USDA LTAR Common Experiment measurement: Butterfly diversity and abundance
Nick M. Haddad, Jamie D. Smith
Long-Term Agroecosystem Research
LTAR
USDA LTAR
Common Experiment
butterflies
citizen scientists
pollinators
insect abundance
surveys
Disclaimer
This research is a contribution from the Long-Term Agroecosystem Research (LTAR) network. LTAR is supported by the United States Department of Agriculture. The use of trade, firm, or corporation names in this publication is for the information and convenience of the reader. Such use does not constitute an official endorsement or approval by the United States Department of Agriculture or the Agricultural Research Service of any product or service to the exclusion of others that may be suitable. USDA is an equal opportunity provider and employer.
Abstract
We will use a measure of butterfly abundance and diversity collected using a method standard across the US and the globe (Wepprich, et al. 2019). This method is used in regional studies, for example, the states of Ohio, Illinois, and Michigan in the US and the countries of the UK, the Netherlands, and Spain. Within these regions across hundreds of sites, data collection is performed by scientists and citizen scientists. Butterflies are the easiest-to-measure indicator of pollinator and insect abundance; metrics from large-scale butterfly surveys have proven consistent with long-term responses of pollinators and other insects across the globe (Van Klink, et al. 2020). The recommended method is easily employed by people not trained with insects and is not burdensome in terms of time. For every replicate, it involves walking a 1 km line transect at least once a month and recording the number and species of butterflies within a set distance of the transect.
Before start
This protocol has been altered from https://faculty.jmcl.wwu.edu/Wildlife/bfly_walk.pdf and https://ukbms.org/methods. The presence and abundance of butterflies are sampled using transect counts, modified from Pollard (1977).
Steps
Pollard walks
Transect routes are established as appropriate for each site.
The recommended transect length is 1 km and the walking duration is 1 hour and 30 minutes.
Transects are marked before surveys commence, and the same transect is sampled throughout the research effort.
Sampling transects are surveyed by walking from one end to the other without doubling back. Walk at a steady pace.
Observers record butterflies seen within a 5 m radius on both sides and above them while walking a transect at a slow and steady pace (see the figure embedded below).
#尊敬的用户，由于网络监管政策的限制，部分内容暂时无法在本网站直接浏览。我们已经为您准备了相关原始数据和链接，感谢您的理解与支持。
https://lh7-rt.googleusercontent.com/docsz/AD_4nXeKHLIGzyBoRj2vuJ01MDvDviEdRqERD4LbzQBP4n0iWSXyUbQxMMChpJXnwwFD6Yv3SzoOEWmtWg2AzguR4HqnKi7IANFuLAZNNfKjyHMGcG1xqBNhs4UjGDaJpuwrgA7cD5Mvqm4zD6-kF_4fk2guxF2Dc5DhAEg0sVBB6iInl0aNF5jLj4g?key=xNFUDOQA8iujL4NlwyrOSw
To avoid double-counting, butterflies behind the observer are not counted.
Species identifications are made visually, using binoculars when needed.
Record species or higher taxa using common names and keep a tally per butterfly taxon.
Record each individual once only. If the survey pauses to capture, identify, photograph, or take notes on a butterfly, stop the timer.
Transects are walked during midday hours (10:00 am − 4:00 pm) when it is warm and insects are flying.
Sampling is restricted to relatively calm conditions and times when air temperature exceeds 11°C in sunny weather or 17°C in cloudy weather.
Ideally, treatments are sampled weekly, and LTAR Sites should aspire to this frequency of sampling.
Samples will occur during the growing season, with samples beginning 14 days after the first butterflies are observed.
The rate at which plots or fields are sampled within an LTAR Site will be consistent across all treatments and replicates.
The rate may vary across LTAR Sites for multiple reasons (e.g., staffing or the nature of additional questions being addressed).
Measures will still be comparable across sites using standard meta-analytic approaches.
Pollard walks - Covariate metrics
Immediately before surveying each transect, record the date, plot (treatment, replicate), surveyor initials, and estimated cloud cover (%).
Obtain temperature from weather station records recorded on-site or nearby.
Record the time of day and start the stopwatch at the beginning of each survey.
Collect data needed on plant diversity using the USDA LTAR Common Experiment measurement: Cropland plant diversity protocol.
Pollard walks - Location
All replicates of prevailing and alternative treatments.
Pollard walks - Species identification
Butterflies are a strong indicator for other insects because they are readily identified on the wing.
Data collectors are trained in one survey day by a lead investigator knowledgeable about butterfly species of the region.
Most species can be identified and learned quickly using a field guide. Some species, especially skippers, can be identified on the wing but with some difficulty. For these species, individuals can be captured and released or can be photographed.
Calculations
An index of abundance is calculated by:
- Calculating the mean count per transect each week, and then
- Summing the mean weekly counts (Pollard 1977).
Quality assessment and quality control
People conducting surveys will be trained in butterfly identification.
This training will occur through field guides and focus on the most common butterflies in a site’s experiment. Then, a new investigator will accompany a trained investigator on at least one day of butterfly surveys until the trained investigator knows the new person can identify butterflies.
Datasheets will be checked for completeness when submitted at the end of a survey day and then entered into a database.
A different person will conduct a second review, checking the entered data against the datasheet.
Finally, data will be reviewed for accuracy. This review includes any necessary adjustments to species/taxa identifications based on spelling errors, the likelihood of species occurrence in a given habitat or geographic location, and difficult identification scenarios, such as species challenging to distinguish morphologically or known to hybridize. This review also includes a check for possible miscounts or typos. For example, if a species typically occurring at low abundance has a very high count in a survey, the reviewer should check the count against the original datasheet.
Recommendations for data collection
Table 1. Summary of recommendations for measuring butterfly abundance and diversity.
| A | B | C |
|---|---|---|
| Attribute | Preferred | Minimum |
| Spatial scale | All prevailing and alternative treatments | |
| Frequency | Once per week | Once per month |
| Covariate metrics | Temperature, cloud cover, plant diversity, plant abundance by species |
