Dung beetle assemblages within Amazonian selectively logged forests

  • Filipe França (Creator)
  • Jos Barlow (Contributor)
  • Vanesca Korasaki (Contributor)
  • Julio Louzada (Contributor)



This dataset results from the MSc research in Applied Ecology, entitled "Selective logging in Amazonian forests: post-logging recovery in dung beetle communities", by Filipe França at the Federal University of Lavras. It includes data on [1] dung beetle species richness, abundances and biomass; [2] dung beetle-mediated ecological processes of dung removal, seed dispersal and soil bioturbation; and [3] environmental conditions related to canopy openness, soil texture (sand, silt and clay content), and disturbance intensity (number of trees and volume of timber removed). Datasets were surveyed using a space-for-time experimental design, between January and March 2010, across 49 10-ha logging management units distributed along a gradient of selective logging intensities within two areas (Bituba and Gueti) in the Jari region of the Brazilian Amazon, state of Pará. Surveyed areas had different times since logging: Bituba was managed in 2005 and Gueti in 2009, with logging intensities varying between 0-76 and 0-55 trees removed, and 0-643 and 0-669 timber volume (m3/10ha), respectively.
We followed the harvest operational plans of the timber concession, which adopts 250×400m logging management units, to establish six sampling points spaced 100m apart in a 2×3 rectangular grid and around 75 m from the unit edges. We sampled dung beetles using baited pitfall traps (40g of dung; 19 cm width, 11 cm depth; 1 trap/sampling point, 6 traps/unit). We surveyed ecological processes using two 1-m diameter mesocosm arenas placed at the middle sampling points of the rectangular grid 24h before installing pitfall traps. Each arena was a circular plot delimited by a 15-cm high nylon net fence (area of ~0.79m2) and received a single 200-g experimental dung deposit mixed with seed mimics of three sizes: 10 large (15.5-mm diameter), 20 medium (8.6-mm) and 50 small (3.5-mm). All dung used had a 4:1 pig to human ratio to increase attractiveness and allow enough baits to carry out all field experiments. After 24h of exposure, we weighted the remaining dung and the loose soil found above the soil surface (if present) to assess rates of dung removal and soil bioturbation. Rates of seed dispersal are based on the difference between total seed mimics placed in each dung deposit and the number of seed mimics found in the remaining dung (when present). At each sampling point, we took a (1) soil sample (up to 10cm depth) and (2) hemispherical photograph using a Nikon FC-E8 fisheye lens attached to a Nikon D40 camera levelled ∼1.20m from the ground. We formed a composite soil sample to represent the soil granulometry (clay, silt, fine sand, and coarse sand; g/Kg) within each sample unit, which was analysed in the soil laboratory of Jari Celulose S.A. We used the Gap Analyser software to estimate the canopy openness (%), which represents the total open area of the hemispherical photographs. For further details, please see related links and 'README' in the attached spreadsheet.
Date made available13 May 2022
PublisherMendeley Data
Date of data production5 Feb 2010 - 19 Mar 2010

Cite this