Wild bees in temperate deciduous forests: how does management affect bee communities, and can we use active remote sensing data to model bee biodiversity?

Chase, Marissa Helene

Permalink

https://hdl.handle.net/2142/122108 Copy

Description

Title

Wild bees in temperate deciduous forests: how does management affect bee communities, and can we use active remote sensing data to model bee biodiversity?

Author(s)

Chase, Marissa Helene

Issue Date

2023-11-14

Director of Research (if dissertation) or Advisor (if thesis)

• Fraterrigo, Jennifer
• Harmon-Threatt, Alexandra

Doctoral Committee Chair(s)

Benson, Thomas J

Committee Member(s)

Diao, Chunyuan

Department of Study

Natural Res & Env Sci

Discipline

Natural Res & Env Sciences

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• wild bees
• forest management
• pollination services
• functional traits
• LiDAR

Abstract

Forests worldwide support insect biodiversity and the critical ecosystem services provided by insects, such as pollination. In many temperate deciduous forests, long-term management suppression has led to mesophication, resulting in a habitat with homogenous composition and structure. As forest health continues to decline, management approaches that emulate historical disturbance regimes (e.g., prescribed fire) have the potential to restore habitat heterogeneity, but effects on biodiversity remain poorly understood, especially for insects. Specifically, the ramifications of forest management practices for key mutualists like bees are unclear and rarely considered apart from other pollinators. Moreover, most bee studies in temperate deciduous forests do not consider how seasonality and management may interact to affect bee functional and taxonomic composition. To address this knowledge gap, I first conducted a meta-regression analysis to determine which bee functional traits have the strongest effect on pollination. I found that bee functional traits have a significant effect on pollination, but data were insufficient, and effect sizes were weak with high variability across studies. I used findings from this study to investigate the functional response of bee communities to temperate deciduous forest management practices that emulate historical disturbances (burning, thinning, and the combination of the two). I found that different management strategies aligned with specific bee functional traits, and these relationships varied throughout the year. Managed and unmanaged areas differed in the distribution of nesting and floral resources and, as a result, altered bee functional trait composition. Thinned plots with increased floral resources promoted smaller bees with limited dispersal capacity, in addition to primitively eusocial and cavity-excavating bees. Unmanaged habitat supported vulnerable bee groups such as cleptoparasites and specialists. Next, I explored how these same practices affect bee community composition and structure and determined which environmental and resource variables best predicted bee diversity and abundance. I found that bee communities varied across management types, with more intensely managed plots (thin+burn) consistently having the highest bee diversity and abundance. On average, more intensely managed plots had three times as many bees in comparison to unmanaged plots. Comparing spring to summer, I found that different floral and nesting resources predicted bee diversity and abundance. Overall, these studies reinforce the importance of using a range of forest management strategies to enhance bee taxonomic and functional diversity and associated ecosystem services. However, I also found that management affected bee communities by altering unmeasured abiotic variables. With this finding, in my last chapter, I set out to determine how forest structural complexity affects wild bee communities. Specifically, I wanted to determine whether remote sensing (light detection and ranging [LiDAR]-derived structural metrics) could predict bee diversity, abundance, and bee functional composition in managed forests. Given that forest canopy structure influences insect biodiversity, I was surprised to find that structural complexity, as measured by LiDAR, did not capture variation in bee community structure and composition, either functionally or taxonomically. I therefore identified the LiDAR-derived structural metrics that were best at predicting bee diversity and abundances across seasons and implore further research to study the link between active remote sensing technologies and modeling bee biodiversity. These findings provide data on the relationship between forest management and wild bee communities and explore ways in which bee functional and taxonomic composition can be measured and modeled across temperate forests.

Graduation Semester

2023-12

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/122108

Copyright and License Information

Copyright 2023 Marissa Chase

Owning Collections