The interplay between defaunation and phylogenetic diversity affect leaf damage by natural enemies in tropical plants
Data files
Jan 22, 2024 version files 881.90 KB
-
data_original.csv
-
README.md
Abstract
- Natural enemies play an important role in controlling plant population growth and vegetation dynamics. Tropical rainforests host the greatest diversity of herbivores, from large mammalian ungulates to microscopic pathogens, generating and maintaining plant diversity.
- By feeding on the same resources, large mammalian herbivores may interfere with plant consumption and leaf damage by important enemy guilds such as invertebrate herbivores and pathogens, triggering indirect trophic cascades. However, the impact of local extinctions of large herbivores on plant-enemy interactions is relatively unknown.
- We experimentally tested the effects of defaunation of large mammalian herbivores (e.g., peccaries, tapirs, brocket deer; hereafter, large herbivores) on leaf damage of 3,350 understory plants in tropical rainforests of Brazil. We examined leaf damage in 10,050 leaves from 333 morphospecies by assigning the area consumed or damaged by five guilds of insect herbivores and leaf pathogens within 86 paired open-closed plots and investigated the joint effects of defaunation and plant phylogenetic diversity.
- Plants released from large herbivores had 9% less leaf damage; this difference was due to the lower leaf pathogens incidence (29%) rather than insect herbivory. Evolutionary Distinctness was similarly and positively correlated with leaf damage in all treatments, suggesting additive effects of defaunation and phylogenetic diversity. Total and pathogenic leaf damage (but not insect damage) decreased with plant richness across treatments, and large herbivores' exclusion resulted in increased plant species richness. This suggests that large herbivores' exclusion leads to a dilution of total and pathogens’ leaf damage by increasing plant species richness.
- Our results suggest that large herbivores' indirect effects decrease the dilution potential of plant communities against pathogens and rather reinforce their top-down impact on vegetation, demonstrating a previously overlooked cascading effect of large herbivore extinction on forest ecosystems.
- Synthesis: The extinction of large mammalian herbivores can lead to a decrease in pathogen-driven leaf damage, a previously unknown indirect effect in forest ecosystems, which might have consequences for plant fitness and ultimately for plant diversity. Large herbivores and plant pathogens might have synergistic effects in regulating the diversity of plant communities in some of the most diverse ecosystems on Earth.
README: The interplay between defaunation and phylogenetic diversity affect leaf damage by natural enemies in tropical plants
https://doi.org/10.5061/dryad.7m0cfxq2h
This dataset is the original data sampled from a defaunation experiment, in which large mammalian herbivores were excluded, using a paired design of open-closed plots. The dataset contains the raw records of leaf damage caused by plants' natural enemies, using the herbivory index described in the main text, and before any averaging and/or analyses.
The variables included in the dataset are as follows:
- local: protected area in which the data was collected, ie., Itamambuca [ITA], Ilha do Cardoso State Park [CAR]; Carlos Botelho State Park [CBO]; Vargem Grande [VGM].
- plot: sequential number (1 to 86) identifying the pairwise plots studied within each local.
- treatment: each pairwise plot is composed of two treatments = open (or the control plot, in which large mammals have open access) and closed (fenced, in which large mammals are excluded).
- plotID = unique code for each plot, combining the local, plot and treatment information.
- ID_plant = unique code for each individual seedling or sapling within each plot. This code is part of DEFAU-BIOTA long-term experimental study, and it is maintained by Prof. Galetti' group.
- N_leaf = sequential number (1 to 3) to identify the leaf sampled in each individual plant.
- binary_data = indicating whether the N_leaf has any type of damage y natural enemies (1), or not (0). NA = leaf not present in the plant.
- perc_leaves_wdamage = percentage of leaves with any type of leaf damage per individual plant, considering the total leaves sampled, i.e., if there were either three, two or one leaf available to be sampled and all of then have damage, then it equals 100% and the value is 1; if two leaves out of three have damage, then the value is equal 0.66%; if one leaf out of three, then is 0.33%; one leave out of two = 0.5%; and if no leaf has damage, it is 0%.
- leaf_damage = categorical value corresponding to the leaf damage category assigned for each sampled leaf: 0 (intact), 1 (1 - 6% of lost and/or damaged leaf area), 2 (6 - 12%), 3 (12 - 25%), 4 (25 - 50 %), 5 (50 - 75%) and 6 (75 - 99%). NA = leaf not present in the plant.
- ME = leaf damage category assigned to external chewers. NA = leaf not present in the plant.
- bin_me = presence (1) or absence (0) of damage by external chewers in that leaf.
- perc_me = percentage of leaves with ME in the corresponding individual plant.
- MI = leaf damage category assigned to internal chewers. NA = leaf not present in the plant.
- bin_mi = presence (1) or absence (0) of damage by internal chewers in that leaf.
- perc_mi = percentage of leaves with MI in the corresponding individual plant.
- PAT = leaf damage category assigned to pathogens. NA = leaf not present in the plant.
- bin_pat = presence (1) or absence (0) of damage by pathogens in that leaf.
- perc_pat = percentage of leaves with PAT in the corresponding individual plant.
- MIN = leaf damage category assigned to leaf miners. NA = leaf not present in the plant.
- bin_min = presence (1) or absence (0) of damage by leaf miners in that leaf.
- perc_min = percentage of leaves with MIN in the corresponding individual plant.
- GAL = leaf damage category assigned to galls. NA = leaf not present in the plant.
- bin_gal = presence (1) or absence (0) of damage by galls in that leaf.
- perc_gal = percentage of leaves with GAL in the corresponding individual plant.
Methods
Our study was carried out in a multi-site, long-term experiment (DEFAU-BIOTA), distributed at four study sites located in protected areas of the Atlantic Forest in the state of São Paulo, Brazil. The four sites have different compositions of large herbivore communities, but nevertheless, a rich assemblage of species can be found in all of them. Due to different human pressures, sites differ in the presence of the largest ground-dwelling mammal species, the white-lipped peccary (Tayassu pecari) and the tapir (Tapirus terrestris) (Galetti et al., 2017). Although both species are present at site Itamambuca [ITA], only white-lipped peccaries are present at Ilha do Cardoso State Park [CAR]; at Carlos Botelho State Park [CBO] only tapirs are found, and neither of the species are present at Vargem Grande [VGM].
In each site, we installed 15 paired plots (5 m long x 3 m wide) in 2009–2010, but some of the plots collapsed over time; as a result, the data in this work were collected in nine paired plots in ITA, 10 in CBO and 12 in CAR and VGM, summing up 86 plots. Each pair consists of a closed, exclusion plot of medium and large mammals (hence simulating defaunation), and an open plot allowing free access to these animals (control). Each plot was subdivided into eight subplots of 1 m2, in which three subplots were randomly chosen and all plants larger than 0,1 m and smaller than 1 m were marked, identified at the lowest possible taxonomic level, and accessed without disturbance to estimate leaf damage.
We evaluated the leaf damage of all marked plants within each selected subplot in July, August, and November 2019. For each individual plant, we randomly chose three mature leaves (when possible; otherwise, we count on the available number per individual, that is, one or two leaves) to estimate their leaf damage according to the percentage of leaf area lost and/or damaged. Next, we visually assigned the following leaf damage categories for each selected leaf: 0 (intact), 1 ( 1–6% of lost and/or damaged leaf area), 2 (6–12%), 3 (12–25%), 4 (25–50 %), 5 (50–75%) and 6 (75–99%). We proceed with this categorization of damage considering four feeding guilds of insect herbivores (internal and external chewers, galls and leaf miners), the guild of pathogens, and the total leaf damage caused by all groups, given that each leaf may be affected by more than one type of damage.