News•March 4, 2013

Spring May Arrive Five Weeks Earlier by 2100, Study Finds

The biological onset of spring could arrive up to five weeks earlier by 2100 in the northern U.S. than it does today, and more than a week earlier in the South, a change that could significantly alter ecosystems from Florida to Maine, according to a paper published in Geophysical Research Letters.

“This is a big deal,” said lead author David Medvigy, an ecologist at Princeton University, in an interview. The reason is that ecosystems have evolved over thousands of years, so that different species live more or less in balance with each other. If spring comes earlier, some species will adapt more easily than others, throwing that balance off by, for example, disturbing the relationship between animals and their food sources.

Credit: promanex/flickr

As with so many disruptions to natural systems, including rising seas, more frequent and intense droughts and heat waves, and more torrential downpours, this projected rollback in the onset of spring — measured in this case by “budburst,” or the annual emergence of leaves on deciduous trees like maples, poplars and birches — is tied to global warming caused by heat-trapping greenhouse gases.

The idea that spring is getting pushed earlier by climate change isn’t new: in fact, scientists have already demonstrated that spring weather has been coming to the U.S. three days earlier during the past 30 years, on average, than it did during the previous 30. Others have documented the shifting, not of weather, but of phenology — that is, biological events of all sorts, including budburst, but also flowering, ovulation, migration and other seasonal changes in plants and animals.

This is the first study that looks in such depth at a single phenomenon. About a quarter of the CO2 emitted through human activity is re-absorbed by the land, mostly by plants (another quarter goes into the sea, and half remains in the atmosphere where it traps the Sun’s energy).

Much of that absorption happens in spring and summer, when plants are most actively growing — if you look at a chart of atmospheric CO2, you see it rise over the years in a sawtooth pattern, with a slight drop every year during the growing season, followed by an even bigger increase in the fall, as leaves fall and plants die.

In order to understand that cycle better, Medvigy and his co-authors tapped into the National Phenology Network, an organization that uses citizen-scientists to go out and report on the timing of biological phenomena. “Until now,” he said, “we’ve had very few data sets on budburst,” and to gather information in the conventional way, he said, “would require huge number of grad students out watching trees.”

Using an index for the onset of spring developed by Mark D. Schwartz and collegues, Climate Central found that spring temperatures have been arriving approximately 3 days earlier, moving from March 20 (1951-1980 average) to March 17 now (1981-2010 average).
For more details: State-by-State Look at How Early Spring Has Arrived.

Armed with reports on budburst and local temperature from the network, and funded by a grant from the National Oceanic and Atmospheric Administration (NOAA), Medvigy and his colleagues used state-of-the-art climate models to look at how changes in temperature would likely affect leaf emergence over the next nine decades or so. “The main result is the obvious one: it’s going to be warmer, so budburst is going to be earlier,” he said.

But the scale of that change is striking, he said. “The typical value is two weeks, but in some cases, it’s a month or more, while in others it’s seven days.” It depends partly on what species you look at, but in general, Medvigy said, the changes are greater in states like New York, Michigan, Wisconsin and Maine. In those states and others, the growing season will get longer, with one possible result that deciduous (i.e., leaf-dropping) trees will start to out-compete pines and other conifers. Changing the mix of trees in a long-established forest could have ripple effects no one can really predict.

What they can predict is that “the North is going to become more South-like,” Medvigy said. “It could lead to a homogenizing of ecosystems,” in which regional ecosystems that now look very different would begin to look alike. That might, in turn, alter the migratory behavior many species of birds and insects — a ripple effect that could lead to further changes in ecosystems.

And that’s just based on what happens in spring, which as Medvigy said “is interesting, but it’s only half the question.” If autumn starts to come later at the other end of the year (and there are hints that this is already happening), the growing season will be even longer, multiplying the chance of ecosystem disruption.

Medvigy and his colleagues therefore plan to look at fall as well. Meanwhile, they’ve partnered with scientists at NOAA’s Geophysical Fluid Dynamics Laboratory to incorporate the new results into the lab’s own climate models. Medvigy said that ideally, this would help the models do a better job of simulating the movement of carbon from the atmosphere to the land and back again.

And that, in turn, could help drive down uncertainties that remain in climate models, allowing scientist to say with relative precision, rather than just approximately, where temperatures are headed for the rest of the century. A project that started out trying to understand the effect of climate on ecosystems could well turn out to be important, in short, in understanding the effect of ecosystems on climate.