NCAR's FY1998 Science Highlights

A selection of scientific results from FY1998. Please refer to the individual divisions for additional research results and findings.

High Altitude Observatory

• On February 26, 1998, HAO continued its long history of coronal science at total eclipse with both ground-based observations at Curacao and, in collaboration with several other institutions, observations from an NCAR aircraft. At Curacao, two investigations were completed. The POISE-98 (POlarimetric Instrument for Solar Eclipse-98) recorded a white-light image with very high photometric and polarimetric precision. The other Curacao experiment, Photometric Eclipse Polar Plume Imager (PEPPI) was optimized to achieve both high angular resolution and low noise images of the corona during the brief 3.5 minutes of totality. The goal of this experiment was to look for changes in the corona which might provide clues to the heating of the corona and the acceleration of the solar wind. A team of scientists led by R. MacQueen (Rhodes College) worked with the ATD division of NCAR to obtain unique observations of the corona at infrared wavelengths, from an airborne platform (the NCAR C-130 aircraft) over the Pacific, during the 26 February 1998 total solar eclipse. Phil Judge worked with Jeff Kuhn and Haosheng Lin (National Solar Observatory) to detect for the first time a potentially strong coronal line of Si IX at 3.934 microns. The effort was successful- the Si IX line- predicted by Judge in a theoretical article in 1998 was positively detected, and analysis of the data indicate that this line offers great potential as a diagnostic of the coronal magnetic field strength.

• A detailed analysis of SUMER full-disk observations (1) and (2) obtained by a 30 hour rastering of the 300 arcsec spectrograph slit, has revealed exciting and largely unanticipated information about the net radial motions of transition region and low coronal plasmas. The results, obtained by Hardi Peter (Visitor, Max-Planck-Institut für Aeronomie), reproduce the well-known net disk-center redshift (downflow) of 6 km/sec in the 154.8 nm C IV line which forms at a temperature of 105 K. On the other hand, the same analysis techniques when applied to the 77.0 nm Ne VIII line, which forms at a higher temperature of 6.5 x 105 K, indicate a disk-center blueshift (upflow) velocity of 2 km/sec. This finding has profound implications for our understanding of momentum and energy transport across the boundary that separates chromospheric from coronal material, and it lays to rest a great deal of uncertainty and speculation about the net motion of upper transition region material. From his analysis, Peter also demonstrated that the nonthermal line widths of these emission lines decrease from the Sun center to the solar limb, suggesting that the unresolved macroturbulent motion are anisotropic in the sense that horizontal motions are suppressed relative to those aligned with the local radial direction.

Climate and Global Dynamics Division

• The first Climate System Model (CSM) simulation of the 20th century climate was completed by Byron Boville (CSM Co-chair) and the Chemistry and Climate Working Group. The globally averaged temperature increases by about 0.6 K between the late 19th century and the 1990s, with most of the increase occuring since 1970, in agreement with observations.

• Frank Bryan (Oceanography Section, OS) and Rick Smith (visitor, Los Alamos National Laboratory, LANL) have demonstrated that a resolution of 10 km or smaller is necessary to resolve adequately mesoscale eddies in a numerical model of the North Atlantic Ocean.

Atmospheric Chemistry Division

• Air samples collected in different parts of the world during several field campaigns (NARE-97, SONEX, GASEX-98, Polar Sunrise Experiment 98, FIRETRAC, WiFE) have been analyzed in the laboratory, providing mixing ratios of about 70 chemical species including chlorofluorocarbons and other halogenated organic compounds. The analysis and interpretation of these measurements and of observations made during previous campaigns (e.g, STRAT, STERAO, PEM-Tropics, and POLARIS) have provided new information on the distributions and budgets of a variety of chemical species as well as on the chemistry and transport processes in the upper troposphere/lower stratosphere at different timescales.
• The HANK regional nested chemical-transport model of the troposphere was completed and used to extensively analyze model runs performed for the Pacific Basin in conjunction with the NCAR-led MLOPEX field campaigns. Version 1 of the NCAR MOZART global chemical transport model of the troposphere was completed, validated, and released to the scientific community. The model was used to determine changes in the chemical composition of the troposphere since the pre-industrial era and to assess the impact of human-induced perturbations including aircraft operations, fossil fuel combustion, and biomass burning.

Scientific Computing Division

SCD's science highlights for FY1998 include upgrading the world's best atmospheric and oceanic datasets and enabling the steady flow of scientific insights through visualization technology.

• The NCEP/NCAR Global Atmospheric Reanalysis Project is an effort to reanalyze a long period of historical data using a single state-of-the-art atmospheric model. Artificial anomalies in the data time series can occur as the models evolve. Furthermore, under the time constraints, only rapidly available non-delayed data are used. The Reanalysis Project is a major effort designed to overcome these limitations in the operational analyses.

  This project is a cooperative effort between NCAR's Data Support Section (DSS) and the National Centers for Environmental Prediction (NCEP) of NOAA. The project goal is to reanalyze the previous 50 years of atmospheric data, and this was completed on July 23, 1998. The dataset provides output each six hours. The analyses are done at a resolution of T62 (208 Km) and 28 levels in the vertical. The project started in 1991, based on many earlier years of data gathering, model development, and related experience. This work and its extensions are also helping other reanalysis projects around the world.

  Comparison of sea surface temperatures in the tropical oceans.

  

  

  The top panel shows El Niño conditions at the ocean surface, and the bottom panel shows La Niña conditions. (Red indicates warmer water and blue cooler.) The Comprehensive Ocean-Atmosphere Data Set (COADS) and satellite data were combined to create these global views, courtesy of the Coupled Model Project at the National Meteorological Center.

  The Comprehensive Ocean-Atmosphere Data Set (COADS) has been created by combining, editing, and summarizing global in situ marine data from many different sources. Merchant ship observations back to 1854 have been supplemented in more recent years by automated measurements, e.g., from drifting and moored buoys. This project with other labs started in 1981 to expand and update the world's best surface ocean dataset. These data include temperature, pressure, wind, clouds, etc. They provide much of what the world knows about ocean surface temperature changes during the past 144 years.

• SCD's Visualization Group (VG) works to advance atmospheric and related science through state-of-the-art visualization. VG staff explore and develop new technologies that improve understanding of complex phenomena by making key processes visible. The intent of visual science is to use computers to supply the human brain with the kind of information that it uses best for exploration and understanding: visual input.

  The ability to explore and understand complex simulated and observed worlds will be of great importance not just to atmospheric science, but to all science. It will be vital to the researcher and formative to school children learning about physics or chemistry. Scientists will explore their world in ways they never could before; children will learn in ways we didn't imagine just a few years ago.

  Our simulations and our observational datasets grow larger and more complex by the day. And, as our computational capabilities continue to track the exponential curve, it may soon be the case that the primary limiting factor for research is not the technology, but the human. One must be able to digest the data and ask questions of it. There are many questions we can ask of these complex simulations -- many questions that in the past made no sense to ask or had no answer that could be easily conveyed. The Visualization Lab is aimed at helping to usher in this new era of visual science -- at providing the ability to freely explore vast dataspaces -- and produce materials that can communicate the results of such efforts to peers and public.

  Cloud system evolution in the eastern tropical Atlantic.

  

  This image (click for larger version) is composed of three isosurfaces and two 2D slices. The white isosurface represents a 0.5 g/kg ice water isovalue, the yellow isosurface represents a 0.5 g/kg liquid water isovalue, and the red isosurface represents a 0.5 g/kg rain water isovalue. The vertical slice at the rear of the model domain shows the east-west wind component (U), and the horizontal colored slice at the bottom of the domain shows contours of rain water at the surface. A Cloud Resolving Model was run using data from a unique 3D experiment performed during September 1-7, 1974, during Phase III of the Global Atmospheric Research Programme Atlantic Tropical Experiment (GATE). The visualization was generated using an NCAR stereo-enhanced version of Vis5D.

  At present, we have well-developed capabilities and experience across the breadth of NCAR science. Extant visualization work encompasses climate, chemistry, ocean, mesoscale systems, forest fires, geophysical and astrophysical turbulence, clear air turbulence, tropical storms, and more. We have developed both interactive and production visualization environments that allow us to create and record both mono and stereo 3D visualizations of very large, very complex multivariate datasets. Highlights of our work in FY1998 are included in the Visualization Lab Research Gallery.

Mesoscale and Microscale Meteorology Division

Weather Research and Forecast (WRF) joint research and operational model: Joseph Klemp, William Skamarock, and Jimy Dudhia of the MMM Division continued to work on the development of the Weather Research and Forecast (WRF) joint research and operational model with colleagues from NCEP, NOAA/FSL, CAPS, and university scientists. Development includes a thorough analysis of treatment of the lower boundary in the vicinity of mountains (e.g., a stepped approach verses a terrain following formulation), minimization of pressure gradient force errors near steep mountains, exploration of various approaches to using only conservative quantities as prognostic variables, and examination of the advantages of a hybrid vertical coordinate where the information surfaces become isentropic surfaces away from the ground. In addition, a prototype framework is being developed that will allow easy portability of model code to a wide range of computing platforms including distributed shared memory machines as well as workstations and vector machines. Early prototypes of the WRF model will be available in calendar year 1999.

• Development of cloud-resolving convection parameterization methods: MMM Division scientists, Wojciech Grabowski and Piotr Smolarkiewicz, developed a method referred to as the cloud-resolving convection parameterization (CRCP) to explicitly represent the effects of tropical convection at large scales in climate models. The approach takes advantage of the nature of tropical convection in environments with significant shear that can often be described by two-dimensional processes. A two-dimensional cloud scale model is imbedded within each column of a large-scale three-dimensional model where the large-scale model provides "ambient forcings" to the cloud scale model and the cloud scale model provides the convective response to the large-scale model. Initial tests are very promising.

Research Applications Program

• The Auto-nowcastor system provides one-hour nowcasts of thunderstorms and strong winds and was originally developed under FAA funding. Demonstrations of the Auto-nowcastor system were held at: 1) Washington/Baltimore National Weather Service (NWS) Forecast office in Sterling, Virginia as part of the System for Convection Analysis and Nowcasting (SCAN), 2) the Army Forecast Offices at White Sands Missile Range in New Mexico, and 3) the Army Forecast Offices at the Aberdeen Proving Grounds in Maryland. In addition, a National Convective Weather Forecast (NCWF) product was demonstrated at the NWS Aviation Weather Center in Kansas City, MO, with products going to Delta, Northwest and Atlantic Coast Airlines. These efforts were sponsored by the FAA, Army, NWS/OSF and NSF under the U.S. Weather Research Program. The demonstrations were highly successful as indicated by the extensive use of the products by operational personnel. At the Sterling National Weather Service Forecast Office (WFO), forecasters utilized output from the Auto-nowcastor to assist them in issuing convective storm advisories and severe storm warnings. The Sterling WFO severe storm warnings for 1998 were far more accurate than any previous year, and they give partial credit to the Auto-nowcastor system for the improvement.
• The multi-parameter radar detection of precipitation type using fuzzy logic is a joint effort between RAP, ATD and NOAA scientists. The algorithm automatically determines precipitation types such as hail, snow pellets, ice pellets, snow, wet snow and dendrites using reflectivity, differential reflectivity, linear depolarization ratio, and differential phase measurements from S-band radars (10 cm wavelength). Because the values of the radar observables that delineate different particle types often overlap and are not sharply defined, a fuzzy logic approach is used. A real-time version of the algorithm was successfully demonstrated this past summer during the PRECIP 98 field program in Florida. Preliminary verification results for the algorithm indicate good skill, however, more verification is needed. Anticipated uses for the algorithm are to aid forecastors in determining the precipitation type, enable researchers to easily determine precipitation type to verify microphysical models, and to increase our knowledge of microphysical storm structure in general.

Atmospheric Technology Division

• Surface Heat Budget of the Arctic Ocean (SHEBA) The Surface Heat Budget of the Arctic Ocean (SHEBA) project was without any doubt one of the most challenging field projects ever supported by ATD staff and instrumentation. In early fall of 1997, SSSF engineers and technicians traveled to the SHEBA ice station about 300 miles north of Prudoe Bay to set up four of ATD's PAM -III stations and a GLASS system for a thirteen month period. The SSSF facilities were only a small part of a much larger array of observational instruments set up within a 62-mile radius of the SHEBA base station to measure a multitude of physical parameters for a full annual cycle. While the GLASS measured the standard meteorological measurements of wind, pressure, temperature, and humidity, the PAM-III stations measured additional parameters such as turbulent fluxes of momentum and heat, incoming and outgoing fluxes of long-wave and short-wave radiation, and the surface heat flux at the snow/ice boundary. The PAM-IIIs were heavily modified before going into the field to withstand the harsh Arctic conditions. Electrical power was produced with propane thermoelectric generators rather than solar panels, and the electronics were housed with the generators to prevent built-up of ice. GPS receivers and electronic compasses were used to continuously monitor station location and orientation, as well as to provide accurate time-keeping. The instruments had to be serviced regularly by ATD staff, who traveled in pairs on snowmobiles, carried rifles in case of polar bear encounters, and relied on survival suits, life jackets, coast guard ships and helicopters once the ice sheet had thinned during the summer months. The Canadian ice breaker "Les Grosseilliers", which was deliberately frozen into the sea ice, served as the base camp and sheltered approximately 50 scientists at the SHEBA ice camp.

  During May and July/August 1998, the NCAR C-130 and RAF and RSF staff conducted joint SHEBA operations out of Fairbanks, Alaska. ATD's Scanning Aerosol Backscatter Lidar (SABL), Airborne Imaging Microwave Radiometer (AIMR) and Multichannel Radiometer (MCR) were part of the aircraft's extensive instrumentation load. AIMR, a dual-channel (37 GHz and 90 GHz) dual-polarization microwave radiometer loaned to ATD by Canada's AES, records the characteristics of surface sea ice. MCR is a seven channel scanning radiometer originally built by NASA Goddard used to map surface emissions in visible and infrared portions of the electromagnetic spectrum. Both of these instruments underwent through significant modifications and upgrades before operations by ATD.

  In the end, the observational component of this $19.5 million project, jointly funded by the National Science Foundation, the Office of Naval Research and various international organizations, was a full success. The data collected will allow scientists to understand interactive processes involving mass changes of the sea ice, storage and retrieval of heat in the mixed layer of the ocean, and the influence of clouds on the surface energy balance. In the long run, results from SHEBA will help us better understand the role of high latitudes in global climate and assist in predicting future climate change and assessing the impact of global warming.

• GPS Dropsondes Development From a technology point of view, the development of the GPS dropsonde system was one of ATD's main highlights this year. In just over two years, the GPS dropsonde system has gone from a prototype developed within ATD to one of the most important observing system in atmospheric sciences. The GPS dropsonde system is now installed on 18 research aircraft within the U.S., Germany, and Canada and routinely used in hurricane reconnaissance and other research missions. The concept of the dropsonde seems to be simple: a sonde attached to a parachute is dropped from an aircraft and measures temperature, humidity, and pressure as it falls. From its position during descent, derived from the sonde's communication with the global navigation systems, winds are calculated. The actual development and deployment however is quite difficult. A dropsonde needs to provide laboratory-quality data in an inexpensive package that one can safely toss out of an airplane moving at greater than 100 meters per second. Faced with this challenge, ATD's dropsonde development team combined new sensor and GPS technology from Vaisala with a completely new structural, electronic, and transmission system of their own design. In each and every application, the GPS dropsonde has provided order-of-magnitude improvements in the quality of data and reliability of operation over previous systems. The vertical resolution for wind-speed measurements has increased nearly a hundredfold. As Hurricane Guillermo raged across the Pacific in August 1997, the sondes helped provide the first-ever high-resolution data on hurricane eyewall structure. This spring, ATD helped equip nine Air Force WC-130s that carry out routine reconnaissance flights for tracking tropical cyclones.

  ATD also improved the accuracy and range of the pressure, temperature, and humidity sensors; upgraded the electronics; and built a completely new in-flight processing system. Up to four sondes can now be launched and tracked as closely as 20 seconds apart. The success of this state-of-the-art instrument was widely noticed. Schematics and descriptions of the dropsonde showed up on the science and weather pages of newspapers and magazines across the country. NASA and NOAA proudly showed the dropsonde during their press conferences, and television stations ran video clips of weather officers launching the dropsondes from USAF C130s. Hurricane forecasters evaluated landfall models on the basis of how much dropsonde data their models assimilated. The user community praised the unprecedented high vertical resolution, excellent performance in bad weather and wind measurements were they counted the most, in the eye of a hurricane and near the surface.

  Specifications for the GPS sonde were developed in collaboration with NOAA and the German Aerospace Research Establishment, both of which funded, with NCAR, the sonde's development. UCAR has licensed the technology to Vaisala for sonde manufacture since 1996.

Environmental and Societal Impacts Group

• La Niņa Summit: Review of the Causes and Consequences of Cold Events

  This event was convened by Michael Glantz in Boulder, Colorado, from 15-17 July 1998. The purpose for convening such a meeting was to identify what is known, what is not known, and what societies need to know about cold events in order to forecast their onset, growth, and decay several months in advance and to prepare for their societal impacts. (The terms "La Niņa" and "cold event" are used interchangeably.) An executive summary of the deliberations and presentations of the participants went on line in late September 1998 at http://www.dir.ucar.edu/esig/lanina/exec_summ.html. The full report has been on line since mid-October 1998. Both reports are available in hard copy. The workshop was supported by the United Nations University (Tokyo), NCAR, the UN Environment Programme (UNEP), and the NSF. The workshop was coordinated by an ESIG team led by D. Jan Stewart. Stewart was assisted by Baat Enosh (University of Colorado-Boulder), Ben Rasmussen (Carleton College), and Hanna Gilbert (University of Colorado-Boulder).

• Nested Regional Climate Change Scenario

  Linda Mearns, along with Larry McDaniel, Elena Tsvetsinskaya, Theo Mavromatis (with William Easterling at Penn State University and Cynthia Hays at the University of Nebraska), completed work on a four-year NIGEC (National Institute for Global Environmental Change) project, "Development of a Nested Regional Climate Change Scenario with an Application to Crop Models." The project involved regional climate modeling with RegCM2 by Filippo Giorgi and Christine Shields (CGD), detailed climate model evaluation, and application to crop models. Each stage of the work focused on some kind of uncertainty analysis: of (1) spatial scale of climate change, (2) the type of downscaling technique, (3) effect of scale differences on agricultural impacts, and (4) choice of impact model. Numerous publications have resulted from this work (see FY97 and FY98 publications).

  (1) A high-resolution climate change scenario (control and doubled carbon dioxide) was formed using the RegCM2 at 50 km grid point spacing of the western two-thirds of the United States. A coarse resolution scenario was formed from the output of the CSIRO general circulation model, which provided the boundary conditions for the regional model runs.

  (2) The climate change scenario from the regional model was compared with a semi-empirical downscaling (SDS) method. The comparison indicates that the climate changes in the RegCM2 are more pronounced than those of the SDS. Also, different directions of change in precipitation were found between the two methods.

  (3) We apply the two different scales of climate change scenarios to the EPIC corn, wheat, and soybean crop models for the GCM grid boxes in the central Great Plains. We found that the different scale scenarios produced substantial differences in the impacts of climate change on these agricultural crops (see figures).

  

  (4) We went on to compare the results from EPIC corn and wheat models for parts of the Great Plains with those of the CERES corn and wheat models. We found that the CERES model produced very different changes in yield from those of the EPIC model (see figures).

Advanced Study Program

• The Thompson Lecture Series
  A new series of visits and lectures was established to foster interaction between prominent scientists and the postdoctoral fellows and other junior scientists at NCAR. Three visitors were brought to NCAR under this program for meetings and discussions focussed on the research of the ASP postdoctoral fellows.

• Geophysical Turbulence Program(GTP)
  FY-98 was a busy year for GTP, which hosted two workshops and numerous scientific visitors.

  An international symposium on Developments in Geophysical Turbulence was sponsored by GTP, The International Union for Theoretical and Applied Mechanics, The International Association of Meteorology and Atmospheric Sciences of the International Union of Geodesy and Geophysics, and coordinated by Robert Kerr of NCAR and Yoshifumi Kimura of Nagoya University, Japan. Theworkshop brought together a broad spectrum of scientists to discuss turbulence modeling, statistics of small scales and coherent structures, convective turbulence, stratified turbulence, and historical developments.

  The second workshop on the topic of "Observations, Experiments and LES-A Triad for Geophysical Turbulence Studies" was hosted by GTP in August 1998 and coordinated by Donald Lenschow (MMM/ATD), Jackson Herring (MMM), Mary Barth (MMM/ACD), Chin-Hoh Moeng (MMM), Peter Sullivan (MMM), Bjorn Stevens (MMM), William Large (CGD) and Steve Oncley (ATD) locally as well as Robert Weller (WHOI), John Wyngaard (Penn State University), Bruce Albrecht (U. Miami), K. R. Sreenivasan (Yale), and Jim McWilliams (UCLA).

  In accordance with recent recommendations by the ASP review panel the university members of the organizing committee played a major role in planning and coordinating the workshop. Fifty-seven scientists representing 27 institutions from the U.S. and 3 other countries attended.

Educational Highlights

• Students from the states of Colorado and Wyoming entered 68 projects into the Colorado Computational Science Fair. Group and individual projects were submitted in the areas of Computational Science and Information Technology. Two computational science projects were sent to the national Adventures in Supercomputing Expo. The "Recursive Topographic Cost Analysis Project" from George Washington High School in Denver took first place in the Advanced Mathematics Category.

• Charlie Knight (MMM), Rajul Pandya (ASP) and Kevin Petty (ASP) were awarded the 1998 UCAR Outstanding Performance Award for Education for their work in support of Project LEARN.