Woody phenology and weather data related to Trelease Woods, Urbana, IL, USA 1883-1912, 1949-1964, and 1993-2023

In this study, weekly phenological observations throughout spring and autumn were made annually over 31 years for 22 woody plant taxa of multiple growth forms in Trelease Woods, a mature old-growth deciduous forest remnant near Urbana, Illinois, USA. The growth forms included canopy trees, saplings, sub canopy treelets, shrubs, and vines. Data collection began in the spring of 1993 and continued through the end of autumn 2023. The phenological data set is paired with weather data collected from a nearby weather station in Champaign, Illinois. These two data sets were used to determine if dates of phenological events and durations of phenophases were changing over time, possibly in connection with changing weather related to global climate change. To supplement these phenological data, two older phenological data sets are made available here. First, Charles Smith, Woods Custodian at the University of Illinois, made comparable observations from 1949 to 1964 of seven of the same phenological events on six of the same woody species. Data were collected from Trelease Woods and Brownfield Woods. The latter is an old-growth forest separated by 2 km from Trelease Woods, and the two sites were connected in a larger forest before the mid-19th century. Second, Thomas Mikesell, a farmer from Ohio, USA, made comparable observations from 1883 to 1912 of six of the same phenological events on 11 of the same woody species. Mikesell’s data were collected in a deciduous forest near Wauseon, Ohio, USA. Because of the distance between the Mikesell study site and the site of the current study in Urbana, Illinois, to compare the two data sets, adjustments were made to the event dates in the Mikesell data set based on bioclimatic laws. Comparison of the current data to those collected by Charles Smith and Thomas Mikesell provides evidence as to whether the trends in phenological event dates observed from 1993 to 2023 held consistent when investigated at longer time scales.

Methods
Field observations of spring phenology of mature individuals of 22 woody taxa (21 species and 1 genus which may have represented multiple species) were made in Trelease Woods, Champaign Co., Illinois, USA from 1993-2023. Also, observations were made of saplings of three of the canopy tree species. The study site is the north half of a 24.5 ha fragment of temperate mature deciduous forest, dominated by sugar maple near Urbana, Illinois, USA.  Elevation varies by less than 5 m across the study area.

The phenological status of each individual was recorded by the same observer (CKA) weekly from February to June of each year. Measurements were recorded and later analyzed, at the individual plant level. Within each of three vegetative phases, Bud Swell (BS), Bud Burst (BB), and Leaf Expansion (LE), three dates were noted (e.g., BS1 (1/3 of units exhibit event), BS2 (2/3 completed event), BS3 (all completed event), each date being one event); the exception was flowering for which only beginning (FL1) and ending of flowering (FL2) were recorded. Species classified as Late Flowering were not observed in all years.  Middle events (e.g., BS2) were not used in any analyses but were critical in determining, by simple linear interpolation, estimated dates of non-observed events that passed in the one week between observations.

From these data, durations (number of days) of phenophases for each individual were calculated: Bud Swell, Bud Burst, Leaf Expansion, Flowering, and Overall Spring Phenology.

Species were analyzed together in seasonal groups. For most events, species were grouped as early or late. Aesculus glabra (canopy trees and saplings) were usually analyzed as a separate group(s) because of their very early phenology.

We used linear mixed-effects models to analyze data from 1993 to 2021 to test for phenological changes through time, and the weather variables that had the most predictive power of phenological events.