Changes in timing of wildflower growth affecting insect food supplies
By Bob Berwyn
SUMMIT COUNTY — Bronze-colored fritillary butterflies visiting pink and yellow fleabane daisies is a sure sign of late spring in the Colorado high country. But as the climate warms — April temps in particular have increased dramatically — it may become a much less common sight.
A trend toward earlier snowmelt, documented at the Rocky Mountain Biological Laboratory in Crested Butte, means the common wildflowers are more susceptible to late frosts. That kills the flower buds and reducea the amount of nectar available for the insects, according to the University of Maryland’s David Inouye and Stanford University’s Carol Boggs, who have been studying alpine flora and insects at the lab near Gothic for several decades.
Stream-flow measurements from the U.S. Geological Survey show a clear trend toward earlier spring runoff — by several weeks at some locations — but the actual date of snow melt has been harder to pin down.
But Billy Barr, the lab’s business manager, has been tracking the melt-off date since 1974, noting the first day that the ground is entirely snow-free. Especially in the early years, there was no clear trend.
“It was highly variable — for the first few decades there was no trend,” Inouye said. Despite that variablity, the record now shows that, over the long-term, the snow is melting earlier, with huge implications for flora and fauna. The snow-melt dates are correlated with stream-flow data from a gage downstream at Alma, Inouye said.
He attributes the earlier snowmelt to the steep climb in April temperatures, warming faster than any other month, as well as an increase in spring dust storms that darken the snow, causing it to melt more quickly.
Reaching the conclusion that the changes in climate are affecting butterfly populations resulted from the combination of two long-term research projects.
In 1973, Inouye started monitoring two-meter square study plots at the lab, counting all the flowers every second day, documenting year-to-year changes about 100 species.
“Back in 1973, nobody had heard of climate change. I didn’t start the study for that reason,” he said. “I wanted to know what are the environmental changes that are causing the variability.”
About the same time, Stanford researcher Carol Boggs started studying insects, and the Mormon fritillary butterfly in particular.
“At one point we realized we were studying two species that interacted … the variation in flowering is a large part of what’s causing the variation in butterflies,” Inouye explained, adding that Boggs had already shown in the lab how changes in the available amount of nectar affects butterfly reproduction.
“That flower has huge ups and downs because it’s susceptible to frost,” he said.
In years when the snow melts early, the flowers start to grow, but there’s a very high likelihood the buds will be killed by late frosts. The interval between the snowmelt date and the typical last frost is a critical part of the equation, he explained.
“Historically, the snow melted at the end of May, and we typically get our last frost around June 10,” he said.
That’s not enough time for the flower buds to develop, so the impact of the frost was minimal. But when the snow melts earlier, the buds are farther along when the last frost hits, so they freeze and die.
“It’s becoming a problem for agriculture, too,” he said, explaining that earlier and longer warm spells are causing cherry and apricot trees to blossom earlier, but the date of the last frost hasn’t change significantly.
“Predicting effects of climate change on organisms’ population sizes will be difficult in some cases due to lack of knowledge of the species’ biology,” said Boggs, lead author of a paper reporting the results online in this week’s journal Ecology Letters.
Taking into account the butterfly’s life cycle and the factors determining egg production was important to the research.
Butterflies lay eggs (then die) in their first summer; the caterpillars from those eggs over-winter without eating and develop into adults in the second summer.
In laboratory experiments, the amount of nectar a female butterfly ate determined the number of eggs she laid. This suggested that flower availability might be important to changes in population size.
Early snowmelt in the second year of the butterfly life cycle worsened the effect, probably through direct killing of caterpillars during early-season frosts.
The combined effects of snowmelt in the two consecutive years explained more than four-fifths of the variation in population growth rate.
“Because species in natural communities are interconnected, the effects of climate change on any single species can easily be underestimated,” said Saran Twombly, program director in the National Science Foundation’s Division of Environmental Biology, which funded the research.
“This study combines long-term, data models, and an understanding of species interactions to underscore the complex effects climate change has on natural populations.”
“One climate parameter can have multiple effects on an organism’s population growth,” Boggs said. “This was previously not recognized for species such as butterflies that live for only one year.
“We can already predict that this coming summer will be a difficult one for the butterflies,” she said, “because the very low snowpack in the mountains this winter makes it likely that there will be significant frost damage.”
“Long-term studies such as ours are important to understanding the ‘ecology of place,’ and the effects of weather and possible climate change on population numbers,” said Inouye.
“This research is critical to assessing the broader effects of weather on an ever-changing Earth,” he said. “By facilitating long-term studies, field stations such as the Rocky Mountain Biological Laboratory are an invaluable asset.”
Stanford University’s Vice Provost for Undergraduate Education also funded the work.