Does the munch affect the bunch? Using community science to explore insect herbivory and fruit production in an understory plant
Data files
Jun 03, 2025 version files 118.74 KB
Abstract
Digital, community-sourced natural history records are valuable for understanding species attributes such as phenology and geographic distribution. When these records include photographs, they can also be analysed for individual phenotypes and species interactions to develop or test ecological hypotheses. Here, we use observational and experimental approaches to assess how insect herbivory affects reproductive success in a widespread forest plant, bunchberry (Cornus canadensis). We queried the Global Biodiversity Information Facility (GBIF) and assembled a dataset of 2,578 photographic records of fruiting plants. Of these, 891 showed evidence of insect herbivory, but herbivory was not significantly associated with fruit production. In a field study we monitored 200 plants over five weeks. Herbivory was widespread (78% of plants showed insect feeding), but damage was generally low—only 5% of plants experienced herbivory ≥40% of total leaf area. No relationship was found between natural herbivory and fruit production. In a second study, we simulated high herbivory by mechanically removing 40% of leaf area in early and mid-season. Early-season herbivory reduced fruit production by 100%, while mid-season herbivory reduced fruit production by 87.3%. These results suggest that although high herbivory early in the season can substantially reduce fruiting, natural levels of herbivory exert limited top-down control on reproduction in bunchberry. By combining large-scale community-sourced records with controlled field experiments, this study demonstrates the value of mixed-methods approaches for testing ecological hypotheses and gaining insight into the processes that shape plant–insect interactions.
Dataset DOI: 10.5061/dryad.w6m905r15
Description of the data and file structure
All data is provided as CSV files.
GBIFSurvey data was visually recorded as herbivore ID, internal and external herbivory intensity, and fruit count from photos accessed through the GBIF database (https://doi.org/10.15468/dl.cj78zz)
LongitudinalFieldSurvey data was collected over 5 weeks spanning flowering to fruit ripening. External and internal herbivory intensities were visually estimated, and fruit counts were taken at the end.
SimulatedHerbivory data was collected as a fruit count after fruit ripening, following the manual defoliation of the two treatments. This was a 40% defoliation, performed during flowering for one treatment and during fruit-set for the second treatment.
Files and variables
File: LongitudinalFieldSurvey_CumulativeHerbivory_FruitCounts.csv
Description: Cumulative herbivory intensities after a five-week survey of 200 ramets, with paired fruit counts.
Variables
- herbivory: Type of herbivory, denoted as External, Internal (leaf mining or galling), Both (both external and internal herbivory), and None
- intensity: Percentage of leaf area herbivorized at the time of fruit count, visually estimated
- fruit_no: Number of fruits counted on the plant once ripe
File: SimulatedHerbivory_FruitCounts.csv
Description: Number of fruits recorded for plants used in a simulated herbivory test with two defoliation timings. Missing data (blank cells) indicates that the plant incurred herbivory during the fruiting period and was discarded.
Variables
- During flowering: Number of fruits counted on the plants once ripe. Plants were manually defoliated of 40% of leaf area during the flowering period.
- Early fruit-set: Number of fruits counted on the plants once ripe. Plants were manually defoliated of 40% of leaf area during the early fruit-set period.
- No stimulated herbivory: Number of fruits counted on the plants once ripe. Plants were not manually defoliated.
File: GBIFSurvey_InsectID_HerbivoryRecords_FruitCounts.csv
Description: Data visually recorded from community-science sourced photos, including the identity of the internal herbivore, the number of fruits, and the intensity of herbivory
Variables
- herbivory: Type of herbivory, denoted as External, Internal (leaf mining or galling), Both (both external and internal herbivory), and None
- external_intensity: Percentage of leaf area herbivorized by an external herbivore (does not leave a visible mine or gall), visually estimated
- internal_intensity: Percentage of leaf area herbivorized by an internal herbivore (mine or gall), visually estimated
- intensity: Percentage of leaf area herbivorized, visually estimated (external_intensity + internal_intensity)
- fruit_no: Number of fruits counted per plant
- place: State of observation, if observed in a location where the species is conservation listed as Threatened or above. Blank cell indicates the plant is not in a threatened range.
- internal_herb_id: Visually identified internal herbivore. Recorded to species or in a broader group. Some plants include multiple IDs. Acronyms identified below:
- PA: Phytomyza agromyzina
- AF: Antispila freemani
- AC: Antispila cornifoliella
- PS: Parallelodiplosis subtruncata
- LR or LF: Family Tortricidae moth (suspected)
- CT: Caloptilia sp.
- CO: Coleophora sp.
File: LongitudinalFieldSurvey_HerbivoryRecords_FruitCounts.csv
Description: Weekly data for the five-week survey of ramets. Herbivory was recorded weekly as visual estimates, and fruits were counted upon ripening. Trampled or mechanically damaged ramets were discarded. Some ramets were unincluded temporarily due to errors in measurement, these are represented as ND and the ramets were reincluded if those weeks could be reasonably interpolated by surrounding data.
Variables
- Plant: Identifier of the plant
- Herb.W1: External herbivory visually estimated on week 1 (18 June 2023), recorded as percentage of leaf area herbivorized
- Mine.W1 : Internal herbivory visually estimated on week 1 (18 June 2023), recorded as percentage of leaf area herbivorized
- Herb.W2: External herbivory visually estimated on week 2, recorded as percentage of leaf area herbivorized
- Mine.W2: Internal herbivory visually estimated on week 2, recorded as percentage of leaf area herbivorized
- Herb.W3: External herbivory visually estimated on week 3, recorded as percentage of leaf area herbivorized
- Mine.W3: Internal herbivory visually estimated on week 3, recorded as percentage of leaf area herbivorized
- Herb.W4: External herbivory visually estimated on week 4, recorded as percentage of leaf area herbivorized
- Mine.W4: Internal herbivory visually estimated on week 4, recorded as percentage of leaf area herbivorized
- Herb.W5: External herbivory visually estimated on week 5, recorded as percentage of leaf area herbivorized
- Mine.W5: Internal herbivory visually estimated on week 5, recorded as percentage of leaf area herbivorized
- Berries: Count of ripe fruits on the ramet
- Veg: Count of new vegetative shoots observed on the ramet
Code/software
Data was reviewed and analyzed in R Studio, but can also be reviewed in Excel or similar programs for other uses. It is uploaded as CSV files.
R Core Team. (2025). R: A Language and Environment for Statistical Computing, https://www.R-project.org/
Access information
Other publicly accessible locations of the data:
-
https://doi.org/10.15468/dl.cj78zz
Community-sourced data was recorded from iNaturalist observations accessed through the above occurrence download. These records are scrubbed of identifying information, and only the visually-collected data has been presented in this Dryad dataset.
Data was derived from the following sources:
- GBIF.org (10 June 2024) GBIF Occurrence Download. https://doi.org/10.15468/dl.cj78zz.