Bees, Wasps, and Sawflies: Comparing the Efficacy of Two Collection Methods in Estimating Hymenopteran Diversity in Grasslands Morgan P

Home , Sawfly

Bees, wasps, and sawflies: comparing the efficacy of two collection methods in estimating hymenopteran diversity in grasslands Morgan P. Rondinelli1,2, Sam Droege3, Abigail A.R. Kula4, David R. Smith5 and Robert R. Kula5 1Department of Entomology, Smithsonian Institution, National Museum of Natural History, Washington, DC, 2University of Michigan, Ann Arbor, MI, 3USGS, Beltsville, MD, 4Department of Science, Mount St. Mary's University, Emmitsburg, MD, 5Systematic Entomology Laboratory, USDA-ARS, Washington, DC

Introduction Materials and Methods Results

Bee Species Accumulation: Elevated Bee Species Accumulation: Ground • Grasslands are among the most endangered • Sampled sites in northern Virginia 90 90

80 80

cie 70 cie 70

• Jones Nature Preserve: 32 ha ecosystems in North America due to conversion to 60 60 50 50 of Sp of Sp

crop fields or habitat loss through urbanization. primarily warm-season grass (WSG) r 40 r 40 30 30 • Documenting biodiversity in grasslands is essential meadow, burned and mowed mb 20 mb 20 Nu 10 Nu 10 • Oxbow Farm: three 1.5 ha primarily for successful conservation and restoration. 284.5 569 853.5 1138 1422.5 1707 232.67 465.33 698 930.67 1163.33 1396 • Pan traps are small, yellow bowls filled with soapy WSG meadows, mowed Number of Individuals Number of Individuals water used to estimate hymenopteran diversity. • Sampled every two weeks April- Sawfly Species Accumulation: Elevated Sawfly Species Accumulation: Ground 35 12

• Compared efficacy of pan traps placed on ground or October 2014; 6 hr/sampling event 30 10 cie cie

elevated fixed distance from ground for estimating • Three pairs of transects at each site; 20 of Sp of Sp r 15 r

bee, sawfly/wood wasp, and braconid wasp diversity each transect contained 15 yellow 10 Figure 2. elevated pan traps (A); ground pan traps (B). mb mb

• pans placed on ground or elevated Nu Nu These groups represent pollinator (bees), herbivore 1.22 m from ground 16.5 33 49.5 66 82.5 99 3.5 7 10.5 14 17.5 21 (sawflies), and parasitoid (braconid wasps) guilds; Number of Individuals Number of Individuals • Pans within one transect were pooled; 12 provide services critical to ecosystem function Braconid Species Accumulation: Elevated Braconid Species Accumulation: Ground samples/sampling event 60 90

80 50

cie cie 70

• Specimens identified to genus and determined 40 60 50

of Sp 30 of Sp

to species or sorted into morphospecies r r 40 20 30 •  mb mb 20 ANOVA ( =0.05) and diversity metrics using 10 Nu Nu 10 RStudio and EstimateS 9.1.0 11.17 22.33 33.5 44.67 55.83 67 32.83 65.67 98.5 131.33 164.17 197 Figure 3. Setting out pan traps. Number of Individuals Number of Individuals Photo by Morgan Rondinelli Figure 5. Species accumulation curves for each guild and treatment. Lines represent S(est) 95% CI Upper Bound, S(est), and S(est) 95% CI Lower Bound. Results • Collected 3103 bee, 120 sawfly, and 264 braconid specimens Discussion

Photo from Photo by Dave Smith Tom Murray • Bees: species richness differed Species Richness: • Elevated pans should be sufficient if only interested in significantly between treatments Elevated vs. Ground Pan Traps sampling to estimate bee diversity. Figure 1. Examples of a bee (A), sawfly (B), and braconid wasp (C). elevated ground • Ground-placed and elevated pans should be used (P=0.0124); diversity was similar between ed treatments (Morista-Horn=0.76) ct 100 when sampling breadth of hymenopteran taxa; both lle co 80 are necessary for sampling braconids. • Sawflies: species richness differed 81 • Species collected not saturated so further sampling is Objectives 73 70 significantly between treatments species 60 necessary. (P=0.0178); similarity in diversity was • Part of broader project assessing efficacy of several 40 46 moderate (Morista-Horn=0.56)

methods (i.e., pan trap, Malaise trap, sweep net) for number of 20 25 sampling hymenopterans al Acknowledgements ot

• Braconids: species richness did not differ • Goal: discern which method, or combination of I thank Bruce Jones (Jones Nature Preserve) and Beatrice von Gontard (Oxbow Farm) for significantly between treatments allowing sampling on their property. Wayne Boo, Elizabeth Garcia, and Dejen Mengis methods, most accurately estimates bee, sawfly, and bees sawflies braconids (technicians, USGS bee lab) processed specimens. Danielle Tarry sorted samples, Marion (P=0.343); diversity was dissimilar Leménager helped with specimen identification, and Noah Winters provided statistical braconid diversity in grasslands Figure 4. Total number of species collected support (Interns, Smithsonian Institution). I am grateful to Dr. Elizabeth Cottrell and Dr. between treatments (Morista-Horn=0.14) per guild, separated by treatment. Gene Hunt (Directors, Smithsonian Institution NHRE program), and Ms. Virginia Power (NHRE program Administrator) for mentorship and professional advice.