Developing clean energy sources, using energy more efficiently, and understanding the effects of increased energy-source emissions on regional climate and forest productivity are targets of ORNL research.

Clean Energy and Climate Effects

Satisfying the world's growing appetite for energy could set the stage for global climate change. Much of our electrical power and heat comes from the combustion of fossil fuels, which release carbon dioxide (CO₂) into the atmosphere. CO₂ is one of the primary gases that contribute to the "greenhouse effect"—the phenomenon in which certain trace gases in the atmosphere trap the earth's radiated energy, causing a gradual warming of its surface. Significant global warming might lead to regional shifts in agricultural and forest productivity and cause the spread of disease and the relocation of coastal populations.

To delay the onset of significant global warming, the developed nations may change their portfolio of energy sources. Some are considering replacing coal with natural gas because combustion of gas emits almost half as much CO₂ as the combustion of coal. Other options are to make greater use of renewable energy sources (including hydropower facilities) and nuclear power because they do not produce CO₂.

A second approach to cutting CO₂ emissions is to develop technologies, such as "smart" cars, buildings, and appliances, that use energy more efficiently (see the article, "Driving the Transportation Revolution."). ORNL also has contributed in this area by developing more efficient refrigerators and heat pumps. A third approach is to focus on both understanding the effects of rising levels of atmospheric CO₂ and pre- venting it from building up to undesirable concentrations. Computer modeling experts are predicting the impacts of increasing emissions of CO₂ on climate, and ecologists are studying the effects of elevated atmospheric CO₂ concentrations on forest productivity. Other scientists are exploring the emerging science and technology of carbon sequestration—the capture and secure storage of CO₂ emitted from the combustion of fossil fuels. The U.S. Department of Energy supports all these approaches. In 1998 ORNL researchers had some outstanding achievements in these areas.

ORNL Recommends Building
Three Hydro Projects

Small mountain streams in northwestern Washington have been proposed as sites for hydroelectric projects.

Small dams that generate electricity are needed to help meet growing power demands in the Pacific Northwest, but the benefits of proposed hydroelectric projects must outweigh their environmental costs. To balance power needs with potential environmental impacts, as required by the Federal Power Act and National Environmental Policy Act, the Federal Energy Regulatory Commission (FERC) conducts environmental assessments of new and existing projects proposed for licensing.

In April 1998, ORNL completed for FERC the final environmental impact statement for eight new hydroelectric projects proposed for the Skagit River Basin in Washington. The ORNL group—Bo Saulsbury, Rich McLean, and Bill Staub (Energy Division) and Warren Webb, Glenn Cada, and Mark Bevelhimer (Environmental Sciences Division)—recommended that three of the eight projects be licensed for construction and operation, provided that the applicants implement certain mitigation measures. If constructed, these three projects would generate about 72 gigawatt hours of electricity annually.

Complex environmental issues arise around the plan to clear land and dam streams for hydroelectric projects in the Pacific Northwest. Will old-growth forest be cleared or protected? Will project construction or an accidental rupture of the project pipeline adversely affect slope stability and result in erosion that could affect water quality? Will water quality degradation or changes in stream flows further threaten Pacific salmon stocks? Can threatened and endangered species, such as the spotted owl and marbled murrelet, still be protected? Will Native American treaty rights and cultural practices be respected and preserved? What will be the socioeconomic effects of construction and operation, such as the impact on housing and schools of workers and their families moving into the community?

ORNL ecologist Warren Webb stands in an old-growth forest near a proposed hydroelectric project site.

The ORNL team recommended these mitigation measures for the three projects: (1) prevent erosion and control sediment to protect water quality; (2) increase stream flows and restock resident fish populations in the projects' bypassed reaches; and (3) acquire and preserve between 7 and 10 acres of forest along other streams to replace the forest habitat cleared for the projects. These mitigation measures would be implemented along with other environmental measures proposed by the applicants.

"We did not recommend licensing the five other proposed projects," Saulsbury says, "because they would pose significant environmental impacts even with available mitigation measures."

Computational Tool Could Aid
Search for Oil and Gas

To solve the problem of faulty seismic image focusing, which plagues the oil and gas exploration industry, (from left) Jacob Barhen, Edward Oblow, Vladimir Protopopescu, and David Reister developed TRUST, a computational method for global optimization. For their work on TRUST, they received an R&D 100 Award in 1998.

The propagation of sound waves underground may contain relevant information about the presence of oil and gas. Therefore, many petroleum exploration companies use seismic analysis for hydrocarbon exploration. Seismic data are obtained by recording the energy returning to the earth's surface from an underground source of acoustic waves. These waves propagated into the earth are reflected back whenever they encounter a change in acoustic impedance (e.g., passing from a dense shale into a porous sandstone layer that may contain oil). An array of receivers on land or underwater picks up sound waves from each reflected signal. Petroleum industry researchers plug the recorded data into a computer code that provides an image of the subsurface geological structure.

Unfortunately, the reflected signals carrying useful information are often buried in the noise from the sensor electronics and from disturbances arising from the degradation and misalignment of some seismic signals. Misalignment is caused by unpredictable delays in the recorded travel time of the seismic waves (which pass more quickly through solid rock layers compressed deep underground than through less rigid rock layers near the surface). As a result, the image of subsurface structures is highly distorted. For large-scale seismic surveys, this problem typically had been considered intractable by industry experts, until ORNL came up with a mathematical and computational solution.

To address the challenge of faulty seismic image focusing, Jacob Barhen, David Reister, Vladimir Protopopescu, and Edward Oblow, all of ORNL's Computer Science and Mathematics Division, developed Terminal Repeller Unconstrained Subenergy Tunneling (TRUST), a computational method for global optimization. This fast, powerful, and robust tool could be used with a petroleum industry computer code to combine and correlate relevant data from all the receivers to get the sharpest possible image. By enabling a multisensor fusion algorithm to identify the meaningful reflections by separating them from the noise, the TRUST algorithm solves the seismic-image focusing problem plaguing the oil and gas industry, potentially reducing exploration costs.

"TRUST rapidly and reliably eliminates large, useless regions of the search space before they are actually searched," says Barhen, an ORNL corporate fellow. "Hence, it increases the overall efficiency up to 45 times higher than any competitive approach."

The development of TRUST was sponsored by the Engineering Research Program of DOE's Office of Science. Its application for geophysical imaging was funded by DOE's Office of Fossil Energy in conjunction with the DeepLock petroleum industry consortium. In 1998 the TRUST developers received an R&D 100 Award.

A More Efficient Gas-Fired Heat Pump

ORNL, in a cost-shared program with York International, has developed a triple-effect absorption chiller, an advanced natural-gas-fired heat pump that is 30 to 40% more energy efficient than other heat pumps. The device will be used to provide space cooling for large commercial buildings. In comparison with the double-effect chiller developed more than 40 years ago, the ORNL chiller's emissions of CO₂ are 99.9% lower. In addition, its emissions of sulfur dioxide and total particulate solids are reduced by 73% and 99% respectively.

From left in front of the field test model of the triple-effect absorption chiller at the Clark County Government Center in Nevada are Ronald Fiskum, DOE program manager and, from ORNL, Bob DeVault, Patti Garland, Abdi Zaltash, and Tony Schaffhauser. The five were celebrating an agreement signed by various partners October 27, 1998, to proceed with the test deployment of the world's first triple-effect absorption chiller at the center.

Bob DeVault of the Energy Division, a co-inventor of the triple-effect absorption chiller, says the "triple effect" comes from feeding a refrigerant-containing absorbent solution through high-, medium-, and low-temperature generators. "The high-temperature condenser receiving vaporous refrigerant from the high-temperature generator is coupled to both the medium-temperature and low-temperature generators," DeVault says. "As a result, the internal recovery of heat within the system is boosted, increasing its thermal efficiency, improving indoor comfort and indoor air quality, and greatly reducing CO₂ and other emissions."

In an October 1998 speech, Secretary of Energy Bill Richardson noted that laboratory testing of the triple-effect chiller prototype at York International showed that it uses 40% less energy than other types of heat pumps. He also announced that the first field test of a full-size triple-effect chiller will be conducted in Clark County, Nevada, in 1999 and 2000. Patti Garland of the Energy Division is leading ORNL's participation in the test.

ORNL's Role in Energy Savings
Performance Contracting

This map shows the Super ESPC regions in the United States. ORNL has played a major role in energy savings performance contracts for the Southeast region.

By 2005, all U.S. federal agencies must use 30% less energy in their buildings than they consumed in 1985. That's the mandate of the Energy Policy Act of 1992 and Executive Order 12902. But energy efficiency improvements cost money, so how can federal government agencies reduce their energy use when their capital expenditures budgets are so tight? One solution is an alternative financing arrangement called energy savings performance contracting (ESPC).

Instead of relying on traditional congressional appropriations of capital funds to finance energy efficiency improvements in federal buildings, federal agencies sign contracts with private energy service companies that agree to pay up-front costs for identifying building energy cost-saving measures and acquiring, designing, installing, operating, and maintaining the energy-efficient equipment. In exchange, the contractor receives fixed payments from the cost savings resulting from these improvements until the contract period expires, up to 25 years later. At that time, the federal government retains all the savings and equipment.

ORNL and DOE's Oak Ridge Operations (ORO) are participating in the Super ESPC program for DOE's Federal Energy Management Program. Super ESPCs are regional "all-purpose" or national "technology-specific" contracts that allow agencies to negotiate ESPC delivery orders with an energy service company without having to start the contracting process from scratch.

The Oak Ridge team has awarded contracts potentially worth $750 million to six private companies for "all-purpose" ESPC in the Southeast. The team has also awarded contracts potentially worth $500 million to five private companies for geothermal heat pump "technology specific" ESPC nationwide. Key technical participants on the Oak Ridge team are Patrick Hughes and George Courville, both of ORNL's Energy Division, and Angela Carroll and Wayne Lin, both of ORO.

Energy services companies are being contracted to help government agencies reduce their energy costs, meet federal energy savings requirements, and eliminate the maintenance and repair costs of aging or obsolete energy-consuming equipment. The contractors also are responsible for operating and maintaining the new energy-saving equipment during the contract term if the federal site so desires.

For example, under an ESPC agreement, energy-efficient lighting, variable-speed motor drives, and an energy management control system are being installed at the Statue of Liberty. For ORNL a contract has been signed with Duke Solutions, Inc., to work with Hicks & Ingle Corporation to quickly replace a failed water chiller with a more efficient one in an Environmental Sciences Division building. In other federal complexes, lighting retrofits, additional insulation, cogeneration systems, and geothermal heat pumps (to replace conventional heating and air conditioning units) are being installed.

Angela Carroll at DOE-ORO is the contracting officer for the six Southeast region and five geothermal heat pump contracts. The DOE contracting officer's representatives for all 11 contracts are Doug Culbreth and David Waldrop of the DOE Atlanta Regional Support Office. Patrick Hughes and a team of project facilitators from ORNL's Energy and Engineering divisions lead acquisition teams at federal agency sites through the delivery order process and provide technical assistance.

"Our team," says Hughes, "will verify that each project's annual cost savings from reduced need for energy and maintenance will exceed the agency's annual payments to the energy service company for providing the energy efficiency improvements and negotiated services." The ORNL-ORO team will also support an appropriate integration of advanced technologies sponsored by DOE's Office of Energy Efficiency and Renewable Energy into the Super ESPC program, such as was accomplished for the geothermal heat pump.

Superconducting Cable and Transformer

Jonathan Demko (left) and Winston Lue check out components of the superconducting cable at the test facility at the Southwire Company plant in Georgia.

A newer way to use energy efficiently is to harness superconducting wire chilled by liquid nitrogen because the high-temperature superconductor offers no resistance to electrical flow. Researchers in ORNL's Fusion Energy Division have been involved in developments that would use this wire to transmit higher amounts of electrical current underground and to change voltage and current levels.

ORNL has entered a partnership with Southwire Company to develop a 30-meter, high-temperature superconducting (HTS) cable at the company's Georgia headquarters. The cable will carry enough energy to power a small city. In 1998 ORNL staff researchers J. Winston Lue, Michael J. Gouge, and Jonathan A. Demko designed, completed, and operated a research facility to support the successful development and testing of the first U.S. system prototype of an HTS power transmission cable. The goal is to retrofit HTS cables in existing underground ducts with cables that can carry 3 to 5 times more current.

The Laboratory played a significant role in the fabrication and testing in 1998 of the first U.S. experimental electric transformer made from HTS wire. The team included ORNL, Waukesha Electric Systems, and Intermagnetics General Corporation. The ORNL team members were Bill Schwenterly, Jonathan Demko, Andy Fadnek, Randy James, Ben McConnell, and Isidor Sauers.

"We were responsible for the innovative cryogenic system that allowed the 1-million-volt-ampere (MVA)-rated transformer to be cooled to 20 Kelvin without liquid helium," Schwenterly says. "Our work set the standard for cryogenic cooling systems for emerging electrical applications based on HTS technology."

Compared with traditional paper-oil-insulated transformers wound with copper wire, HTS power transformers will increase efficiency of power delivery, eliminate the use of oil (a fire hazard and environmental contaminant), and upgrade the capability to handle power overloads. The team is now participating in the development of a 5-MVA transformer to be operated on the utility grid at Waukesha Electric's factory in Wisconsin.

Regional Climate Modeling and
Assessment at ORNL

First, think globally. What are the effects on future climate of rising concentrations of CO₂ from increased fossil fuel combustion? No one knows for sure, but global climate models now being developed for parallel supercomputers may predict these effects accurately someday. Now, think locally, or at least, regionally. If significant global climate changes are expected, what are the implications for the southeastern United States? The answer depends on the ability of computer specialists to predict changes in regional climate based on results of global scenarios.

The challenge is to present these changes on a much finer spatial and temporal scale. If such "downscaling" could be done, it might be possible, for example, to predict accurately whether East Tennessee will have less precipitation and more tornadoes in the next decade. Or whether the Carolinas will endure more hurricanes in the first two decades of the next century than they did during the last two decades in this century. Or whether the sea will rise and inundate the coast of Florida in the middle of the next century.

John B. Drake of ORNL's Computer Science and Mathematics Division and three investigators in ORNL's Environmental Sciences Division—Mac Post, Tony King, and Mike Sale—recently received funding for a regional climate modeling and assessment project. The source was ORNL's internally funded Laboratory Directed Research and Development Program.

"We have developed a statistical conceptual framework for simulating regional climate," Drake says. "The framework will house a variety of models, compare models, and combine results of different models. It uses physically based weather models, results of ecological experiments, and historical climate observations. Eventually, we will be able to predict temperature and precipitation data for any 1-kilometer grid for a particular decade or longer.

"But, what our customers want is predictions of extreme events. Our goal is to be able to predict that a region in the Southeast during a certain decade will experience, for example, 30% more tornadoes, or 20% fewer hurricanes, or 25% more big storms that cause major floods than it did in the 1980s. Of course, we will also provide error bounds because we cannot make such predictions with 100% certainty."

Regional modeling requires many highly resolved data fields, such as terrain (elevation), land use category, vegetation categories, soil categories, and ground temperature. The color scheme for land use for the Southeast (shown here) is as follows: black—urban land; yellow—agriculture; purple—treeless grassland; green—deciduous forest; dark green—coniferous forest; red—mixed forest and wetland; blue—water; and light blue—marsh or wetland.

The ORNL group has been studying the ability of today's global circulation models to predict the fate of rainfall in the Southeast and keep an accurate freshwater budget. "Our model tells us how much rain goes east of the continental divide to the Atlantic Ocean and how much goes west to the Mississippi River or to the Gulf of Mexico coast," Drake says. "But we want to improve the model's resolution by partitioning rainfall so we can predict how much actually goes into each of the major rivers, such as the Ohio and Tennessee rivers."

Making predictions for the Southeast will require scientific discovery of the relationship between local biogeochemical processes and large-scale weather and climate shifts. "If the climate becomes warmer and drier," says King, "the growth of smaller, shallow-rooted trees in a region's forests may be reduced, decreasing the region's uptake of carbon. The rise in temperature could increase the rates of tree respiration and decomposition of soil and litter, resulting in greater releases of carbon to the atmosphere that could bring increased climatic warming.

"In addition, w1e need to discover the effects on climate of changes in water availability to regional forests. These changes affect transpiration, the way in which trees transfer rainwater back to the atmosphere. We also must determine the effects on forests of seasonal changes such as early springs—which could result in early leaf production, increased growth, and greater carbon uptake—and early springs punctuated with later freezes that could hamper reproduction, reducing the long-term productivity of the forest.

"We will look at the impacts on forests of summer and winter droughts, which are expected to have different effects on forest growth," King adds. "Our models will run different climate scenarios to determine how they affect the ability of forests, crops, and other plants to take up carbon and influence future climate."

The purpose of the regional climate model is to provide scientifically grounded information for modelers in the assessment community. These researchers seek to predict the impacts of climate change on health, food production, the environment, and the economy.

"Suppose that our model predicts a slightly warmer and drier climate for the Southeast in the next few decades," Drake says. "These results could be plugged into models used to determine the effects of temperature and precipitation changes on mosquito proliferation and the spread of malaria."

Other modelers will look at climate impacts on agricultural production, growth of forest trees, and reproduction of wildlife species. Some modelers will try to determine if a climatic warming could have immediate economic impacts, such as severe coastal flooding from a rise in sea level and a higher frequency of hurricanes.

The ORNL modelers are expecting to examine the impact on regional climate of various CO₂ emission levels in the Southeast. They may be running different scenarios in which regional firms burning fossil fuels pay other nations for the right to exceed limits in emitting carbon. They will also look at the effects on climate of enhancing the natural sequestration of carbon by improved management of land, forests, and agriculture.

"By predicting future climate for the Southeast," Drake says, "our community of ORNL researchers could become an important link between the modeling community and the policy-oriented impact assessment communities who are devising strategies to deal with increasing atmospheric CO₂ and the predicted impacts of global and regional warming."

For that reason, as part of the U.S. Scientific Simulation Initiative, ORNL is proposing to serve as the regional climate prediction center for the Southeast. ORNL researchers use global circulation models, but for climate prediction, they are starting to think regionally.

High-Performance Storage System
and Climate Data Archive

This ARM millimeter cloud radar instrument in Oklahoma cattle country enables scientists to determine whether a cloud contains mostly ice crystals or liquid water. Such measurements help scientists predict the degree to which the cloud reflects, absorbs, or transmits sunlight.

A scientist needs data about how different types of clouds reflect, absorb, and transmit the energy of sunlight. The data, based on measurements taken by instruments on the ground and aboard airplanes and satellites, will help the scientist improve the accuracy of a computer model in predicting the influence of human activities on climate.

The scientist accesses a web-based interface and requests 100 files of data from DOE's Atmospheric Radiation Measurement (ARM) data archive, located at ORNL. In this archive are more than three million files containing more than 15 terabytes of data. Three robots retrieve the tapes on which the requested files are stored and load them for copying on the disk drive of the ARM web-site server. Within an hour, the scientist can access the requested files.

For the past two years, the ARM data archive has been using the High-Performance Storage System (HPSS), storage-system software that leads the computer industry in capacity and transfer speeds. HPSS was developed by a consortium of DOE national laboratories and IBM. The DOE participants are ORNL, Sandia, Lawrence Berkeley, Los Alamos, and Lawrence Livermore national laboratories. HPSS, which received an R&D 100 Award in 1997, is marketed by IBM.

Deployed at about 20 sites and used productively for more than two years, HPSS is now the standard for storage systems in the high-performance computing community. The HPSS community has been joined by two new industrial partners, Sun Microsystems and Storage Technology Corporation.

HPSS 3.2 has been the version in production use at most sites for more than a year. In December 1998, HPSS 4.1 was released by the collaboration and is expected soon to become the production storage system software at most sites. Version 4.1 provides significant improvements in scalability, performance, end-user access, small-file support, and input-output support for massively parallel supercomputers.

ORNL's primary customer for HPSS is the ARM project; the Laboratory's role is to provide and support the data archive. The ORNL HPSS system manages the hierarchy of devices storing more than 3.5 billion measurements. It can place 2000 new files a day into storage. It will eventually be able to routinely find and retrieve up to 5000 files an hour to meet the growing requests for information related to global change.

Facing a Future of More
Carbon Dioxide for Forest Trees

Take an eastern deciduous forest—the type that displays brilliantly colored leaves in the fall. Expose it to air enriched in 50% more CO₂ than is present in the atmosphere. Reduce the amount of water normally available to this forest.

Is this a recipe for slower or faster forest growth? Because of the expected rise in the combustion of fossil fuels to satisfy the world's growing energy appetite, scientists want to know if additional emissions of CO₂ might significantly affect the growth of forest trees. What about feedbacks from the forest to the atmosphere? If a forest is affected by increases in CO₂ concentrations that can influence the climate, could the forest itself affect the climate?

Standing tall in a small sweetgum plantation are a tower and vent pipes that provide the trees with additional amounts of carbon dioxide. This hardware is part of ORNL's novel free-air CO2 enrichment system. During the first year of exposure to the increased CO2 concentrations, the trees grew faster and conserved water.

To face these tough questions, ORNL has a world-class user facility in a forest that features free-air CO₂ enrichment (FACE) technology. The hardware for the selected hardwoods—a 10-year-old sweetgum plantation in the Oak Ridge National Environmental Research Park—elevates the air's CO₂ concentration across the plantation's 25-meter-diameter plots. Because the plantation has no walls, the effects of elevated CO₂ can be studied under natural field conditions. The facility is open not only to nature but also to researchers from universities and other laboratories across the nation who wish to study the response of forests to atmospheric CO₂ enrichment.

"Plant physiologists and ecologists have learned a great deal about how small trees and other plants will respond to increasing CO₂ concentrations in the atmosphere, but it is much harder to say how a whole forest will respond," says Richard J. Norby, leader of the collaboration at the FACE facility. "Understanding the response of forests is challenging because they are tall and biologically complex. Fortunately, next-generation technology in the FACE facility should help us better evaluate the sensitivity of forests to global change and, in turn, understand the dynamic role played by forested ecosystems in the earth's climate system."

It is known that forests provide a critical "biotic" feedback between the earth's terrestrial vegetation and our ever-changing climatic system. Each is dependent on the other largely because forests and the atmosphere are sources of water and CO₂ to each other. Large-scale studies of these interdependencies are needed for accurate climate predictions and for understanding the structure and function of our future forest resources. These interdependencies were underscored by the first-year results at the FACE facility—forest growth was increased and the limited supply of water was conserved in the CO₂-enriched plots.

During the first year of the experiment, the ORNL scientists observed that the tree leaf pores (stomata) that allow CO₂ to enter and water vapor to escape were not open as wide in plots receiving the extra CO₂. As a result, trees in the CO₂-enriched atmosphere conserved water, while maintaining much higher rates of photosynthesis—the process by which plants use the energy from sunlight to convert CO₂ and water into the sugars needed for growth. The researchers also detected a significant increase in the production of wood in the tree trunks and very fine roots in the soil. Evaluation of changes in the nitrogen content in trees and soil will help scientists determine if these important growth responses will be sustained for many years.

The FACE facility complements the Throughfall Displacement Experiment in Walker Branch Watershed, which allows study of the responses of forest trees to not only ambient but also above ambient and below ambient levels of precipitation that may be typical of a changing climate. ORNL scientists are setting the standard for large-scale ecological research that could provide a recipe for success in predicting correctly the impact of future climate on forest productivity.

Capturing and Isolating Carbon

In April 1999, DOE released a 200-page "working draft" describing research paths that could lead to long-term technologies that might slow or stop the buildup of CO₂ in the atmosphere and delay possibly undesirable climatic effects. This "research and development roadmap" identifies key research needed to allow development of a variety of carbon sequestration technologies. These technologies might separate and capture CO₂ from energy systems, make products from some of the carbon, and sequester the rest in oceans, geological formations, and terrestrial ecosystems such as forests, vegetation, soils, and crops. Just recently DOE awarded a research contract for a collaborative team of ORNL, Pacific Northwest Laboratory, Argonne National Laboratory, and several universities to form a center to perform research on ways to enhance uptake and long-term sequestration of atmospheric CO₂ by terrestrial ecosystems. DOE also awarded a center to Lawrence Livermore National Laboratory and Lawrence Berkeley National Laboratory to perform research on ocean sequestration of carbon.

The draft was compiled, edited, and printed at ORNL. Its chapters were coauthored by experts from DOE national laboratories and universities throughout the nation. One of the leaders for this DOE effort was ORNL Associate Director David Reichle, and key chapter authors included ORNL researchers Rod Judkins, Gary Jacobs, Allen Croff, and others. The expected need for carbon sequestration technologies is likely to open up new research opportunities for ORNL scientists and engineers in clean energy technologies and climate effects research.