Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/njlyon0/lyon-ms-thesis_bee-project

Native bee project exploration and analysis
https://github.com/njlyon0/lyon-ms-thesis_bee-project

data-science ecology

Last synced: 14 days ago
JSON representation

Native bee project exploration and analysis

Awesome Lists containing this project

README 

        
# Lyon MS Thesis - Native Bee Project



## Project Overview



This repository contains analysis of native bee and floral resource community data. These data were collected between May and August of 2017 and 2018 in the Grand River Grasslands (GRG) of Ringgold Co. Iowa and Harrison Co. Missouri. Data tidying are performed in a dedicated repository that tidies data across several other projects conducted in the GRG. The [tidying repository](https://github.com/njlyon0/lyon-ms-thesis_field-tidy) is set to "Private" but if you need access to it for some reason, please contact Nick Lyon to begin that conversation.



## Data Availability



While the data are embedded directly in this repository, the metadata is not as consistently recorded as I would like (sorry about that!). To help facilitate others in using these data, I have published the dataset on the [Environmental Data Initiative](https://environmentaldatainitiative.org/)'s (EDI) Data Portal with full metadata documentation. See [here](https://portal.edirepository.org/nis/mapbrowse?scope=edi&identifier=1210&revision=1) for the full dataset with metadata and citation information.



## Experimental Design



This project was meant to explore native bee response within two different management methods. For both of the sub-projects, each "site" (i.e., full pasture) was divided into three equal "patches" and this is the scale at which management intervention was applied.



Native bees were sampled via "bee bowls" (small cups with soapy water) mounted onto fence posts at 20m intervals (5 posts for 100m / patch) and identified to species-level *post hoc*. Given that both management methods affected vegetation height (see below for management description), each post had a bowl mounted at the top (~1m from ground) and at the bottom (~2cm from ground).



Bee bowls can be either blue, white, or yellow to appeal to the visual spectra of different bee species. In 2017, all bees collected in any bowl on a transect were added together and specific bowl metrics were not retained. In 2018--learning from this mistake--I preserved bowl metrics (e.g., color, volume, "retrieval status") for later consideration--likely as random effects in analysis.



Finally, I surveyed nectar-producing plants (floral resources) by counting every inflorescence within 1m radius of the bee bowl post (in 2018) or every inflorescence in a 100m x 1m transect along the transect where bowls were placed (in 2017).



2017 data are (briefly) described above, *but* given the improvements made to the methodology for the 2018 season, 2017 data are left largely unexplored here.



### Patch-Burn Grazing Sub-Project



Patch-burn grazing (PBG) is a method where on a given site, one third of the area is burned in the spring on a rotating basis such that no part of the site goes more than three years without a burn. This is meant to encourage cattle with free access to the entire site to forage primarily on the most recently burned "patch" which should lead to spatial heterogeneity at the site level and from there to increased diversity of resources for native flora and fauna.



In this sub-project, I was curious whether native bee communities differed among patches that had burned at different times. Vegetation height and bare ground visibility differ among patches so I hypothesized that bee species (and thus community assemblages) might also differ.



### Spray-and-Seed Sub-Project



In addition to the PBG sites, some sites were simply 'graze and burn' where the entire site is burned once every three years and cattle have free access to the site every year. Within these sites tall fescue (*Schedonorus arundinaceus*) was a problematic, invasive grass species that was managed against at the patch level. One patch was left as a control, another patch was sprayed with the herbicide glyphosate in fall 2014, and the final patch was both sprayed with glyphosate and seeded with a prairie plant seedmix the following spring.



Vegetation characteristics differ among these patches, and I hypothesized that native bee community might also differ in substantive ways.