Study area and site selection
The Taï National Park covers an area of about 508,186 hectares, in south-western Côte d'Ivoire and extends between 5°08'192'' and 6°23'790'' N, and between 6°46'579'' and 7°26'006'' W (Fig. 1).
It is located on the borders of the departments of Guiglo, Taï, Tabou, San-Pedro, Méagui, Soubré and Buyo (Décret n° 2018-496 du 23 mai 2018). It represents the last and largest intact block of primary rainforest in West Africa and represents more than half of the total area of West African forest under high protection status. It is characterized by a subequatorial climate; hot and humid with four seasons: two rainy seasons, a long one from March to June and a short one from September to November, that are alternated by a long dry season from December to February and a short dry season from July to August (Chatelain et al. 2001). Rainfall varies from 1700 mm in the north-east to 2200 mm in the south-west of the park. The annual variation in temperature is around 4°C, between 24 to 28°C. Relative humidity is always very high and varies from 70 to 90% (Chatelain et al. 2001). During the 2019 study year, the average rainfall and temperature were 1877 mm and 25.2°C, respectively. The soil of the Taï National Park is of two types: highly desaturated and fragile ferralitic soils and more fertile hydromorphic gley and pseudogley soils drained by numerous permanent water streams which are shared between two large catchment areas and two basins of small coastal rivers (Perraud 1971). About 1,300 species of vascular plants have been recorded in the TNP, 12% of which are endemic (Adou et al. 2005). Vegetation is homogenous overall with two large plant formations, namely the hyper humid forest with "Diospyros spp. and Mapania spp. or pelohygrophilous forest, and the evergreen forest with "Eremospatha macrocarpa and Diospyros mannii" (Adou et al. 2005).
A floristic and structural survey of chimpanzee community habitats in northwestern TNP by Goné bi et al. (2016) identified 264 plant species belonging to 52 botanical families dominated by Diospyros mannii (Ebenaceae), Calpocalyx brevibracteatus (Fabaceae) and Coula edulis (Olacaceae).
These chimpanzees territories have high densities and land areas comparable to most tropical forests, in general, and forests that support chimpanzee communities, in particular (Goné bi et al. 2016). We collected data in the research area of the Taï Chimpanzee Project, where the chimpanzees of four neighrbouring communities are followed and observed every day from sun rise to sun set (Wittig and Boesch 2019). We determinded the Study plots by identifying the chimpanzee nesting sites of the last two months prior to the start of the study.
Experimental design
We established seven plots of 40 x 40 m in each of the chimpanzee nesting sites (North, South and East), thus resulting in a total of 21 plots. The main "pitfall trap" device for harvesting dung beetles was set up in center of each plot. To imitate the spatial pattern of aggregated defecation of chimpanzees in the nesting sites, we installed five devices horizontally on either side of the main device following the orientation 290° West - 70° East (Fig. 2).
Plant species studied
The seeds of four plant species, namely Uapaca guineensis Muell. Arg. (Phyllantacaea), Diospyros mannii Hiern (Ebenaceae), Dacryodes klaineana (Pierre) H.J. Lam (Burseraceae) and Pycnanthus angolensis (Welw.) Warb. (Myristicaceae) were selected for the study of dispersal by dung beetles. These seeds are regularly consumed by chimpanzees during the November-February period (Kasparek 2000). Further exploration of the droppings during the study period showed that 66% of the total seeds found in the droppings belonged to the four plant species chosen above. The average seed is compiled in the table 1.
Table 1
Seed dimensions (in cm) of the plant species studied (average ± SD, N = 75)
|
|
Plant
|
species
|
|
|
Dacryodes klaineana
|
Diospyros mannii
|
Pycnanthus angolensis
|
Uapaca guineensis
|
Length
|
1.65 ± 0.18
|
3.07 ± 0.21
|
1.66 ± 0.15
|
2.59 ± 0.1
|
Width
|
1.13 ± 0.13
|
1.49 ± 0.12
|
1.02 ± 0.07
|
1.27 ± 0.1
|
Thickness
|
0.82 ± 0.08
|
0.9 ± 0.1
|
1.07 ± 0.09
|
0.92 ± 0.05
|
Dacryodes klaineana (Pierre) H.J. Lam (Burseraceae): A medium sized, dioecious, evergreen, hypogeous germinating tree reaching 25-30 m high often branched a few meters above the ground. The shaft is branchless over 10 m, up to 60-120 cm in diameter. The leaves are composed of imperipinnate, opposite leaflets with inflorescent flowers in a raceme fascicle in the axils of the terminal leaves. The fruits are ovoid edible drupes of about 2 cm x 1 cm with a lemony taste (Kasparek 2000).
Diospyros mannii Hiern (Ebenaceae): A small tree with simple, alternate leaves. The flowers are unisexual, fasciculate and axillary. The calyx is hairy brown. The fruits are curious berries completely covered with brown hairs. The genus Diospyros has a wide range of distribution from Côte d’Ivoire to Gabon. There are six species within the study area: Diospyros chevalieri De Wild, Diospyros heudelotii, Diospyros mannii, Diospyros sanza-minika A. Chev, Diospyros soubreana F. White and Diospyros vignei F. White (Kasparek 2000).
Pycnanthus angolensis (Welw.) Warb. (Myristicaceae): An evergreen tree, monoecious or dioecious, medium to large in size and up to 25 – 35 (-40) m high. The straight, cylindrical shaft is branchless up to 15 (-25) m, reaching 120 (-150) cm in diameter. This species usually bears buttresses. The outer bark is greyish-brown with orange-brown exudate. The cyme is small, the alternate leaves are disticticulate, simple and whole, without stipules. The fruit is an ellipsoid to oblong or globular, yellowish-orange drupe with a fairly hard and solid wall. The seeds are ellipsoid, aromatic, dark brown with pink to red arils, laciniated almost to the base (Kasparek 2000).
Uapaca guineensis Muell. Arg. (Euphorbiaceae): This genus has a distribution that extends from Sierra Leone to Congo. It grows along rivers and wet, swampy ground. It has large, arched aerial roots (the stilt-roots) which support the tree. The leaves are large in terminal rosettes of simple leaves with small white to fairly large yellow flowers, dioecious tree. The fruit is a drupe of different sizes, ovoid. The genus Uapaca is particularly well represented within the Taï National Park, the species Uapaca esculenta A. Chev. Ex Aubrév. & Leandri, Uapaca guineensis Muell. Arg. and Uapaca paludosa Aubrév. & Leandri (Kasparek 2000).
Secondary seed dispersal
Determination of faecal material to be exposed
For better reproduction of a chimpanzee faecal sample, we evaluated the average mass of chimpanzee faecal matter (N = 200, average ± SD: 114.33 ± 49.46 g), the average diameter (N = 40, average ± SD: 10 ± 2.12 cm) and the average number of seeds by faecal sample (N = 75, average ± SD: 28.24 ± 3.06 seeds). We exposed 28 seeds, i.e seven seeds per plant species contained in 85g of faecal matters in the forest for the study, taking into account the average diameter measured. The average mass of the 28 seeds was estimed (N = 75, average ± SD: 29.47 ± 3.64 g). The whole faecal matter and seeds weighed 115 g.
Methodology for the study of seed dispersal
For each of North, South and East territories, we exposed 70 chimpanzee faeces insisde which were inserted 28 seeds each, ie ten faeces per nesting site. In total, we exposed 210 dropping over the whole of this study.
We directly deposited the faecal samples on the forest floor early in the morning for a 48-hour exposure. We indelibly marked each seed in order to avoid confusing them with seeds possibly present at the site of the devices or which would have been forgotten inside the faecal matter before the introduction of the marked seeds. A plastic sheet placed at 30 cm over each device prevented the seeds from being displaced by the rainwater. We estimated the proportion of faecal matter remaining on the surface after 48 h and counted the number of seeds on the surface. When it was lower than the number of seeds exposed initially, the soil was excavated following the orientation of the pedotrophic nests or over a maximum depth of approximately 50 cm and within a maximum radius of 1 m around a flag initially placed in the center faecal matter. Seeds are indeed rarely moved over a horizontal distance exceeding 15 cm (Andresen 2001). Since the average radius of a faecal sample is 5 cm (N = 40, average ± SD: 5 ± 1.06 cm), we reduced the horizontal distances to be measured in the field by 5 cm in order to obtain an estimate of the actual value of horizontal distances. Seeds remaining on the surface after 48 h and located within 5 cm of the center are considered not to have been dispersed (Petre et al. 2015). Those buried were systematically considered dispersed.
The competitiveness of the species of dung beetles harvested in the different habitats is given in table 2.
Table 2
Abundance of individuals (percentage in brackets) belonging to the three competitiveness categories and result of the statistical test
|
|
Habitats
|
|
|
Kruskal test
|
|
North
|
South
|
East
|
|
χ²
|
df
|
p
|
Rollers
|
633 (21,41)
|
740 (35,34)
|
991 (36,41)
|
|
|
|
|
Fast burying tunnelers
|
209 (7,07)
|
276 (13,18)
|
243 (8,93)
|
|
|
|
|
Total category 1
|
842 (28,48)
|
1016 (48,52)
|
1234 (45,33)
|
|
|
|
|
Slow burrowying tunnelers
|
0 (0)
|
0 (0)
|
0 (0)
|
|
6,543
|
2
|
0,038*
|
Small tunnelers
|
2105 (71,21)
|
1056 (50,43)
|
1466 (53,86)
|
|
|
|
|
Total category 2
|
2105 (71,21)
|
1056 (50,43)
|
1466 (53,86)
|
|
|
|
|
Dwellers (category 3)
|
9 (0,3)
|
22 (1,05)
|
22 (0,8)
|
|
|
|
|
Individuals were prioritized into three categories, from most competitive to least competitive. The first category includes telecoprids ("Rollers"; genera Sisyphus, Neosisyphus and Garetta) as well as large nocturnal paracoprids burying faeces rapidly ("Fast burying tunnelers"; genera Catharsius, Heliocopris and Diastellopalpus). The second category includes small paracoprids (Tribe Onthophagini). The last category is constituted by the endocoprids (Liatongus sjoestedti (Felsche, 1904) and Oniticellus pseudoplanatus Balthasar, 1964).
Overall, the proportion of individuals belonging to the most competitive category is greater in the eastern nesting sites, but “fast burying tunnelers” of the genus Heliocopris, Catharsius and Diastellopalpus are more abundant in the southern nesting sites. Conversely, "rollers" are predominant in those of the East. We observe a significant dependence of the distribution of individuals in the three habitat competitiveness categories.
Soil compaction
The rate of faecal matter burial can be greatly influenced by soil structure and compaction. For this, we evaluated the dry bulk density of the surface horizons, as an indexe of soil compaction, using the so-called “cylinder method”, described below.
We took three soil samples exactly 30 cm from the center point of each device that allowed secondary seed quantification. Here, we used a cylinder 10 cm high and 3 cm in diameter. It was driven into the ground with a hammer. We carefully cleaned the resulting soil sample of tiny roots and its contents poured into numbered and sealed plastic bags (Petre et al. 2015). We heated this content in an oven at 105°C. We weighed the soil sample sudjected to heating every hour until the mass stabilized perfectly, meaning that the soil sample was dry. The evaluation of soil compaction was calculated using the following formula:
Litter quantification
We sample the litter an area of 1m² around each device that allowed the quantification of the seeds. We sieved using a 1 × 1 cm mesh and the resulting particles, namely the sieve and the large particles were weighed (Petre et al. 2015).
2.5- Data analysis
All statistical analyses were performed with the statistical software R version 4.1.0, at the 5% significance level.
The software allowed the verification of the normality and variances homogeneity of the insect abundances classified by nesting behavior and of the number of dispersed seeds on the one hand, but also of the litter data and of the soil compaction on the other hand. The Kruskal-wallis test (non-parametric test) and ANOVA (parametric test) were used for the comparison of means. The Kruskal-wallis post-hoc test was used to establish significant differences between the dispersion parameters.
Pearson correlation matrices tests were used to test correlations between the different parameters quantifying seed dispersal and dung beetle community parameters. When a significant correlation between two parameters was observed, simple linear regression was used to determine a statistical model linking the two variables in question.