II. SCIENCE AND FACILITIES -- THE VISION

---

"The Center contributes uniquely and significantly to the scientific infrastructure of the national and international programs in atmospheric science through a well-designed and multifaceted set of UCAR/NCAR service/outreach activities."

--from the review panel report to the National Science Foundation on UCAR/NCAR management, May 1, 1997

The research is increasingly interdisciplinary. There is an increasing need for broad approaches that combine expertise in diverse areas. This is illustrated well by the development of the Climate System Model, a set of co upled models that represent the current state of our knowledge of the atmosphere, the oceans, the land surface, and sea ice and how they interact. Studies of interactions among gas-phase chemistry, aerosols, clouds, and radiation also span disciplinary bo undaries. Other examples are efforts to couple models of the upper and lower regions of the atmosphere and the increasing inclusion of atmospheric chemistry into dynamic models.

There is increasing focus on components of the hydrological cycle. Mesoscale studies emphasize precipitation processes and forecasting, and studies of clouds seek to understand mechanisms of precipitation formation and to represent them in models at all scales. The Research Applications Program has established a focus on water resources to conduct research and develop technologies for precipitation forecasting and estimation; RAP is also establishing a community surface-hy drologic modeling facility. The Environmental and Societal Impacts Group is studying water resource management; societal vulnerabilities to hydrological extremes and management options in dealing with them are being identified. Instruments are being devel oped to improve ways of measuring water vapor. Field experiments are focusing on components of the hydrological cycle.

Interactions between modeling and observations continue to increase. This trend is exemplified by data assimilation, retrieval techniques applied to observations, the use of adjoint models to characterize model sensitivity to observational error, assessments of the value of targeted observations, and observing system simulation experiments. Field experiments and modeling are becoming more closely linked, so that now most field experiments have strong modeling components th at help focus the experiments and are used after the experiment to aid in analysis of the observations.

Modeling of the atmosphere has reached a level of maturity that supports an expanded range of scientific investigations. Community models for global, mesoscale, and cloud-scale simulations are widely available and used as research tools. A new community mesoscale model is being developed as a framework to serve both operational and research needs (see Community Models).

New facilities and instrumentation are opening new areas of scientific research. The airborne ELDORA (Electra Doppler radar) is tested and reliable and will be a key tool supporting mesoscale process studies in the near fu ture. Plans for new aircraft (the WB-57F and a mid-sized jet) will open new regions of the atmosphere and the globe for investigation by NCAR and university researchers, while the NSF-owned NCAR C-130 will continue to provide valuable large-payload suppor t for field experiments, such as in atmospheric chemistry. New chemical instrumentation is making it possible to collect routine measurements of key chemical constituents (such as OH) that not long ago were extremely difficult to obtain. The High-Resoluti on Dynamics Limb Sounder, part of NASA's Earth Observing System, will provide global measurements of a number of chemical species important to global change research (see NCAR -- Science, Facilities, and Service). The LOWL (Low-Degree Oscillations) instrument has produced breakthrough observations of the sun's interior rotation profile, soon to be augmented with the installation of a second instrument that will increase the time coverage by roughly 90%.

There is an increasing focus, especially in community-wide named programs, on topics of relevance to society. Studies in global change address anthropogenic influences on climate and the impacts of atmospheric changes on h uman societies and on the natural resources on which societies depend. Mesoscale weather researchers aim to improve forecast performance to warn of hazards and assist in practical planning, while human dimensions researchers address efficient communicatio n of forecasts and management of weather risks. Atmospheric chemists turn increasingly to the effects of air pollution. Studies of the solar interior have potential bearing on forecasts of magnetic storms and associated upper-atmospheric conditions that c an affect power transmission and communications.

There are increasing and more-regular interactions between university groups and NCAR programs. Large community-wide activities include the development of the Climate System Model and the Solar Magnetism Initiative, both c entered at NCAR, and the U.S. Weather Research Program, which has strong NCAR as well as university involvement. There is strong collaborative involvement in all aspects of these programs, including planning, research, analysis, and model and instrument d evelopment. All have mechanisms for shared governance.

These trends underscore the importance of maintaining the breadth of NCAR's research program. The composite program described here results from an important community investment in intellectual talent and scientific infrastructure. Through its leadership in programs like the U.S. Weather Research Program and the Climate System Model, NCAR will continue to serve as a focal point allowing the scientific community to address the most important, challenging, and difficult scientific and societal problems.

---

"We place great value on our UCAR participation and membership. We look forward to many more years of productive association."

--Richard D. Preslin, President, Drexel University, letter of reapplication for membership, 1992

A. FUNDAMENTAL RESEARCH: A NECESSARY FIRST PRIORITY

-- Anaxagoras, Greek philosopher, 500-428 B.C.

National Center for Atmospheric Research Advanced Study Program (ASP) Sponsors recent Ph.D. scientists and students in Ph.D. programs, bringing them to NCAR to expand their own studies and enrich the center's research. ASP also offers seminars, workshops, and colloquia on areas of particular importance to the atmospheric sc iences. Atmospheric Chemistry Division (ACD) Focuses on global- and regional-scale air quality and problems related to the complex interactions among the oceans, ecosystems, and atmosphere. Researchers study the cycles of chemicals in the atmosphere, ways in which the composition of the air evolves, and the impact of human activities on atmospheric chemistry. Atmospheric Technology Division (ATD) Provides observing facilities and instrumentation to the atmospheric and related sciences, including advanced radars, research aircraft, flux measurement systems, and integrated sounding systems. These facilities allow investigators around the globe to ga ther data required for their research programs. Climate and Global Dynamics Division (CGD) Contributes to better prediction of weather and climate through the development of models that promote understanding of the physical causes of past, present, and future climates and large-scale atmospheric and oceanic dynamics. Environmental and Societal Impacts Group (ESIG) Improves understanding of the interactions among the atmosphere, environmental processes, and society, and communicates atmospheric science-related information to a broad community of researchers and decisionmakers, through research and workshops. ESIG st aff assess how societies might better understand and cope with severe weather and climate shifts. High Altitude Observatory (HAO) Studies the variable nature of the sun and the impacts of solar outputs on the terrestrial environment, particularly the earth's upper atmosphere. This research is relevant to prediction of the lifetime of satellite orbits, disruption of high-latitude com munications networks, damage to high-latitude power grids from solar-produced magnetic storms, variability of stratospheric ozone, and possible changes in the earth's climate. Mesoscale and Microscale Meteorology Division (MMM) Investigates the basic physical processes that govern the weather: how the atmosphere and the earth receive incoming radiation, scatter and absorb it, and retransmit it; how weather and climate are affected by terrain and the characteristics of soil and v egetation; how severe storm systems develop and die; and how precipitation processes occur. Research Applications Program (RAP) Contributes to improved aviation safety through research on thunderstorms, icing conditions, snowstorms, wind shear, and turbulence. RAP develops early warning systems that help save lives and aircraft at U.S. airports and increase the efficiency and capa city of the nation's air space. Scientific Computing Division (SCD) Provides high-performance computers, data archives, and associated parallel processing capabilities to the atmospheric and related sciences. The division emphasizes large simulations, large datasets, and archives of interest to the geosciences.

In order to interact with and serve the universities, NCAR will maintain a balanced program of focused and fundamental research. New insights and understanding are fostered by a broad program of research that allows scientists to pursue the most fruitful lines of study. Knowledge gained in fundamental studies (e.g., cloud physics and precipitation formation) will become the basis for future improvements in forecasts of weather and climate. Climate predictions will become more reliable and useful with a cl earer understanding of natural variability. Improvements in air quality will depend on better understanding of the formation of urban ozone. An understanding of the role of solar variability in climate change will be possible only through fundamental stud ies of solar magnetism.

The foundation of NCAR's program in the future will therefore remain fundamental research. Because it is difficult to predict where and when progress will be made, NCAR's base program will be broad and sufficiently flexible to enable unpredictable discove ries. The five selected areas below exemplify some major thrusts in NCAR's fundamental research.

1. Precipitation Physics

Recent fundamental studies of the physics of precipitation formation have focused on two themes: testing theories of how precipitation forms through the warm-rain process and determining how ice forms and grows to precipitation. In addition, improved repr esentations of precipitation have been incorporated into mesoscale models through the use of adaptive grids and nesting techniques and via improved parameterizations of microphysical processes. Future plans target the following opportunities:
• The Small Cumulus Microphysics Study, conducted in 1995, resulted in a dataset of uniquely high quality that will be used to test current theoretical understanding of warm-rain formation. Collaborators include the New Mexico Institute of Min ing and Technology and the Universities of Illinois, Wyoming, and Chicago.

• The microstructure of the upper regions of tropical cirrus clouds is largely unexplored. Altitude and payload capabilities such as those available on a WB-57 aircraft will provide a significant new opportunity to document the composition of these clou ds and link it to the characteristics of the precursor aerosol.

• Results from recent studies provide the basis for revised parameterizations of precipitation processes. Most parameterizations lag behind current understanding of the warm-rain precipitation process and the origin of ice in clouds.

• Ongoing studies at NCAR address the formation of particles in the atmosphere, the detection of cloud-active particles, the role of aerosols in the radiation balance of the earth, and the links between aerosols and precipitation. In some current field experiments, particles are being introduced into clouds in order to study the relationship between the precursor aerosol and subsequent precipitation development.

Precipitation Physics Studies of warm rain production in small cumulus clouds in Florida during the Convection and Precipitation/ Electrification Experiment (1991) and Small Cumulus Microphysics Study (1995) reveal both hydrometeor and Bragg scattering signals in the data from the NCAR/CP-2 X- and S-band radar (3- and 10-cm wavelengths). Bragg scattering comes from cm-scale variations in the refractive index due to turbulent mixing between cloudy and drier environmental air and can produce strong mantle echoes around the sides and tops of the clouds. The mantle echo is most prominent at S-band, as illustrated by DZS in the upper panel of the figure. The strong X-band echo (DZX in the bottom panel) in the cloud interior comes from cloud and small drizzle drops. The prominent DZ S weak-echo region just above cloud base coincides with the updraft axis and extends upward into the interior X-band hydrometeor echo. The flatter X-band pattern--contours are more horizontal--is likely to be from cloud droplets activated at cloud base th at grow as the updraft air ascends adiabatically. Large values of reflectivity below 1 km come from insects and other material in the boundary layer. At this early stage, the cloud top was a little above 3 km AGL, and the cloud remained entirely below the freezing level at about 4.6 km. This cloud went on to produce a brief burst of light precipitation with equivalent reflectivity factor attaining about 6-10 dBZ.

2. Natural Variability of the Earth System

Earth's climate system varies naturally over a range of space and time scales. Ice core records that span 400,000 years indicate that the variability has a rich structure. Short-term variations, such as El Niño, are of great societal importance. De tecting human influences on earth's climate system requires an understanding of the magnitude of natural variability on all time scales.

The internal and external mechanisms that cause the climate system to vary are a fundamental issue in earth system research. External processes include solar variability, volcanic activity, and other geological changes. Internal processes include the chan ging chemical composition of the atmosphere and changes in the biosphere. Other sources of internal variability are nonlinear processes within the climate system that yield rich, chaotic structures. The complexity of the earth system as a whole creates a tremendous capacity for variability arising from nonlinear interactions.

Understanding natural variability requires observations, models, and theory. Basic research to understand the causes, sources, and predictability of these variations spans the scientific divisions within NCAR. The Climate and Global Dynamics Division will study variability on space scales greater than 100 km and time scales greater than hours with a hierarchy of models. They will also use global analyses and satellite data to look at dominant modes of variability. The Mesoscale and Microscale Meteorology Division will use observations and model simulations to address smaller space and time scales. Within the Atmospheric Chemistry Division, both in-situ observations and chemical transport models will be used to study natural variability in the chemical sta te of the atmosphere on a diverse range of space and time scales.

3. Atmospheric Kinetics and Photochemistry

The atmosphere consists of a complex, ever-evolving mixture of gases and particles, and its composition is determined by the chemical reactions among these components. For example, the oxidizing capacity of the troposphere (which is largely controlled by the abundance of the hydroxl radical) is determined through a complex interplay between hydrocarbons, nitrogen oxides, and hydrogen-containing radicals. Stratospheric ozone depletion is caused by reactions involving members of the nitrogen, halogen, and h ydrogen families. Understanding how the future atmosphere will evolve as the result of changing burdens of trace gases requires a fundamental and quantitative knowledge of the individual photochemical processes.

The major source of OH in the lower atmosphere is photolysis of ozone. Despite the central importance of the photochemistry of this and other related molecules, at least 60 years of fundamental research have yet to reveal all the quantitative details of O H-related processes. A major effort at NCAR in the forthcoming years will be to study key processes controlling the production and destruction of OH radicals, including the oxidation of biogenic and anthropogenic hydrocarbons and the photodecomposition of their degradation products.

---

"Washington State University has developed strong ties with NCAR during the seven years we have been a member in UCAR. . . .WSU faculty have collaborated with Atmospheric Chemistry Division scientists on numerous field studies. . . . There ha ve been several scientist visitations between our institution and NCAR/UCAR as well. . . . I wish to confirm Washington State University's commitment to provide an environment for addressing important national and global research issues in th e atmospheric sciences and related fields."

--Samuel H. Smith, President,Washington State University, letter of reapplication for membership, 1994

---

Studies of the destruction of stratospheric ozone have focused in part on the chemistry of inorganic forms of chlorine, the interconversion of which leads to net ozone destruction. In the forthcoming years, NCAR, in partnership with university colleagues, will address a number of remaining fundamental problems, specifically those related to the role of bromine and the importance of heterogeneous chemical conversion on liquid and solid atmospheric particles.

4. Solar Magnetism

The outputs of radiation and magnetized plasma from the sun are the principal physical agents controlling earth's atmosphere, oceans, and space environment. The solar magnetic field is the underlying cause of the variability in these outputs. Understandin g the structure and behavior of the solar magnetic field is therefore a scientific prerequisite for developing any reliable forecast of variations in solar outputs that cause changes in our terrestrial and space environments.

Stokes Polarimeter Observations of a Young Active Region These observations from the High Altitude Observatory/ National Solar Observatory Advanced Stokes Polarimeter show the vector magnetic field of a solar active region on its first day of appearance. The panels display this active region as if it were seen from an oblique angle. From bottom to top, the planes represent the visible light image (Ic), the magnetic flux (corresponding to an ordinary magnetograph observation), the field strength (B) with some arrows superimposed to show the orientation of the fi eld, and the vertical velocity as revealed by the Doppler shift of the spectrum lines. Note that the vector field is nearly horizontal in the emerging flux region between the two sunspots and that this emergence is accompanied by upward motions (patches o f blue in the top image). The emerging flux is nearly invisible to ordinary magnetographs. It is expected that detailed studies of emerging regions such as this one will reveal important clues both to the nature of the magnetic fields within the sun and t o the processes which govern solar variability. Such studies play a central role in the Solar Magnetism Initiative.

NCAR has contributed to several important developments in the study of the sun in the past three decades, bringing solar physics to the point where the crucial questions for understanding solar magnetism can now be identified and answered. Helioseismology is providing new information about the solar interior, where the solar magnetic field originates. This new information can be combined with the growing wealth of observations of solar magnetism in the solar atmosphere out to the far reaches of the solar corona and into the interplanetary medium to build an integrated picture of the whole of solar magnetism.

Fundamental scientific questions that may now be answerable from a synergistic combination of new observations, analytical theory, and numerical modeling include

• the processes by which magnetic fields are stored at the base of the solar convection zone long enough to be regenerated by dynamo action;

• the injection of magnetic fields into the convection zone and through it into the photosphere, chromosphere, and corona above;

• the interrelations of field structures on an enormous range of spatial scales and the effects of these structures on the physics of variations in the radiative and particle output; and

• the production of major perturbations, such as flares and coronal mass ejections, produced from the great variety of coronal magnetic structures.

The Solar Magnetism Initiative (SMI), is being proposed by NCAR's High Altitude Observatory (HAO) and the solar physics community to accelerate progress on all of these questions. The emphasis of this initiative will be on fundamental science that is high ly integrated across different layers of the sun, across several scientific disciplines, and across activities from observations and interpretive studies to theory and modeling. SMI will be built on a firm foundation of scientific results from the core sc ientific program at HAO as well as at other institutions in the United States, Europe, and Japan.

5. Dynamical and Turbulent Systems

The original scientific leaders of NCAR recognized that in order to understand the dynamics of the atmosphere, the oceans, and the sun, understanding relevant turbulent and dynamic processes was essential. Thus, there has been a sustained emphasis on geop hysical turbulence and astrophysical fluid dynamics at NCAR, in research, visitors, seminars, and workshops, since the center's inception.

Because of the nonlinearity of the processes that govern the atmosphere, ocean, and sun, scale interactions play important roles in each of these fluids. For example, in the atmosphere, synoptic-scale fluxes play an essential role in maintaining the plane tary processes that dominate climate variability. In turn, mesoscale phenomena such as fronts, convection, and rainbands form against a background of synoptic-scale eddies and maintain and amplify synoptic-scale processes. In the ocean and sun, smaller-sc ale, turbulent motions have analogous influences on large-scale phenomena. A heightened awareness of the importance of these mechanisms has led to concerted, continuing efforts within CGD, MMM, and HAO to understand the role and nature of cross-scale inte ractions in a number of areas: the general circulation of the ocean, the breakdown of the polar stratospheric vortex, the maintenance of blocking regimes in the troposphere, the Madden-Julian oscillation in the tropics, mesoscale convective complexes, and magnetic field variations on the sun. A significant accomplishment to date in this area has been the development of the Gent-McWilliams ocean eddy transport parameterization.

Modeling Laboratory Convection Volume rendering of temperature from a calculation of classical incompressible Rayleigh-Benard convection between two rigid plates at two temperatures at Ra=20 million. Hot fluid is represented by yellow-brown; cold fluid by green-blue. Note the dominant large-scale diagonal pattern, which persists for all Rayleigh numbers simulated and all times. The fine-scale patterns on the surface are reminiscent of solar granulation and consist of smaller plumes being swept into the larger-scale patterns. The combin ation of large-scale patterns and small-scale turbulence is now understood to be a property of a variety of convective flows in the atmosphere, laboratory, and sun. How this might change the scaling properties of convection is being studied. Figure courte sy of Bob Kerr, High Altitude Observatory.

In addition to scale interactions, dynamics researchers are also studying the influence of dynamically active fields of moisture and salinity in the atmosphere and ocean, respectively, and of magnetic fields in the sun. Because dynamic and thermodynamic i nfluences can profoundly alter the behavior of the fluids, studies are focused on understanding these influences on explosive cyclone development, double diffusive convection, thermohaline overturning, and magnetically enhanced shear flow instabilities.

One catalyst for this research is the Geophysical Turbulence Program (GTP). The GTP is strongly oriented toward basic science and is inclusive and outreaching. Its members, now numbering more than 20 NCAR scientific staff and visitors, employ a range of s cientific methods, including analytic theory, numerical simulations, and the gathering and interpreting of laboratory and observational data. The GTP workshops (held approximately once a year) serve as a particularly useful method for NCAR and university scientists to keep abreast of new research in geophysical and astrophysical turbulent flows. Two recent workshops focused on geophysical and astrophysical convection, and stratified and rotating turbulence. In each case, the workshops brought together a b roader community for the purpose of better understanding similarities between flows in very different regimes through an intercomparison of theories, simulations, and experiments.

A particularly significant and practical outgrowth of GTP has been joint research by several members to assess the parameterizations of the planetary boundary layer and upper ocean. Results are being incorporated in the latest version of the Community Cli mate Model and in the Climate System Model.

Building on the success of these collaborations, the GTP members plan to investigate a set of scientific problems important to geophysical and astrophysical flows. Examples include: the role of vortex reconnection in turbulent cascade and the analogous pr oblem of magnetic field reconnection and ohmic dissipation in the solar atmosphere; the mechanism of entrainment of quiescent fluid into turbulence, with quantitative estimates of mixing of different chemical species including water, as well as reactive a nd ionized gases; and the physical and statistical basis of parameterization of subgridscale processes in numerical models.

B. UNDERSTANDING AND PREDICTING THE EARTH SYSTEM

1. Predicting the Weather on Short Time and Space Scales

As we enter the 21st century, the prospects are excellent for greatly improved forecasts of the weather, including severe weather events such as thunderstorms, tornadoes, and hurricanes as well as day-to-day changes in wind, temperature, and precipitation . New observational platforms also offer great promise. Satellite and radar systems, surface-based wind profilers, commercial aircraft, and ground- and space-based GPS systems, combined with new techniques for assembling the diverse resulting datasets in dynamically consistent ways, offer the potential for defining initial states of the atmosphere at resolutions of 10 km or better in some regions of the globe. With such initial states, accurate predictions of the onset and magnitude of changes in many wea ther variables are possible.

Refining the models to achieve forecasting advances will require better techniques to account for unresolved processes within the models. These include convective, cloud microphysical, and boundary-layer processes and turbulence. Better models require bet ter understanding of how these processes influence the physics of weather systems. Accordingly much of the basic research in weather systems at NCAR will include work on the physics of cyclones and fronts, orographic effects on organized convection, and m esoscale structures in cyclones.

NCAR is developing a diverse new program designed to address the goals of the U.S. Weather Research Program (USWRP), particularly the improvement of short-range weather forecasting. The center has been at the forefront in revitalizing the national USWRP e ffort, and these activities will complement those of the universities and government labs. NCAR's activities are designed to: improve remote-sensing techniques for measuring profiles of water vapor; develop comprehensive mesoscale numerical models, includ ing a community mesoscale data assimilation system; create techniques for targeting observations into regions where they would have maximum impact on weather forecasts; use model- and radar-based precipitation estimates in forecasting flash flooding in co mplex terrain; improve techniques for estimating precipitation from radar and satellite observations; develop nowcasting techniques for thunderstorms and quantitative precipitation; and research the impact of improved forecasts on society.

These activities involve extensive interactions among four divisions of NCAR--MMM, CGD, RAP, and ATD--as well as extensive contributions from the university community. Societal concerns will be integrated with the scientific enterprise through an ESIG-led component focusing on forecasting impacts. We will also work closely with the operational community to transfer research results.

Climate is the integral of weather systems over a long time, yet processes that have much shorter time scales have a profound impact on the atmospheric systems that determine the climate. A considerable part of NCAR's future effort in mesoscale meteorolog y will be aimed at understanding the impacts of these processes on climate and how they can influence climate change. In cases where very-long-term integrations cannot account directly for short-term processes (such as effects of clouds, cloud systems, an d the boundary layer) better paramaterizations will be developed.

2. Climate--Natural Variability and Anthropogenic Change

The temperature record over the past century varies on a range of time scales and suggests enhanced human-induced greenhouse warming. Key questions for research are to what extent natural variability explains the observed record, how to separate the effec ts of human activity from natural variability, the relative roles of the different forcing changes over the last century, and how human-induced climate change alters the modes of natural variability, such as the North Atlantic Oscillation and the El Ni&nt ilde;o/Southern Oscillation (ENSO).
a. The Coupled Ocean-Atmosphere

Two areas of special interest at NCAR are variability in the North Atlantic and Arctic and variability in the tropical oceans and its effects on the global atmosphere. The North Atlantic/Arctic system is important for climate variability on a variety of t ime scales. Decadal variability in the North Atlantic and in sea ice have been documented and are related to atmospheric variability. Thermohaline fluctuations produce important climate variations on 50-year and longer time scales. Future scientific focus es include the North Atlantic Oscillation and its causes, the variations in the extent of Arctic sea ice and its causes, and the interaction between Arctic sea ice and the North Atlantic meridional overturning circulation and related causal mechanisms. An overarching goal of this research is to develop the components of the Climate System Model (CSM) so that it is capable of simulating these phenomena. Collaborators with NCAR include scientists at the University of Washington, Dartmouth College, Los Alamo s National Laboratory, and NOAA's Geophysical Fluid Dynamics Laboratory.

The Earth System A modified "Bretherton diagram" (based on a concept of Francis Bretherton) highlighting some of the linkages between social systems, biogeochemical systems, and the physical climate system. Courtesy Guy Brasseur, NCAR Atmospheric Chemistry Division.

Building on the Tropical Ocean and Global Atmosphere program, NCAR research on the tropical oceans will initially focus on seasonal-to-interannual variability in the Pacific associated with ENSO and on the coupled evolution of the atmosphere and ocean in this region. Extended research will include decadal variability, possible influences of anthropogenic climate change, the role of other tropical oceans, and the tropical and extratropical teleconnections.

b. Climate of the 20th Century and Beyond

NCAR scientists and collaborators are committed to helping answer the question of how and why climate has changed in the recent past and may change in the future. The earliest runs with the CSM have shown that a substantial amount of the change over the p ast century can be explained as natural variability (see Figure 3).

Figure 3. Results from first CSM CO2 experiment. Global Temperature Projections with the CSM The first Climate System Model experiment performed with CO2 increasing at 1% per year shows a global surface temperature increase of approximately 1.4 degree K when the CO2 doubles, which is near the low end of the range of estimate s from earlier work. Here, the top line depicts global average temperatures when carbon dioxide increases at 1% a year starting in year 10. The bottom line shows temperatures obtained from a control run that holds CO2 levels constant. The horiz ontal lines above and below the control indicate the range of natural climate variation in the control run. Illustration courtesy Maurice Blackmon, NCAR Climate and Global Dynamics Division.

Ensembles of simulations with climate models will be required to separate human effects from natural variability. The extent to which the models can reproduce observed global and regional trends will determine how credible the models will be in studying f uture climate change. The economic and social impacts of this change are enormous. Therefore, it is critical to have reliable information on future greenhouse scenarios and how various possible limits to greenhouse emissions would affect the climate.

NCAR scientists, in collaboration with the NOAA Aeronomy Laboratory, are working toward a series of simulations with the CSM to study the climate of the 20th century and beyond. This group has already defined the anthropogenic changes in the chemical comp osition of the atmosphere for the period 1870 to 1990. Three preliminary simulations based on a limited number of forcing factors and including the effects of methane on stratospheric water vapor will be carried out--the first with changes to greenhouse g ases, the second with changes to sulfate aerosols, and the third with changes due to solar variability and volcanic aerosols. Human activities related to climate forcing are changes in land use, stratospheric aerosols linked to aircraft, and aerosols prod uced by biomass burning. Additional simulations are planned in the next five years to account for these factors. Scientists at the University of Wisconsin will provide land-use information for the climate simulations. The goal is to create accurate forcin g factors, based as much as possible on observational data, and to apply these sequentially to allow for attribution of climate signatures to specific forcing agents. The series of simulations will also greatly aid in the climate detection and pattern rec ognition techniques currently being used to understand the role of humans in climate change.

The result of the 20th century simulations will serve as the initial condition for the studies of potential future climate change. These studies will address what types of climate changes to expect under various CO₂ stabilization scenarios. We will begin with three experiments, involving three levels of stabilization of greenhouse gas emissions: little intervention, moderate intervention, and rigorous intervention. Such experiments will be at the forefront of global climate change modeling, and , if successful, will provide information of relevance for the Intergovernmental Panel on Climate Change and for other studies that use such scenarios.

Land-surface characteristics have been altered significantly in the 20th century. Natural vegetation throughout the extratropics and tropics has decreased dramatically as it has been converted to urban development or to agriculture. Tropical deforestation is particularly significant, but extratropical land use, including reforestation, may also be important. Land-use practices are implicated in the desertification of the Sahel region of Africa. Land-surface hydrology and the recycling of water may be impo rtant determinants of persistent droughts and floods during the 20th century. Numerous modeling studies have demonstrated the sensitivity of the climate to changes in land use and soil moisture. In addition, natural climate variability has led to numerous droughts and floods throughout the century. However, the role of these forcings in determining the climate of the 20th century has not yet been studied, and this will be an objective for our research.

c. Chemistry and Climate

Changing atmospheric composition is a key driver of climate on long time scales. Sources, sinks, and chemical reactions respond to the state of the atmosphere and influence it, requiring that a comprehensive climate model include interactions with atmosph eric constituents and the controlling biological and chemical mechanisms. Key questions to be addressed in future research include the causes and the climatic consequences of the observed trends in greenhouse gases; how climate and the carbon cycle have i nteracted over the past hundred years; how perturbations in the global nitrogen cycle will affect ecosystems and climate; and the role of various types of aerosols in earth's climate; the climate forcing associated with human-induced changes in the concen tration of tropospheric and stratospheric ozone.

3. Atmospheric Chemistry and Air Quality

The atmosphere is a chemically complex system interacting internally--principally within the troposphere and stratosphere--and externally with the rest of the earth system. Its composition is known to have changed continuously since the formation of the e arth some 4.6 billion years ago. Changes have accelerated during the last century as the result of human activities. For example, it is well established that the ozone hole over Antarctica since the late 1970s has been caused by emission of industrially m anufactured chlorofluorocarbons (CFCs). Ozone depletion is leading to increased UV-B levels at the earth's surface, with potential biological and health effects. Reactive compounds from fossil fuel combustion and biomass burning affect the oxidizing capac ity of the atmosphere, that is, its ability to convert chemical species (including pollutants) into compounds that can be more readily removed from the atmosphere. These reactive compounds include nitrogen oxides, carbon monoxide, and a variety of hydroca rbons. There is an increase of climatically important greenhouse gases such as carbon dioxide, methane, and nitrous oxides, all of which may contribute to global warming and other climatic changes.

A major challenge for NCAR in the next decade will be to understand the processes that affect these global changes, to assess their importance for the future evolution of the atmosphere, and to predict their consequences for other components of the earth system and for society. These challenges, which will require complex field campaigns and the further development of an integrated earth system model, will be met through close collaborations among NCAR divisions and through university partnerships.

An important scientific objective of NCAR is to identify and quantify the processes that regulate the chemical composition of the troposphere and middle atmosphere and to assess future changes caused by human activities. Key to these questions is our unde rstanding of the role of the biosphere in producing and consuming trace gases and of chemical and photochemical processes occurring in different atmospheric environments. These societally important questions will be addressed within the framework of the N SF Global Tropospheric Chemistry and Climate Modeling and Analysis Programs.

a. Tropospheric Ozone and Other Photo-oxidants

Photochemical oxidation, transport, and eventual deposition of chemical products to the earth's surface are primarily responsible for maintaining an atmosphere benign to human life. It is critical to understand how biogenic and ever-increasing anthropogen ic emissions will interact to influence these processes. On regional scales, depending upon meteorological conditions, oxidation processes involving hydrocarbons, carbon monoxide, odd hydrogen, and odd nitrogen can lead to ozone concentrations at levels h armful to plants and animals, including humans. Determining how to mitigate these episodes most effectively remains an important research goal. On the global scale the primary oxidant--the hydroxyl radical--transforms many primary sulfur, nitrogen, hydroc arbon, and halogen emissions to products that are more readily removed from the atmosphere. Understanding these gas-phase and heterogeneous processes is necessary to predict the effects of changes in emissions caused by human activity. To address scientif ic issues related to tropospheric photo-oxidants, NCAR scientists, in partnership with university colleagues, will plan, develop, and implement field campaigns focusing on specific chemical and photochemical processes in the free troposphere. Likely colla borators include the University of Colorado, San Diego State University, Duke University, the University of Sao Paolo (Brazil), and the University of Witsatersrand (South Africa). These campaigns will rely heavily on the capabilities of high-altitude airc raft and other airborne platforms and will use models as planning and diagnostic tools.

Modeling Ozone Distribution Distribution of ozone at the earth's surface calculated by the NCAR IMAGES chemical transport model (July conditions). The high concentrations over North America, Europe, and Southeast Asia are associated with high levels of pollution (i.e., fossil fuel c ombustion) in these regions. High concentrations in Africa and South America result from trace gas emissions by biomass burning. Concentrations are expressed in parts per billion in volume (ppbv).

Key issues to be studied include

• What are the mechanisms responsible for regional (summertime) ozone episodes? Can we make reliable predictions of these events? What is the impact of regional ozone formation on tropospheric ozone at the global (hemispheric) scale?

• Is the abundance of tropospheric ozone changing globally, and, if so, what are the mechanisms involved? What are the impacts on the climate system? What are the changes expected in the future?

• What are the effects of human activities (eg., fossil fuel burning, aviation emissions) on the abundance of ozone and other chemical constituents in the atmosphere? How will changing climate affect the abundance and distribution of photochemical oxida nts in the atmosphere?

b. Biosphere-Atmosphere Interactions

Quantifying recent changes in trace gas fluxes due to changes in land use (see Chemistry and Climate above) requires an understanding of the controls over biogeochemical cycles. These fluxes are poorly quantified but are probably increasing due to increasing population and industrial development. NCAR scientists, with collaborators from Washington State Univers ity, the University of California at Berkeley, and elsewhere, will address several key issues related to biogeochemical cycles. The emphasis in these studies will be on the measurement of emission fluxes of biogenic hydrocarbons and other gases, and their degradation in the atmosphere.

Key issues to be studied include

• What are the roles of vegetation, soils, and marine organisms in regulating the chemical composition of the atmosphere?

• How are changes in land use, including biomass burning and deforestation, affecting trace gas fluxes?

• How do changes in atmospheric composition influence the functioning of the biosphere?

• How does biosphere/atmosphere exchange of chemical species influence tropospheric oxidants, aerosols, stratospheric chemistry, and climate change?

• Can we close the global biogeochemical cycles of carbon, nitrogen, and sulfur? What are the interactions between the carbon and the nitrogen cycles?

NCAR and university scientists will participate in field campaigns and will use earth system models as tools for integration and synthesis to address these questions.

c. The Role of Aerosols

Aerosols play an important role in determining the radiative balance of the earth's climate, both directly by scattering light and indirectly by generating cloud condensation nuclei. Heterogeneous reactions on aerosol particles can dramatically influence atmospheric processes such as ozone destruction in the polar stratosphere and acid rain production in the troposphere. The ability of aerosols to significantly alter chemical pathways in the stratosphere is newly understood and merits examination in the t roposphere as well. The routes by which aerosols and clouds might influence tropospheric chemistry and climate are highly speculative at present but cover a broad range of interactions including biogenic and anthropogenic oxidants, and carbon, sulfur, and nitrogen compounds.

Key issues to be studied include

• What are the processes that lead to the formation, growth, hygroscopic qualities, and chemical composition of aerosols? What are the global distributions of aerosol properties, and what controls these distributions?

• Does the large variety of particles (sulfate, nitrate, organic, soot, ice, etc.) significantly perturb gas-phase photo-oxidation processes in the troposphere? How are aerosols removed from the atmosphere?

• How do aerosols affect earth's climate, both directly through their effects on radiation and indirectly through their effects on cloud formation?

NCAR expects to work with the University of Minnesota and the University of Hawaii to include comprehensive aerosol models into the NCAR Climate System Model and Chemical Transport Model. Satellite observations will also be used to estimate aerosol type, distributions, properties, and sources for use in global models.

4. Solar-Upper Atmosphere Coupling

The effects of the variable sun on earth's atmosphere are manifest in a variety of physical and chemical processes in the mesosphere, thermosphere, and ionosphere. NCAR's efforts to model the earth's upper atmosphere are focused around continuing to perfe ct the Thermosphere-Ionosphere-Mesosphere-Electrodynamics general circulation model (TIME-GCM), completing and refining the representation of key physical and chemical processes. Investigating the response of this coupled system to external forcing throug h solar radiative and plasma variability is the basis for understanding the role of these layers in the earth's atmospheric system as a whole. Understanding the coupling of the troposphere to the higher layers and to the solar forcing is crucial for the s uccess of the U.S. Global Change Research Program. There is an important need to represent better the couplings between the regions modeled in TIME-GCM and the earth's magnetosphere for future studies relating to the behavior of the upper atmosphere durin g geomagnetic storms. With this research, to be done in collaboration with many groups in the university community, NCAR will make substantial contributions to the basic science underlying "space weather" phenomena targeted in the National Space Weather I nitiative.

TIME-GCM Two model runs from the Thermosphere-Ionosphere-Mesosphere-Electrodynamics general circulation model (TIME-GCM) are depicted below. In the figure at left, the TIME-GCM is flux-coupled to the NCAR Community Climate Model version 3 near 30 km altitude, intr oducing lower atmospheric variability into the upper atmospheric system. The contours represent meridional wind (positive northward in m/s) along the 17.5 degrees N latitude circle. The figure at right illustrates TIME-GCM's ability to model solar-terrest rial interactions; shown is modeled difference in total electron content with respect to quiet-day background at the peak of the substorm on October 18, 1995, illustrating enhancements in the equatorial Appleton anomaly and in the high-latitude auroral re gions. Courtesy Ray Roble, NCAR High Altitude Observatory.

C. ADVANCED SCIENTIFIC FACILITIES

---

"Our membership in UCAR and the facilities available to us through membership have been a vital part of our programs."

--William Kirwan, President, University of Maryland at College Park, letter of reapplication for membership, 1997

1. Computing and Data Handling Facilities

Modeling capability at NCAR has evolved in step with the center's aggregate computing power over the past decade and will continue to do so in the next decade (see Figure 4). UCAR and the community it serves currently enjoy world leadership in several are as of atmospheric research that depend on high-performance computing. Extraordinary computational capability--along with storage and analysis of huge datasets that are not widely available--is required to maintain this leadership.

Co-evolution of Climate Modeling and Computing Capbility at NCAR Figure 4. Computing/Modeling evolution. CCM 0 ran on a Cray 1 with simplified parameterizations of convection, survace energy exchange, clouds and diffusion. CCM 1 ran on a Cray X-MP with improved radiative transfer, annual cycle capability, and vertical and horizontal diffusion. CCM 2 ran on a Cray Y-MP with major improvements to radiative transfer, cloud formation, complete diuranl cycle and exp,.icit boundary layer treatment. CCM 3 runs on a Cray C90 as a semi-Lagrangian model with cloud optical properties, incorporating a sophisticated land surface model, a slab ocena model and all-sky radiative transfer.

a. Computing Requirements

Present and future computational and data-handling requirements are typified by the NCAR Climate System Model (CSM). On our current primary computer for executing the CSM, the Cray C90, using all 16 processors in parallel, it takes approximately 16 days t o simulate 100 years of climate, with the model running 24 hours a day at five gigaflops. Scientists routinely need to simulate several climate scenarios and at least four sensitivity studies for each scenario. Thus, a single 100-year study may involve 20 or more 100-year simulations. Each 100-year simulation produces hundreds of gigabytes of data that must be archived and analyzed. Within the next two years, climate modelers will need to increase spatial resolution and the sophistication with which physi cal, chemical, and biological processes are treated in the models. At that point, at least one machine that can sustain 25 or more gigaflops will be required to run the CSM.

Table 2 summarizes the level of computing capability that some of UCAR's peer organizations will have by the beginning of 1998. While many of these organizations provide operational forecasting services, this table shows that the global atmospheric scienc es community is dramatically increasing its computational capability. All of these centers have aggressive plans to augment their capabilities in 1999. In order to maintain leadership, UCAR must have comparable computing capability.

What's Happening at Forecast Centers Organization Systems Number of Processors Sustained Performance (Gflops) Comments European Center for Medium-Range Weather Forecasts Fujitsu/VPP 116 60-80 (1) Canadian Meteorological Center NEC SX-4 48 30-40 UK Meteorological Office Cray T3E 700 35 Australian Bureau of Meteorology NEC SX-4 32 20-25 (3) Meteo-France Fujitsu/VPP 26 15-20 (1) Geophysical Fluid Dynamics Laboratory Cray T90 26 15 (2) Danish Meteorological Center NEC SX-4 16 12 Japanese Meteorological Center Hitachi S-3800 4 6.9 Japan's Marine Science and Technology Centre NEC SX-4 20 12-15 NCEP Cray C90 16 5 (1) The next generation of the Fujitsu VPP will be available in 1999 and it is expected to sustain 2-3 Gflops per processor. Both the European Center and Meteo-France may acquire a system with 30-40 processors giving each an additional 60-100 Gflops of ca pability. (2) Assumes sustained performance of 0.6 Gflops per processor. (3) About 20 Gflops of sustained performance will be added in late 1999. Table 1. Computing systems at forecast centers.

Following a recent Department of Commerce determination that Japanese NEC supercomputers were being offered to NCAR at less than fair value, NSF declined to approve the NEC procurement, and NCAR has withdrawn the request. Within the past year we have inst alled two Cray computers, a C90 and a J9se, and a Hewlitt-Packard Exemplar X-Class system (called an SPP), giving us computing capability comparable to any other U.S. supercomputer center.

Two parallel architectures can provide atmospheric scientists with the level of performance they need--parallel vector systems and highly parallel, microprocessor (nonvector) systems. Emerging highly parallel, microprocessor systems offer very attractive performance per unit of cost, and NCAR has substantial experience with and expertise in this technology. We believe it can provide the U.S. atmospheric sciences with computing capability second to none. Consequently, we will intensify our evaluation of th is technology and make the transition to it by the end of the century, if not before.

b. Data archiving and management system

As computing capacity at NCAR continues to grow, the need for archival storage will continue to grow concomitantly. For example, in mid-97, the NCAR mass storage system contained over 100 terabytes and had a net growth rate of more than three terabytes pe r month.

CHALLENGE We will maintain a world leadership position in providing supercomputing capability to the community.

Data management involves much broader issues than simply archiving growing volumes of data. The gap between servers' capability to generate output and bandwidth to move data is widening at an alarming rate. Thus, we must enhance our ability to manage grow th of the archive. At the same time, we must provide higher levels of functionality. We can no longer treat data as a collection of bytes stored in sequentially accessed files. We must develop intelligent data management services that can access only the bytes of data needed to satisfy a request. The need to archive flat files will continue, but an archive should not handle active or temporary data. Those datasets must be supported by front-end file servers. To address this need for an intermediate data r epository, we will investigate the technology of tightly coupling active and temporary data with data analysis services in a "data park" system, where very large quantities of data can be stored, retrieved via high-speed connections, processed by high-per formance computational systems, and hierarchically managed. File services in this kind of system must be capable of caching terabytes of data on high-performance storage devices and of moving data to local tape library systems as well as the archive. Data parks will be the foundation for higher-level data management systems capable of supporting data warehousing and data mining applications.

We must provide these capabilities in a heterogeneous distributed computing environment where network connectivity and performance will be the highest hurdles to clear. In summary, in the next five years the community will require extraordinary computatio nal and data handling capability.

2. Atmospheric Observing Systems

NCAR and university research will continue to cover a broad range of topics, with several areas of emphasis: atmospheric chemistry (including increased attention to chemical effects of aircraft emissions), earth's radiation budget, the hydrological cycle, and coastal severe weather. Researchers will increasingly address problems that cross disciplinary boundaries and will design field programs that need multiple systems to gather necessary data. These programs will often focus on regions far from the cont inental United States. At the same time, investigators will continue to request observing systems for relatively small domestic projects and for support of academic teaching programs. Therefore, the lower atmospheric observing facilities at NCAR should ha ve mobility and flexibility to support a wide range of community needs and should provide highly accurate basic measurements (e.g., of water vapor and solar radiation) that relate to these needs.

Getting the most from numerical models will increasingly drive the need for observations. Mesoscale models will increase their ability to assimilate observational data in real time, which will intensify the need for real-time production and transmission o f data from, for example, satellites and radars. Coupled models will focus observations toward atmosphere-ocean, atmosphere-land, and atmosphere-ocean-sea-ice processes, and will require expanded spatial coverage and extended seasonal and annual measureme nt periods.

The U.S. National Weather Service modernization will continue to improve standard domestic observational capabilities, especially radar coverage. At the same time, however, the number of global sites launching rawinsondes as part of the World Weather Watc h network and the frequency of launches from each site will decline. Field projects in remote regions will have greater need for quality data from NCAR systems. NCAR observing systems will have to support the research necessary to develop and implement fl exible observing strategies as U.S. and other weather services increasingly look to "adaptive" observing systems.

a. Plans for aircraft facilities

The community depends on the NCAR aircraft fleet to provide a reliable and affordable combination of payload and range over the full troposphere. Payloads should accommodate multiple instruments from multiple researchers. Range should allow extensive oper ations over oceans and polar regions. At a minimum, the aircraft operated by NCAR will support research throughout the troposphere. Ideally, at least one aircraft should carry research payloads into the stratosphere. The NSF/NCAR C-130, coupled with a new high-performance jet, and the University of Wyoming King Air constitute an effective fleet to meet a large portion of NSF research needs long term (see Figure 5). However, an overall NSF fleet with only those three aircraft leaves two urgent needs unmet: extreme (greater than 50,000 feet) high-altitude capability, and a platform to carry NCAR's airborne Doppler radar, ELDORA.

Figure 5. NSF aircraft capabilities.

Observing Facilities at the Millennium AIRCRAFT SURFACE SYSTEMS SOLAR OBSERVING SYSTEMS C-130 S-Pol Radar HIAPER Integrated Sounding Systems Stokes polarimeter U. of Wyoming King Air Wind Profilers Precision Solar WB-57F Lidar water vapor profilers Photometric Telescope ELDORA platform GPS sounding systems LOWL Integrated flux-PAM Advanced Coronal Observing System

C-130. The NSF/NCAR C-130 will provide reliable support for a wide variety of payloads and a broad range of missions up to altitudes of 28,000 feet for at least the next two decades. The C-130 will carry large payloads and provide unique infrastructure to studies of atmospheric chemistry, cloud physics and radiation, large-scale surface fluxes, and various remote-sensing missions. NCAR will upgrade the C-130 to cover, at a minimum, the requirements for additional inlets an d apertures at specific locations, improvements to the basic sensor suite, and improvements in data and communication systems.

High-performance research aircraft. NSF/ATM is seeking funds for a new high-performance instrumented airborne platform for environmental research (HIAPER), a mid-sized jet capable of providing 8,000 pounds of payload, work ing extensively at 50,000 feet, and covering 7,500 miles in a single flight. If the NSF funding request succeeds, the HIAPER and the C-130 will become the major platforms in a long-term, base-funded, two-aircraft fleet. The HIAPER will be able to operate reliably and efficiently from a variety of sites worldwide and to access the upper troposphere and remote regions anywhere on earth. Like the C-130, the HIAPER will carry a wide range of payloads and support a broad spectrum of research covering atmospher ic chemistry, clouds and radiation, large-scale and oceanic severe weather, and many remote sensing applications. As HIAPER moves from possibility to reality, NCAR will institute a joint NCAR-university instrumentation initiative to modernize, miniaturize , and customize a wide variety of instruments and sensors to match the capabilities of the HIAPER.

WB-57F. NCAR will seek an operational partnership with the National Aeronautics and Space Administration's Johnson Space Center (JSC) to bring lower stratospheric capabilities to NSF and national researchers. The WB-57F pr ovides unique payload and altitude capabilities, and is likely to remain valuable for the next five to ten years. A partnership with JSC would combine their operational expertise with NCAR expertise in instrumentation and project support to enable communi ty use of this important asset.

ELDORA platform. NCAR will seek cooperative ways to continue to provide the Electra Doppler radar (ELDORA) to NSF researchers. A sturdy P-3 is a better platform than the Electra to carry ELDORA for hurricane and other seve re-weather missions. NCAR will explore with NOAA ways to share P-3 and ELDORA support, and will carefully evaluate other ELDORA options, including installation on the NSF/NCAR C-130 and access to a P-3 from a source other than NOAA.

b. Plans for surface systems

NCAR will support a mix of existing and new remote and in-situ sensors based on two frameworks: the transportable S-band Doppler dual-polarization (S-Pol) radar and the transportable Integrated Sounding System (ISS). The Atmospheric Technology Division (A TD) will move toward integrating all other ground-based systems into the infrastructures for power, data processing, and communications of the S-Pol or ISS, and will make those two infrastructural nodes more compatible. Individual units--such as a flux po rtable automated mesonet station (Flux PAM) or Cross-chain Loran atmospheric sounding system (CLASS)--will stand alone but will have hardware and software consistent with the larger system.

S-Pol. NCAR will improve and maintain the S-Pol radar to provide a fully transportable, multiparameter radar to meet a variety of community needs, with special applications to hydrological research and forecasting (see NCAR -- Science, Facilities, and Service). A series of field studies involving concurrent radar and in-situ precipitation measurements will be conducted from a variety of sites. The studies will help validate satellite-based precipitation e stimates and assist in determining the cost-effectiveness of adding dual polarization to NEXRAD, the national network of WSR-88Ds. NCAR will supplement the S-Pol system with a set of bistatic receivers, developed jointly with the University of Oklahoma, t o allow real-time retrieval of multiple-Doppler wind fields. We expect development of remote-control capabilities and real-time data distribution systems to allow universities to cooperate with each other on use of the radar for instructional purposes.

S-Pol Radar NCAR's S-band Doppler dual-polarization radar (S-Pol) is capable of significantly improved signal processing and polarization measurements compared with earlier radars. Its design greatly simplifies transporting and significantly lowers shipping costs.

ISS. NCAR will upgrade the heavily used and highly flexible Integrated Sounding System to include improvements to specific components: incorporation of Flux-PAM technology into the surface measurement component; conversion of some existing container-based systems to systems that can also operate from trucks; regular inclusion of NCAR's aerosol lidar; and a complete system for remote-control, remote-monitoring, remote-QC, and integrated data and communications. ISS units co uld reside at universities between research deployments, to offer students a wealth of hands-on experience with a variety of sensors.

Wind profilers. NCAR will continue to emphasize development of spaced-antenna technologies to replace the 915-MHz beam-swinging wind profiler systems that now form the centerpiece of the ISS, to allow much-higher-resolutio n wind measurements and perhaps direct measurements of turbulent fluxes from the profilers.

Water vapor profilers. NCAR, in collaboration with NOAA's Environmental Technology Laboratory, will develop and deploy compact, relatively low-cost, surface-based lidar systems. Deployed alone or as part of the ISS and in conjunction with ground-based global positioning system (GPS) receivers, these systems should provide accurate profiles of near-surface water vapor and be an important supplement to GPS-based measurements.

Sounding systems. NCAR will implement a GPS version of the rawinsonde, patterned after ATD's successful GPS dropwindsonde. This innovation will allow ATD to provide rawinsonde systems for operations throughout the world fo llowing the demise of the Omega navigation system in October 1997.

Surface systems. In response to demand for both intensive and extensive surface-exchange flux measurements, often within a single field project, NCAR will continue development of the integrated surface-exchange/micrometeor ology Flux PAM system. This system will provide reliable, automated, solar-powered surface meteorology and surface flux measurements in environments from the tropics to the ice caps. At sites with generated or line electrical power, the Flux-PAM systems w ill support additional chemical sensors.

3. Solar Observing Systems

Solar research relies on indirect means to infer velocity and magnetic fields and the thermodynamic state of the solar plasma through the analysis of emitted radiation. In the past decade, helioseismology--the remote determination of global pulsation freq uencies of the sun--has opened a new window to the solar interior. The Solar Magnetism Initiative, an effort to understand how and why the sun's magnetic field induces radiative and particulate variability (see NCAR -- Science, Facilities, and Service), involves acquisition and quick-access archiving of improved solar observations coupled to developments in theory and numerical simulation.

NCAR will pursue all options to place a Stokes polarimeter in space to obtain extended periods of high-quality views of magnetic flux in the solar photosphere. NCAR will continue to acquire and archive data from the advanced coronal observing system and f rom the chromospheric helium imaging photometer, which provide indirect information on the topology of larger-scale solar magnetic fields. In cooperation with the National Solar Observatory, NCAR will assist with deployment and operations of the precision solar photometric telescope at the Mauna Loa Solar Observatory. This telescope forms the centerpiece of an NSF program on radiative inputs from sun to Earth by acquiring full-disk images of the chromosphere in the Ca II K-line and two images of the photo sphere. Finally, NCAR will operate a second Low-Degree Oscillations Experiment (LOWL) instrument in the Canary Islands. The Canary Islands and Mauna Loa instruments together will provide more observations of solar acoustic oscillations each day and more a ccurate information on structure and dynamics of the solar interior. NCAR will provide community access to data from these systems.

4. Community Models

NCAR develops and supports several complex models for community use, with services including cooperative research runs of a model on NCAR computers, transfer of code to universities for research by individual or groups outside of NCAR, tutorials, workshop s, user groups, and collaboration on evaluation and improvement. Widely-used models supported by NCAR include the NCAR/ Penn State Mesoscale Model 5 (MM5) (see NCAR -- Science, Facilities, and Service),the Climate System Model (CSM), the Chemical Transport Model (CTM), and the Thermosphere-Ionosphere-Mesosphere-Electrodynamics general circulation model (TIME-GCM) (see NCAR -- Science, Facilities, and Service).

Community Models at the Millennium Integrated mesoscale model for research and forecasting Comprehensive Earth System Model Comprehensive Chemical Transport Models

a. Mesoscale Model, version 5 and beyond

NCAR will develop a complete system of four-dimensional variational data assimilation for community use, based on the MM5 and its full-physics adjoint. This assimilation system will allow incorporation of, for example, GPS-based meteorological data or NEX RAD radar data. NCAR will cooperate with the NWS National Centers for Environmental Prediction, NOAA's Forecast Systems Laboratory, and the University of Oklahoma to produce a next-generation forecast model. This model will be designed for both research a nd forecasting, and will eventually replace the MM5, providing the community with an integrated mesoscale numerical prediction system capable of resolution in the 1 to 10 km range. NCAR will contribute particularly to representing convection in this natio nal forecast and research model. An initial research version is planned for two years in the future, with the operational version expected to be in use in five years.

Additions to Mauna Loa The extension of NCAR/HAO's Mauna Loa Observatory will house the Precision Photometry Telescope of NSF's SunRise program to measure aspects of the sun's radiative variability. The extension of the building will allow for additional lab space hosting a var iety of solar experiments conducted at Mauna Loa.

The new community model will feature the same broad input from users that has been the hallmark of the MM5. The model's users will contribute to its refinement through continual interaction with the model developers and through an annual workshop where us ers discuss results, problems, and suggestions for improvement. Because of the close connection to the operational models, the new model is expected to attract even more users than the MM5, which has over 200 worldwide.

b. Climate System Model

NCAR is committed to maintaining and improving the CSM, and to supporting it with documentation, workshops, and improved access to data. With many university investigators, we will explore improvements to all components of the CSM and use it to conduct co mmunity experiments and to contribute to the Intergovernmental Panel on Climate Change assessments and to the Paleoclimate Model Intercomparison Project. Collaborators will include the Geophysical Fluid Dynamics Laboratory, University of Washington, Color ado State University, University of Colorado, University of California-Los Angeles, Ohio State University, University of Arizona, and Los Alamos National Laboratory. Other members of the community will also be involved through annual CSM workshops and thr ough various CSM working groups. Eventually, with the inclusion of biogeochemistry and coupling to NCAR's upper atmosphere models, the CSM will form the basis for a more comprehensive earth system model.

c. Thermosphere-Ionosphere-Mesosphere Energetics and Dynamics General Circulation Model

NCAR will refine and augment the representation of physical and chemical processes within the TIME-GCM, with particular attention to couplings between atmospheric regions. NCAR will also improve representation of solar inputs into these models and interac tions with the magnetosphere to ensure better representation of upper atmospheric response to solar variability. Eventually, a physically consistent model combining TIME-GCM, the Community Climate Model, version 3, and models of global tropospheric chemis try will lead to comprehensive earth system modeling focusing here on the impacts of solar variability. NCAR will provide results and analyses from these model simulations for wide community use. The existing TIME-GCM is an essential tool within NSF's Cou pling, Energetics, and Dynamics of Atmospheric Regions (CEDAR), Geospace Environment Modeling (GEM), and Space Weather efforts as well as in various NASA programs, and it is in use by many collaborators in some 20 universities nationwide. Offsprings of th e TIME-GCM are used in modeling atmospheres of Mars and Venus (recently in support of the NASA Mars Pathfinder mission).

d. Chemical transport modeling

NCAR will further develop comprehensive chemical transport models (CTMs) to establish the global and regional budgets of chemical compounds in the atmosphere and to assess the importance of surface emissions and deposition, chemical transformations, and t ransport processes on the distribution of these chemical species. These models will be used to integrate observational data and to predict the future evolution of the atmospheric composition, for example in response to anthropogenic perturbations. The emp hasis will be on regional models focusing primarily on air quality and on global models addressing global change. These latter atmospheric models will progressively be coupled to detailed models of the continental biosphere and the ocean, in order to trea t the global cycles of chemical elements (carbon, nitrogen, and sulfur) in the earth system.

As instruments and modeling tools become more powerful, NCAR increasingly plans to quantify and predict the complex interactions among aerosols, nucleation, and cloud-phase chemistry, and their impact upon climate. As an example, we plan to collaborate wi th Georgia Institute of Technology and others to measure and model the dynamic growth and dissipation of sulfate aerosols in convective systems.

5. Datasets

NCAR makes available to the university research community a wide variety of datasets from model runs, field programs, and special activities such as reanalysis projects. This collective archive represents an invaluable resource to the community, much of i t maintained by the researchers around the center who are its informal curators; they are intimately involved with their data and are available to guide other researchers directly in use of the data. In addition, NCAR's Scientific Computing Division maint ains over 490 distinct datasets in a computer-accessible research archive that is available to scientists around the world. This archive represents a unique and irreplaceable store of observed data and analyses, and is used for major national and internat ional atmospheric and oceanic research projects. As resolutions of field observations and of model runs increase, the size of these datasets more often requires special handling available only from a national center. Large multidisciplinary field programs and more-extensive community use of NCAR models will increase needs for reliable central access and archiving services with data available via the Internet, through file or media transfers, or through NCAR computers.
a. Plans for data delivery and support

Through the Information Infrastructure Technology and Application project in the UCAR Office of Programs, and through cooperative efforts among many NCAR divisions, NCAR will provide easy and flexible access to the full range of its data holdings. This im proved access should allow investigators from outside the center to easily search field project datasets and model output, without necessarily knowing beforehand the specific field project or model run sources. For observational datasets, NCAR will contin ue its efforts to provide more near-real-time access to quality-controlled datasets to enable remote use and data assimilation.

b. Reanalysis project: toward a comprehensive global dataset

NCAR is working to maintain the rich legacy of observations reflected in research and operational datasets from the past 50 years. The National Centers for Environmental Prediction (NCEP)/ NCAR Reanalysis Project will produce a research-quality set of glo bal observed data and a complete set of analyzed fields from the analysis/forecast system (see NCAR -- Science, Facilities, and Service. The Reanalysis Project will process data covering 1948-1997. NCAR has prepared many datasets of global observations for use in reanalysis, including land-surface, ocean, upper-air, aircraft, and sat ellite data. The Reanalysis Project uses these data as input into a state-of-the-art data assimilation system to produce a consistent set of global analyzed fields at 28 levels in the atmosphere. NCAR will also collaborate with the European Centre for Med ium-Range Weather Forecasts to produce an independent 50-year global reanalysis; this will be complete in about 2002.

NCAR will maintain and distribute output from both reanalysis projects. The output consists of a full global dataset each six hours, including temperature, winds, precipitation, snow, clouds, and radiation. These datasets already support a wide range of r esearch on weather, climate, and oceanic topics. In March 1997, the project completed 29 years of reanalysis (1968-1996). Starting in about 1999, it will produce a new set of analyses, using the best of methods and models available at that time. NCAR will continue its support of the reanalysis and the subsequent data access.

CHALLENGE We will advocate effectively for the principle of free and open exchange of scientific data.

6. Networking and Data Communications

NCAR will provide cost-effective Local Area Network capability that meets UCAR and NCAR needs, Metropolitan Area Networking to support work off-site, and Wide Area Networking that meets the need for national and international links to all UCAR entities. T hese goals present a considerable challenge in the current technological environment. Increases in the speed and disk capacity of desktop computers will drive increases in researchers' ability to generate and process information, leading to requirements f or increased networking and data communications. Use of the World Wide Web, as well as new applications of it, will continue to grow, with corresponding demands on network technology.

Advances in information technology will lead to a new era in electronic communication, collaboration, data collection and analysis, and management. This environment will require high-performance networking at all levels--local, metropolitan, national, and international. Through partnerships and collaborations with peer organizations and industry, NCAR will provide the cost-effective networking and data communications needed by its community.

D. HUMAN DIMENSIONS AND SOCIETAL IMPACTS

NCAR's research program on human dimensions and societal impacts promotes the use of atmospheric science information in service to society and is designed to improve understanding of the complex interactions among human societies, natural resource systems , and atmospheric processes. The mission of this program is to conduct scientific research on the impacts of climate and meteorological processes on society, both directly and through impacts on the natural systems that humans use and value; on anthropoge nic influences on the climate system; on societies' ability to cope with climate-related impacts; and on the use and value of meteorological information in a variety of contexts.

1. Human Dimensions Initiative

NCAR will develop an institution-wide research theme on the human dimensions of the atmospheric sciences. This new focus will entail an increased emphasis on multidisciplinary research and an assessment of institutional policies in order to encourage cros s-divisional and cross-program interactions.

The expanded human dimensions research program within UCAR will build on the leadership of NCAR's Environmental and Societal Impacts Group (ESIG). Other divisions will participate by giving increased attention to the implications of their research for hum an society and by incorporating social science-based analyses of human activities in their assessments of anthropogenic impacts on the atmosphere. ESIG will continue to provide intellectual leadership, and its staff will work as partners and collaborators with climate impacts researchers as well as with researchers in NCAR, the UCAR Office of Programs, and the university community.

The Human Dimensions Initiative will provide a forum for the development of cross-divisional research initiatives in this area. NCAR scientists and their university collaborators will identify research projects and activities aimed at answering specific, tractable questions regarding the atmosphere-society interface. Joint research on these questions, and the preparation of proposals, may be supported by the NCAR director through a special fund to supplement this cross-divisional initiative. ESIG's role i n fostering the Human Dimensions Initiative will be to organize meetings with scientists in the other divisions to discuss shared research interests and to help to identify the intellectual resources and social science methodologies needed to address the research questions.

2. Current Human-Dimensions Themes

ESIG's own research plans for the human dimensions initiative may be grouped under three major scientific themes: (1) interactions among atmospheric processes, environment, and society, (2) use of weather- and climate-related information in decision makin g, and (3) development and application of impact assessment methodologies. As the Human Dimensions Initiative takes shape, ESIG and the other participants will coordinate their efforts and explore potential new research themes.
a. Interactions among atmospheric processes, environment, and society

A particular focus in the next few years will be on the impacts of climate variability on natural resource systems. These include systems that are intensively managed by the private sector, such as agriculture, and publicly managed systems, such as fisher ies, water resources, and public lands, where multiple interests and conflicting objectives complicate the responses of managers and resource users to the effects of climatic variations. Research in this area will focus on how decisions about long-term in vestments and institutional factors influence the robustness, resilience, and flexibility of responses to ongoing climate variability. Because this work is inherently multidisciplinary, it will involve collaborations between ESIG, scientists in other divi sions, and members of the university community. Collaborators include the University of British Columbia; Carleton College; the National Institute of Water and Atmospheric Research (New Zealand); Clark, Colorado State, Columbia, Duke, Harvard, Johns Hopkin s, Oregon State, Penn State, and Stanford Universities; and the Universities of Colorado, Montana, Nebraska, South Carolina, and Washington.

Other work will focus on furthering our understanding of adaptation to climate change. For example, ESIG will use climate model scenarios to study possible adaptations to climate change in agriculture in selected countries. Additional projects will focus on other resource sectors. This area is expected to provide opportunities for both quantitative and qualitative research, with cross-divisional collaboration.

Finally, as the Human Dimensions Initiative develops, ESIG may interact increasingly with researchers in the Climate and Global Dynamics Division, and particularly with the ACACIA program (A Consortium for the Application of Climate Impact Assessments), t o understand the socioeconomic factors driving the impacts of human activities on the climate system. Such an improved understanding will help to better forecast those impacts and will contribute to policy analyses related to emissions and land-use manage ment.

b. Use of weather- and climate-related information in decision making

Use of weather-related information. Society appears to be increasingly vulnerable to weather, suggesting a pressing need to improve our abilities to prepare for, respond to, and mitigate the effects of high-impact weather events such as hurricanes, floods, and blizzards. Many of the severe weather problems society faces are due more to social and demographic changes than to increases in the frequency and magnitude of the events themselves. In addition, everyday phenomena s uch as lightning, precipitation, temperature fluctuation, and wind can significantly affect the costs of doing business. Improved information can reduce the impacts of these typical weather events, which in the aggregate are economically very significant.

The use and value of weather information to decisionmakers will be addressed within the human dimensions side of the U.S. Weather Research Program. This project will focus on how weather information is used in policy responses to extreme events, as well a s on the impacts of routinely disruptive weather and the potential value of improved forecasts for management of those impacts. The project will bring together physical and social scientists with users of weather information to foster a closer link betwee n the production and use of meteorological research.

Use of climate-related information. The use of information on the El Niño Southern Oscillation (ENSO) will be investigated in two planned activities. (1) An ENSO information audit will review the use of ENSO informa tion in decision-making within specific government agencies, seeking to enhance the use of the increasing body of ENSO-related information, including forecasts, in decision-making and interannual planning. (2) Scientists have identified conflicting sets o f years as ENSO years, possibly causing errors in assessments of ENSO impacts. This second activity will focus on approaches to identifying ENSO and non-ENSO years, in order to improve assessments of the societal aspects of ENSO.

Studies of the use of climate information in natural resource management will focus on management of common property or transboundary resources. In particular, fluctuations in the size of migrating fish stocks frequently lead to highly destructive "fish w ars." An example is the international management of migratory fish populations. Climate-driven variations in stock abundance affect the international management of certain transmigrating fishery resources. A goal of this research is to find ways of using information about the effects of climate variations on fish populations to facilitate cooperative management and avoid fish wars.

CHALLENGE We will contribute meaningfully to the formulation of national and international scientific assessments and other policy issues.

c. Development and application of impact assessment methodologies

Application of climate model results to environmental and societal problems. The integration of global and regional climate model runs, climate change scenarios, and climate impacts is an important goal receiving increased attention in the international modeling and impacts communities. Such integration maximizes the value of climate model results. The spatial resolution of general circulation models is too coarse to address many environmental issues. Climate model results must be downscaled, either dynamically by applying regional climate models, or statistically, to bring the scale of model results closer to the needs of impacts research.

Figure 6. Flood damages and deaths.

Several multidisciplinary projects will address the issues of integration and spatial resolution. These projects include: (1) the effect of spatial scale in climate change on agricultural impacts in the Great Plains, (2) the response of vegetation to meso scale climate variability in the mountainous West of the United States, (3) the effect of changes in mean and variability of climate on crop production and regional economics of the southeastern United States, (4) the integrated social and economic effect s of extreme climatic fluctuations on forests in the northeastern United States, and (5) international projects focused on wheat production in Italy and agricultural development in the Yangzte Delta of China.

Development and application of statistical methods in climate impact assessment. There is uncertainty in all aspects of impact assessment, including numerical models of the climate system and response functions linking cli mate to its impacts. These uncertainties are exacerbated by the mismatch of spatial and temporal scales in attempts to integrate components. Research on downscaling will include statistical methods and development of a physically realistic mechanism for i mproving stochastic generators used to create scenarios of climate variability. It will also compare dynamic and statistical downscaling methods.

Extreme events have ramifications in many aspects of impacts assessment. Extremes are not treated rigorously in either climate research or climate impacts research. The inadequate treatment of extremes in the 1995 Intergovernmental Panel on Climate Change reports is now well recognized. We will apply the well developed statistical theory of extremes, which is not widely used in the atmospheric sciences, as an example for other climate and impacts researchers.

E. EDUCATION AND TRAINING

the fact that such an Institute offers the possibility of preserving the natural alliance of research and education without unbalancing the university programs (UCAR, 1959).

Objectives of ASP are to

• encourage the development of young scientists in the field of atmospheric science,

• direct attention to timely scientific areas needing special emphasis,

• help organize new science initiatives,

• support interactions with universities, and

• promote continuing education at NCAR.

Through ASP and other programs, NCAR's research and facility programs have also entrained students as graduate fellows, research assistants, and in special focused activities such as colloquia and summer employment programs. NCAR scientists routinely hold adjoint appointments in university departments and serve on university thesis committees. NCAR has published 160 NCAR Cooperative Theses documenting the Ph.D. research of graduate students who have worked collaboratively with NCAR scientists.

In recent years, UCAR and NCAR have added an important new focus to their education activities and programs, contributing to the nation's science and mathematics education through programs aimed at undergraduates, grades K-12, and the general public. The nation's commitment to improving math and science education has been highlighted in recent years, as test scores of U.S. students in these subjects indicate that our students are less accomplished than their peers around the world.

There have been several successful programs at NCAR and UCAR dedicated to improving math and science education. These include Project LEARN (Laboratory Experience in Atmospheric Research at NCAR), Skymath, and the Global Change Instruction Program (see NCAR -- Science, Facilities, and Service). NCAR's outreach and tour program is known throughout the region, especially among the elementary schools in Colorado and neighboring states.

A new initiative will likely be the focus in the future for UCAR's extramural education programs. This is the Program for Advancement of Geophysical Education (PAGE), which will work with university faculty across the country to determine the best ways to deploy the resources of NCAR and UCAR in developing multimedia educational tools for use in the undergraduate geoscience curriculum. PAGE will build on the successful models developed in the UCAR Office of Programs (UOP) at Unidata and the Cooperative Pr ogram for Operational Meteorology, Education, and Training (COMET). PAGE will use raw materials developed by COMET for training NWS forecasters.

Another very important new activity at UCAR is the Significant Opportunities in Atmospheric Research and Science (SOARS), whose goal is to produce a marked increase in graduate degrees in the atmospheric and related sciences held by seriously underreprese nted groups (see NCAR -- Science, Facilities, and Service). This program combines summer research activities, mostly though not exclusively at NCAR, with graduate school support. SOARS is funded jointly by NSF, NASA, and the UCAR member universities, and is evidence of the commitment of the sponsors, the center, and UCAR to be leaders in increasing diversity in our field.

NCAR scientists will continue to serve as lecturers and specialty experts for COMET, which is a part of the NWS modernization to upgrade forecasters' performance and level of understanding of physical processes. The COMET program has made some important i nroads internationally, and we anticipate that its international outreach and impact, and that of other UCAR activities, will expand in the future.

LEARN students with Neal Lane A group of middle school students demonstrate some hands-on science while NSF director Neal Lane (far right) cheers them on during his visit to Colorado in March of this year. (Photo by Carlye Calvin.)

The outcry for sound information for political leaders in Congress and the executive branch, and for the general public, has increased in recent years, particularly as budget pressures are increasingly forcing difficult choices. The scientific consensus o n matters such as ozone depletion and climate change is being challenged in many quarters at the same time that public interest in environmental issues is at an all-time high. The people and products of the research enterprise must be used to translate sc ientific expertise into terms that policymakers, legislators, and private citizens can understand and use. Through its Office of Government Affairs and Office of Communications, we have already expanded our visibility both on Capitol Hill and in the press and will continue to leverage our resources to communicate with Congress and the public, directly and through the print and broadcast media.

CHALLENGE We will increase communication with policymakers and the public about the importance of science, involving colleagues from the community.

UCAR enthusiastically embraces its responsibility to contribute to the nation's education programs and will continue to undertake activities that are germane to the atmospheric and related sciences, that appropriately draw on the expertise within our inst itution and constituent community, and that make appropriate use of the resources in NCAR and UOP.

F. APPLICATIONS AND TECHNOLOGY/INFORMATION TRANSFER

1. Direct Transfer of Information and Technology

Much direct transfer provides weather- and climate-related information to aid in decision making. For example, in the Environmental and Societal Impacts Group, researchers inform policy- and decisionmakers about the societal aspects of atmospheric process es. The use of El Niño information to anticipate and react to droughts in Africa is an excellent example of what has been termed "usable science."

Societal impacts are also inherent in the applications work in NCAR's Research Applications Program (RAP), which receives almost all of its funding from government entities and private companies that have vested interests in RAP's technology transfer. In the aviation weather research program, RAP collaborates with universities, government labs, and NCAR divisions around two general themes: improving the detection and forecasting of hazardous weather, and working to modernize the delivery of such informati on to the decisionmakers involved. This work impacts the airlines, air traffic controllers, and pilots, though of course the flying public is the main beneficiary. A major emphasis for the future involves providing graphical weather information directly t o airplane cockpits, so that pilots can act to avoid such hazards as turbulence, icing, and thunderstorms.

In certain situations the most important practical result of a directed research effort is not an improved algorithm or a data delivery system but an education and training effort. In the case of RAP's work on low-level windshear, for example, training le d to a reduction in accidents long before any detection and warning systems were in place. Other more recent examples include educating pilots about the dangers associated with large-drop icing and about the relationship between visibility and snowfall.

The aviation weather work will also be pursued through support from governments abroad. For instance, RAP and the Mesoscale and Microscale Meteorology Division, working with Hong Kong University of Science and Technology and the University of Wyoming, are currently in the process of delivering to the government of Hong Kong a turn-key system for the detection and forecast of

terrain-induced windshear and turbulence at the new Hong Kong airport. (This activity has been carried out through Weather Information Technology, Inc., a for-profit corporation owned primarily by the UCAR Foundation.) This is the only system in the world that integrates the detection and forecasting of both terrain-induced turbulence and convective windshear. The underlying windshear technology was developed by RAP with funds from the Federal Aviation Administration for the protection of U.S. commercial airports. This combined technology is expected to provide the basis for an operational solution to a severe turbulence and windshear problem in Juneau, Alaska, where RAP and the Atmospheric Technology Division (ATD) are now gathering data to better unders tand the local phenomenon. This case is an interesting example of a government abroad paying for research and development that significantly enhances the original technology developed with U.S. government funds and which will now be deployed in the United States.

In association with the Universities of Colorado. Wyoming and Illinois, Pennsylvania State University, and others, RAP and ATD will be involved with aspects of hydrometeorology and the transfer to NOAA and others of improved methods for remote rainfall me asurement, thunderstorm nowcasting, flood forecasting, and other capabilities. They will also seek to transfer specialized weather information to other weather-sensitive sectors of the economy, such as agriculture, construction, energy, and recreation. In ternet technology is one possible vehicle for providing such information.

2. Transfers Involving Public-Domain Access

Some technology is made available by placing it in the public domain and providing ready access to it over the Internet. Numerical modeling codes that simulate and predict complex processes in the atmosphere are prime examples. These range from the CSM to MM5, systems with a long history of development and use by hundreds of university scientists worldwide.

Other good examples of software made available by public domain access are the large-eddy simulation code used to study boundary-layerprocesses and the Zebra software, which is unique in its ability to integrate and display diverse and disparate datasets.

3. Licensing Activities

NCAR will continue to transfer many of its technologies through the use of licensing agreements. The UCAR Foundation provides the necessary legal, business and marketing support, and to date has issued more than 30 licenses to non-UCAR organizations (see NCAR -- Science, Facilities, and Service).

New Hong Kong Airport Through Weather Information Technologies, Inc. (WITI), a for-profit subsidiary of the UCAR Foundation, UCAR is sharing its expertise in predicting and warning of weather hazards and terrain-induced turbulence around airports. WITI has a $16 million contra ct with the Observatory of Hong Kong to provide an operational windshear warning system for Hong Kong's new airport on Chek Lap Kok Island. The system draws on expertise from ATD, MMM, and RAP.

Licensing has two principal advantages. First, it makes the technology available widely in a manner that controls the quality. Many technologies, particularly hardware, would not be readily and economically available to our constituents without licensing agreements that facilitate their commercial manufacture and sale. The GPS dropsonde and the Low-Level Windshear Alert System are two examples of transfers through licensing agreements with commercial firms. The second reason for using licensing is to gene rate revenue to offset the costs of the licensing activities (including intellectual property protection) and to reward the inventors and developers and support UCAR science and education programs.

UCAR also grants nonexclusive, no-cost licenses. This approach helps ensure that technology is properly replicated and used and where derivative works are developed that they are licensed back to UCAR for the benefit of the entire community. Licenses of t his type are frequently granted to universities. Two past examples are the Integrated Sounding System and a short-path IR absorption hygrometer. This approach will continue to be used when it best serves the mutual needs of NCAR and university investigato rs.