Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network
Data files
Jul 01, 2025 version files 102.13 KB
-
README.md
9.36 KB
-
Rezendeetal_JEcol_dataset.csv
14.41 KB
-
Rezendeetal_JEcol_dataset.txt
14.41 KB
-
Rezendeetal_JEcol_Rscript_network.R
27.27 KB
-
Rezendeetal_JEcol_Rscript_SEM.R
36.67 KB
Abstract
Indirect interactions are major forces structuring communities. In terrestrial ecosystems, changes in plant traits caused by herbivory have the potential to mediate cascades of indirect interactions that can profoundly affect community structure. We have begun to understand the mechanisms behind indirect effects mediated by trait changes in plants, but we still know little about how these interaction-level mechanisms affect higher levels of community organization, in particular network structure, and how this is dependent on community context. In this study, we aimed to understand how herbivory affects floral traits, flower visitation and the position of plant individuals in a plant-flower visitor network. To do this, we assessed the natural incidence of leaf and flowerhead herbivory in a tropical shrub and tested how it affects flowerhead density and the interactions between flowers and Lepidoptera, Hymenoptera, and Diptera at the pairwise and network levels. We also tested how the local density of conspecific plants adds context dependency to these interactions by affecting both herbivory and flowerhead visitation in focal plants. Increased leaf herbivory reduced flowerhead density per plant and indirectly reduced flowerhead visitation by Hymenoptera, while for Diptera direct effects of herbivory were important and Lepidoptera were not affected. At the network level, plants with more flowerheads were more connected and central. By reducing flowerhead density, leaf herbivory caused individual plants to lose centrality, becoming less connected and therefore peripheral in the plant-flower visitor network. This dynamic is context dependent since increased plant conspecific local density reduced leaf herbivory in the focal plant and indirectly benefited its flowerhead density, flowerhead visitation and network centrality. Synthesis: We demonstrate how the effects of herbivory on a key plant trait can cascade to shape the interactions between individual plants and their flower visitors and how the indirect effects between herbivores and pollinators can modulate the structure of interaction networks in plant-based ecosystems.
Dataset DOI: 10.5061/dryad.7h44j106f
Description of the data and file structure
In December 2015, we marked all T. adenantha individuals found along a 2.5 km trail at the Reserva Biológica Municipal da Serra do Japi, Jundiaí, São Paulo, Brazil (23°14'01.3"S 46°56'09.2"W), totaling 150 plants. Of these, 61 either died or did not flower during the study, leaving 89 individuals included in the analysis. Between December 2015 and May 2016, we randomly sampled plants with at least one reproductive branch (approximately five individuals per day, totaling 15 per month over 21 sampling days). Each plant served as an independent unit for data collection, which included:
- Observation of flowerhead visitation (1h observation of a single focal reproductive branch per plant)
- Flowerhead density (total flowerheads in anthesis per plant)
- Flowerhead herbivory (estimated for all flowerheads in anthesis on the focal branch per plant)
- Leaf herbivory (estimated for all expanded leaves on the focal branch and 30 randomly selected fully expanded leaves from other branches per plant)
- Nearby density of conspecific plants (number of all the reproductive individuals within a 3 m radius of the focal plant)
- Plant size (the length of the longest branch per plant)
Each plant was sampled only once. For interaction network analysis, we included only those plants that interacted with flowerhead visitors during observation (41 out of the 89 sampled).
For a detailed description of the sampling methods and effects, see Rezende et al. (2025). Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network. Journal of Ecology.
Files and variables
DATA & FILE OVERVIEW
-
File List:
- Rezendeetal_JEcol_dataset.txt - File containing data on Trichogoniopsis adenantha herbivory and flowerhead visitation (.txt file).
- Rezendeetal_JEcol_dataset.csv - File containing data on Trichogoniopsis adenantha herbivory and flowerhead visitation (CSV file).
(NOTE: the two data files above are identical and are provided in different formats to maximise accessibility) - Rezendeetal_JEcol_Rscript_network.R - R environment file containing code for the network analysis (See Rezende et al (2025). Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network. Journal of Ecology)
- Rezendeetal_JEcol_Rscript_SEM.R - R environment file containing code for the structural equation modelling (See Rezende et al (2025). Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network. Journal of Ecology)
-
Relationship between files, if important: N/A
-
Additional related data collected that was not included in the current data package: N/A
-
Are there multiple versions of the dataset? No
METHODOLOGICAL INFORMATION
-
Description of methods used for collection/generation of data: See Rezende et al (2025). Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network. Journal of Ecology.
-
Methods for processing the data: See Rezende et al (2025). Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network. Journal of Ecology.
-
Instrument- or software-specific information needed to interpret the data: See Rezende et al (2025). Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network. Journal of Ecology.
-
Standards and calibration information, if appropriate: N/A
-
Environmental/experimental conditions: N/A
-
Describe any quality-assurance procedures performed on the data: N/A
-
People involved with sample collection, processing, analysis and/or submission: Gabriela Gomes and João Suzigan helped with the field work. André V.L. Freitas, Viviane Ferro, João Robazzi and Sidnei Mateus helped with flowerhead visitor identification. Karoline Ceron helped with the R code for the interaction network figure.
DATA-SPECIFIC INFORMATION FOR: Rezendeetal_JEcol_dataset.txt
-
Number of variables: 47
-
Number of cases/rows: 90
-
Variable List:
- plant: individual plant identity
- month: sampling year and month
- hour: sampling hour interval
- conspecific_neighbors: number of reprodutive plants neighboring the focal plant
- plant_length_cm: length of the longest branch of the plant in centimeters (cm)
- reproductive_branches: number of reproductive branches of the plant
- plant_flowerheads: number of receptive flowerheads on the plant
- branch_flowerheads: number of receptive flowerheads on the focal branch
- damg_flowerheads: proportion of damaged flowerheads on the focal branch
- leaves: number of leaves sampled on the plant
- damg_leaves: proportion of damaged leaves regardless of herbivore guild
- damg_leaves_category: categorization of plants into less damaged (minor_damage:
- damg_leaves less than or equal to 0.55) and more damaged (major_damage: damg_leaves greater than 0.55)
- damg_ps: proportion of damaged leaves by piercer-suckers
- damg_ch: proportion of damaged leaves by chewers
- damg_mn: proportion of damaged leaves by miners
- damg_only_ps: proportion of damaged leaves only by piercer-suckers
- damg_only_ch: proportion of damaged leaves only by chewers
- damg_only_mn: proportion of damaged leaves only by miners
- damg_only_ps_ch: proportion of damaged leaves only by piercer-suckers and chewers
- damg_only_ps_mn: proportion of damaged leaves only by piercer-suckers and miners
- damg_only_ch_mn: proportion of damaged leaves only by chewers and miners
- damg_only_ps_ch_mn: proportion of damaged leaves only by piercer-suckers, chewers, and miners
- visit_total: number of flowerhead visits regardless of insect order
- visit_dip: number of flowerhead visits by all insects of the order Diptera
- visit_hym: number of flowerhead visits by all insects of the order Hymenoptera
- visit_lep: number of flowerhead visits by all insects of the order Lepidoptera
- visit_lep_ni: number of flowerhead visits by all insects identified only up to the order Lepidoptera
- visit_hym_ni: number of flowerhead visits by all insects identified only up to the order Hymenoptera
- visit_dip_ni: number of flowerhead visits by all insects identified only up to the order Diptera
- visit_nymphalidae_ni: number of flowerhead visits by all insects identified only up to the family Nymphalidae
- visit_halictidae_ni: number of flowerhead visits by all insects identified only up to the family Halictidae
- vists_aeria_olena_olena: number of flowerhead visits by Aeria olena olena
- vists_episcada_sylvo: number of flowerhead visits by Episcada sylvo
- vists_eumenogaster_sp: number of flowerhead visits by Eumenogaster sp
- vists_heterosais_edessa: number of flowerhead visits by Heterosais_edessa
- vists_ithomia_drymo: number of flowerhead visits by Ithomia drymo
- vists_nridela_acroxantha: number of flowerhead visits by Nridela acroxantha
- vists_pseudoscada_erruca: number of flowerhead visits by Pseudoscada erruca
- vists_rhynchopyga_sp: number of flowerhead visits by Rhynchopyga sp
- vists_sphecopsyche_sp: number of flowerhead visits by Sphecopsyche sp
- vists_plebeia_droryana: number of flowerhead visits by Plebeia droryana
- vists_pseudaugochlora_sp: number of flowerhead visits by Pseudaugochlora sp
- vists_trigona_hyalinata: number of flowerhead visits by Trigona hyalinata
- vists_syrphidae_sp1: number of flowerhead visits by Syrphidae sp1
- vists_syrphidae_sp2: number of flowerhead visits by Syrphidae sp2
- vists_syrphidae_sp3: number of flowerhead visits by Syrphidae sp3
- vists_syrphidae_sp4: number of flowerhead visits by Syrphidae sp4
-
Missing data codes: N/A
-
Specialized formats or other abbreviations used: N/A
Code/software
We performed all the network and statistical analyses using the R Environment (version 4.0.4; R Core Team 2021). Specifically, we used the following R package:
- igraph R package (version 2.0.3; Csardi & Nepusz, 2006): to build the interaction network and identified the shortest network distance
- stats R package (version 4.4.1), prcomp function: to perform the Principal Component Analysis (PCA)
- piecewiseSEM R package (version 2.3.0.1, Lefcheck, 2016): to perform the Structural Equation Modelling (SEM) analysis
- lme4 R package (version 1.1-35.4; Bates et al., 2015): to build the Generalised Linear Mixed Models (GLMMs)
- DHARMa diagnostics (Hartig, 2016) and vif function of the car R package (version 3.1-2): to assess base model suitability
We provide the commented R scripts for Structural Equation Modelling (SEM) and Interaction Network analysis at the Rezendeetal_JEcol_Rscript_SEM.R and Rezendeetal_JEcol_Rscript_network.R., respectively.
For a detailed description of the data analysis, see Rezende et al. (2025). Foliar herbivory pushes plant individuals towards the periphery of a plant-floral visitor interaction network. Journal of Ecology.
