TY - JOUR

T1 - Scheimpflug lidar range profiling of bee activity patterns and spatial distributions

AU - Rydhmer, Klas

AU - Prangsma, Jord

AU - Brydegaard, Mikkel

AU - Smith, Henrik G.

AU - Kirkeby, Carsten

AU - Kappel Schmidt, Inger

AU - Boelt, Birte

N1 - Funding Information: This study was supported by a grant by Idagaardfonden, Denmark, Innovation Foundation, Denmark, the Swedish Research Council, Norsk Elektro Optikk AS, Norway, Formas Sweden and 15. Juni and Aage V. Jensen Nature Foundations, Denmark. Publisher Copyright: © 2022, The Author(s).

PY - 2022/12

Y1 - 2022/12

N2 - Background: Recent declines of honeybees and simplifications of wild bee communities, at least partly attributed to changes of agricultural landscapes, have worried both the public and the scientific community. To understand how wild and managed bees respond to landscape structure it is essential to investigate their spatial use of foraging habitats. However, such studies are challenging since the foraging behaviour of bees differs between species and can be highly dynamic. Consequently, the necessary data collection is laborious using conventional methods and there is a need for novel methods that allow for automated and continuous monitoring of bees. In this work, we deployed an entomological lidar in a homogenous white clover seed crop and profiled the activity of honeybees and other ambient insects in relation to a cluster of beehives. Results: In total, 566,609 insect observations were recorded by the lidar. The total measured range distribution was separated into three groups, out of which two were centered around the beehives and considered to be honeybees, while the remaining group was considered to be wild insects. The validity of this model in separating honeybees from wild insects was verified by the average wing modulation frequency spectra in the dominating range interval for each group. The temporal variation in measured activity of the assumed honeybee observations was well correlated with honeybee activity indirectly estimated using hive scales as well as directly observed using transect counts. Additional insight regarding the three-dimensional distribution of bees close to the hive was provided by alternating the beam between two heights, revealing a “funnel like” distribution around the beehives, widening with height. Conclusions: We demonstrate how lidar can record very high numbers of insects during a short time period. In this work, a spatial model, derived from the detection limit of the lidar and two Gaussian distributions of honeybees centered around their hives was sufficient to reproduce the observations of honeybees and background insects. This methodology can in the future provide valuable new information on how external factors influence pollination services and foraging habitat selection and range of both managed bees and wild pollinators.

AB - Background: Recent declines of honeybees and simplifications of wild bee communities, at least partly attributed to changes of agricultural landscapes, have worried both the public and the scientific community. To understand how wild and managed bees respond to landscape structure it is essential to investigate their spatial use of foraging habitats. However, such studies are challenging since the foraging behaviour of bees differs between species and can be highly dynamic. Consequently, the necessary data collection is laborious using conventional methods and there is a need for novel methods that allow for automated and continuous monitoring of bees. In this work, we deployed an entomological lidar in a homogenous white clover seed crop and profiled the activity of honeybees and other ambient insects in relation to a cluster of beehives. Results: In total, 566,609 insect observations were recorded by the lidar. The total measured range distribution was separated into three groups, out of which two were centered around the beehives and considered to be honeybees, while the remaining group was considered to be wild insects. The validity of this model in separating honeybees from wild insects was verified by the average wing modulation frequency spectra in the dominating range interval for each group. The temporal variation in measured activity of the assumed honeybee observations was well correlated with honeybee activity indirectly estimated using hive scales as well as directly observed using transect counts. Additional insight regarding the three-dimensional distribution of bees close to the hive was provided by alternating the beam between two heights, revealing a “funnel like” distribution around the beehives, widening with height. Conclusions: We demonstrate how lidar can record very high numbers of insects during a short time period. In this work, a spatial model, derived from the detection limit of the lidar and two Gaussian distributions of honeybees centered around their hives was sufficient to reproduce the observations of honeybees and background insects. This methodology can in the future provide valuable new information on how external factors influence pollination services and foraging habitat selection and range of both managed bees and wild pollinators.

KW - Entomology

KW - Honeybees

KW - Landscape ecology

KW - Lidar

KW - Pollination

KW - Remote sensing

U2 - 10.1186/s40317-022-00285-z

DO - 10.1186/s40317-022-00285-z

M3 - Journal article

AN - SCOPUS:85128483871

VL - 10

JO - Animal Biotelemetry

JF - Animal Biotelemetry

SN - 2050-3385

IS - 1

M1 - 14

ER -