Methods:
Understory species occurence & abundance:
Plants were considered part of the understory if they were less than 1m in height. At each of the 16 1-m2 quadrats, a 1×1m PVC quadrat was placed oriented north to south with the northwest corner aligned with the permanent plot marker. Percent cover of each species was visually estimated by the area of the quadrat that all individuals of that species cumulatively occupied, from 0% to 100%. Since there was vertical variation in where species occurred due to plant heights, the total percent cover in a quadrat could exceed 100%.
Plant functional traits:
The most common species in each site were determined from the percent cover and frequency of occurrence in the plots. The 3-5 representative species selected in each site account for on average >80% of the total plant cover, excluding bryophytes and conifers due to their sporadic occurrences across the sites. In each site, 5 individuals of each representative species were collected, resulting in 15-25 individuals per site and a total of 165-180 plant specimens per sampling period for further data collection and analysis. Plants were collected near the plots, not within, to maintain the integrity of the monitoring plots for long-term study.
Maximum height was measured in the field prior to collection. Each plant was then clipped at ground level, placed it into a labeled plastic bag, and transported it back to the lab in a large, dark, closed bag to minimize water loss from leaf tissue. To measure leaf area, leaves were individually plucked from each plant and scanned using an Epson Perfection V39 flatbed scanner (Epson America, California, USA). LA was calculated from scanned images using the LeafArea package in R (v0.1.7, Katabuchi 2015). Lamina and midrib thickenss were measured using digital calipers. Samples were then dried in either a Grieve 343 oven (Grieve, Illinois, USA) or a Heratherm OGS100 oven (Thermo Fisher Scientific, Massachusetts, USA) at 60°C for at least 24 hours prior to weighing for dry leaf mass (ML). Specific leaf area was calculated from leaf area and dry leaf mass data using the equation SLA = LA/ML.
All trait data were paired with the corresponding data collected in 2022, which followed similar methodologies, to allow for an analysis of inter-annual variation in functional trait values. Additionally, all trait data were collected three times during summer 2023 (early June, mid-July, and early August) to allow for comparisons of trait values throughout the growing season, except for Indian Point which only has trait data at one time point (peak growth in July).
Environmental/Climatic variables:
Air (+15cm), surface (+2cm), and soil (-6cm) temperature, soil moisture, leaf area index (LAI), and photosynthetically active radiation (PAR) at ground level were measured during the same three time points as plant trait collection. Temperature and soil moisture data were logged by TOMST TMS-4 data loggers every 15 minutes (Wild et al. 2019), between mid-June 2022 (mid-July 2022 for IP) and August 10th, 2023. In this case, LAI is considered an environmental condition because overstory structure is a primary driver of light availability for understory plants (Angelini et al. 2015). Four canopy photos per site were taken at 1.3m above ground level with a Insta360 ONE X 360º camera (Arashi Vision Inc., Guangdong, China) and analyzed with Hemisfer software to estimate average LAI in each site (Schleppi et al. 2007). PAR was measured 10 times at ground level in each plot using a MQ-200X full spectrum quantum PAR sensor (Apogee Instruments, Utah, USA) and then averaged at the plot level in each site. PAR was unable to be measured in August due to logistical constraints of the field season.