A 39-year study of wildflower blooms in a Colorado Rocky Mountain meadow shows that more than two-thirds of alpine flowers have changed their blooming patterns in response to climate change.
Not only are half the flowers beginning to bloom weeks earlier, but more than a third are reaching their peak blooms earlier, and others are producing their last blooms later in the year.
The blooming season, which used to run from late May through early September, now lasts from late April to late September, according to University of Maryland ecologist David Inouye.
The wildflower records, made up of more than two million blooms, suggest that flowering plants' responses to climate change are more complex than previously believed, with different species responding in unexpected ways.
The combinations of flowering species that bloom together are changing, too, with potential effects on insects and birds.
Studies that focus only on the date of flowers' first blooms--as most do--understate these changes, says Inouye, co-author of a paper published in this week's issue of the journal Proceedings of the National Academy of Sciences (PNAS).
"Long-term data are essential to understanding every environmental challenge the world faces," says Saran Twombly, a program director in the National Science Foundation's (NSF) Division of Environmental Biology, which funded the research.
"This study relies on long-term data to drive home the fact that species' responses to climate change are complex."
Phenology, the study of the timing of seasonal events, is crucial to knowing how climate change affects plants, animals and the relationships that bind them into natural communities.
To answer these questions, phenologists are collecting new data and poring through old records, such as amateur naturalists' notebooks.
"Most studies rely on first dates of events like flowering or migration because they use historical data sets that were not intended as scientific studies," Inouye says.
"First flowering is easy to observe. You don't have to take the time to count flowers. So that's often the only information available.
"It's taken a lot of effort to get the comprehensive insights needed for this analysis, which helps us understand how ecological communities may change in the future."
By counting blooms in each of 30 plots every other day, up to five months per year, for four decades, Inouye and colleagues amassed a data set including more than two million flowers.
For the study, University of Arizona biologist Paul CaraDonna, University of Maryland biologist Amy Iler and Inouye analyzed data on the 60 most common species.
Bloom times are changing fast, the researchers found.
The date the first spring flower appears has advanced by more than six days per decade over the course of the study.
The spring peak, when masses of wildflowers burst into bloom, has moved up by five days per decade.
And the last flower of fall occurred about three days later every decade. "The flowering season is about one month longer than it used to be," Iler says, "which is a big change for a mountain ecosystem with a short growing season."
Of all the species that have changed their flowering schedules in some way, only 17 percent shifted their entire bloom cycles earlier. The rest showed more complicated changes.
"First flowering isn't always the best predictor of all the changes we find," CaraDonna says.
"There's a lot more going on than you can get from this single, simple measure. So, it's important to take a closer look to understand all the ways climate change affects these wildflower communities."
As the plants' bloom patterns continue to change, researchers expect that some plants that bloomed simultaneously will no longer overlap, and others will start blooming together for the first time.
Ecologists refer to these new combinations as "'no-analog' communities."
"We usually think of no-analog communities as something that happens when plants or animals move into areas where they haven't lived before, creating novel combinations of species," Iler says.
"For example, we have red foxes at our study site now. It used to be too cold for them in winter. Now the marmots that live there have to deal with a new predator.
"But this study shows that even when species don't actually move, changes in the timing of key events in their life cycles may also result in no-analog communities, where species may interact differently than before."
The changes are likely to have a strong effect--for better or worse--on pollinating insects and migratory birds.
For example, Inouye says, hummingbirds that summer in the Rocky Mountains time their nesting so their eggs hatch at peak wildflower bloom, when there is plenty of flower nectar for hungry chicks.
But as the bloom season lengthens, the plants are not producing more flowers. The same number of blooms is spread out over more days, so at peak bloom there may be fewer flowers.
Will there be enough food for the hummingbirds' young?
To find out, Inouye plans to fit adult hummingbirds with radio transmitters to study how they interact with this summer's blooms.