Most of Vermont’s forests are made up of what are known as northern hardwood trees, like maple and beech. As you climb to higher elevations, the temperature generally drops. Eventually, it becomes too cold for the northern hardwoods to thrive, and conifers like fir and spruce take over.
But there are areas where this pattern is inverted — valleys and basins where cold air pools and which are populated by plants living far below their usual elevation.
This inversion of vegetation comes through a process called cold air pooling, explained Melissa Pastore, a research ecologist with the U.S. Forest Service who led a recently-published study on the phenomenon. Throughout the day, the surface of the Earth is heated by solar radiation from the Sun, which it then emits back as longwave radiation. At nighttime, the ground no longer receives solar radiation, but continues to emit longwave radiation. This causes the ground to cool, in turn cooling the air in contact with it.
This phenomenon, where cool air lies underneath warm air, is called a temperature inversion.
“When this cooling happens on mountain slopes, the air becomes denser and heavier, which causes the air to sink downslope in streams called katabatic winds. And when the cold air flows into a relatively flat or protected area, like a valley or basin, the streams of cold air slow down and can pool,” Pastore said. “This can cause the valley to fill with cold air, kind of like a stream filling a lake.”
More than a curiosity, cold-air-pooled areas could serve as refuges for cold-adapted plants in a warming world, Pastore said. But before she and her team could study these areas, they had to find them first. And the best way to do that is from space.
Eye in the sky
Jane Foster, a research ecologist with the U.S. Forest Service, identified potential study areas using data from two artificial satellites. The satellites have gathered over 20 years of data on a wide variety of metrics, including surface temperature, using a sensor called Moderate Resolution Imaging Spectroradiometer (MODIS).
But satellite imagery alone isn’t enough: MODIS’ imagery is coarse-grained, meaning it can only assign a temperature to a square kilometer; in mountainous areas, temperatures can vary a lot in that small range. MODIS also can’t measure beneath the canopy, where most organisms live, and it can’t see through clouds, so Foster relied on cloud-free periods in the 20-year MODIS dataset.
“We can get large-scale measurements and look at watershed-scale patterns of some of the temperature distributions that we’re interested in. We use those initially to look for cold air pooling events happening,” Foster said. "Then we can compare those to elevation data from a digital elevation model and see that we have colder temperatures in the valley.”
Ultimately, the team decided on three areas: one in the Nulhegan Basin in the Northeast Kingdom, one near the New Hampshire-Maine border, and one in the northern Green Mountains, at Camel's Hump and Little River state parks.
Measuring temperatures beneath the canopy
The team set up circular plots along elevational transects, a series of areas along a straight path up a mountain slope. They recorded the trees in the plots and measured various properties of the soil. And they set up hourly temperature sensors and let them run for months.
The team found the areas with the most frequent and strongest temperature inversions had the clearest inverted vegetation patterns, with more high-elevation trees growing at lower elevations in the study area. But there was considerable variance.
Take the Nulhegan Basin, a 16-mile-wide, crater-like basin in the Northeast Kingdom. Unlike the two valleys in the study, the basin site saw much colder temperatures at lower elevations, along with many more spruce and fir trees.
“When you have an area where things are really protected and you don’t have outflow channels for the air, you might see a lot more cold-air pooling,” Pastore said. “Even though the elevation change was pretty small, we saw a ton of cold-air pooling I think just because of that topography, just because it’s that protected basin.”
On the other hand, the sites with the least frequent cold air pooling, located in the northern Green Mountains, at Camel's Hump and Little River state parks, had little to no conifers at lower elevations.
“The way that valley is oriented is actually similar to the direction that the prevailing winds flow,” Pastore said. “So it’s possible that that was creating a kind of outflow for the colder air pools, and that may be why it was less common.”
Soil properties did not vary much within sites and could not explain vegetation patterns, Pastore said, lending credence to the theory that cold-air pooling was the deciding factor.
To the team’s surprise, they observed cold air pooling year-round and during the day, whereas historically it’s been considered a nocturnal, wintertime phenomenon.
Improving models
Anthony D’Amato, a professor of forestry and director of the forestry program at the University of Vermont, said part of the impetus of the study was to improve data modeling.
When scientists try to project how a given species might survive in a warmer future, D’Amato said, they tend to look at a broad scale — a 100-mile-by-100-mile area, for example — and predict what the climate in that area will look like and how it might affect a species. But this broad-scale approach can overgeneralize, missing out on important environments.
“What that broad scale projection might overlook is that there could be some local places that even if at that hundred-by-hundred mile scale, that species at large is not going to do well, there still could be some spots at the half-mile-by-half-mile or mile-by-mile scale that will still provide habitat that allows that species to persist in the landscape,” D’Amato said.
In other words, cold-aired microclimates like in the Nulhegan Basin could serve as refuges for plants adapted to colder weather in a warmer world. It wouldn’t be the first time this has happened. During the Last Glacial Maximum, when ice sheets were at their greatest extent, plant and animal populations persisted in areas where climate change was buffered; when the ice sheets retreated, those populations were able to disperse and colonize the land, Pastore explained.
Another potential benefit of this research is the reintroduction of species that were lost in Vermont’s less-green past.
“Some of the work we’re doing [at two of the sites] really involves not just managing forests for future conditions, but also including some of these species that have been lost from that landscape due to past land use,” D’Amato said.
Red spruce, in particular, could sustain itself in these areas, D’Amato said.
Have questions, comments or tips? Send us a message.