These data were used as part of a study analyzing the patterns of changing phenology and abundance. Species day of year observations were analyzed using quantile regression (of the 0.1 and 0.9 quantiles) of over years. Species abundance trends were measured using list length analysis, the result being the annual change in the proportion of trips on which a species was observed. Both phenology and abundance metrics were correlated and analyzed to understand thier potential relationship. Please refer to the manuscript, appendix, and metadata for more information.

Data were collected by skilled volunteer observers alone or in small groups. The number of each species seen on each outing was recorded, as was the date, observers, and location. Each file contains data for a species of butterfly which can be identified by its four letter code (e.g. 'MONA' = Monarch Danaus plexippus: see 'SPECIESCODES.csv', which includes the code, common name, scientific name, that the species was included or combined with another due to similarities or changes in reporting, classifications for voltinism and range type relative to the latitude of Boston, MA).

Each trip was assigned retroactively. A 'trip' was considered a single observation, or set of observations, made by a single person, or the same set of people, on a single day in a single year at a certain location (within a single township). Each row of each file consitutes a single trip and for each trip the following are included:

'listlength', the list length of a trip (i.e. the total number of species were observed on this trip)

'present', whether the species was observed on that trip

'listyear', the year of the trip

'yearday', the numeric day of year of the trip (i.e. 1 = Jan. 1st, and 365 = Dec. 31st)

'listplace', a numeric code corresponding to the township in which the trip occurred (see 'TOWNSHIPS.csv' which has only the towns and the corresponding codes)

Data were cleaned prior to processing though there remain some 'NA' values.