California society was built on the assumption that we would always have access to an abundant supply of clean, inexpensive water. Today the state faces a water-supply crisis. Each year, the public agencies that control our water juggle the competing needs of thirsty residents, agriculture, and industry, as well as electrical-power generation, flood control, and environmental requirements. With each passing year, their complex balancing act becomes more precarious as our population increases, our environmental problems intensify, and climate change alters the long-standing climate patterns that we had come to expect.
The state's high mountain ranges are the heart of our water system, wringing precipitation out of the clouds that blow in off the Pacific, storing that moisture as either snowpack or groundwater, and slowly releasing it during the long, dry summer months that characterize our Mediterranean climate. The snowpack in the Sierra Nevada mountains is especially critical, providing roughly 40 percent of the runoff that sustains the state through the dry period of the year.
Historically, Californians have never been content to watch this runoff flow "wastefully" to the ocean. We've spent much of the last two centuries diverting water for our own purposes, constructing a vast network of dams and aqueducts that has transformed arid desert regions into rich farmlands and sprawling cities. Increasingly, however, it's becoming apparent that this approach is no longer sustainable. We have already used most of the state's obvious dam sites, and we hope to avoid the environmental consequences of continued construction and diversion.
Global climate change is also altering California's weather patterns and hydrological regime in ways that are having repercussions up and down the state. The best available models predict that moisture that once fell in the Sierra as snow will increasingly fall as rain. And the smaller snowpack will melt more quickly or be absorbed by soils parched by more frequent drought years. Changing weather patterns are already having a dramatic impact on plant communities throughout the state. Tinder-dry forests are more and more prone to insect infestations and catastrophic wildfires.
Many feel it's time to reassess how our state deals with water. Despite the critical role it plays in our lives, we know very little about how water moves through a watershed once it has fallen from the sky. Though we have good historical records on how much rain has fallen and how much water flows down the major streams, we can't fill in the details between input and output. How long do watersheds store water in the subsurface? What combination of flowpaths does water follow to reach a stream? What factors determine how much water returns to the atmosphere via plant transpiration? How is freshwater resupplied and recycled?
The Keck HydroWatch Center at UC Berkeley was established to answer these basic questions, while also developing technology and processes for monitoring watersheds across the country and around the globe. Now completing its second year, the four-year project funded by the W. M. Keck Foundation is focusing its efforts on two NRS reserves: the Angelo Coast Range Reserve in Mendocino County and the Sagehen Creek Field Station in Nevada County north of Lake Tahoe. These two reserves present dramatically different water cycles, one dominated by heavy rains and the other by heavy snowfall.
Inez Fung, a professor of Atmospheric Science with a dual appointment in Earth and Planetary Science and in Environmental Science Policy and Management, leads an interdisciplinary team of researchers from the Keck HydroWatch Center. She recalls that the genesis for the project was a simple question that no one could answer: "I asked what I thought was a fundamental question: how old is the water in the stream? Is it from yesterday's rain? Or last year's rain? Is it this season's rain, or is it 100,000 years old?"
Looking for an answer, Fung sought out colleagues who were working on different aspects of the water cycle. Together they came up with the proposal for the HydroWatch Center, which Fung believes will revolutionize our understanding of the water cycle. "What was weather prediction like before we had satellites?" she asks. "A local forecaster would look up at the sky and say, This is what will happen with the weather tomorrow. Today satellites have given us the ability to see storm systems develop and move. So we've developed not only a better understanding of the weather, but also the ability to predict the weather. What we hope to do with the Keck HydroWatch Center is to develop a similar capability for water. We want to be able to follow the movement of water from the ocean to the land, to see it evaporate and transpire, and move across continents, into streams, and back to the ocean."
A Scientific Dream Team
Fung's associates in the Keck Hydro- Watch Center are all UC Berkeley faculty as well as national leaders in their respec-tive fields. With tongue firmly in cheek, she refers to each one by his stage of the hydrologic cycle. The "air guy," for example, is Ron Cohen, a chemistry professor, director of the Berkeley Atmospheric Science Center, and a leading authority on the changing chemical composition of the Earth's atmosphere. Cohen monitors water in the air, water that cannot yet be characterized as precipitation, whether rain, snow, hail, sleet, or fog.
Once precipitation falls from the sky and hits the soil, it becomes the focus of the "mud guy," geomorphologist Bill Dietrich. Elected to the National Academy of Sciences in 2003, Dietrich is currently a principal investigator in the National Center for Airborne Laser Mapping (NCALM) and in the Science and Technology Center (STC), National Center for Earth-surface Dynamics (NCED).
As soil moisture is taken up by vegetation, it moves into the domain of the "tree guy," plant physiologist Todd Dawson. Dawson's specialty is the interaction between plants and their environment. His research has taken him from the California redwoods to the Amazon rain forest. Dawson, a professor in the Department of Integrative Biology, serves as faculty manager for the NRS's new Blue Oak Ranch Reserve; he is also the director of UC Berkeley's Stable Isotope Laboratory (see sidebar, page 5), a facility that is pioneering new techniques for determining the source of water and the paths it travels across a landscape.
Once the water reaches a creek or river, it flows to the "stream guy," hydrologist Jim Kirchner. Kirchner is an expert in watershed hydrology and geochemistry. He also serves as the faculty director of the Sagehen Creek Field Station. Throughout his career, Kirchner has shown an uncanny ability to analyze large datasets and make breakthrough discoveries in fields as wide-ranging as paleontology and environmental science.
Innovative new sensors - "motes" - will be required to track the water through each stage of the cycle. The development of these motes falls to "sensor guy" David Culler, a professor of electrical engineering and computer science who has been at the forefront in the development of wireless networks for environmental science over the last 20 years. He works with CITRIS, the Center for Information Technology Research in the Interest of Society, a research partnership between industry, the state, and the University (Berkeley, Davis, Merced, and Santa Cruz), created to apply information technology solutions to California's biggest challenges in energy conservation, transportation, seismic safety, education, health care, and environmental monitoring (<http://www.citris-uc.org>).
Fung herself might be called the "planet guy." She is the co-director of UC Berkeley's Institute of the Environment and has been a leader in global climate change research for 25 years. Her focus is on developing increasingly accurate models for tracking global climate change; lack of accurate information on the water cycle has presented her with a major barrier. Fung is also a member of the National Academy of Sciences and a contributor to the United Nations-created Intergovernmental Panel on Climate Change (IPCC), which shared the 2007 Nobel Peace Prize with Al Gore.
Supporting these principal investigators, of course, is a large group of talented faculty collaborators, research scientists, technical staff, graduate students, and postdoctoral researchers. For example, Mary Power, faculty manager of the Angelo Reserve, has hosted the Keck team at that site and contributed valuable insights on how river food webs are affected by changes in the hydrologic cycle. NRS reserve managers also play key roles. Peter Steel, manager at the Angelo Reserve, and Jeff Brown and Faerthen Felix, manager and assistant manager at the Sagehen Creek Field Station, have contributed to the installation and maintenance of the equipment and networks at their respective reserves. Steel, for example, has worked with Collin Bode, from UC Berkeley's California Biodiversity Center, to install solar panels high up in a number of 60-meter-tall Douglas-firs to power the sensor network. The two have also installed radios on each tree, creating a self-monitoring, wireless network that will carry data from the forest back to campus.
With this wide range of expertise and cutting-edge technical wizardry, the biweekly meetings of the "Keckians," held on Berkeley campus at McCone Hall, take on the flavor of a scientific United Nations, as each participant tries to interpret the languages of the others' specialized knowledge areas. Where geomorphologists speak of "soil composition" and "rock porosities," the plant physiologists are more interested in "microclimates" and "transpiration rates." Hydrologists and atmospheric scientists speak in terms of "parts per million of contaminants" or "maximum streamflow rates," while for engineers it's all about "sensors," "motes," and "network protocols." Aided by large doses of high-quality chocolate (courtesy of Fung), the scientists struggle to understand and integrate their colleagues' perspectives as they troubleshoot balky sensors, interpret preliminary datasets, generate new field questions, and brainstorm potential ways to achieve answers.
Their discussions make it clear that tracking water as it moves through the watershed is technically extremely challenging. A large portion of one meeting, for example, focused on surprising first readings from boreholes drilled into the Angelo watershed. Why did the moisture content in the bedrock go off the chart with the first light rains of the season? Was this a normal response? Was something concentrating the moisture, or did something go wrong with the sensors? Did the material used to fill in the boreholes create a "wet-diaper effect" that produced errant readings? How could they find out? How could they fix it?
Welcome to Rivendell
These theoretical discussions become real at the Angelo Reserve, where the Keck team is installing its first sensor network. The team has focused its attention on a small watershed on Elder Creek and dubbed it "Rivendell," after the Middle-earth Elven outpost in J. R. R. Tolkien's Lord of the Rings. They've also named the individual trees that support the wireless network and solar-power panels. One particularly old and difficult-to-climb Douglas-fir is known as Treebeard, named after the oldest Ent, one of Tolkien's race of humanoid treelike creatures.
And, indeed, Rivendell bears some resemblance to a scene from a scientifically focused Middle-earth. To protect the fragile soils of the steep hillside, Reserve Manager Steel has installed a network of aluminum ladders that connect with fallen tree trunks to climb from the creekbed to the top of the ridge. Researchers move over this maze as they string wire and install sensors. Other scientists hoist themselves 50 meters up into the forest canopy to install rain gauges, sap sensors, and solar panels. Wire-filled PVC pipes jut up from the ground, indicating the locations of sensor-filled boreholes that reach down to the bedrock.
Rohit Salve, a research scientist at Lawrence Berkeley National Laboratory, is leading the team that is installing the underground sensors. He pauses from his work connecting a data-logging panel in order to explain the Keck team's strategy: "To get a handle on what's going on belowground, we put in a series of soil-moisture sensors. Some are traditional sensors, and some are experimental time-domain reflectometers (TDRs), designed to give us really high spatial/temporal resolution. For example, over here we have a resistance probe that goes down 2 meters with sensors at 10-centimeter intervals. This will give us a detailed pattern of how water is wetting or drying up in the rock below. And since we're very close to trees and roots, w'’re also trying to work out how the roots that are sucking up the water are actually influencing what's happening to the water around it."
With a wave of his hand, Salve indicates another pipe protruding from the ground. "We also have seven wells that go down about 30 meters to track the response of the deep water table when the rain comes. Our preliminary findings from last winter were quite interesting. It turns out that fractured rock is a good medium for transporting water quickly, so the rocks deep down in the profile get wet much faster than those near the surface."
From the day they started working in the watershed, the research team members were puzzled by how its shallow, steep soils could provide enough moisture for the forest throughout the area's long, hot summers. Rohit Salve, Todd Dawson, Bill Dietrich, and others are now investigating the idea that this deep "rock moisture" might be the key. They propose that, on sunny days, the large trees use both their deep and shallow root systems to extract moisture from the ground. Once the moisture reaches the trees' leaves, it is transpired into the atmosphere. At night, however, the stomata on the trees close, transpiration stops, and the flow of moisture in the shallow root systems reverses. The trees continue to use their deep roots to take up water, which then flows, via the shallow root systems, into the shallow soils. The next day when transpiration resumes, the trees have two water sources to draw on, one shallow and one deep (see illustration on page 15).
This phenomenon, known as "hydraulic lift," was discovered by Todd Dawson and his research group and has now been seen in many plants, including sugar maples, African acacias, deep-rooted plants from Australia and California, and even Amazonian trees. "Hydraulic lift provides water not only for the trees, but also for the plants that grow around the trees and have their roots commingling in and around the tree roots," he explains. "When I published my first paper on this, it created a big ripple effect, because plant ecologists had always supposed that one plant growing up next to another plant was a bad thing. But it turned out that living next to a tree that does hydraulic lift is not bad at all. There's actually a facilitation going on. This new understanding shocked a lot of ecologists, because they had always thought of plant-to-plant interactions as being competitive. But, in this case, it's facilitation. And this has become an area where ecological research has really taken off - the competition/facilitation tradeoffs that plants have."
To determine whether hydraulic lift is happening in Rivendell's Douglas-firs, the team will need to position sap-flow sensors on the trees’ shallow root systems to determine whether the flow of moisture reverses at night. The soils have been too dry and too hard this fall to install the sensors without damaging the delicate roots. Their current plan is to wait for the winter rains to soften the soils before they install their sensors. While their idea remains a hypothesis at this point, confirmation of their theory would have major implications for land managers, emphasizing the important role that large, old trees play in sustaining the other plants in the forest.
For Dawson, this line of research confirms the value of working in interdisciplinary groups. "Bill [Dietrich] is an earth scientist, but he appreciates how biology impacts the earth systems he studies. Inez also. Five years ago, when I started showing her some of the things we were learning about transpiration and trees, she was amazed: 'Really? Plants are doing this? This is going to have such a huge impact on our global climate models.' Now, she's one of the biggest advocates of the roles trees play in hydrology. For a long time, I'd been thinking about rock moisture and how plant roots get into deep rock cracks, but I'd never known how to approach it. It was really Bill who said, 'We've got to measure this,' and then Rohit got involved. Those guys have the expertise in earth sciences to know how to approach the problem."
Others on the research team have also been busy this summer, installing a wide array of different sensors. Although the plant physiology team could not install sap-flow sensors in the tree roots, they did climb into the forest canopy at Angelo to install 16 sap sensors to determine where the trees are getting their water and how much they're drawing in each day. Other scientists set up hyper-accurate rain gauges and weather stations. Jim Bishop of Earth and Planetary Science at UC Berkeley and Todd Wood of Lawrence Berkeley National Laboratory (who together developed robotic sensors in the open ocean) automated water sampling in the creek and in a nearby well so they could analyze the chemistry of the water. Bill Dietrich and his students have installed time-delay cameras for the hydrology team and installed water-chemistry samplers in the creek and in a nearby well, as well as time-delay cameras for monitoring sediment transport in the seasonal tributaries. Finally, Jay Taneja, a graduate student on David Culler's sensor team, has designed and produced several dozen custom motes for recording microclimate data throughout the watershed, both on the ground and in the trees. Eventually the system will include more than 200 motes.
Currently, most of these data collection points are linked by wires to a series of central control panels. To access the data, researchers must make the three-and-one-half-hour drive from the San Francisco Bay Area to the Angelo Reserve to download the information stored on data loggers. No wonder the ultimate vision is to create a "wireless watershed" with sensors that transmit data to central data loggers, then up into the wireless network Steel and Bode have installed in the treetops, out to an Internet node on a nearby peak, and into the researchers' office computers a couple hundred miles away.
There, Fung and her team of statisticians and modelers are gearing up to make sense of the large volume of data the system will produce. "I'm not an outdoors person," admits Fung, "so it will be incredible to be able to sit at our office computers and view real-time data as rain falls in the Elder Creek Watershed."
Contrasting Watersheds
The value of the Natural Reserve System's ecosystem diversity, from reserve to reserve, became evident early this summer when scientists from the Keck HydroWatch team traveled from Angelo, near Northern California's coast, to visit the Sagehen Creek Field Station, in the Sierra Nevada, and begin planning the layout for their second watershed-monitoring system. Located in a valley high on the eastern side near the Sierra crest, Sagehen receives some of the heaviest snowfall in the state. The watershed is broad and stretches from high-altitude evergreen forests to sagebrush at the reserve's eastern edge. Broad meadows and fens line the creek.
As was the case at Angelo Reserve, a lot of scientific data has already been collected at Sagehen. Weather and streamflow records stretch back to the 1950s, and 10 weather stations currently collect data at different elevations in the basin. Detailed, laser-based LIDAR maps lay out the topography in great detail (see Transect 23:2, page 1), and a 2007 field class from UC Santa Barbara, under the direction of emeritus professor Art Silvester, created a geologic map of the entire watershed (see Transect 25:2, page 11). Researchers from the University of Nevada, Reno, the USDA Forest Service' s Pacific Southwest Research Station, the Desert Research Institute, and a number of other institutions have also installed a range of sensors throughout the basin.
"Stream guy" Jim Kirchner knows this area and its hydrological characteristics well. As faculty manager of the Sagehen Creek Field Station, he has hiked all of the nearby watersheds (see sidebar, page 6), taking water samples and recording streamflow data. "Understanding how water cycles through the natural environment," he explains, "is the key to understanding a host of things we need a better handle on, like the availability of streamflow and groundwater for ecosystems and people, as well as the availability of soil moisture for the trees on the landscape, keeping the forest moist enough so that it doesn't burn down. There are many issues in managing this kind of landscape, and how water is processed at the surface of the Earth is pivotal for all of them."
Kirchner is also very aware of some of the missing links in our understanding of the water cycle. Plant transpiration is one of the factors he has in mind. More than half of the precipitation that lands in the Sierra evaporates back to the atmosphere. If water managers want to know how much rainfall will end up in a stream, they need to understand what controls evaporation across a landscape. Scientists have a general understanding of what affects transpiration rates. Transpiration goes faster when the atmosphere is dry, the weather is warm, and the landscape is wet, but slower when the atmosphere is humid, the sky is cloudy, or the landscape is dry. But they don't know the details, and so they find it difficult to make accurate future predictions.
Kirchner provides some good examples. "What will happen to evaporation rates if we cut down, or lose, half the trees in a forest?" he asks. "What will happen over the next 50 years if 50 to 80 percent of the snowfall comes as rain instead? How will that change low flows in the summertime?"
These are the kinds of questions Kirchner and his partners in the HydroWatch Project hope to answer. "Right now, our answers would be little better than guesses," he admits. "And those are things we need to know in order to forecast the impact of climate change. If we're wrong about water yield from the Sierra Nevada by 20 percent, that's roughly equal to all that the cities of California use each year. (Agriculture uses most of the other 80 percent.) California's water managers need to know how the landscape is going to respond to the weather, how much water won't be available because it's going to evaporate back into the atmosphere, and how this number will change 20 years down the road."
One HydroWatch goal is to develop new ways to visualize and measure water flow through the ground with new types of sensors and communications technology. The time-domain reflectometers, or TDR probes, installed at Angelo's Rivendell study site, for example, provide a new level of detail on the moisture content of soil. But will the new technology reveal new information? Kirchner is confident it will. "This is not just a shot in the dark," he asserts. "We have good reason to think there are some interesting things out there, based on tantalizing data we've collected with existing instruments. The measurements we're already getting from the meadow here at the [Sagehen] field station are telling us something about how the system works. We just need to unpack the story in more detail to get a much richer understanding of the basic processes that control hydrologic behavior. More precise probes and sensors might allow us to do that."
Streams as Landscape Sensors
Beyond the obvious management implications of the Keck Center's findings for California's water and land managers, the group is also focusing on questions that are interesting for their own sake. Fung's original question — "How old is the water in the stream?" - is a perfect example.
Kirchner acknowledges he doesn't have an answer yet to that one. He does know, however, that a landscape can hold water for weeks or months and then, when a heavy rainstorm hits, release that water into a stream in a matter of hours or even minutes. Many hydrologists have tried to model this behavior, but testing the validity of these models will require new types of technology.
In the meantime, Kirchner is pursuing a separate line of inquiry that might provide a valuable new perspective. Rather than relying on TDR probes or boreholes that record data in a single spot, he’s exploring ways to use streams to get a better sense of what’s happening across a landscape. In a sense, he’s back to looking at input and output signals again, but this time in a more sophisticated form. What is the rainflow rate in and what is the streamflow rate out? What are the chemistry and isotopic composition of the rain, and what are the chemistry and isotopic composition of the streamflow? His concept is to use streams to synthesize a broader spectrum about what's happening underground.
"I'm fascinated with the idea that what happens in the stream reflects what's going on across the landscape," he explains, "so I've been developing techniques for watching the fluctuations in the stream on time scales of minutes and hours, and using that to infer what's happening on the landscape. How hard is it raining? - or, rather, how fast is the forest transpiring water back to the atmosphere? And I've found that, under some circumstances, I can figure out what's going on in the landscape by looking at the stream. You can use the stream as a kind of electrocardiogram for the physiology of the landscape."
Kirchner's inspiration for this approach came from his long-term daily monitoring of Sagehen Creek. During the spring, the streamflow depended on snowmelt. Each morning as the sun rose, the stream would begin to rise and would continue to rise through the day, peaking at sunset. Then, in the evening when the snow stopped melting, the streamflow would decline as it drained the landscape of water. The next day, a new pulse of water would surge into the landscape, and the process would start again.
During the summer, Kirchner noted, the pattern was reversed, and that reversed pattern depended on plant transpiration. When the sun rose, plant transpiration pulled water out of the landscape and streamflow declined. This process continued all day until the evening, when the sun went down and the plants stopped transpiring. At that point, the stream would start to rise again. Kirchner is now building on these observations to develop technique for modeling a stream's future behavior: "If it's raining [this hard], then we expect the stream to rise by [this much], [this many] hours later." He's also using his modeling technique to look backwards: "If the stream rose by [this much], then [this amount] of rain must have fallen or [this amount] of snow must have melted."
"So what we've done is develop techniques for quantitatively looking at that relationship both forwards and backwards," Kirchner explains. "We can know whether a summer day was cloudy, because plant transpiration will be lower, and hence the decrease in streamflow will not be as great. You can absolutely see this. I can download the streamflow record from Sagehen Creek and know what the weather is up here - then I can check the weather station and verify it. That's big fun to look at!"
High-Risk Science
Inez Fung once referred to the HydroWatch Project as "high-risk science," because no one could predict the value of the data the team would collect until after they were collected. As the team completes its first full sensor deployment, answers will begin to emerge. Will attempts to pry open the black box of long-standing hydraulic models reveal critical new information - or simply confirm the validity of existing models? Will the new probes and techniques developed by the team improve the accuracy of future streamflow models? Will the Keckians get the data they need to make more accurate predictions on global climate change?
Kirchner is optimistic that the project will pay rich dividends. "Almost by definition," he explains, "when you're investigating the unknown, you don't know what you're going to find, sometimes even after you've found it. And you don't know necessarily how it's going to be useful. Yet we have seen, over and over and over again, that if you let smart, curious people follow their instincts and try to figure out how the world works, you discover something that turns out to be useful."
Fung echoes Kirchner's confidence: "Once we're successful with the sensor technology, we'll be able to develop an understanding of how water goes up and down trees - when it goes up and down into the root system, how roots access the water at various depths in the soil, whether the roots fracture the rocks, and what goes back into the stream. If we're successful and have the system replicated around the country and around the world, then we'll have developed an underground system for tracking water under our feet. That will be cool."
Whatever its outcome, the project illustrates the value of the University's Natural Reserve System. As Kirchner notes: "Trying to find out how the forest landscape works biologically, and hydraulically, and geochemically, is the best reason I can think of for places like Sagehen and Angelo to exist. Such research stations are rare. California is lucky to have 36 of them." - JB
For more information:
Visit the Keck HydroWatch Center homepage at: <http://hydrowatch.cs.berkeley.edu/Welcome.html>.
And see page 13 of the NRS publication, Special Research Projects: National Centers & Other Landscape-scale Projects that Utilize NRS Reserves, online at: <http://nrs.ucop.edu/program_reports/Research_Projects.pdf> - also available in hard copy, upon request, from the NRS Systemwide Office: 510-987-0150.
|
