Data on arthropod abundance in tropical forest restoration plots with high or low plant phylogenetic diversity
Data files
Feb 23, 2024 version files 141.02 KB
-
1._Arthropod_abundance.xlsx
11.85 KB
-
2._Species_Rank_Abundance.xlsx
21.22 KB
-
3._Arthropod_Species_Abundance_(by_guild).xlsx
30.94 KB
-
4._Species_composition_(NMDS_for_Morphospecies_and_OTUs).xlsx
70.11 KB
-
README.md
6.90 KB
Abstract
Consideration of plant phylogenetic diversity in ecological restoration carries substantial potential, as communities with a greater diversity of lineages with older evolutionary histories can increase the diversity of niches and thus are likely to recover larger species networks than communities clustered in specific clades with reduced variation in functional traits. In this study, we experimentally assessed how arthropod communities were affected by the phylogenetic diversity of a set of tropical tree species. We established 12 experimental restoration plots with either high or low plant phylogenetic diversity, while maintaining constant the number of plant species. After one and three years, arthropods with different feeding habits (herbivores, predators, pollinators, and detritivores) were collected and identified as morphospecies or operational taxonomic units using metabarcoding techniques. We provide insights on the influence of plant phylogenetic diversity on arthropod abundance and species diversity, particularly among predator, pollinator, and detritivore common and dominant species, which increased with plant phylogenetic diversity. The trend, however, was the opposite for the diversity of herbivore common and dominant species, which decreased as plant phylogenetic diversity increased. These findings highlight the importance of considering plant species richness when designing restoration strategies, but also their evolutionary histories, as the same number of plant species can produce different outcomes for higher trophic levels, as a function of their phylogenetic relationships.
README: Data on arthropod abundance in tropical forest restoration plots with high or low plant phylogenetic diversity
https://doi.org/10.5061/dryad.qnk98sfqb
Description of the data and file structure
Data sets contents
- Abundance of arthropods (predators, herbivores, pollinators and detritivores)
- Arthropod Species Abundance by Rank (to calculate species rank abundance curves)
- Arthropod Species Abundance by treatment and year (by guild: Predators, herbivores, Pollinators and Detritivores)- To calculate number of Hills
Arthropod Species Abundance by treatment and year (to calculate species composition differences for both, Morphospecies and OTUs)
*in all cases: HPD=High Phylogenetic Diversity and LPD = Low Phylogenetic Diversity
Experimental Procedure:
The study was carried out within 'Los Amigos' cattle ranch, located in the buffer zone of the Los Tuxtlas biosphere reserve (18°32’56” N, 94°59’58” W) in Veracruz, México. The climate in this area is warm and humid with mean temperatures ranging from 24 to 26°C and total precipitation ranging from 2,000 to 4,000 mm (Soto, 2004). The vegetation in this region includes evergreen tropical forest with a high level of fragmentation surrounded by an agricultural and cattle ranching landscape (Von Thaden, et al., 2018). The ranch itself has cattle pastures, bordered by remnants forests
Between July 2018 and January 2019, we established 12 restoration plantings (15 × 15 m) within the active cattle pastures, with a minimum distance of 50 m among the plots. To manipulate the phylogenetic diversity of the plantings, we used a total of 43 tropical tree species. In half of the plots, we established plant assemblages with high phylogenetic diversity (HPD), incorporating 27 phylogenetically distant species from 23 families, widely distributed across the phylogeny of the regional pool of species (see below). In the remaining plots, we established plantings with low plant phylogenetic diversity (LPD), with 27 species from 10 families clustered within the phylogeny and with more recent common ancestors (Supplementary information Table S1). In each plot we planted between 1 and 7 seedlings (between 30 - 70 cm tall) from each species in a 1 × 1 m grid, resulting in a total of 196 individuals per plot. However, due to plant mortality, between 21 and 28 species (15 to 19 non-pioneer species and 10 to 12 pioneer species) remained in each plot, but contrasts in phylogenetic diversity between both treatments were maintained (Supplementary information Table S1, Santos-Gally and Boege, 2022).
We collected arthropods from the experimental plots in September 2019 and 2021 (~ 1 and 3 years after establishment of restoration plots), using distinct methods for each arthropod guild. To capture pollinators, we used pan traps comprising three plastic containers painted in blue, yellow, and white, each filled with 200 ml of soapy water. This variety of colors is recommended to attract the greatest diversity of pollinators, as different taxa have distinct preferences for different colors (Leong, et al., 1999, Padrón, et al., 2020). Two sets of traps were placed on each plot, one in the middle of the upper right quadrant and the other at the plot´s center. To collect adult lepidopterans, we used one Van Someren- Rydon trap with a bait made with a mixture of pineapple fermented with beer, positioned at the plot´s center. Sweep nets and manual surveys were employed along four perpendicular transects positioned every two meters within each plot to collect herbivores and predators from the canopies of saplings and surrounding vegetation. For detritivores and other crawling arthropods, two pitfall traps were placed in the middle of the lower left quadrant, as well as in the center of each plot. These diverse traps enabled the collection of resident organisms colonizing the plots but also those passing through the vegetation islands. After 72 hours, specimens were collected from all traps and stored with 70% undenatured ethanol, except for lepidopterans which were collected after 24 hours in the Van Someren-Rydon traps. These specimens were subsequently frozen and stored in glassine paper bags. After taxonomic and/or molecular identification (see below), organisms were classified based on their feeding habits, using available ecological literature for the most specific taxonomic level. We classified organisms that consumed any part of plants (including seeds) as herbivores, except for adult Lepidoptera specimens and other floral visitors, which were classified as pollinators. Organisms that feed on other animals were classified as predators, and those involved in the consumption of decomposing organic matter or associated with the cycling of soil nutrients were classified as detritivores. Organisms that could not be classified were excluded from subsequent analyses.
We systematically grouped all collected organisms based on their respective order, family, and when feasible, assigned them to genus and species. In cases where precise taxonomic identification was unattainable, we categorized the specimens as morphospecies. For molecular identification, we sorted specimens into two distinct groups: i) those obtained from color and Van Someren-Rydon traps, and ii) those captured with pitfall traps, manual surveys, and sweep nets. Specimens were categorized into small (*< *3 mm), medium (> 3 mm and < 15 mm) and large (> 15mm) size classes, according to the criteria outlined by Galvez-Reyes, et al. (2021). Whereas the entire body was used for small arthropods, medium-sized arthropods were dissected to select the thorax (including the head), and only two legs were selected for large arthropods. After placing specimens or their parts in falcon tubes according to their size class, we extracted DNA from the composite arthropod samples of each size category and group, to perform a metabarcoding analysis (i.e. sensu Yu et al., 2012), following a modified protocol by Galvez-Reyes, et al. (2021) using a DNeasy® Blood and Tissue Kit (Qiagen). We used primers B_F 5’ CCIGAYATRGCITTYCCICG 3' (Shokralla, et al., 2015) and Fol-degen-R 5’ TANACYTCNGGRTGNCCRAARAAYCA 3' (Yu, et al., 2012) to amplify a 418-base pair region located at the 5’ end of the mitochondrial COI gene. This specific region is part of the standard barcode region for metazoans. For the construction of metabarcoding libraries, individual samples (n =104) were pooled and grouped in equimolar proportions based on their designated category (pollinators or crawling arthropods), resulting in the generation of 60 final libraries (one library per sample). DNA samples, along with negative control, were sequenced using an Illumina Miseq 2x300 bp platform at Illumina (RTL Genomics). Additional details on bioinformatic read processing are elaborated in Supplementary Information A1.
Methods
Between July 2018 and January 2019, we established 12 restoration plantings (15 × 15 m) within the active cattle pastures, with a minimum distance of 50 m among the plots. To manipulate the phylogenetic diversity of the plantings, we used a total of 43 tropical tree species. In half of the plots, we established plant assemblages with high phylogenetic diversity (HPD), incorporating 27 phylogenetically distant species from 23 families, widely distributed across the phylogeny of the regional pool of species (see below). In the remaining plots, we established plantings with low plant phylogenetic diversity (LPD), with 27 species from 10 families clustered within the phylogeny and with more recent common ancestors.
We collected arthropods from the experimental plots in September 2019 and 2021 (~ 1 and 3 years after establishment of restoration plots), using distinct methods for each arthropod guild. To capture pollinators, we used pan traps comprising three plastic containers painted in blue, yellow, and white, each filled with 200 ml of soapy water. This variety of colors is recommended to attract the greatest diversity of pollinators, as different taxa have distinct preferences for different colors (Leong, et al., 1999, Padrón, et al., 2020). Two sets of traps were placed on each plot, one in the middle of the upper right quadrant and the other at the plot´s center. To collect adult lepidopterans, we used one Van Someren- Rydon trap with a bait made with a mixture of pineapple fermented with beer, positioned at the plot´s center. Sweep nets and manual surveys were employed along four perpendicular transects positioned every two meters within each plot to collect herbivores and predators from the canopies of saplings and surrounding vegetation. For detritivores and other crawling arthropods, two pitfall traps were placed in the middle of the lower left quadrant, as well as in the center of each plot. These diverse traps enabled the collection of resident organisms colonizing the plots but also those passing through the vegetation islands. After 72 hours, specimens were collected from all traps and stored with 70% undenatured ethanol, except for lepidopterans which were collected after 24 hours in the Van Someren-Rydon traps. These specimens were subsequently frozen and stored in glassine paper bags. After taxonomic and/or molecular identification (see below), organisms were classified based on their feeding habits, using available ecological literature for the most specific taxonomic level. We classified organisms that consumed any part of plants (including seeds) as herbivores, except for adult Lepidoptera specimens and other floral visitors, which were classified as pollinators. Organisms that feed on other animals were classified as predators, and those involved in the consumption of decomposing organic matter or associated with the cycling of soil nutrients were classified as detritivores. Organisms that could not be classified were excluded from subsequent analyses.
We systematically grouped all collected organisms based on their respective order, family, and when feasible, assigned them to genus and species. In cases where precise taxonomic identification was unattainable, we categorized the specimens as morphospecies. For molecular identification, we sorted specimens into two distinct groups: i) those obtained from color and Van Someren-Rydon traps, and ii) those captured with pitfall traps, manual surveys, and sweep nets. Specimens were categorized into small (< 3 mm), medium (> 3 mm and < 15 mm) and large (> 15mm) size classes, according to the criteria outlined by Galvez-Reyes, et al. (2021). Whereas the entire body was used for small arthropods, medium-sized arthropods were dissected to select the thorax (including the head), and only two legs were selected for large arthropods. After placing specimens or their parts in falcon tubes according to their size class, we extracted DNA from the composite arthropod samples of each size category and group, to perform a metabarcoding analysis (i.e. sensu Yu et al., 2012), following a modified protocol by Galvez-Reyes, et al. (2021) using a DNeasy® Blood and Tissue Kit (Qiagen). We used primers B_F 5’ CCIGAYATRGCITTYCCICG 3' (Shokralla, et al., 2015) and Fol-degen-R 5’ TANACYTCNGGRTGNCCRAARAAYCA 3' (Yu, et al., 2012) to amplify a 418-base pair region located at the 5’ end of the mitochondrial COI gene. This specific region is part of the standard barcode region for metazoans. For the construction of metabarcoding libraries, individual samples (n =104) were pooled and grouped in equimolar proportions based on their designated category (pollinators or crawling arthropods), resulting in the generation of 60 final libraries (one library per sample). DNA samples, along with negative control, were sequenced using an Illumina Miseq 2x300 bp platform at Illumina (RTL Genomics). Additional details on bioinformatic read processing are elaborated in Supplementary Information A1.