As global temperatures rise, ecosystems face new pressures and often multiple challenges simultaneously. This was the case in 2016 in areas of the northeast that experienced a one-two punch of extreme drought and an onslaught of spongy moth caterpillars that feasted on a massive portion of the region’s oak leaves.

Eastern Connecticut, much of Rhode Island, and large swaths of Central Massachusetts were hit hard, says UConn Department of Natural Resources and the Environment Associate Professor Robert Fahey. This stacking of disturbances is expected to increase with climate change, and it is important to understand how forests are responding.

Fahey and his collaborators Danielle Tanzer ’21 MS, now at the University of Wisconsin; UConn Department of Ecology and Evolutionary Biology Associate Professor Robert Bagchi; Audrey Barker Plotkin at the Harvad Forest; James Mickley ’17 Ph.D., now at Oregon State University; Keenan Rivers ’20 (CAHNR), now at Michigan Technological University; researcher Maya Sagarin, now at the University of California; and UConn Department of Natural Resources and the Environment Assistant Professor Chandi Witharana saw the opportunity to study these interactions and their impacts on defoliation and tree mortality and their findings are published in the International Journal of Applied Earth Observation and Geoinformation, and Forest Ecology and Management.

“When disturbances overlap in their effects on an ecosystem, we often call that compounding disturbance, where sometimes there is more influence on the ecosystem than you would get from either of those disturbances independently. It’s this additive or multiplicative effect,” says Fahey.

The team developed a proposal to study these multiplicative effects with a National Science Foundation RAPID grant, which streamlined the funding process and helped them jumpstart the project.

Fahey explains they applied experimental and observational methods to assess the interactions of the disturbances by collecting increment cores from tree trunks to estimate biomass accumulation before and after the disturbances and by surveying the mortality of trees across the study sites.

Then the researchers compared their field data with satellite imagery in hope of developing a method to remotely assess mortality that was not only accurate but also faster and less labor-intensive than taking field samples.

The Landsat satellite collects images on an almost bi-weekly basis, and finding a method to analyze these vast quantities of data can be tricky. Besides being labor-intensive and time-consuming, current methods also rely on costly aerial overflights.

“One of the things we were trying to do is compare what we can see in the remote sensing imagery and use machine learning models to take the known mortality and map mortality across the landscape, and then compare that to the aerial documentation,” says Fahey.

The method they developed was able to predict between 60% and 80% of the mortality within Landsat’s resolution of a 30-by-30-meter pixel. Fahey says the method could be a useful tool, enabling land managers to quickly and easily assess the landscape.

To better understand the frequency and timing of the defoliation relative to the drought conditions, Fahey teamed up with Bagchi, whose research group had been studying caterpillars and their interactions within the food web in the region. They hoped to study the characteristics that led to different outcomes and levels of severity across the region.

Fahey’s group sampled and surveyed sites around Eastern Connecticut where Bagchi’s lab had already sampled for spongy moth caterpillars.

One curious observation was the timing of the defoliation differed across the landscape and the researchers wondered if these timing differences led to variations in mortality, says Fahey.

“The question is if that’s because there were fewer caterpillars in some places,” he says. “Is it because the drought differed in its severity across the landscape? Is it because there were fewer oaks available as host species across different forests, across the landscape, or is it something to do with the environment?”

They found the factor that mattered the most was whether a site experienced multiple years of defoliation, which Fahey says is not a novel or surprising result, but it is interesting because it showed the severity of the drought, and the timing of the defoliation also did not seem to matter as much as frequency.

“The drought definitely impacted the defoliation, but it didn’t seem to impact the mortality outcomes relating to the defoliation. The drought is probably associated with the severity of the defoliation in multiple ways,” says Fahey.

For example, one of the main controls of the spongy moth caterpillars is a fungus that doesn’t get established when there’s a drought; therefore, in an extremely dry year like 2016, the spongy moth population was able to explode across the landscape.

That extremely dry weather also stressed the trees, rendering them less capable of fighting defoliation. The 2016-17 drought was possibly the most severe New England has experienced since the 1960s, says Fahey, and we have had multiple such “100-year” droughts in the last decade.

“Obviously, things are changing, but that 2016 drought was severe enough across the landscape that there wasn’t enough variation for us to pick up a signal, and it probably affected the outcomes of defoliation and led to higher mortality across the landscape. We can’t say for sure because we don’t have anything to control it against, because there wasn’t a place that didn’t have drought,” he says.

Moving forward, Fahey says they are evaluating the response of the overall forest to the disturbance by looking at productivity, carbon sequestration, and any changes that occurred. The researchers are also trying to understand how growth prior to the disturbances impacted mortality outcomes. Did fast or slow-growing oaks fare better, and why? These questions are the focus of ongoing research that will help us understand how the region’s forests will fare as the climate continues to change. With thousands of increment cores from trees across Eastern Connecticut and from the Harvard Forest in Massachusetts yet to analyze, Fahey says it will take some time before they have answers.

“The frequency, severity, and nature of the disturbances that affect our forests is changing as a result of the impacts of climate change and other stressors, such as invasive pests and pathogens,” says Fahey. “These changes are leading to more frequent interactions between disturbances and understanding how compounding disturbance affects our forests will be an essential part of predicting the future of our region and its ecosystems.”