridge University Press.
McGrew WC (2010). In search of the last common ancestor: new findings on wild chimpanzees. Philosophical transactions of the Royal Society 365(1556): 3267–3276.
McHenry HM (1992). Body size and proportions in early hominids. American Journal of Physical Anthropology 87: 407–431.
McLennan MR, Hill CM (2010). Chimpanzee responses to researchers in a disturbed forest-farm mosaic at Bulindi, western Uganda. American Journal of Primatology 72(10): 907–918.
McLennan MR (2018). Tie one on: ‘Nest tying’ by wild chimpanzees at Bulindi-a variant of a universal great ape behavior? Primates 59(3): 227–233.
McLennan MR, Howell CP, Bardi M, Heistermann M (2019). Are human-dominated landscapes stressful for wild chimpanzees (Pan troglodytes)? Biological Conservation 233: 73–82.
Mehlman PT, Doran DM (2002). Influencing western gorilla nest construction at Mondika research center. International Journal of Primatology 23: 1257–1285. https://doi.org/10.1023/A:1021126920753.
Morgan D, Sanz C, Onononga JR, Strindberg S (2006). Ape abundance and habitat use in the Goualougo Triangle, Republic of Congo. International Journal of Primatology 27(1): 147–179. https://doi.org/10.1007/s10764-005-9013-0.
Nakagawa S, Johnson PCD, Schielzeth H (2017). The coefficient of determination R2 and intra-class correlation coefficient from generalized linear mixed-effects models revisited and expanded. Journal of Royal Society Interface 14(134). https://doi.org/10.1098/rsif.2017.0213.
Nakamura M, Hosaka K, Itoh N, Matsumoto T, Matsusaka T, Nakazawa N, Nishie H, Sakamaki T, Shimada M, Takahata Y, Yamagami M, Zamma K (2019). Wild chimpanzees deprived a leopard of its kill: implications for the origin of hominin confrontational scavenging. Journal of Human Evolution 131: 129–138. https://doi.org/10.1016/j.jhevol.2019.03.011.
Nieuwenhuis R, Manfred G, Ben P (2012). influence.ME: tools for detecting influential data in mixed effects models. The R-Journal 4: 38–47.
Oelze VM, Head JS, Robbins MM, Richards M, Boesch C (2014). Niche differentiation and dietary seasonality among sympatric gorillas and chimpanzees in Loango National Park (Gabon) revealed by stable isotope analysis. Journal of Human Evolution 66: 95–106.
Pruetz JD, Fulton SJ, Marchant LF, McGrew WC, Schiel M, Waller M (2008). Arboreal nesting as anti-predator adaptation by savanna chimpanzees (Pan troglodytes verus) in southeastern Senegal. American Journal of Primatology 70: 393–401. https://doi.org/10.1002/ajp.20508.
Prüfer K, Munch K, Hellmann I, Akagi K, Miller JR, Walenz B, et al. (2012). The bonobo genome compared with the chimpanzee and human genomes. Nature 486: 527–531. https://doi.org/10.1038/nature11128.
R Core Team (2020). R: a Language and Environment for Statistical Computing. Vienna, R Foundation for Statistical Computing.
Reynolds V, Reynolds F (1965). Chimpanzees of the Budongo Forest. In Primate Behavior – Field Studies of Monkeys and Apes (De Vore I, ed.), pp. 368–424. New York, Holt, Rinehart and Winston.
Robbins MM, Ando C, Fawcett KA, Grueter CC, Hedwig D, Iwata Y, Lodwick JL, Masi S, Salmi R, Stoinski TS, Todd A, Vercellio V, Yamagiwa J (2016). Behavioral variation in gorillas: evidence of potential cultural traits. PLoS One 11(9): e0160483. https://doi.org/10.1371/journal.pone.0160483.
Romani T, Mundry R, Mayanja Shaban G, Konarzewski M, Namaganda M, Hobaiter C, Gruber T, Hicks TC (in press). Decay rates of arboreal and terrestrial nests of eastern chimpanzees (Pan troglodytes sweinfurthii) in the Bugoma Central Forest Reserve, Uganda: implications for population size estimates. American Journal of Primatology.
Sanz C, Morgan D, Strindberg S, Onononga JR (2007). Distinguishing between the nests of sympatric chimpanzees and gorillas: distinguishing sympatric ape nests. Journal of Applied Ecology 44(2): 263–272.
Sanz CM, Morgan DB (2007). Chimpanzee tool technology in the Goualougo Triangle, Republic of Congo. Journal of Human Evolution 52: 420–433.
Sarmiento EE, Butynski TM, Kalina J (1996). Gorillas of Bwindi-Impenetrable Forest and the Virunga Volcanoes: taxonomic implications of morphological and ecological differences. American Journal of Primatology 40: 1–21.
Scally A, Dutheil JY, Hillier LW, Jordan GE, Goodhead I, Herrero J, et al. (2012). Insights into hominid evolution from the gorilla genome sequence. Nature 483(7388): 169–175. https://doi.org/10.1038/nature10842.
Serckx A, Huynen MC, Beudels-Jamar RC, Vimond M, Bogaert J, Kühl HS (2016). Bonobo nest site selection and the importance of predictor scales in primate ecology. American Journal of Primatology 78(12): 1326–1343. https://doi.org/10.1002/ajp.22585.
Southern LM, Deschner T, Pika S (2021). Lethal coalitionary attacks of chimpanzees (Pan troglodytes troglodytes) on gorillas (Gorilla gorilla gorilla) in the wild. Scientific Reports 11(1): 14673. https://doi.org/10.1038/s41598-021-93829-x.
Stern JT (2000). Climbing to the top: a personal memoir of Australopithecus afarensis. Evolutionary Anthropology: Issues, News, and Reviews 9(3): 113–133. https://doi.org/10.1002/1520-6505(2000)9:3<113::aid-evan2>3.0.co;2-w.
Stewart FA (2011b). The evolution of shelter: ecology and ethology of chimpanzee nest building. PhD dissertation, University of Cambridge.
Stewart FA, Pruetz JD (2013). Do chimpanzee nests serve an anti-predatory function? American Journal of Primatology 75: 593–604. https://doi.org/10.1002/ajp.22138.
Stewart FA, Piel AK, Azkarate JC, Pruetz JD (2018). Savanna chimpanzees adjust sleeping nest architecture in response to local weather conditions. American Journal of Physical Anthropology 166(3): 549–562. https://doi.org/10.1002/ajpa.23461.
Tagg N, Willie J, Petre CA, Haggis O (2013). Ground night nesting in chimpanzees: new insights from central chimpanzees (Pan troglodytes troglodytes) in South-East Cameroon. Folia Primatologica 84: 362–383. https://doi.org/10.1159/000353172.
Tédonzong LRD, Willie J, Makengveu ST, Lens L, Tagg N (2020). Variation in behavioral traits of two frugivorous mammals may lead to differential responses to human disturbance. Ecology and Evolution 10(8): 3798–3813. https://doi.org/10.1002/ece3.6178.
Treves A, Palmqvist P (2007). Reconstructing hominin interactions with mammalian carnivores (6.0-1.8 Ma). In Primate Anti-Predator Strategies (Gursky SL, Nekaris KAI, eds.), pp. 355–381. Boston, Springer. https://doi.org/10.1007/978-0-387-34810-0_17.
Tsukahara T (1993). Lions eat chimpanzees: the first evidence of predation by lions on wild chimpanzees. American Journal of Primatology 29: 1–11.
Turner A (1990). The evolution of the guild of larger terrestrial carnivores during the Plio-Pleistocene in Africa. Geobios 23: 349–368.
Tutin CEG, Fernandez M (1984). Nationwide census of gorilla (gorilla g. Gorilla) and chimpanzee (Pan t. troglodytes) populations in Gabon. American Journal of Primatology 6: 313–336.
van Casteren A, Sellers WI, Thorpe SKS, Coward S, Crompton RH, Myatt JP, Ennos AR (2012). Nest-building orangutans demonstrate engineering know-how to produce safe, comfortable beds. Proceedings of the National Academy of Sciences 109(18): 6873–6877. https://doi.org/10.1073/pnas.1200902109.
van Dijk K, Cibot M, McLennan MR (2021). Chimpanzees (Pan troglodytes) adapt their nesting behavior after large-scale forest clearance and community decline. American Journal of Primatology 83(10): e23323. https://doi.org/10.1002/ajp.23323.
van Lawick-Goodall J (1968). The behaviour of free-living chimpanzees in the Gombe Stream Reserve. Animal Behaviour Monographs 1: 161–311, IN1-IN12. https://doi.org/10.1016/S0066-1856(68)80003-2.
van Schaik CP (2002). Fragility of traditions: the disturbance hypothesis for the loss of local traditions in orangutans. International Journal of Primatology 12: 527–538.
van Schaik CP (2003). Orangutan cultures and the evolution of material culture. Science 299(5603): 102–105. https://doi.org/10.1126/science.1078004.
Ward CV (2002). Interpreting the posture and locomotion ofAustralopithecus afarensis: where do we stand? American Journal of Physical Anthropology 119(S35): 185–215. https://doi.org/10.1002/ajpa.10185.
White LJT, Edwards A (2000). Conservation Research in the African Rain Forests: a Technical Handbook. New York, Wildlife Conservation Society.
White TD, Gen S (1987). Hominid footprints at Laetoli: facts and interpretations. American Journal of Physical Anthropology 72: 485–514.
Whiten A, Goodall J, McGrew WC, Nishida T, Reynolds V, Sugiyama Y, et al. (2001). Charting cultural variation in chimpanzees. Behaviour 138: 1481–1516.
Willie J, Tagg N, Petre CA, Pereboom Z, Lens L (2014). Plant selection for nest building by western lowland gorillas in Cameroon. Primates 55(1): 41–49.
Wrangham R (2009). Catching Fire: How Cooking Made us Humans. New York, Basic Books.
Yamagiwa J (2001). Factors influencing the formation of ground nests by eastern lowland gorillas in Kahuzi-Biega National Park: some evolutionary implications of nesting behavior. Journal of Human Evolution 40(2): 99–109. https://doi.org/10.1006/jhev.2000.044.