Data from: Evaluating LiDAR-derived structural metrics for predicting bee assemblages in managed forests
Data files
Mar 25, 2025 version files 16.36 KB
-
BeeSpeciesTraits.csv
2.28 KB
-
LiDARStructuralMetrics.csv
8.78 KB
-
README.md
4.42 KB
-
site_traits_full_21_22_abundance_match_LiDAR_data.csv
886 B
Abstract
Aim: Globally, insects depend on forest habitats for shelter from disturbances and critical nesting and floral resources. Forest structural complexity can affect the distribution of these resources and likewise alter insect assemblages within forests. Despite the importance of temperate deciduous forests for bees and their outsized contribution to pollination services within forests and beyond, the relationship between forest structure and bees has received scant attention. This is especially true in managed temperate deciduous forests, where management strategies alter forest structural complexity and may therefore affect bee communities.
Location: Illinois, United States of America
Methods: We investigated whether structural metrics derived from light detection and ranging (LiDAR) data could predict bee diversity and abundance, as well as bee functional trait composition within managed forest lands. We addressed three specific questions: 1) How does forest management affect structural complexity; 2) Can structural metrics predict bee diversity and abundance in spring and summer; and 3) How are structural metrics related to bee functional trait composition?
Results: We found that LiDAR-derived structural metrics could not differentiate between management types and were weak predictors of bee diversity and abundance and bee functional trait composition. Metrics related to the understory and midstory vegetation structure showed the strongest association with forest bee community patterns. Specifically, vegetation density in the understory (0 - 2 m) had a positive effect on bee diversity and abundance in spring, while in summer, vegetation density in the mid-canopy (2 - 5 m) negatively affected bee communities.
Main conclusions: Our findings suggest mid- and understory vegetation structure may have an important influence on forest bee communities. Future studies should focus on the structural elements of these forest strata to improve understanding of how structural complexity influences bee communities within managed forests and evaluate the potential for using LiDAR-derived structural metrics to monitor and predict biodiversity patterns.
[https://doi.org/10.5061/dryad.j6q573nq2](https://doi.org/10.5061/dryad.j6q573nq2)
Description of the data and file structure
We investigated whether structural metrics derived from light detection and ranging (LiDAR) data could predict bee diversity and abundance, as well as bee functional trait composition within managed forest lands. We addressed three specific questions: 1) How does forest management affect structural complexity; 2) Can structural metrics predict bee diversity and abundance in spring and summer; and 3) How are structural metrics related to bee functional trait composition?
Files and variables
File: site_traits_full_21_22_abundance_match_LiDAR_data.csv
Variables
* site: site name
* polylectic: lecty - generalist (# of generalists)
* oligolectic: lecty - specialist (# of specialists)
* cleptoparasitic: whether a bee is a cleptoparasite (# of cleptoparasitic bees)
* non-cleptoparasitic: whether a bee is a cleptoparasite (# of non-cleptoparasitic bees bees)
* solitary: sociality (# of solitary bees)
* subsocial: sociality (# of subsocial bees)
* social: sociality (# of social bees)
* ground-excavator: nesting (# of ground-excavating bees)
* cavity-excavator: nesting (# of cavity-excavating bees)
* cavity-renter: nesting (# of cavity-renting bees)
* diverse-nesting: nesting (# of diverse-nesting bees)
File: BeeSpeciesTraits.csv
Variables
* species: bee species
* body size: intertegular span measurement in millimeters
* lecty: specialization designation (oligolectic or polylectic)
* nesting: nesting strategy (ground, cavity-renting, cavity-excavating, diverse)
* parasitism: cleptoparsitic bee or non-cleptoparasitic bee
* sociality: sociality (solitary, subsocial, primitively eusocial)
* scopa location: # of distinct locations on a bee’s body where a scopa is located
* tongue length: tongue length of a bee in millimeters
* flight phenology: # of a months a bee is foraging on the landscape as an adult
** NA values in dataset do not have associated data
File: LiDARStructuralMetrics.csv
Variables
* ORIG_FID: original unique identifier
* FID_: unique identifier after GIS analyses
* plot: plot #
* site: site name
* lat: latitude
* long: longitude
* treatment: management type
* bee_div: bee diversity overall
* bee_div_spring: spring bee diversity
* bee_div_summer: summer bee diversity
* bee_abun: bee abundnace
* bee_abun_spring: spring bee abundance
* bee_abun_summer: summer bee abundance
* Dist_Ag_KM: distance to ag in kilometers
* total_bloom_spring: total bloom abundance in spring
* total_bloom_summer: total bloom abundance in summer
* total_bloom: total bloom abudnance
* Perc_Under: percentage of vegetation in understory
* Perc_Can_1: percent vegetation in the canopy
* Perc_Mix: percent of vegetation in both the understory and canopy, as well as mid-story
* VHmean_new: mean vertical height in meters
* VHsd_new: standard deviation of vertical height
* Dist_Ag_m: distance to ag in meters
* Perc_Canop: percent canopy
* Perc_Gap: percent of canopy gaps
* VD_CV: vegetation density coefficient of variation
* VD_MEAN: mean vegetation density
* VH_CV: vegetation height coefficient of variation (meters)
* VH_Mean: mean vegetation height in meters
* DEM_CV: DEM coefficient of variation
* DEM_MEAN: mean DEM
* VD_2m_CV: vegetation density 0-2 meters coefficient of variation
* VD_2m_MEAN: mean vegetation density 0-2 meters
* VD_2_5m_CV: vegetation density 2-5 meters coefficient of variation
* VD_2_5m_MEAN: mean vegetation density 2-5 meters
* VD_5_15m_CV: vegetation density 5-15 meters coefficient of variation
* VD_5_15m_MEAN: mean vegetation density 5-15 meters
* VD_Above_15m_CV: vegetation density above 15 meters coefficient of variation
* VD_Above_15m_MEAN: mean vegetation density above 15 meters
Code/software
Statistical analyses were conducted with R 4.2.1 (R Core Team 2022).
Access information
Other publicly accessible locations of the data:
* Illinois Geospatial Data Clearinghouse
Data were derived from the following sources:
* Illinois Geospatial Data Clearinghouse