Impact of Hemlock Wooley Adelgid Infestation on Forest-Atmosphere Fluxes in a New England Hemlock Stand

The Hemlock Wooley Adelgid (HWA) is an insect pest that was introduced to North America in the early 1950’s and has been responsible for loss of eastern hemlock across large areas of its range in the eastern U.S. Extensive work has documented the pace of hemlock death and the rate that HWA expands northward. The northern limit of HWA expansion reached southern New England in 1985, and many hemlock stands across Connecticut have been completely replaced by black birch. The transition from mature hemlock dominated stand to young black-birch stand is anticipated to significantly alter canopy-atmosphere exchanges. The hemlock canopy progressively thins with consequences for light interception, photosynthesis, and canopy roughness. Declining hemlock photosynthesis is offset by increasing photosynthesis by the newly established birch seedlings and other ground vegetation stimulated by increasing light at the forest floor. Ultimately the mature evergreen canopy is replaced by a young deciduous canopy, which alters the stand phenology. Large stores of carbon in the deep organic-rich soils and dead hemlock roots become vulnerable to oxidation. Stand-level biometric observations from previously infested hemlock stands do not provide necessary detail to predict the net consequence to canopy-atmosphere exchange from HWA infestation over time. With the expectation that HWA would eventually reach the Harvard Forest in north central Massachusetts, a flux tower (US-Ha2) was established in a healthy hemlock stand and became fully operational in 2004. Observations from 2004 until the arrival of HWA in 2009 define the baseline for fluxes of CO₂, H₂O, and momentum from healthy hemlock forest canopies. Once HWA reached the site the combination of flux measurements, environmental observations in the subcanopy and forest floor, and continuous soil respiration chambers will provide new insights on how canopy-atmosphere exchange from hemlock stands changes as HWA infestation progresses.

Over the years since 2009, we have observed a thinning canopy, death of hundreds of trees, and establishment of black birch seedlings in forest gaps where light levels have increased. Overall, the magnitude of CO₂ uptake during the peak growing season has decreased, but with some evidence of recovery due to a pause in HWA infestation or increased activity in the subcanopy. In order to account for confounding influence from variations in light and temperature we examine we focus on defining functional relationships between canopy-atmosphere fluxes and observed temperature and light. Comparison of light curves that relate CO₂ fluxes to the incident photosynthetically active light illustrate how net CO₂ exchange has been affected. Here we divide the observational record into a pre-HWA period prior to 2010 (Pre), an initial infestation phase from 2010 to 2015 (HWA1), and a second stage from 2015 to 2021 (HWA2). the year is broken into 2-month intervals initialized on January 15 of each year. The January 15-March 15 and November 15 -January 15 periods correspond to predominantly winter conditions, though a few periods with above freezing temperatures have detectable photosynthesis. March 15 – May 15 is a spring transition with progressively warmer temperatures and photosynthesis. September 15 – November transition is a fall transition. The two intervals spanning May 15 – September 15 are summer growing season. CO₂ flux observations are fit to an empirical model that includes a respiration term consisting of mean CO₂ flux at the mean soil temperature for the period and a linear response to deviations in soil temperature from the mean. A light response term based on Michaelis-Menton equation (a₃*PPFD)/(a₄+PPFD) defines the photosynthetic term. A freeze/thaw index based on air temperature being above or below 0° C modifies the photosynthetic term to exclude observations when freezing conditions shut down photosynthesis. Overall the empirical model accounts for nearly 80% of variance in the mid-summer interval, but the r² declines to 10% for mid winter.

Comparison of observed CO₂ flux vs light across the pre, HWA1, and HWA2 periods illustrates a progressive decline in net CO₂ uptake for given light level over the course of infestation so far. The reduction in net CO₂ uptake is most pronounced at moderate light intensities. However, in summer and fall the net CO₂ fluxes are comparable over the entire period at the highest light levels when photosynthetic response saturates.

Observations over the first 12 years since HWA was detected at the Harvard Forest have demonstrated a perceptible decline in photosynthetic capacity. We anticipate even more significant changes as an increasing fraction of the hemlock trees finally die. How the increasing leaf area of rapidly growing sapling layer and decomposition of soil organic matter and dead roots will offset the loss of canopy hemlocks will be the focus of our continued observation at the site.