Seeds go up and down: The role of dung beetles in soil seed movement in the Southern Atlantic Forest of Argentina
Abstract
The vertical movement of seeds performed by dung beetles from the soil bank and in animal feces influences seed germination and the temporal dynamic of forest regeneration. While this process has been explored at the community level, the individual role of species is less understood. Here, we investigated the role of dung beetle size and seed size in this vertical movement under experimental conditions.
We performed experiments using a gradient of dung beetle sizes and three sizes of artificial seeds (plastic beads) in two situations: inside feces (secondary dispersal) and buried in the ground (soil bank). Through regression analysis, we related dung beetle size, seed size, and the initial position of seeds on the soil bank to the final position of seeds and their potential germination.
For seeds in the soil bank, upward movement and exhumation were mainly of medium and large seeds, initially located at shallower depths, with larger beetles being primarily responsible for this movement. The downward movement was similar for all seed sizes. In dung seeds, the percentage of small seeds buried gradually increased with beetle size, while larger beetles made the main contribution for medium and large seeds. Besides, all seed sizes were buried at an average maximum bury depth of nearly 4 cm (the limit of the germination zone).
The relative contribution of species depended on the interaction between dung beetles and seed sizes. Moreover, large dung beetles were essential for burying large seeds in the southern Atlantic Forest.
README: Seeds go up and down: The role of dung beetles in soil seed movement in the Southern Atlantic Forest of Argentina
https://doi.org/10.5061/dryad.wh70rxwxb
Description of the data and file structure
Files and variables
File: Data.xlsx
Description: On the first sheet, the file contains the species of dung beetle used in each replicate, the initial location of the plastic beads, their size (small: 2 millimeters (mm), medium: 6 mm and large: 8 mm) and the final depth (centimeters). On the second sheet, the file contains the metadata.
Variables
- Number of counted plastic beads
Code/software
You need excel or similar software.
Methods
Study area and collection of dung beetles
We conducted this study with dung beetle species native to the Southern Atlantic Forest of Argentina (25° 44’ S and 54° 30’ W). The region is characterized by a warm seasonal climate, with annual temperatures between 17 and 22°C and a mean annual precipitation of 2000 mm, with no dry season (Oliveira & Fontes, 2000). We worked with eight copro-necrophagous beetle species representing the weight and size gradient for the whole assemblage (ranging from 0.007 g to 1.7 g.) (Table 1) (Giménez Gómez et al., 2022).
We capture live dung beetles in late spring of 2015 (December) using ten pitfall traps in two habitat types: native forests in protected areas and nearby mature pine plantations. Traps were a plastic cup of 12 cm in diameter and depth. We baited five traps with human dung and five with decomposing chicken because they are the most efficient resource for capturing dung beetles (Gardner et al., 2008). We separated the traps by 50 m between them to minimize the interference among them (Larsen & Forsyth, 2005). We used the same sites and traps to capture individuals until we finished the experiments. Once captured, we took the beetles to the laboratory, keeping them alive until we performed experiments. The collected individuals were weighed and measured in the laboratory.
Manipulative experiments
The objective of the experiments was to determine the role of dung beetles in the vertical movement of seeds in the soil, considering the initial location and size of the seeds and the body size of the beetles. An illustrative figure of the experiments can be found in Figure 2 and Figure S1. The experiments were composed of two parts:
1) Soil seeds, we explored the role of dung beetles in the vertical movement (up and down) of the seeds in the seed bank;
2) Dung seeds, we explore the role of dung beetles in the burial of seeds contained in feces.
For the experiments, we used cylinders 40 cm deep and 30 cm in diameter filled with 35 cm soil (Figure S1). In all cases, we used spherical plastic beads as artificial seeds. Plastic beads as a replacement for natural seeds have been used successfully in other studies (Griffiths et al., 2015; Koike et al., 2012). The sizes of the plastic beads used in this study were 2, 6 and 8 mm in diameter, representing the size range of 84% of the seeds of woody plant species dispersed by animals in the region, excluding very large seeds (Vespa et al., 2014) (Figure 1).
To study the vertical movement of seeds in the seed bank (soil seeds), we buried 240 g of plastic beads between 0 – 8 cm; 60 g of orange spherical plastic beads between 0 to 2 cm, 60 g of green beads between 2 to 4 cm, 60 g of red beads between 4 to 6 cm, and 60 g of blue beads between 6 to 8 cm (Figure 2, left). For each colour/level, the 60 g were divided into three groups of 20 g using beads of different sizes: 1746 small beads of 2 mm, 215 medium beads of 6 mm, and 73 large beads of 8 mm (Figure 2, left). In total, each cylinder contained 8136 beads in the artificial seed bank. We used this number of beads to ensure that differences, if there were, in the role of dung beetles in vertical seed movement based on beetle size could be observed.To study the burial of seeds in the feces (dung seeds), we placed 60 g of white spherical plastic beads (also divided into three groups of 20 g as above) in 100 g of howler monkey dung on the ground surface (Figure 2, left).
Then, we placed ten individuals of the same dung beetle species in each experiment. All individuals were not fed for 24 hours before the experiment. We covered the system with a mesh with 2 mm pores for 72 hours to keep the beetles inside.
At the end of this time, we moved the beetles and the soil surface, centimetre by centimetre, recording the depth at which we found each plastic bead, its size and colour (Figure 2, right). We performed between two and three replicates for each dung beetle species depending on the number of individuals available (19 replicates in total) (Table 1). Beetles were not reused between replicates.
Data analysis
Soil seeds experiment
To analyse seed movement in the soil, we first calculated the percentage of beads moved upward and downward for each size and initial depth; then, we compared these percentages using a chi-square analysis. We performed a two-factor ANOVA to compare the percentage of beads exhumed (emerged to the surface) according to initial depth and bead size.
To analyse the effect of vertical seed movement on the potential probability of germination and subsequent seedling emergence, we considered a threshold of 4 cm as the critical depth. Seeds buried at greater depths have a low probability of germination and seedling emergence (Urrea-Galeano et al., 2019). Considering this, we defined two zones: 1) viable zone, above 4 cm depth and 2) non-viable zone, below 4 cm depth. Then, we performed two regression analyses using the total length of the beetles as explanatory variable and as response variables: 1) the percentage of beads moved from the viable zone to the non-viable zone, and 2) the percentage of beads moved from the non-viable zone to the viable zone. We fitted all data to linear, exponential, and sigmoidal models. Then, we selected the best fit for each case based on the Akaike criterion of information. In cases that did not meet the assumptions of normality and homoscedasticity, the data were ln transformed.
Dung Seeds experiment
To explore the role of dung beetles on seed burial, we first performed a regression analysis using the percentage of buried seeds according to size as the dependent variable and the total length of the beetles as the independent variable. We also performed a regression analysis for the depth to which the seeds are buried; in this case, we used the average depth to which the seeds were buried for each seed size as the dependent variable.