Climate change's impacts on crop yields may force as many as seven million Mexicans to emigrate to the U.S. over the next 70 years, according to research published July 26 in the Proceedings of the National Academy of Sciences. The study is among the first to attempt to put hard numbers on questions about "environmental refugees" that may be caused by climate change.

"There is a significant response of emigration from Mexico to past climate variations," says atmospheric scientist Michael Oppenheimer of Princeton University, an author of the study. "Climate changes predicted by the global circulation models would cause several percent of the Mexican population to move north [if] all other factors are held constant."

Based on historical patterns of emigration, crop yield and climate change between 1995 and 2005, the researchers project that as much as 10 percent of Mexico's population could be forced to migrate in coming decades. In essence, for every 10 percent reduction in crop yield as a result of climate change an additional 2 percent of Mexicans would emigrate. The U.S. National Research Council (NRC) estimates that every degree Celsius of warming in global average temperatures means a 5 to 15 percent drop in yield, particularly for corn, in North America.

The number of climate refugees could be significant, anywhere from 1.4 million to 6.7 million Mexicans, depending on how much warming actually occurs. The researchers attempted to account for other variables that governed migration in the past—from the passage of the North American Free Trade Agreement (NAFTA) to U.S. border control policy—and isolate the impact of climate change. "For NAFTA, we took special pains to ensure the robustness of the result by comparing Mexican states that were greatly affected by NAFTA with those which were not," Oppenheimer says.

But the study relies on census data to estimate actual emigration, and it assumes that climate change's impact on maize production—for example, via drought—caused migration in the past and will cause it again in future. "Migration decisions, like all livelihood decisions, are about much more than material quality of life," argues geographer Edward Carr of the University of South Carolina, who studies human migration in countries such as Ghana and was not involved in the Mexico emigration research. "What I am seeing in sub-Saharan Africa are very complex patterns in which environmental change is but one of several causal factors."

The growth in those causal factors—from a population boom to reduced economic opportunity in the countryside—has driven the greatest migration worldwide in recent history. As of 2005, roughly three percent of the world's population lived outside their country of birth—the largest proportion ever recorded by the United Nations. And this represents but a small fraction of migration, since the bulk of that occurs within a given nation from rural to urban areas. "Most often international migration is not an option and rural residents migrate to urban areas, contributing to urbanization and urban poverty in developing countries," says sociologist Elizabeth Fussell of Washington State University.

That is certainly the case in Mexico, according to population and migration researcher Haydea Izazola of the Universidad Autónoma Metropolitana-Xochimilco, also not part of Oppenheimer's team for the new study. "The great majority of the rural population who grow maize—rain-fed agriculture—for their own consumption are the poorest of the poor and lack the means to invest in the very expensive and risky migration venture."

In fact, recent trends indicate it is city residents and educated elites who are driving migration from Mexico today, Izazola notes, suggesting a possible shift from the period in this study. That may be because Mexico experienced a severe drought starting in 1994 that persisted throughout this period, making farmers more "susceptible to these climate effects since there was no time to adapt to them," Fussell speculates. "Migration [is] one of several options rural households face when they confront diminishing livelihoods such as a decline in crop yields."

Once that process begins, however, it can be self-reinforcing, as community members share their experience—and any wealth gained in the new country—with folks back home. Regardless, "if our study results are valid, the effect could be significant in many regions" of the world, Oppenheimer argues.

Of course, the worst-case scenario in this analysis ignores the possibility of new crop varieties that maintain (or improve) yields under warmer average temperatures. "How much can we adapt?" asked climate scientist Susan Solomon of the U.S. National Oceanic and Atmospheric Administration at the release of the NRC report on July 16 "Look at corn. Maybe we can choose to grow something else or genetically engineer that corn to make it more robust."

And climate change is not confined to Mexico. Crop yields in the U.S. will likely suffer as well. "People do not move blindly; they move to greater opportunity," Carr notes. "So we should probably be using [these economic and climate] models to examine the impact of future climate change on various migrant-employing sectors of the southwestern U.S. economy."

In fact, that is exactly what Oppenheimer and his colleagues are currently working on next: predicting how crop yields in the U.S. may impact the immigration debate.