The scientists' findings reveal the impressive adaptive engineering of deep-rooted trees in adjusting the size and structure of their piping, or xylem, to maximize water uptake, maintain their integrity and avoid flow-blocking embolisms. In particular, the findings reveal how the deepest roots develop the largest conduits in order to draw deep water most efficiently.
The scientists' technique of studying tree roots in place offers a far more realistic look at how trees manage their water circulation than researchers have had before. The approach is like studying the plumbing of a house by carefully tracing its pipes, valves and faucets. In contrast, the usual approach -- drilling core samples and separating tree root samples for study -- is like taking a wrecking ball to the house and sifting through the debris to understand the plumbing system.
The researchers, led by biologist Robert Jackson, published their findings in the September 2004 issue of the journal "New Phytologist." Lead author on the paper was Andrew McElrone, formerly of Duke and now at St. Joseph's University in Philadelphia. Other co-authors on the paper were William Pockman of the University of New Mexico and Jordi Martínez Vilalta of the University of Edinburgh in the United Kingdom. Jackson is a professor of environmental sciences and biology in the Nicholas School of the Environment and Earth Sciences and a professor in the Department of Biology at Duke. The scientists' work was supported by the National Science Foundation, the Mellon Foundation and the U. S. Department of Agriculture.
In their study, the researchers took advantage of the fact that a well-mapped system of limestone caves runs beneath the Edwards Plateau in central Texas. Thrusting into those caves from above are the roots of trees, some of which reach depths of 60 feet.
The Texas state government is interested in understanding the effects of such trees on water uptake because the Edwards aquifer supplies a large area of central Texas with drinking water, including the city of San Antonio. The water supply from this aquifer has been affected by the change in the landscape above from a savanna to forest, due to grazing and human prevention of natural wild fires over the last 150 years.
In their experiments, Jackson and his colleagues descended into the caves and sampled roots of four tree species -- a juniper, an evergreen oak, a deciduous oak and the deciduous gum bumelia, which is also known as the chittamwood.
"The evergreen oak we studied is the dominant oak on the Edwards Plateau," said Jackson. "Because it needs a lot of water to maintain its year-round foliage, it has the deepest roots, going some 60 to 75 feet into the bedrock to find water."
In their studies, the researchers did genetic fingerprinting on both the roots and above ground trees to match the roots to specific trees. The fingerprinting is the same technique used in criminal forensics.
They then measured the diameters and wall thicknesses of the water-conducting root xylems. And in laboratory experiments, they determined how vulnerable specific sections of root were to "cavitation," in which mechanical stress or drought induces bubbles that block water flow.
The researchers found that trees adjust their root anatomy very effectively to work in different environments. Shallow roots and especially stems have thicker walls and smaller interior diameters because they must support the tree more, and because they experience more drying and temperature extremes that cause cavitation. Deeper roots, more protected from such demands, can have larger diameters and thinner walls to maximize the conduit size and enhance water flow.
"The data are interesting because they show that the same individual plant can adjust its anatomy and physiology to maximize water transport deep underground where the plant is less likely to experience drought or freezing conditions which cause cavitation," said Jackson.
"The surprise was the extent to which individual trees can make these developmental changes," he said. "There has long been data showing that roots tend to have larger conduits than shoots. But there has not been an analysis like this looking across the gradient of differences within the same plant. And nobody has analyzed roots that go as deep as these," said Jackson. According to Jackson, such insights will improve theoretical models of water transport within trees.
"If you're building a model of water transport and you don't take into account these differences among roots, you could severely underestimate the importance of those deep roots," he said. "Their biomass isn't that great, but they are extremely efficient in transporting water."
In further studies, Jackson and his colleagues are installing instruments on trees to measure their water flow in real time to determine the dynamic effects of the anatomy they have discovered.
"These studies will allow us to ask what proportion of the water in trees comes from different depths and how that proportion changes during drought conditions," said Jackson. "Ultimately, the question behind this work is how much of the water balance of the Edwards Plateau is driven by the vegetation. And for that question, we need more than static physiological measurements. We're already teaming up with hydrologists and geologists in the region to find the answer."
Note to editors: Rob Jackson may be reached at 919-660-7408, e-mail Jackson@duke.edu.
Journal
New Phytologist