Methods

Summary

Between May and September 2022, I will conduct surveys at ~10 New  Hampshire peatland sites. Each site will be sampled three times throughout the season, once in each season (Spring, mid-Summer, and late Summer). Both pan trapping and netting will be conducted to collect information on the bee assemblages. Pan trap bee bowls (7 cm diameter) will consist of white, fluorescent yellow,  and fluorescent blue souffle cups (ProPak©️ 4oz. clear polystyrene portion cups,  Scarborough, Ontario ) filled with soapy water (blue Dawn©️ dishwashing liquid)  (Droege, 2016). Following the USGS  Native Bee Inventory and Monitoring Lab’s standard protocol, bowls will be placed, alternating colors, every 5m along a transect running from the peatland up to 30m into forest edge (Droege, 2016). A single transect, or two in large sites, will aim to opportunistically circumvent open water in the style of Stephenson and  Dowling (2017), to minimize potential damage to delicate peatlands. To further reduce impact on delicate peat and account for traps over water/damp areas, the bee bowls will be supported by  36” garden stakes (Vigoro©️ Circular Plant Support,Northbrook, IL), the stakes will be inserted into the substrate to the point that bowls are no taller than the surrounding vegetation and still protected from any changes in water level (at least 10 cm above substrate). Bee bowls will be left out for 24 hours, and collection surveys will be conducted only on sunny, warm days. After 24  hours, bees will be removed from the traps, strained, and stored in 70% isopropyl alcohol. Site conditions such as cloud cover, temperature, humidity, and wind will be recorded each visit.  Collection will be on a per-bowl basis per transect, to account for distance from the forest edge. 

To accurately assess the greatest diversity and abundance of bees at each site, net sampling will also be used during each visit (Popic et al., 2013). Net sampling effort, modified from Lerman & Milam (2016), will consist of opportunistic sampling on flowers with 15-minute  surveys. Netting surveys will focus on rare and unique fen and wetland plant species if present.  Upon capture, netted bees will be placed in vials containing soapy water and will then be processed and stored in the same manner as bees collected from bee bowls prior to pinning for identification. GPS location, time, and host plant species will be recorded for all netted bees. 

To prevent the spread of invasive plants and pathogens boots and any field equipment that comes into contact with wetlands will be self-inspected and cleared of any deposits (mud, plants,  algae) following a modified USFWS “Clean.Drain.Dry” protocol. Boots/waders and bee bowl stakes will be placed in plastic shopping bags for transport. Upon return from the field,  equipment will be treated with a 3% bleach solution following the Northeast Partners for  Amphibian and Reptile Conservation Disinfection of Field Equipment to Minimize Risk of  Spread of Chytridiomycosis and Ranavirus protocol (February 2014). 

All bees will be washed, dried, pinned and labeled following LeBuhn et al. (2003). Bees will be identified to the species level when possible, using taxonomic keys (Mitchell 1960, 1962;  Gibbs 2010, 2011; Gibbs et al. 2012). Joan Milam (Univ. of Massachusetts) will assist with bee identification, and we will send additional bees of uncertain identification to other specialists for  ID. Specimens will be deposited to Antioch University (reference collection), Harvard’s  Museum of Comparative Zoology, or the American Museum of Natural History. Specimens collected at US Army National Guard sites may be deposited with the US Army Corps of  Engineers.  

In addition to records from opportunistic net sampling, I will also quantify vegetation within each peatland via point intercept vegetation sampling in the manner of James and Shugart  (1971). pH of any standing water will be assessed at each point along the transect (every 5m) once during the season and then be averaged to provide the average peatland pH of each site.

Challenges

In the previous season, the weather has proven the biggest hindrance. Remaining flexible and patient are the best ways to persevere in the face of persistent rain, flooding, etc.

Pre Analysis Plan

Data Analysis 

All statistical analyses will be performed in R (R Core Team 2020). We will use Generalized Linear Mixed Models to examine bee assemblage diversity and abundance (Shannon’s Diversity Index) among sites and peatland types, in relation to floral abundance/diversity data, and between the peatland and adjacent forest. We will account for potentially confounding predictor variables such as year, survey period (early, mid, late), and site via designation as a combination of fixed and random effects as appropriate (e.g. peatland  class=fixed, site=random, etc) (Harrison et al., 2018). Bee abundance, diversity, and richness will be examined with potential to also examine specific bee groups (Polylectic vs. Oligolectic,  Eusocial vs. Solitary, nesting type, size, species with high capture rates) via GLMMS in the style of Roberts et al. (2017). Statistical significance will be determined at alpha ≤ 0.05 unless otherwise noted. 

Protocols

This project has not yet shared any protocols.