Research summaries


Investigating Conspecific Negative Density Dependence in Ngel Nyaki Forest

Iveren Abiem

Understanding how diversity is maintained is essential in making informed decisions about the management of natural areas. One of the leading mechanisms for explaining the patterns of diversity and species coexistence in forests is conspecific negative density dependence (CNDD). CNDD proposes that tree seeds and seedlings often experience high mortality near conspecific adults because of hostspecific natural enemies (herbivores, pathogens) or strong intraspecific competition and so, dispersal of seeds away from their maternal trees leads to increased survival and growth of offspring (Janzen, 1970, Connell, 1971). In addition, CNDD is hypothesized to promote species coexistence by giving rare species an advantage over common species because common species would have elevated mortality rates (Umaña et al., 2017). While many studies have examined the role of NDD in plant community structuring and species coexistence in tropical forests, these studies have been concentrated in the Neotropics and Asian tropics.

While African tropical forests may superficially appear similar to Asian and South American forests, their ecological structure is very different. This may reflect different processes underlying patterns in their species composition. African forests have low alpha diversity compared to their Asian and South American counterparts but have high carbon stocks, even higher than the Amazon (Sullivan et al., 2017). Moreover, African tropical montane forests are very different from African lowland tropical forests. Thus, quantifying density dependence in Afromontane forests is important in order to fill the knowledge gap of how plant species coexist.

My focus this year was investigating CNDD in Ngel Nyaki forest. I monitored the survival of 10,741 seedlings of 93 species for two years in the long-term forest study plot. Seedlings were recorded from 8,430 one metre squared quadrats throughout the 20.28 hectare plot. For every seedling, we calculated conspecific and heterospecific seedling neighbour densities from within 1 m2 quadrats and conspecific and heterospecific adult densities within a 20 metre radius of the focal seedling. At the community level, neighbour density significantly affected seedling survival. Seedling neighbours negatively affected survival while adult neighbours positively influenced survival. The relationship between seedling survival and neighbourhood densities varied across growth forms and shade tolerance guilds.

Conspecific densities had a negative effect on the survival of canopy and light-demanding species while total adult densities had a significant positive effect on survival of emergent, understory and liana species, and total seedling densities had a significant negative and marginally significant negative effect on the survival of understorey tree and liana species respectively. Survival of large seedlings was most likely to be significantly affected negatively by conspecific adult density and total seedling density while survival of small seedlings was affected negatively by conspecific seedling density and positively by total adult density. Survival for newly recruited seedlings was not significantly affected by the biotic neighbourhood but was significantly affected by the season when the seedling recruited into the community.

Seedlings that recruited in the dry season had significantly higher survival rates than those that recruited in the wet season. We conclude that conspecific negative density dependence may not be a key factor explaining species composition and coexistence patterns in the Afromontane plant community and the drivers of community coexistence in the unique environment will need to be investigated using detailed experiments.

The value of forest fragments in maintaining avian functional diversity and associated ecosystem services within a West African agricultural landscape

Murna Tela PhD NZ Development Scholarship

Field assistants Yakubu Vugeh, Ahamadu Usman and Usman Bashiru

Forest-agricultural mosaics are common ecosystems in Nigeria and in montane habitats the forest fragments support a rich diversity of bird species. While some of these species may be providing pest control services to agricultural crops, the future of the forest fragments is threatened because they are being farmed. Therefore, it is important that we understand whether small forest fragments provide ecosystem services to surrounding crop species by being habitat for birds which control insect pests of crops.

As part of my thesis, I quantified the pest control services provided by birds to farmers and identified their contribution to crop productivity in Zea mays, the most common crop on the Mambilla Plateau. Since last year, I have continued to carry out fieldwork to monitor bird species across 36 farmlands and 22 forest fragments and classified each species into a feeding guild. I used exclosure experiments (using a series of caged vs open maize plots across the Mambilla Plateau) to test my hypothesis.

My hypothesis is that (i) if birds are excluded from contact with the maize plants, more insects will attack the crop and productivity will be reduced. (ii the retention of forest fragments close to maize farms will have effects on crop yield in Mambilla farmlands via increasing pest control services provided by birds’ diversity and richness

Key findings this year;

  1. Crop yield was higher in the control treatments where insectivorous birds were present than in the bird excluded treatments.
  2. Crop yield increased with an increase in insectivorous species in the control treatment but decreased when birds were excluded.
  3. The retention of forest fragment close to farmlands have no positive effect on crop yield via increasing pest control by insectivorous birds.
  4. Crop yield increased with increase in distance of farmlands from forest fragments for both treatments.

Overall, my exclosure experiment showed that insectivorous bird abundance played an important role in pest control in Mambilla maize farmlands, leading to an increase in crop yield. Although forest proximity to farmlands can be essential for enhancing crop productivity, this might not be so for Nigerian farmlands. For farmlands at Mambilla, the retention of forest fragments away from these farms (e.g. land sparing in which land conservation is held separate, and away from crop production) may be the best overall conservation strategy.


Tool and anvil use by a highland population of Pan troglodytes ellioti in Ngel Nyaki Forest Reserve, Nigeria

Dr Paul Dutton
Primary field assistant: Alfred Christopher
Submitted to Primates (July 2013)

In this study, carried out as part of Paul Dutton’s PhD thesis on the general ecology of Ngel Nyaki chimpanzees, we found that the Ngel Nyaki chimpanzee community has their own unique toolkit consisting of several different tool types. We describe a tool type that has never been recorded before, a tool type that has rarely been observed, a tool type that has never been recorded for this chimpanzee subspecies, and another tool type that requires its defined target species to be reviewed. When comparing tool types between Ngel Nyaki and neighbouring Kwano communities we found that of the total eight tool types from both communities only three were common to both: stingless bee digging sticks, stingless bee probing sticks and ant dips; however differences were found in their dimensions and secondary modifications. Our results suggest that there is fine-scale variation in tool use among populations of P. t. ellioti and that these variations may reflect both ecological constraints and cultural variation.

Dietary preferences of chimpanzees (Pan troglodytes ellioti) and food availability in Ngel Nyaki Forest Reserve

Dr Paul Dutton
Primary field assistant: Alfred Christopher
Details submitted to Folia Primatologia (October 2013)

The dietary preferences of chimpanzees residing in Ngel Nyaki Forest Reserve was investigated using faecal analysis, observations of feeding remains, evidence of tool use and fruiting phenological data between April 2010 and March 2011. A total of 495 faecal samples were collected, with 75 food items identified, of which 52 items were seeds. Ficus spp. were the most common species identified, occurring in 61.2% of all faecal samples. Based on faecal analysis and phenological data, Ngel Nyaki chimpanzees do not solely consume fruits based on their availability within the habitat. Detailed results are presented in the paper.

Tantalus monkeys and seed dispersal

Fiona Agmen BSc (Hons)

Along with birds, primates are key dispersers in Afromontane forests, making up between 25 and 40 % of frugivore biomass. Primates are important because they have a wider gape than most bird species, and so are theoretically able to disperse (through swallowing) larger seed than even wide-gaped bird species. However primate populations are in decline, and so for conservation management it are important to highlight why their continued survival is important for forest ecosystem function.

Ngel Nyaki Forest Reserve and the surrounding riverside forests (from here on referred to as forest fragments) are degraded and increasingly fragmented due to recent increases in human and cattle populations. Hunting has been rife, especially during the 1980’s so that the only primate species now regularly crossing open grassland and visiting forest fragments is Chlorocebus tantalus (tantalus monkey). In the past other primates such as chimpanzee, putty nose and mona monkeys were common visitors to the fragments.

The aims of this research are to:

  1. Identify which fruit tree species C. tantalus feeds on in the fragments
  2. Determine the relative importance (effectiveness) of C. tantalus in the dispersal of each of these species compared with their other frugivores.
  3. Investigate the potential for C. tantalus to move seed from Ngel Nyaki forest back into depleted fragments.

The results of this work will contribute to our understanding of the vulnerability of forest fragments to frugivore decline and to the importance (or not) of C. tantalus in their survival.

Aspects of the ecology of the Cameroon Olive Pigeon (Columba sjostedti) at Ngel-Nyaki (Adang, Kombe Lucas, PhD)

The Cameroon Olive Pigeon (Columba sjostedti) is one of the endemic bird species found in the Montane Forests at Ngel Nyaki. Little is known of the ecology of this species, thus the proposed study is aimed at providing information on some aspects of the ecology of this species at Ngel Nyaki Forests. The ecological aspects of interest are outlined below.

Aspects of interest

Food habits

  1. Visual observations of the bird’s feeding and noting the plant or tree species on which it feeds.
  2. Mist-netting the bird to collect crop contents (regurgitates) for analysis of possible or identifiable food items.
  3. Collection of faecal samples for analysis of possible or identifiable food items.
  4. Possible feeding regimes of the bird on a particular tree species viz a viz competition with other bird species on such food item.

Breeding Biology

Try to identify or locate nests of the bird and monitor its breeding biology from courtship or territoriality to fledging.

Habitat preference

  1. Note where the bird is commonly sighted, i.e. in the fragment or main forest and to postulate reasons for such preference.
  2. Record the number of sightings of the bird in a day within the fragments and main forest.
  3. Perform ringing experiments to determine the territorial range of the bird within the plateau and its environs, especially the Gashaka Gumti National park.

The role of sunbirds in pollination (Kerry-Anne Weston MSc)

When addressing conservation issues currently facing areas of afromontane forest such as the Ngel Nyaki forest reserve, anthropogenic impacts upon the flora and fauna within the forest must be addressed not only in terms of species loss, but also in terms of disruption to the mutualisms between these species which may be indirectly precipitating or compounding this species loss. An increased understanding of plant-animal interactions such as pollination services may be vital in advising conservation efforts. The first step towards this understanding is to identify the web of interactions linking plants species and their pollinators. Sunbirds are small, largely nectarivorous passerines with long, slender, sharply pointed, decurved bills. There are 131 species of sunbird worldwide, 28 of which occur in Nigeria. The sunbirds are thought to be implicated in the pollination of several tree species of the Nigerian afromontane forest, however the relative role and consequently the importance of sunbirds as pollinators is largely unknown.

Predicting the main pollinators of flowering plants can be difficult from pollination syndromes alone due to the wide taxonomic range of pollinators servicing most plant species, often resulting in temporally and spatially variable natural selection. Consequently, field observations combined with experimentation are required to test these predictions and identify the role and importance of these interactions in overall ecosystem function.

A preliminary mutualist web in a West African montane forest (Kennedy Poloma Yoriyo (PostDoc), Merodie Beavon (MSc)

In order to begin to understand the role of different seed dispersers and pollinator species/groups at Ngel Nyaki we are going to begin to create a mutualist web. To do this we need to identify all the different seed dispersers and pollinators (at least to family level) on all (or as many as possible) of the tree species in Ngel Nyaki forest.

To do this transects (already cut throughout the forest) will be walked on a regular basis and trees will be observed for seed dispersers and both insect (K. Poloma and D. Campbell) and bird (M. Beavon and D. Campbell) pollinator activity. Because of the difficulties of differentiating between insect/bird visitation and pollination, we shall initially record only visitation. Insects will be collected and identified to genus level wherever possible. We may have to build viewing platforms at several strategic locations along the transects to give us access to higher trees.

The influence of matrix habitat restoration on dung beetle species responses across tropical forest edges (Andrew Barnes, MSc)

Anthropogenic land-use change has led to the fragmentation of habitats, resulting in the pervasive impacts of edge effects in ecosystems. These effects have presented conservation challenges that call for active measures to restore these degraded ecosystems. We experimentally tested the relative influence of matrix-habitat restoration on edge effects in Afromontane forest dung beetle communities by comparing abundance, richness, similarity, and individual species responses across forest edges adjacent to both degraded and regenerating matrix where external anthropogenic threats had been excluded. Dung beetle capture rates showed strong responses to habitat edges with decreasing numbers of beetles from the forest to matrix habitat. Also, capture rates and species richness were higher across the entire edge gradient in regenerating sites compared to degraded sites. These patterns were also apparent in compositional patterns where forest and matrix communities were the most dissimilar and there was clear compositional dissimilarity between communities in regenerating and degraded sites. Moreover, we found that 9 out of 10 species displayed strong individual responses to edge effects and the capture rates of these species were highly elevated in response to adjacent matrix restoration. Our findings indicate that dung beetles exhibit strong responses to edge effects through a decline in numbers as well as changes in relative species abundances due to variation in individual species responses. Importantly, we have also demonstrated that low-intensity restoration efforts in the adjacent matrix habitat can have high-level positive impacts on dung beetle communities through ameliorating edge effects and imposing beneficial off-site impacts on remnant communities in degraded landscapes.

Relationships between the tantalus monkey and forest structure in a West African montane forest (Abby Grassham, MSc)

Many of the world’s tropical forests are under threat, with anthropogenic deforestation and degradation occurring at an alarming rate. Seed dispersal in an important process in forest restoration and regeneration, however seed rain is often low in degraded habitats, hindering reforestation efforts. Up to 90% of tropical fruit are dispersed by vertebrates, animal seed dispersers are incredibly important in maintaining forest health. Additionally, frugivores that disperse seeds into degraded areas may be of great importance in aiding natural reforestation. I therefore, investigated the potential role of the frugivorous monkey, Chlorocebus tantalus tantalus, in forest regeneration via seed dispersal. I assessed its patterns of habitat use, the quality and quantity of seed dispersal it provides, the effectiveness of current conservation management actions and the density of C. t. tantalus at Ngel Nyaki Forest Reserve. I found C. t. tantalus utilised forest, edge and grassland habitats, and dispersed seeds of 28 pioneer and forest edge species into these habitats. Moreover, the number of seeds dispersed per faeces was significantly higher in the grassland than the forest with means of 16.4 +/- 6.1 and 3.4 +/- 0.97 seeds >2 mm in these habitats respectively. Germination of C. t. tantalus dispersed seeds was highest in grazed grassland and lowest in grassland protected from grazing and fire, suggesting the current practice of fencing off grassland to protect from cattle grazing may not be sufficient on its own, due to seed-seedling conflict in habitat suitability. These findings combined with an estimated density of 28 +/- 10.8 C. t. tantalus individuals km2 suggests C. t. tantalus may benefit forest regeneration via its role as a seed disperser, provided appropriate management actions are implemented. This and other frugivorous species may play similar roles in other locations but such roles need to be investigated in order to implement management actions that ensure their seed dispersal benefits are maximised for forest restoration and regeneration.

Habitat Effect on the Behaviour and Condition of the Yellow-breasted Boubou (Laniarius atroflavus) (Samuel Temidayo Osinubi, PhD)

This project was aimed at investigating behaviour and condition of the Yellow-breasted Boubou, Laniarius atroflavus, in response to habitat differences across core, edge and riparian Afro-montane forest habitats. This species is little known and conservation effort will require direction in identifying the habitat of best quality for their survival. The determination of habitat association using correspondence analysis of census data suggested strongest association with the riparian habitat, even though this habitat held the least overall avian biodiversity as determined from a modified Shannon index. Territoriality, vocalisation and time budget showed trends indicating L. atroflavus were most abundant and fared best in the riparian habitat. In this habitat, there was a greater density of territories and a smaller mean territory size, better call quality in frequency bandwidth and duration, and increased displaying and foraging time in the riparian habitat. Difference in size, colour and growth-based measures of condition showed difference between sexes, but did not show a strong habitat effect – males were larger than females, yet females appeared to have better quality of yellow breast feathers for equal carotenoid concentration. The effect of nest predation risk as a predictor of habitat quality revealed nests in the riparian habitat had the greatest daily survival probability, and within this habitat nests established at lower heights survived longest. While the evidence pointed towards the riparian habitat being most suitable for L. atroflavus, sadly, this habitat was most prone to anthropogenic disturbance. L. atroflavus appeared not to hold territories in the core habitat and its IUCN listing as Least Concern was suspected be an over-estimation due to the species’ far-carrying call.

The influence of forest reserve protection on the structure, stability, and functioning of dung-associated invertebrate communities (Kristy Udy, MSc)

Communities are influenced by many factors, with anthropogenic impacts being one of the strongest. In particular, farming practices such as livestock grazing are common threats to the surrounding ecosystem. These effects are often exacerbated by variation across seasons that can significantly influence community structure and may have filtering effects on the response traits, such as body size, of species. This can result in non-random species loss, which in turn can have further flow-on effects that alter ecosystem processes, resulting in ecosystem-level responses to these drivers. Anthropogenic impacts can also alter the effects of abiotic factors that alter resource quality over different temporal scales. This can mediate the intensity of competition within a community, with competition generally being higher when species within a community occur at high densities. The community based around a resource is made up of species that interact together through trophic and non-trophic interactions. Both of these interaction types can alter the effect of the other and are important for structuring the community. Non-trophic interactions are generally measured by counting the number of co-occurrences of species at a shared resource, thereby inferring the existence of competition. However, this is not always an appropriate measure, as different species may use different parts of the resource. Consequently, when measuring actual competition in a community it may be more effective to visually observe which species interact and how often. Additionally, the structure of trophic interactions between species in food webs can in part be predicted by the ratio of predator to prey body size, and this ratio may also be useful for predicting the outcome of non-trophic interactions.

This study tested the effects of protecting forest reserves to exclude the effects anthropogenic activities in adjacent systems and was conducted in an Afromontane forest in Nigeria using the dung-associated community as a focal study system because it contains species guilds that are commonly used as biological indicators of ecosystem health. Taking season into account, I tested how protection of forest from livestock grazing and fire threats affected the structure of the dung-associated community and how these factors affected the ecosystem functions of dung removal and secondary seed dispersal performed by this community. I also measured how dung beetle (Coleoptera: Scarabaeinae) abundance and size responded to forest protection and seasonal variation and how these influenced dung removal and secondary seed dispersal. Dung is a highly ephemeral resource, as it can be entirely exploited by associated invertebrates in a matter of hours and is also subject to rapid desiccation, therefore making this an ideal system for quantifying the abiotic and biotic determinants of community structure over time. To investigate how competition levels within a community vary over short timescales I observed how dung desiccation over three days influenced the communities attracted to this resource. I also tested the influence of competition in this community by experimentally altering the size and numbers of dung beetles, which are the most functionally important taxonomic group in this system. Because environmental changes have dramatic effects on insect communities, experiments were also run in both the wet and dry seasons to better understand the relative influence of desiccation changes across seasons. Furthermore, video analysis was used to determine which species competed and how often, and then trapping was carried out to sample the community. The structure of the community was compared between protected and unprotected areas of the forest and sampling was carried out in both the wet and dry seasons to account for strong temporal variation. Food-web metrics were measured to gain an understanding of the interaction network structure and to test whether co-occurrences can be used to infer the response of competitive interactions to anthropogenic activities, such as habitat loss.

Community structure was influenced by forest protection, and dung beetles were the most functionally important guild within the dung-associated community. The size and abundance of dung beetles increased in the wet season and in protected areas of the forest, which significantly increased the amount of dung removed and the number of seeds secondarily dispersed. Changes across seasons altered desiccation rates of the dung resource, which were significantly higher in the dry season, resulting in almost completely desiccated dung after three days. As a result, there were large compositional changes in the dung-associated invertebrate community over days and among seasons, with the highest overall invertebrate abundance found within the first day of dung deposition and during the wet season. As expected, dung removal increased significantly with increasing dung beetle size and densities. However, there was a reduction in per capita dung removal rates, with increasing dung beetle densities. This was most likely due to competition between individuals over the resource. As a consequence of these density-dependent processes, individual removal rates for beetles of a particular size-class were not additive and therefore could not be used to predict community-level removal rates based on individual level functional efficiency. Thus, using body size is an ineffective predictor of community-level function if utilised without taking into account competitive interactions. Furthermore, when community metrics were measured for the interaction networks nestedness of a network differed between protected and unprotected forest and varied with season in the competition network, however no effect of these factors was detected in the co-occurrence network. Therefore, when determining network structure, measuring the actual rates of competition gives more accurate results than numbers of co-occurrences, and the body-size ratio between competing species can be a useful to tool to aid in predicting the outcome of competitive interactions or to estimate non-trophic effects in food webs. The body-size ratio between winners and losers of a competition was also a good predictor of an interaction outcome, especially when the ratio in body size was greater than four between competing individuals. Thus, the exclusion of anthropogenic threats from habitat adjacent to forest reserves influences the structure of the community and interactions within it, the size of dung beetles, and significantly increases ecosystem functioning. Additionally, the structure and functioning of dung-associated invertebrate communities are strongly influenced by both abiotic (dung desiccation) and biotic (size and density-dependent competition) factors over small and large temporal scales. Therefore, mitigating anthropogenic threats to ecosystems through protection of reserves had an effect on the structure of associated communities, but was only apparent when quantified with the effect of changing seasons.