Location: Providence, RI, U.S.A. A three-hour talk about grasslands may not sound very interesting on paper, but I attended a surprisingly engaging one Thursday morning at the "Climate Change and its Impacts" BIARI. Specifically, Colorado State University Prof. Alan Knapp's lecture was about "Ecosystem Responses to Climate Change: a Grassland Perspective." Knapp's charismatic presentation didn't begin exactly at 9:00 a.m. as scheduled because more than half of the attendees were watching the Argentina-South Korea game in the CIT lobby. His lecture may not have been as intense as the World Cup, but it did cover quite a bit of urgent ground.
We no longer need to be uncertain about the existence or cause of climate change, Knapp said, but instead need to address uncertainty about the consequences. Gradual changes in how we shape the environment do not necessarily translate into gradual responses, he added. He then proposed a three-step process of change for an affected ecosystem: a physiological response, the re-ordering of existing species, and finally the immigration of new species. The problem is that experts can't tell how long each of those steps would take, or how gradual the change would be. Those factors depend upon the sensitivity of the the ecosystem, Knapp said--a redwood forest would change much more slowly than a much shorter-lived invasive species. "This is an ongoing process, [and] we [ecologists] don't have all the answers," he said. 
Knapp then turned to his own research in the tall-grass prairie of eastern Kansas, where two halves of a 17-year climate effects experiment produced dramatically different results. A physiological change occurring over the first eight years gave way to a reordering of the grassland community in the next nine. He repeatedly emphasized that there are many factors at play, but that even in looking at one factor like precipitation, there is plenty of uncertainty about what's going to happen next. Likewise, a similar experiment to Knapp's original conducted by one of his students in the short-grasslands near Colorado State University gave very different results; varied sensitivity in the environment can lead to different consequences for the same sorts of environmental manipulations. Generalizations, therefore, become very difficult for ecological researchers. Knapp's suggested solution to these issues of uncertainty was a simple one: finding more ecologists, and expanding the scope of scientific experiments in the field. For instance, he proposed a networked study stretching across hundreds of ecosystems from coast to coast in the U.S. as a way to coordinate experimentation in settings of different sensitivities. "What we don't get is the big picture approach," he said. "The only way we can do that is by having coordinated research efforts...the issues we are dealing with aren't local, they're global." BIARI convener Osvaldo Sala chimed in, "I believe that this is one of the most important things that society could be focusing on." 
In the course of discussion, Knapp suggested that "natural resource managers" like him haven't been as responsible as other types of scientists in responding to urgent global demands. Sala countered that this may be partly a function of policymakers who pointedly avoid funding policy-relevant science. The reason? Those controversial experiments are more likely to produce results that force political leaders' hands and make them change policies in ways they don't like. Another intriguing detour from the main focus of Knapp's presentation was his claim that "environmentalism is a philosophy, [but] ecology is a science." He wondered aloud if the impact of his research, as funneled through the global media, would be weakened if he were depicted as an environmentalist rather than a scientist, even though the two terms aren't mutually exclusive. "Virtually every ecologist makes up his or her mind on where they're going to fall on that line," Knapp said. Sala added that the ever-increasing speed of contact between scientific institutions and media may be dangerous. The rush to publicize scientific results strips away the time other scientists used to have to check and verify experiments and balance the conclusions. Today, policymakers can act on not-fully-baked conclusions because of this lack of "checks and balances," Sala said. 
After lunch, I headed to a very different sort of presentation: "Quantitative Methods for the Study of Educational Inequality" by New York University's Florencia Torche in Kassar House. As part of the Development & Inequality BIARI, the talk questioned whether education--widely believed to be the great equalizer--actually leads to societal improvements. Unfortunately, most of the details of Torche's presentation was out of my league, dealing with the sophisticated statistical techniques used to illuminate otherwise invisible truths. But I'll offer here what I was able to take away by the end of the day. Some economists are beginning to question the blindly accepted association/correlation between education and gains in both the private and social realms, she said. But her work suggests that this association is much stronger in the developing world. The key for policymakers is differentiating between educational outcome (how much education individuals within the population receive) and opportunity (how much access they have to education in the first place). 
Gains in educational attainment, especially among poorer classes, in developing countries led to substantial declines in inequality; in developed countries like France and the U.K., however, these gains were much less pronounced over the same time period. Torche said a ceiling keeps countries from making an infinite amount of progress in reducing inequality; in other words, educational expansion happens "at the bottom of the distribution." Torche left a large chunk of her conclusions open-ended, citing research that disputes the claim that inequality of opportunity is actually being reduced. The "big take-away," as I am wont to calling it, was that to get the right answers in studying inequality, one has to use the correct models.
