Authors

Carol K. Augspurger, Department of Plant Biology, University of Illinois Urbana-Champaign, carolaug@illinois.edu
Carl F. Salk, Southern Swedish Forest Research Centre, Swedish University of Agricultural Sciences, Alnarp, Sweden and Institute for Globally-Distributed Open Research and Education, Gothenburg, Sweden, carl.salk@slu.se

Abstract

This study used phenological field observations made year round for 15 dominant canopy tree species (1995-2022), understory saplings of three species (1995-2022), and 33 herb species (1995-2017) in Trelease Woods, a mature old-growth deciduous forest remnant near Urbana, Illinois, USA. The phenological data sets are paired with basal-area data, mid-day light transmission data from 21 days in 2002, daily solar radiation data from a NOAA (SURFRAD) site at Bondville, Illinois, and daily temperature data from a nearby weather station in Champaign, Illinois. These datasets were used to parameterize models of canopy light transmittance and understory plant light interception.
These data sets were used to test the mismatch hypothesis, viz, whether canopy trees or understory plants are more sensitive to climate change, thus changing through time the relative amount of transmitted light understory species intercept. We estimated how four factors (understory phenology, cold temperatures, canopy phenology, and sunlight) individually limit the potential light interception of each understory species.

Primary Research Domain

(OECD Fields of Science and Technology classification): 1.6 Biological science

Funding

Only institutional support (University of Illinois Urbana-Champaign)

Keywords

climate change; understory herbs; light limitation; long-term study; mismatch hypothesis; phenological asynchrony between forest strata; phenology; solar radiation; saplings; temperate deciduous forest; temperature

Related Works

The data on canopy tree phenology were collected from the same study site as the herbaceous plant phenology data described by and published in Augspurger and Zaya (2020a, b).
Weather data for Champaign, Illinois, USA were collected from a station that is part of the National Weather Service Cooperative Observer Program (US-COOP, https://www.weather.gov/coop/), and data were obtained through the Midwestern Regional Climate Center (http://mrcc.purdue.edu/).
Solar radiation data (daily total photosynthetically active radiation (PAR) above the canopy level) were collected from the NOAA surface radiation monitoring (SURFRAD) site at Bondville, Illinois, USA.

Relative basal areas of canopy tree species are from a 2005 complete census of Trelease Woods, Urbana, Illinois, USA by J. Edginton, and are modified from 2016 onward to account for the death of Fraxinus (ash) trees because of Agrilus planipennis, the emerald ash borer.

• Augspurger, C. K. and D. N. Zaya. 2020a. Concordance of long-term shifts with climate warming varies among phenological events and herbaceous species. [Dataset]. Dryad. https://doi.org/10.5061/dryad.mcvdncjxh
• Augspurger, C. K. and D. N. Zaya. 2020b. Concordance of long-term shifts with climate warming varies among phenological events and herbaceous species. Ecological Monographs 90: e01421.

Data files and descriptions

Data Set 1

Phenological Observations of Canopy Tree Species: Spring 1993-2022
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1993-2022CanopyTreePhenologySpringDates.csv’

Summary:
Field observations of spring phenology of mature individuals of 15 canopy tree species were made in Trelease Woods, Champaign Co., Illinois, USA from 1993-2022. The study site is the north half of a 24.5 ha fragment of temperate mature deciduous forest, dominated by Acer saccharum (sugar maple) near Urbana, Illinois, USA. Elevation varies by < 5 m across the study area.
The phenological status of each individual was recorded by the same observer (CKA) weekly from February - June of each year. Measurements were recorded at the individual plant level. Within each of three vegetative phases, (Bud Swell (B), Bud Burst/Leafing Out (E), and Leaf Expansion (F)), three stages were noted (e.g., B1 (1/3 of units exhibit stage), B2 (2/3 completed stage), B3 (all completed stage); each date’s observation represents the dominant stage, i.e, the status of a plant on a particular day).

Description of the data and file structure:
Latin names and common names of each canopy tree species are in columns A and B, respectively. The unique number assigned to each individual is on one row of column C: ‘number’. The first observed date of each phenological stage in each year (e.g.,1993-B1) has a separate column of data. B = Bud Swell; E = Bud Burst/Leafing Out; F = Leaf Expansion. A value in a cell is the day of year in one year of a single individual exhibiting one stage. NA indicates that a stage was not observed, usually because that stage was entirely passed between two observation dates (i.e., if an individual tree was in stage E1 on one observation date, and E3 on the next observation, then E2 would have an NA because it was unobserved). Nine columns are used for the nine spring phenological stages of each of the 30 years.

Relationship between data files:
‘Data Set 2’ has the same file structure and information as this data set (‘Data Set 1’), except the observations were made annually from July - December to obtain autumn phenological data on the same individuals.

Links to other publicly accessible locations of the data:
Augspurger, Carol; Zaya, David (2022). Woody phenology and weather data related to Trelease Woods, Urbana, IL, USA 1993-2022 [Dataset]. Dryad. https://doi.org/10.5061/dryad.3j9kd51p1

Was data derived from another source? No

Data Set 2

Phenological Observations of Canopy Tree Species: Autumn 1993-2022
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1993-2022CanopyTreePhenologyAutumnDates.csv’

Summary:
Field observations of autumn phenology of mature individuals of 15 canopy tree species were made in Trelease Woods, Champaign Co., Illinois, USA from 1993-2022. These are the same individuals observed in spring of each year (see Data Set 1). The study site is the north half of a 24.5 ha fragment of temperate mature deciduous forest, dominated by Acer saccharum (sugar maple) near Urbana, Illinois, USA. Elevation varies by < 5 m across the study area.
The phenological status of each individual was recorded by the same observer (CKA) weekly from September - December of each year. Measurements were recorded at the individual plant level. Within each of two vegetative phases, (Senescence (S), and Leaf Drop (D)), three stages were noted (e.g., S1 (1/3 of units exhibit stage), S2 (2/3 completed stage), S3 (all completed stage); a stage represents the overall status of a plant on a particular day).

Description of the data and file structure:
Latin names and common names of each canopy tree species are in columns A and B. The unique number assigned to each individual is on one row of column C: ‘number’. Each unique phenological stage in a specific year (e.g.,1993-S1) is labeled in the top row of its column of data. S = leaf senescence (leaf coloration); D= leaf drop. A value in a cell is the day of a particular year when a single individual exhibiting one stage. NA indicates that a stage was not observed, usually because that stage was entirely passed between two observation dates (i.e. if an individual tree was in stage S1 on one observation date, and S3 on the next observation, then S2 would have an NA because it was unobserved). Six columns are used for the six phenological stages of each of the 30 yr.

Relationship between data files:
The ‘Data Set 1’ file has the same structure and information as this data set (‘Data Set 2’), except the Data Set 1 observations were made annually from February - June to obtain spring phenological data on the same individual trees.

Links to other publicly-accessible locations of the data:
Augspurger, Carol; Zaya, David (2022). Woody phenology and weather data related to Trelease Woods, Urbana, IL, USA 1993-2022 [Dataset]. Dryad. https://doi.org/10.5061/dryad.3j9kd51p1

Was data derived from another source? No

Data Set 3

Phenological Observations of Saplings: Spring 1993-2022
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1993-2022SaplingPhenologySpringDates.csv’

Summary:
Field observations of spring phenology of sapling-sized individuals of three canopy tree species were made in Trelease Woods, Champaign Co., Illinois, USA from 1993-2022. The study site is the north half of a 24.5 ha fragment of temperate mature deciduous forest, dominated by Acer saccharum (sugar maple) near Urbana, Illinois, USA. Elevation varies by < 5 m across the study area.
The phenological status of each sapling was recorded by the same observer (CKA) weekly from February - June of each year. Measurements were recorded, at the individual plant level. Within each of three vegetative phases, (Bud Swell (B), Bud Burst/Leafing Out (E), and Leaf Expansion (F)), three stages were noted (e.g., B1 (1/3 of units exhibit stage), B2 (2/3 completed stage), B3 (all completed stage); each day represents one stage, i.e, the status of a plant on a particular day.

Description of the data and file structure:
Latin names and common names of each canopy tree species are in columns A and B. The unique number assigned to each individual is on one row of column C: ‘original_num’; some individuals have a physical location attached to them in column D: ‘physical_num’. Each unique phenological stage in a specific year (e.g., 1993-B1) is labeled in the top row of its column of data. B = Bud Swell; E = Bud Burst/Leafing Out; F = Leaf Expansion. A value in a cell is the day of year in one year of a single individual exhibiting one stage. NA indicates that a stage was not observed, usually because that stage was entirely passed between two observation dates (i.e., if an individual tree was in stage E1 on one observation date, and E3 on the next observation, then E2 would have an NA because it was unobserved). Nine columns are used for the phenology of each of the 30 years.

Relationship between data files:
‘Data Set 4’ has the same file structure and information as this data set (‘Data Set 3’), except the observations were made annually from July - December to obtain autumn phenological data on the same saplings.

Links to other publicly-accessible locations of the data:
Augspurger, Carol; Zaya, David (2022). Woody phenology and weather data related to Trelease Woods, Urbana, IL, USA 1993-2022 [Dataset]. Dryad. https://doi.org/10.5061/dryad.3j9kd51p1

Was data derived from another source? No

Data Set 4

Phenological Observations of Saplings: Autumn 1995-2022
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1995-2022SaplingPhenologyAutumnDates.csv’

Summary:
Field observations of autumn phenology of sapling individuals of 3 canopy tree species were made in Trelease Woods, Champaign Co., Illinois, USA from 1995-2022. The study site is the north half of a 24.5 ha fragment of mature temperate-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 September - December of each year. Aesculus glabra saplings were observed in July - August for their early leaf coloration and leaf drop. Measurements were recorded at the individual plant level. Within each of two vegetative phases, (Senescence (S), and Leaf Drop (D)), three stages were noted (e.g., D1 (1/3 of units exhibit stage), D2 (2/3 completed stage), D3 (all completed stage); each day represents one stage, i.e, the status of a plant on a particular day).

Description of the data and file structure:
Latin names and common names of each canopy tree species are in columns A and B. The unique number assigned to each individual is on one row of column C: ‘number’. Each unique phenological stage in a specific year (e.g.,1993-S1) is labeled in the top row of its column of data. S = leaf senescence (leaf coloration); D = leaf drop. A value in a cell is the day of year in one year of a single individual exhibiting one stage. NA indicates that a stage was not observed, usually because that stage was entirely passed between two observation dates (i.e., if an individual tree was in stage S1 on one observation date, and S3 on the next observation, then S2 would have an NA because it was unobserved). Six columns are used for the phenology of each of the 30 yr.

Relationship between data files:
‘Data Set 3’ has the same file structure and information as this data set (‘Data Set 4’), except the observations were made annually from February - June to obtain spring phenological data on the same saplings.

Links to other publicly-accessible locations of the data:
Augspurger, Carol; Zaya, David (2022). Woody phenology and weather data related to Trelease Woods, Urbana, IL, USA 1993-2022 [Dataset]. Dryad. https://doi.org/10.5061/dryad.3j9kd51p1

Was data derived from another source? No

Data Set 5

Phenological Observations of Herb Species 1993-2017
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1993-2017HerbPhenologyDates.csv’

Summary:
Field observations of year-round phenology of 33 herb species were made in Trelease Woods, Champaign Co., Illinois, USA from 1993-2017. The study site is the north half of a 24.5 ha fragment of mature temperate-deciduous forest, dominated by Acer saccharum (sugar maple) near Urbana, Illinois, USA. Elevation varies by < 5 m across the study area.
The average phenological status among all individuals of a species in each of 25 quadrats (1 m squares) was recorded by the same observer (CKA) weekly from March - August, biweekly from September - December, and monthly from January - February of each year. Measurements were recorded at the quadrat level. The five vegetative phases recorded were: Emergence (shoot above-ground), Expansion (shoot/leaves fully expanded), Senescence (leaf coloration), Dormancy (no structure remains above-ground), and Cotyledons (emergence of cotyledons of winter-annual species).

Description of the data and file structure:
The species number is in column A. The Latin name of each herb species is in column B.
Special designation of particular leafing cohorts of a species are as follows:
Hydrophyllum appendiculatum 1A = year 1 spring-summer non-reproductive leaf cohort
Hydrophyllum appendiculatum 1B = year 1 autumn non-reproductive leaf cohort
Hydrophyllum appendiculatum 2 = year 2 leaf cohort including flowering
Hydrophyllum virginianum A = spring-summer leaf cohort
Hydrophyllum virginianum B = autumn-winter leaf cohort
Sanicula odorata A = spring-summer leaf cohort
Sanicula odorata B = autumn-winter leaf cohort

Each quadrat is on one row of column C: ‘Quadrat’. The year of observation is in column D: ‘Year’. Each unique phenological stage in a specific year (e.g., 1993) is found in columns E through I under the following headers: Expansion (shoot/leaves fully expanded), Senescence (leaf coloration first observed), Dormancy (no structure remains above-ground), and Cotyledons (emergence of cotyledons of winter annuals). A value in a cell is the day of year representing the first observed date in a year where the average of individuals of species in the quadrat reached a particular stage. NA indicates not observed.

Relationship between data files:
The data in Data Set 5 were collected at the same study site on (most often) the same day as in Data Sets 1, 2, 3, 4.

Links to other publicly accessible-locations of the data:
Augspurger, C. K. and D. N. Zaya. 2020. Concordance of long-term shifts with climate warming varies among phenological events and herbaceous species. [Dataset]. Dryad. https://doi.org/10.5061/dryad.mcvdncjxh

Was data derived from another source? No

Data Set 6

Basal Area data from Trelease Woods Illinois, USA, 1993-2022
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1993-2022BasalArea.csv’

Summary:
Relative basal area among the 15 dominant tree species for 1993-2022 in Trelease Woods, Champaign Co., Illinois, USA. Mean basal area for each species was calculated for stems larger than 22.9 cm (9”) DBH measured at 1.4 m above the soil surface. Data are from a complete tree census in 2005 by John Edgington, and are modified from 2016 onward for the three Fraxinus spp. to account for their deaths caused by Agrilus planipennis, the emerald ash borer.

Description of the data and file structure:
Latin names and common names of each canopy tree species are in columns A and B. A value in a cell is the proportion of basal area in the forest belonging to a given species in a given year. Values for years from 1993 - 2022 are in columns C through AF labeled “y1993”, “y1994”, etc.

Relationship between data files:
Data Set 6, canopy light transmittance data (Data Set 7), solar radiation (Data Set 8) and canopy phenology (Data Sets 1 + 2), were used in building a model of the community-level canopy light interception by canopy leaves. The mean phenological stage on each census date in 2002 (from Data Sets 1 and 2) was calculated for each species. These values were aggregated to the community-level canopy light interception by leaves (Lc) by taking mean phenology values weighted by each species’ relative basal area. Then, a value of I/Io was computed by multiplying the modeled light transmission value for the leafless canopy by 1 - Lc.

Links to other publicly-accessible locations of the data:
The data are not in any other publicly-accessible location.

Was data derived from another source? No

Data Set 7

I/Io data from Trelease Woods, Illinois, USA, 2002
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger2002I-Io.csv’

Summary:
The canopy light interception model has two parts: light interception by stems and branches that are present year-round, and light interception by canopy leaves that are seasonally present. The mid-day canopy light transmittance to the understory (I/Io) is used to parameterize this model for both of these phenomena.
Mid-day observations of light transmission were collected on 21 clear sunny days throughout 2002 using a quantum sensor (LI-190R, Li-Cor, Lincoln, Nebraska, USA). The 21 sampling dates were at 3–4-week intervals, but were more frequent during rapid canopy closure in spring and opening in autumn. I/Io is the proportion of irradiance at mid-day transmitted to the understory, where I = irradiance reaching understory individuals and Io = irradiance in the open. Io readings were made in an open field 25 m west of the forest edge at the beginning and end of each 45-minute sampling period; Io was the mean of the two readings. Irradiance (I) was measured at 50 haphazardly- selected understory sites in Trelease Woods by holding the sensor horizontally at a height of 1.4 m in areas not experiencing sunflecks or overtopped by nearby leaves. The mean value of the 50 values of was used to calculate I/Io for each date. The percentage of light reaching the understory in the absence of leaves as a function of day of year was modeled using the I/Io data from the non-leafy season (see main text for details). The leafy-season difference between the modeled values and observed values was used to estimate the contribution of different stages of leafing phenology to canopy light interception (see main text).

Description of the data and file structure:
A value in column A is the day of 2002 of the I/Io measurement. The reading of I/Io on that given day of year is in column B. The value of I represents the mean of 50 values taken in the understory. The value of Io is the mean of two readings taken 45-min apart at mid-day in an open field.

Relationship between data files:
The canopy light transmittance (Data Set 7), basal area (Data Set 5), and canopy phenology datasets (Data Sets 1 and 2) were used to build a model of daily canopy light transmittance to the understory; it accounts for both light interception by branches and trunks, and light interception by canopy leaves. To calculate the estimated daily sunlight available to understory individuals throughout the year, daily solar radiation (Data Set 8) was multiplied by the estimated daily I/Io.

Links to other publicly accessible locations of the data:
The median daily I/Io values are used in Figure 1b of Augspurger et al. (2005). Otherwise, the data are not in any other publicly accessible location.

Was data derived from another source? No

Data Set 8

Solar radiation data from Bondville, Illinois, USA, 1995-2022
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1995-2022SolarRadiation.csv’
Summary:
The daily photosynthetically active radiation (PAR) from 1995-2022 was obtained from a NOAA (SURFRAD) site at Bondville, Illinois,USA, about 14.8 km west of Trelease Woods. It was assumed that a similar daily amount of radiation is received above the canopy of Trelease Woods.

Description of the data and file structure:
Definitions of Columns A-D follow:
year – the year of the daily observation.
day – the day of year for the daily observation. This was determined by setting January 1st as day 1, and takes into consideration leap day.
par – photosynthetically active radiation. The readings from NOAA are given in units of energy per area (watts/m2) and were reported over 3 min intervals from 1995-2008, and over 1 min intervals from 2009-2022. The raw values were multiplied by their respective time interval and summed over the entire day to estimate total daily incoming PAR in energy/area units (watt- minute/m2). NA: data not available.
allpar – repeated data from column ‘par’. When the column ‘par’ has NA, the missing value for allpar was replaced by the mean value for that particular Julian calendar date over 1995-2022.

Relationship between data files:
The canopy light transmittance (Data Set 7), basal area (Data Set 6), and canopy phenology datasets (Data Sets 1 and 2) were used to build a model of daily canopy light transmittance which accounts for both light interception by branches and trunks, and light interception by canopy leaves. This model, in combination with above-canopy solar radiation (Data Set 8) and estimations of understory plant leaf area, was combined with understory phenology (Data Sets 3, 4, 5) and temperature data (Data Set 9) to calculate the species-specific estimated annual light interception of saplings and herbs.

Links to other publicly-accessible locations of the data:
The data can be obtained from the NOAA surface radiation monitoring (SURFRAD) site at Bondville, Illinois, USA. (gml.noaa.gov/aftp/data/radiation/surfrad/Bondville_IL/)

Was data derived from another source? No

Data Set 9

Temperature data from Champaign, Illinois, USA, 1992-2022
Carol K. Augspurger and Carl F. Salk
File Name: ‘Augspurger1992-2022TemperatureDailyMaxMin.csv’

Summary:
Daily temperature data from Champaign, Illinois, USA are from the weather station approximately 8 km southwest of Trelease Woods. The weather station is part of the National Weather Service Cooperative Observer Program (US-COOP, https://www.weather.gov/coop/). Data were obtained by the authors through the Midwestern Regional Climate Center (http://mrcc.purdue.edu/) using the cli-MATE Online Data Portal. The weather station name is ‘CHAMPAIGN 3S’, and the NWS COOP number is 118740.\
The original data set included maximum daily temperature measured in degrees Fahrenheit and minimum daily temperature measured in degrees Fahrenheit. We provide temperature in metric units (degrees Celsius).

Description of the data and file structure:
Definitions of columns A-D follow:
day – The day of year for the daily observation. This was determined by setting January 1st as day 1, and takes into consideration leap day.
year – the year of the daily observation.
maxT_C – the daily maximum temperature, measured in degrees Celsius. This was calculated from the Fahrenheit value by subtracting 32, and then dividing the result by 1.8.
minT_C – the daily minimum temperature, measured in degrees Celsius. This was calculated from the Fahrenheit value by subtracting 32, and then dividing the result by 1.8.

Relationship between data files:
Data Set 9 was used to evaluate whether to apply certain restrictions on which days’ light interception values (see Data Set 8: Relationship between data files) were accumulated toward seasonal/yearly totals, based on daily maximum temperature. If the daily maximum temperature exceeded 5 °C it was considered suitable for plants to reliably photosynthesize. If the temperature was between 0 and 5 °C, then this was considered suitable if the day was sunny, defined as the daily PAR exceeding the median value for that Julian date. If the temperature was below 0 °C then it was considered unsuited for reliable photosynthesis.

Links to other publicly-accessible locations of the data:
The data can be obtained from the Midwest Regional Climate Center,
(https://mrcc.purdue.edu), using the cli-MATE Online Data Portal.

Was data derived from another source? No