INTRODUCTION

Fossorial and semi-fossorial mammals such as Cape porcupines (Hystrix africaeaustralis) and aardvarks (Oryceteropus afer) are ecosystem engineers and fulfil an important role in making resources such as shelter and food available to other animals through burrowing (Haussman et al., 2018; Jones et al., 1994; Skinner & Taylor, 2010). The burrows are created when the animal digs for shelter (den) and food (foraging holes) (Whittington-Jones et al., 2011). Digging enables porcupines to access foods such as bulbs and roots that may be unavailable to non-fossorial species (Fattorini & Pokheral, 2012). The foraging holes may vary in size (area and depth) and age (new and old) (Khan et al., 2016; Whittington-Jones et al., 2011). The size of the foraging hole may indicate how deep the animal dug for food (such as tree roots or bulbs) or it may also be influenced by the type of food the animal sought (Alkon, 1999; Khan et al., 2016). The roots of trees that porcupines feed on may also vary with the age of the tree. For example, younger trees may likely have shallow roots while old trees may have deep roots (Kadir et al., 1998). This, too, may likely influence the depth of their digs.

Little has been documented about the foraging behaviour of porcupines and the resultant foraging holes in African savannas. This is likely because porcupines are nocturnal, which limits the ability to study and understand their foraging ecology in the wild. However, they leave a mark on the landscape through disturbance of vegetation and the soil, which in turn bears evidence of their foraging activities despite the nocturnal lifestyle (Andersen, 1987; Haussman et al., 2018). Similar to other herbivores, the feeding behaviour and diet of porcupines as generalist feeders may be impacted by and vary with seasons. Herbivores in southern Africa tend to increase their foraging efforts and intake of previously avoided foods in the dry season (Abraham et al., 2019; Owen-Smith, 1994; Owen-Smith & Chafota, 2012). Bark and roots of trees are reported to be important dietary components of the Cape porcupine's diet (de Villiers & van Aarde, 1994; Sweitzer, 1996). A diet of roots may lead to the health of the tree compromised or its death. Uprooting of trees leading to their death by elephants has been reported in African savannas (Guy, 1976; MacGregor & O'Connor, 2004; O'Connor, 2017). Thus, tampering with the roots of a plant may comprise its health and survival. The extent of feeding on roots of trees has rarely been studied for the smaller herbivores such as porcupines in savannas.

Porcupines excavate soils to create holes in order to access perennial plants such as bulbs and roots (Khan et al., 2000; Wilby et al., 2001), which may be staple food for porcupines. In Italy, crested porcupines were recorded to eat bulbs in both warm and cold months (Mori et al., 2017). The diet of Cape porcupines may vary with season and habitat. The digging effects of porcupines have been reported to vary in severity between agricultural and natural systems (Laurenzi et al., 2016). The foraging activities of porcupines may result in tree damage through girdling, uprooting and debarking (Khan et al., 2000; Kraai et al., 2022). The damage exerted on trees may be influenced by age. For example, old trees were reported to be avoided and trees younger than 20 years were debarked but no deaths of trees were recorded despite utilization by porcupines (Laurenzi et al., 2016).

There is little information on the characteristics, abundance and age of foraging holes of porcupines between seasons. Additionally, no information exists on the size of foraging holes associated with digging for roots of trees. We determined the foraging behaviour of Cape porcupines in two savanna ecosystems and between seasons. In this study, the temporal variation in foraging holes was investigated to determine their extent and persistence on the landscape. A better understanding of these aspects of foraging holes of porcupines is important to determine the potential environmental impact of these animals, and the role of Cape porcupines in savanna ecosystems. The study was aimed at evaluating the seasonal digging effects of Cape porcupines on trees. We predicted that there would be an increase in the number of new foraging holes and the utilization of trees as food during the dry season compared to the wet season.

METHODS AND MATERIALS

RESULTS

The number and age of foraging holes varied with time of year in Bisley, with foraging activities increasing in the drier months (May–August) of the year (Table 1). Trees used by porcupines for roots were limited to one or two species per site, and specific sizes of trees at either Roodeplaat or Bisley (Table 2). We recorded 105 foraging holes dug on different trees by porcupines, 64 at Roodeplaat and 41 at Bisley in the dry season (May–August), and 320 foraging holes at Bisley between 2019 and 2021.

The month of October, which is the beginning of the wet season, in both years comprised a greater number of old compared to new foraging holes (Table 1). In contrast, the month of May, June, July and August (dry season) had the greatest number of new foraging holes (Table 1). Thus, the foraging activities of Cape porcupines for tree roots were pronounced in the dry season and decreased in the wet season.

The use of V. nilotica trees by Cape porcupines resulted in the death of most trees (Table 1). The number of trees used and impacted to death increased with the progression of the dry season. Specifically, 15 trees were killed out of 22 used in May and 25 of 29 were killed in August. Overall, 89 trees died out of 144 utilized by porcupines. No death of trees from porcupine foraging activities was recorded for the greater part of the wet season.

Cape porcupines primarily ate the roots of Vachellia robusta and Dombeya rotundifolia to a lesser extent at Roodeplaat, while Vachellia nilotica roots were eaten more than Z. mucronata and C. rudis at Bisley (Table 2). The size of the trees used for roots by Cape porcupines tended to be small in both sites except for larger trees of V. robusta selected in Roodeplaat. The preference for V. nilotica trees made them susceptible to death after use of roots by the Cape porcupines at Bisley (Table 2). No other deaths of trees were recorded.

The surface area of new and old foraging holes of different tree species (tree × age interaction) was significantly different (F_{2, 97} = 5.31, p = 0.006). The surface area of old foraging holes was larger than that of new foraging holes for V. robusta and D. rotundifolia occurring at Roodeplaat (Figure 1a). In addition, both new and old foraging holes of D. rotundifolia were larger than foraging holes of V. robusta and V. nilotica. Similarly, the depth of new and old foraging holes differed significantly for different tree species (tree × age interaction: F_{2, 97} = 4.29, p = 0.016). New and old foraging holes on V. nilotica trees were deeper than foraging holes of V. robusta and new holes of D. rotundifolia (Figure 1b).

The surface area of Cape porcupine foraging holes on V. nilotica trees differed significantly with season (F_{2, 167} = 3.41, p = 0.035). Foraging holes had a larger surface area in the wet season compared to the early wet and early dry seasons (Figure 2a). The depth of the foraging holes differed significantly with interaction of age and season (F_{2, 167} = 3.65, p = 0.028). The depth of new foraging holes increased in the early dry season, while older holes were shallower in the wet season compared to the early wet and early dry seasons (Figure 2b). Ultimately, the foraging holes were wider and shallower in the wet season.

DISCUSSION

The foraging activities associated with digging by Cape porcupines were greater in the dry than wet season. As predicted, we found more new foraging holes with mounds in the dry season, which likely indicated increased utilization of tree roots by Cape porcupines. In contrast, the landscape was comprised of more old foraging holes in the wet season. As a result, the size of the foraging holes shrunk in the wet season. In addition, shallow foraging holes in Bisley in the early wet and wet seasons indicated a decreased utilization of tree roots compared to the dry season. This may indicate that the roots of the trees were abandoned as a food source in the wet season. Bragg et al. (2005) reported that geophytes were likely a primary food source in the diet of porcupines in the wet season in the Northern Cape of South Africa. Seasonality may thus trigger the switch in diet of Cape porcupines. Switching diet is common behaviour in many herbivorous mammals and driven by food quality and quantity (Abraham et al., 2019; Owen-Smith & Chafota, 2012). In addition, Sweitzer (1996) found that porcupine foraging on tree bark, although not included in the current study, was associated with the food-limiting season. Elephants were also reported to feed on bark and roots of trees in the dry season (Owen-Smith & Chafota, 2012). Generally, it appeared that woody plants in the form of roots thus comprised a greater portion of the dry season diet of Cape porcupines.

The diet of tree roots appeared narrow, as only one major species was utilized in each site, which belongs to the Vachellia genus. Similarly, roots of Burkea africana were preferred out of 11 tree species available to Cape porcupines in a savanna habitat (de Villiers & van Aarde, 1994). Thus, Cape porcupines showed selective utilization of roots of Vachellia species in the dry season. Similar behaviour was reported by Bragg (2003) where porcupines selected 27 of 350 species of geophytes. Vachellia trees are preferred for their nutritional leaves by other herbivores (Aganga et al., 1998; Nzimande et al., 2022; Owen-Smith & Chafota, 2012; van der Merwe & Marshall, 2012). Preferred plants may experience extensive herbivory, which in some cases, may result in the death of the plant (Ihwagi et al., 2009; Shannon et al., 2008) and may be exacerbated by the influence of fire in large trees (Das et al., 2022). In the current study, seedlings of V. nilotica died (63.5%) as a result of Cape porcupine feeding. This may in turn reduce the abundance of V. nilotica, an encroaching plant, in Bisley (Muvengwi et al., 2017). In this light, the effects of Cape porcupine feeding on small trees of V. nilotica may be likened to the African elephant, albeit at a smaller scale in savannas.

Cape porcupines likely made successive visits to roots of larger trees, as the resource is unlikely to be depleted over a once-off feeding event. This may explain the shallow and narrow new holes on V. robusta at Roodeplaat. Yet, foraging holes on seedlings were four times deeper and seven times wider at Bisley compared to Roodeplaat. Similar to Bisley, recently dug foraging holes of geophytes were deep (8 cm) and wide with a surface area of 400 cm² in the Negev desert highlands in Israel (Alkon, 1999). In addition, porcupines were shown to dig up to a depth of 21 cm for bulbs in Pakistan (Khan et al., 2016). The roots of old trees are likely deeper and may not be easily accessible as younger trees (Dorval et al., 2016; Ennos, 2000). Porcupines are also solitary foragers, which would make their impact on old trees less severe than social foragers. Roots of larger trees are also likely established and stronger than young trees (Perry, 1989), and thus difficult to kill the tree. Larger trees may be less susceptible to the effects of herbivory due to their firm root structure. Thus, the size of foraging holes may vary depending on how deep and wide the animal dug, which may be influenced by the size and location of the resource.

Interestingly, porcupine diggings and subsequent death of trees may contribute to the management of small encroaching trees in Bisley. The biomass, basal cover and abundance of herbaceous vegetation have been shown to improve after the removal of encroaching trees (Mndela et al., 2022). Promoting the growth of herbaceous instead of woody plants is important in savannas, which are prone to woody plant encroachment, which in turn depreciates the state of savanna ecosystems (Luvuno et al., 2022; O'Connor et al., 2014). In essence, Cape porcupines killed V. nilotica trees by detaching the stem from the root and eating the roots, which may enhance the establishment and growth of herbaceous plants where a tree seedling previously occupied. Thus, porcupine foraging activities may play an important role in facilitating landscape changes, and vegetation composition and structure of savannas.

CONCLUSION

Cape porcupines are important in the savanna ecosystem through their foraging activities, which may shape vegetation communities. We have shown that diggings transform the landscape in the dry season and may promote the growth of herbaceous plants in affected areas in the wet season. Increased digging in the dry season coincided with porcupines' increased intake of tree roots during a period of scarcity of herbaceous food resources. As a result, Cape porcupine foraging resulted in the death of small trees, and potentially old trees through persistent use, or through predisposing the plants to attack by other organisms. Porcupines may also, through foraging, result in the recruitment of young plants thereby affecting the demography and structure of plant communities. Foraging activities of porcupines have cascading effects on the landscape, which influence plant and likely animal communities. Cape porcupines are thus capable of modifying savanna habitats and may contribute, even though to a small extent, to eliminating trees in dense areas. A wider sampling regime is likely to strengthen the observed patterns.

AUTHOR CONTRIBUTIONS

Unathi Masiobi Kraai: Data curation (equal); formal analysis (equal); investigation (lead); methodology (lead); writing – original draft (lead). Manqhai Kraai: Conceptualization (equal); formal analysis (equal); investigation (equal); supervision (equal); writing – review and editing (equal). Zivanai Tsvuura: Conceptualization (equal); investigation (equal); supervision (equal); writing – review and editing (equal). Ntuthuko Raphael Mkhize: Conceptualization (equal); funding acquisition (equal); supervision (equal); writing – review and editing (equal). Tlou Julius Tjelele: Funding acquisition (equal); supervision (equal); writing – review and editing (equal).

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no competing financial or personal interests to influence the work reported in this paper.