Don't it always seem to go
That you don't know what you've got
Till it's gone.
They paved paradise
And put up a parking lot.
-Joni Mitchell, "Big Yellow Taxi"
Going, going, gone.
Since 1960, population biologist Paul Ehrlich and his research group have been conducting a classic study of the population of Jasper Ridge's Bay checkerspot butterflies, subspecies Euphydryas editha bayensis. But now the biologists won't have to muck about the ridge counting insects as they have for decades to characterize rates of births, deaths, immigrations and emigrations. The last two Jasper Ridge populations went extinct in 1991 and 1998. Examining 70 years of rainfall and population data, the researchers now conclude that extreme swings in regional climate hastened extinction of the butterflies. They report their findings in the April 30 issue of the Proceedings of the National Academy of Sciences.
"This is the first time anyone has documented the effects of weather variation linked to climate change on extinction of populations," says Ehrlich, the Bing Professor of Population Studies. Previous studies have looked at shifts in average conditions - say, the climate becomes drier or wetter overall - or the effects of a single extreme climate event, such as a hard freeze. The Stanford study instead examined the effect of climate variability - swings in rainfall highs and lows - on extinction, and tied the extinctions tightly to a mechanism. "People have long been aware of the problem of species extinctions, but are just beginning to realize the importance of population extinctions."
Not only are population extinctions a prelude to species extinctions, but they also threaten the vital ecosystem services upon which human life depends. Stanford's Center for Conservation Biology has pioneered investigations of population extinctions, in part because of the Ehrlich group's observations on the extirpation of checkerspot populations.
"What we've done is develop a model describing how precipitation drives the changes in population numbers of the butterfly," says co-author Carol Boggs, director of the Center for Conservation Biology and an associate professor (teaching) of biological sciences. "We looked at how precipitation has changed locally in the last 70 years and found that the variability in precipitation had increased since 1971." The drastic differences in rainfall created pressures the checkerspots simply could not overcome, she says. "Population extinction in these insects can happen very fast - in a matter of decades - if you have high variation in rainfall."
The Jasper Ridge checkerspots weren't the last checkerspots on Earth. Several other populations of the Bay Area subspecies are known - including a large source population in Morgan Hill with numbers in the tens of thousands to hundreds of thousands, and a very small, extremely endangered population in Edgewood County Park in the hills above Redwood City. The Bay checkerspots were listed as threatened in 1987 under the United States Federal Endangered Species Act.
A series of smaller populations used to exist in areas surrounding the source population on nutrient-poor native soils derived from serpentine rock, common throughout California. But most of these have now been lost to development. "Three hundred years ago, the checkerspots were almost certainly widespread in California," Ehrlich says. But when the Spaniards settled the state in the 1700s and 1800s, in the hay they fed their cattle they inadvertently imported Eurasian grasses that competed with the native plants upon which the butterflies fed.
Because their preferred host plants thrived on serpentine soils that the invading plants had trouble colonizing, the checkerspots' range is now confined to serpentine patches. "Within four kilometers of the big population are areas of suitable habitat that are not inhabited, which gives us some idea about the movement capability of the butterfly," Boggs says. "They're really pretty sedentary. Within an area of suitable habitat, movements of even 50 meters are pretty big." Among the Jasper Ridge populations, mere tens of meters apart, only 1 percent of the animals crossed unsatisfactory habitat to join other populations, Boggs says.
If populations become too small or isolated, fairly rapid extinction is inevitable. Small populations have trouble riding out the waves that can wash out their numbers. Environmental fluctuations loom large, and when fluctuations cause an extinction, isolation prevents recolonization. Human population growth can bring urban sprawl and plant invasions that isolate one population "island" from another and create insurmountable barriers to migration and repopulation.
The checkerspot species as a whole has a larger range - from Baja to Canada - than that of the Bay Area subspecies. But a 1996 study by University of Texas biologist Camille Parmesan showed that checkerspot populations in the south were going extinct faster than populations in the north. It appears that their range is contracting as climate change at global and regional scales alters their distribution and human development wipes out habitat.
Rainfall is the biggest factor in determining checkerspot numbers. Caterpillars hatch from eggs in April but will starve if they can't get big enough before the summer drought comes and their seasonal food plants perish. The longer those plants stay alive, the more time the caterpillars have to get fat enough to reach their fourth instar (the stage between molts). At that point, the caterpillars can go into a resting stage called diapause, when their metabolism shuts down for the summer.
"The biggest source of mortality for the butterflies is failure to get big enough to go into diapause before the plant vanishes," Boggs says.
When the rains come in November, the caterpillars begin feeding again. Now they are only half an inch long, but they must multiply their weight many times before they become chrysalids (pupae), the stage at which caterpillars metamorphose into butterflies.
"This happens in late winter," Ehrlich says. "It takes about 10 days for the feeding-growing machine, the caterpillar, to be converted into the reproducing-dispersing machine, the adult."
Born to lose?
"Population extinctions, like species extinctions, have been going on ever since life first evolved," Ehrlich points out. "What's different now is the rate of extinction is higher than any time in the last 65 million years - since the dinosaurs went extinct." The Jasper Ridge checkerspots were going to go extinct anyway, according to Boggs, Ehrlich and co-authors who ran the computer model, former Stanford postdoctoral researcher John McLaughlin of Western Washington University and former Stanford doctoral student Jessica Hellmann of the University of British Columbia. The model first employed rainfall data from 1932 to 1970 to predict that the first population, which lived on a flat patch of Stanford's biological preserve, would have been gone in 444 years. The second population, which lived on a hill, would have lasted 162 years.
But something happened in 1971 that sped the walk toward extinction. Annual rainfall began to vary wildly. "We have no data to indicate the cause," Boggs says. "It could be anthropogenic. It could be natural. But it does give a warning sign that if something like this can happen naturally, any perturbations added by humans are just going to make things worse."
Plugging weather data from 1972 to 1998 into their model, which began with population numbers from 1932, the scientists found that large variations in rainfall sped extinction times to about 19 years for the flatlanders and 52 years for the hill dwellers.
"Topography buffers weather to some extent, which is why we see differences in the (flatlanders) versus (hill dwellers)," Boggs says. "North-facing slopes are colder than south-facing slopes. That means that on north-facing slopes, caterpillars develop more slowly and plants dry out more slowly. In spring, if it's a really dry spring, the caterpillars should do better on a north-facing slope. If it's a wet spring, then they want to be warmer - they want to take advantage of the sun when it's there - so they'll do fine on a south-facing slope." If their habitat is big enough, female checkerspots will lay eggs on both north- and south-facing slopes to spread the risk, Boggs says. But butterflies inhabiting flatlands have no way to spread the risk.
A caterpillar in a favorable microclimate might grow up two weeks before one in an unfavorable microclimate, and that extra time may mean the difference between life and death.
The future doesn't look good for the Bay Area's remaining checkerspots. The source population probably will not be able to repopulate Jasper Ridge, Boggs says. Morgan Hill is too far away for butterflies to make the journey on their own, given their natural tendency to stay put, physical barriers created by the further paving of paradise, and climate-destabilizing greenhouse gases from automobiles, power plants and other human sources.
"The advantage to having done this work is that we now thoroughly understand the mechanisms causing extinction, which can allow us to start thinking about and implementing interventions to save other populations [of checkerspots] in the Bay Area," says Boggs. "The Jasper Ridge populations were studied for almost 40 years, so we have an unusually good database to draw on to be able to understand what happened."
-by Dawn Levy-
CONTACT: Dawn Levy, News Service 650-725-1944;
dawnlevy@stanford.edu
COMMENT: Paul Ehrlich, Biological Sciences 650-723-3171; pre@stanford.edu
Carol Boggs, Biological Sciences 650-723-5923;
cboggs@stanford.edu
EDITORS:A photo of the Bay checkerspot butterfly is available at http://newsphotos.stanford.edu.
Photo courtesy: Center for Conservation Biology.
Relevant Web URLs:
http://www.stanford.edu/group/CCB/
Journal
Proceedings of the National Academy of Sciences