”]Facing a growing population and increasing demands for fresh water, India is hoping that an engineering fix will help solve its water-scarcity problems. The country’s National Water Development Agency has begun work on the first of a system of 30 canals that would link 46 rivers, in a process known as inter-basin water transfer.

The need for more water is particularly acute in India’s driest regions, some of which receive as little as 4 inches (10 centimeters) of water a year. (For comparison, American deserts receive about 11 inches –28 cm– of rain per year.) An average of 46 inches (117 centimeters) of rain falls annually across the country, meaning some regions are relatively water-rich, a fact that Indian planners would like to exploit. However, very little study has been done on the long-term ecological impacts of creating a 6,000-mile canal network that would link distinct river basins.

The issue drew the interest of Heather Lynch, an assistant research scientist at the University of Maryland in College Park (soon to join Stony Brook University, in Stony Brook, N.Y., as an assistant professor of ecology and evolution) and her research colleagues. Using a computerized database of India’s river system, they developed a geographically-accurate model to investigate the impacts on fish diversity of adding new connections to a branching river system.

They used the model to simulate the effects of the new inter-basin links, focusing on the long-term impact that changing the river network’s geometry would have on the fish diversity. (Because fish are relatively long-lived, the final equilibrium point takes a very long time to develop, so their model simulates patterns of biodiversity tens to hundreds of thousands of years into the future.)

The team ran the model until each of the original fish species had either gone extinct or evolved into newer species. By the end of that process, the researchers found that the canals had increased local fish diversity throughout the interconnected networks, boosting the number of species by as much as 40 percent in some stretches.

The researchers recently published their results in Water Resources Research, an AGU journal.

While so-called ‘neutral’ models, like the one used in this study, are simplistic, they have been shown in other studies to accurately predict real-world patterns of biodiversity over long time scales, Lynch said. (For instance, see co-author Muneepeerakul et al.’s study of the Mississippi-Missouri River Basin published in Nature in 2008).

The team’s model allots a fixed number of designated “spots” for fish, with more spots going to the stretches of the river that accumulate more water from upstream watersheds. Various fish species swim within the model’s river networks, some that are common and others that are rare. Each are allowed to randomly migrate along the river network and compete for spots left by local extinction.

“Using a model that requires few biological details, we can make predictions on how the biogeography changes when we rewire the landscape in which these fish populations live,” Lynch said.

However, the new study steers clear of conclusions about short-term ecological dynamics (those that occur over decades or a few centuries). Accurately addressing those requires detailed information about the biology of a river’s particular fish species, data that is mostly unavailable in this case, Lynch has found.

To work, the model makes some simplifications, including leveling the competition between species so that each has the same chance of winning over another fish species for a certain number of habitat spots in the model. Also gone are special niches that exist in the real world for different types of fish.

The study finds that the geometry of a habitat is an important constraint on fish biodiversity and, as a result, the addition of inter-basin canals can influence local species richness simply through their impact on the geometry of the network.

Lynch explains that it’s like a subway system that handles a fixed number of passengers coming from stations. After an extension is added that connects the system to a local commuter train system, train passengers begin to use the subway and vice-versa, increasing the mixing of passengers from the farthest extent of those lines. People who previously would only travel to areas serviced by one can become regulars at stops along the other system.

“We didn’t expect to see impacts on the other side of the basin,” Lynch said. “In hindsight it makes sense, because it’s over very long time scales – hundreds of thousands of years. So, even if only a tiny, tiny fraction of fish ever travel that far, impacts on species richness will ultimately affect the entire river system.”

The researchers are now working on the next step, she said – a model that has greater “biological realism” that will be able to predict the more immediate impacts of new river connections on a particular fish species. To do so, they are compiling a “huge database” of Indian freshwater fish from museum records and published papers.

“To test this theory, we’d have to wait long beyond our lifespans,” Lynch said. “But as we insert more biology, we’ll be able to look at shorter-term change. In particular, we want to know whether rare species are going to go extinct more quickly in a more heavily-linked system, and which basins may be most vulnerable to India’s plans for inter-basin water transfer.”

–Kathleen O’Neil, AGU science writer

Lynch, H., Campbell Grant, E., Muneepeerakul, R., Arunachalam, M., Rodriguez-Iturbe, I., & Fagan, W. (2011). How restructuring river connectivity changes freshwater fish biodiversity and biogeography Water Resources Research, 47 (5) DOI: 10.1029/2010WR010330