Goal Area: Weather and Climate Information for Improved National Security and Public Safety
Motivation
NCAR has developed several technologies, mostly model–based, over the past decade that are focused on Department of Defense and National Security needs. This work began well before the September 11, 2001 attack on the U.S. but has increased dramatically since that event. Much of this work has been focused on coupled–model concepts, model–based climatologies for data sparse regions and on atmospheric scales ranging from neighborhood to mesoscale. By necessity urban areas, particularly those with high–valued national assets have received much attention. This work provides an excellent scientific and technical basis for strategically planning NCAR research and development over the next five years.
Additionally, requirements for research and development in this area of investigation continue to grow rapidly. As stakeholders gradually incorporate advanced weather and climate-based decision systems, they continually see new application areas. It is imperative that this science base and its associated technologies be pushed even farther in the areas of a) plume modeling in urban areas with a focus on street canyons and effects of flow around buildings, b) overall modeling of the flow regimes and atmospheric conditions in a metropolitan area down to street scale, c) coupling of atmospheric models to other sector models such as agriculture and public health, and d) production of regional climatologies to provide planning information to decision makers with regard to the effect of climate change on their region.
Near-Term Objectives
Development of Customized NWP Solutions (2009–2013)
MM5 24-h forecast of winds (blue vectors), and plumes
from three sources over Greece. The red lines show
observed wind direction. Such predictions were provided
to support the counter-terrorism efforts at the Athens Olympics
RAL has developed and deployed very high fidelity, computer–based weather analysis and forecasting systems for many applications worldwide. For example, new weather–prediction systems have brought the Army Test Ranges and Proving Grounds into the 21st century, in terms of weather services. The improved weather information for test planning has saved tax payers millions of dollars. Numerous other domestic projects include providing general weather support for Navy missile launches in California and Hawaii, and for potential Space Shuttle landings at alternate sites. Other such systems focus on urban areas and urban impacts on the weather. In contrast to most NWP models that don't recognize the existence of cities in a meaningful way, recent RAL models resolve the large–scale effects of the cities, and some even represent the complex winds in street canyons. In addition to these domestic applications, weather systems are being deployed worldwide to support special missions. For example, they have been used by the National Ground Intelligence Center during Operations Enduring Freedom and Iraqi Freedom to assess the consequences of the potential release of hazardous airborne material. Similar models have also been employed for counter-terrorism support by the Defense Threat Reduction Agency at the Salt Lake City, Athens, and Torino Olympics. This work will continue, meeting the needs of a growing number of sponsors with specialized needs.
Actions:
2009–2011 – Continue to explore new ways in which WRF and other models can be applied to produce specialized forecasts on short-range to seasonal time scales.
Targeted Sponsors: Sponsors include those for which RAL already provides specialized forecasts, as well as those in the energy, water–resources, and air–quality industries.
Anticipated Collaborators: Developers of the global atmosphere–ocean modeling software that is needed for downscaling of inter–seasonal predictions.
Specific Measurements of Success: The ability of user–centric metrics of forecast accuracy to show superiority over those associated with forecasts from national weather services and private venders.
Mesoscale Climatologies (2009–2013)
Mesoscale analyses of current climates can be used for many purposes, including optimal siting of wind–energy farms and airports, calculating the most probable direction of the transport of hazardous material at some future date and time, and scheduling the time and season for events that require specific meteorological conditions. To construct such climatologies for the many areas of the world where there are few routine four–dimensional (4D) observations of the atmosphere, RAL has developed a Climate Four–Dimensional Data Assimilation (Climate–FDDA) system that uses WRF to downscale present–day climates from archived global analyses.
The Climate–FDDA system is able to generate a 4D description of the diurnal and seasonal evolution of regional atmospheric processes, with a focus on the boundary layer. Unlike point measurements, the gridded fields define coherent multi–dimensional realizations of complete physical systems. Not only does the Climate–FDDA system define mean values of variables as a function of season and time of day, extremes are also estimated, and example days are produced.
As an example of one Climate–FDDA application, the figure below shows a map of the probability that 30–m above–ground–level (AGL) winds will exceed 10 m s–1 in the month of February in southern Europe, where such an analysis would be valuable for wind–energy prospecting. These statistics are based on a 20–year downscaling from the NCEP–NCAR Reanalysis Project global data set.
C-FDDA-generated map of the probability that the 30-m
AGL winds will exceed 10 m/s in the month of February, for
southern Europe (based on a 20 year data-assimilation period)
Actions:
Targeted Sponsors: RAL currently has several sponsors, including three groups in France, the U.S. National Ground Intelligence Center, the DoD High Performance Computing Modernization Office, The Army Test and Evaluation Command (ATEC), and the Missile Defense Agency (MDA).
Anticipated Collaborators: Univ. of Colorado and other experts on Self–Organizing Maps.
Specific Measurements of Success: Demand for climatologies and the frequency of use for long-term physical–process studies.
Understanding, Modeling, and Forecasting Urban Atmospheres (2009–2013)
Lidar–derived winds over Washington, D.C. Wind vectors for approximately 1424 EDT 7 May 2004, on a horizontal surface about 25 m AGL. Every second vector is shown in the image. The red line corresponding to the 3 m/s isotach, and is taken as the leading edge of the gust front.
Atmospheric processes are influenced by urban complexes over a wide range of scales. The larger metropolitan area produces an aggregate effect on the mesoscale atmosphere, and RAL is developing improved urban–canopy parameterizations for representing the bulk dynamic and thermodynamic effects of buildings for use in the community version of WRF. On the smaller scales of neighborhoods, a variety of models and measurement systems have been used to define boundary–layer structure. For example, Doppler lidar radial–wind data are ingested into a four-dimensional variational data–assimilation (4DVAR) system to produce a three–dimensional dynamically consistent analysis of winds every five minutes, with a horizontal grid increment of 100 m. The figure above shows an example of the analyzed lidar winds over Washington, DC, for a situation where a gust front is passing over the Potomac River from the northeast.
On even–smaller scales, other models calculate street–level winds in urban street canyons. For example, the NCAR EuLag model is a building–aware Large–Eddy Simulation (LES) model that is being used for a variety of projects that require urban winds on the scale of meters. The EuLag LES model has been coupled with the WRF model to produce seamless multi–scale simulations of the mesoscale to the street–canyon scale. For rapid operational generation of building–aware winds in urban areas, RAL has collaborated with the Los Alamos National Laboratory (LANL) to adapt their QUIC-URB software package that algorithmically calculates the effects of buildings on the wind field. The QUIC model is a component of the Urban–Shield system that RAL has deployed in Washington, DC for the calculation of the transport of hazardous material in the city.
Research is aimed at developing an improved general understanding of atmospheric boundary–layer processes and their parameterization. Scientific and technological contributions have been based on the analysis of field–program data and on modeling studies.
Actions:
2009–2013 – We must develop ways of aligning the disparate capabilities that we already have, related to urban atmospheric modeling, and redirect them so that they satisfy the needs of a broader community, encompassing air quality, architectural climatology, public health, emergency response, etc. A first step will be to better coordinate the varied urban–related activities within NCAR.
Targeted Sponsors: We have discussed our urban–scale modeling capabilities with state and local organizations, and will continue to do so. There may also be opportunities related to traditional air–quality sponsors, and non–ATM entities in NSF. Most U.S. agencies involved in homeland security are likely to be very interested in these advances, including the Defense Advanced Research Projects Agency (DARPA), the Department of Homeland Security (DHS), the Defense Threat Reduction Agency (DTRA), and the Pentagon Force Protection Agency (PFPA).
Anticipated Collaborators: LANL should continue to be an important partner in urban modeling at fine scales, especially since they are the principal urban modelers for DHS.
Specific Measurements of Success: Making advances in this area would create new opportunities with our targeted sponsors, which would result in new tasking from them. The development of better organization and coordination of urban activities within NCAR will be an early metric of success. In addition, our ability to reach out to other groups and to garner support from non–traditional sources will be another measure.
Modeling Plumes of Hazardous Material (2009–2013)
The mesoscale and urban–scale meteorological modeling activities provide essential input data for multi–scale capabilities that track the movement of plumes of hazardous material. Numerous Department of Defense and civilian (e.g., Environmental Protection Agency) plume models are employed, depending on the need, where the models have been verified using urban field–program data. As with the LES models, there are quickly executing plume models, such as LANL's QUIC-Plume model that are designed for operational applications such as the previously mentioned Urban Shield project, and there are more complex models that are used for research and for verifying the fast models. One aspect of this work area is the development of methods for characterizing the source of a plume (size, time of release, location, etc.) based on downstream measurements, thus allowing for intervention if material continues to be released. The figure below shows a plume simulation for Denver, based on the QUIC-Plume software.
A simulated plume of hazardous material hypothetically released into the atmosphere in Denver, Colorado about five minutes before the time of the image. Building–aware winds were calculated using a large–scale wind as input to the LANL QUIC–URB software package. The transport and diffusion of the plume were calculated by LANL's QUIC–PLUME software. The opaque red color indicates the highest concentrations and the translucent green and blue show lesser values.
Actions:
Targeted Sponsors: All of our DoD sponsors have continuing interest in seeing progress in this area (ATEC, DTRA, DHS, DARPA, MDA, and PFPA).
Anticipated Collaborators: Our partnership with LANL continues to be important.
Specific Measurements of Success: Success will be measured by significant increase in accuracy of plumes, measured concentrations of contaminants and the ability to extend this technology to address traditional air-quality problems.
Mesoscale Ensemble Prediction (2009–2013)
The decisions made by RAL sponsors, based on mesoscale–model forecasts, can be improved through the availability of probabilistic information. Thus, mesoscale ensemble prediction systems are being developed, and prototype systems are now in operational use. For example, the following figure shows a montage of displays for a nested WRF ensemble system that is being run operationally for a sponsor over western Utah. It is a 30–member ensemble system, with a fine–grid increment of 3.3 km, which generates four 48–h forecasts per day. A challenging and exciting aspect of this effort is working with forecast users to help them better incorporate stochastic information into their decisions.
Three displays of statistical wind and temperature information on the fine grid (3.3 km increment) from a forecast from a triply nested 30–member ensemble system centered over western Utah. Four 48–h forecasts are produced per day.
Actions:
2009–2013 – Develop optimal methods for defining ensemble members and calibrating forecasts for specific user groups. Work with ensemble–forecast users to help them develop improved methods for integrating the statistical information into their decision–making systems
Targeted Sponsors: This is an area of high interest to all of our current DoD sponsors, but we see this effort as being relevant to any end user needing forecast uncertainty information as part of their decision–making process. One example sponsor category is the wind energy sector, but there are many more.
Anticipated Collaborators: Various Programs in RAL, the NCAR Data Assimilation Research Testbed (DART), the National Center for Environmental Prediction, and others.
Specific Measurements of Success: Ability to educate the users of the statistical information about its merits in decision–making processes and ability to optimize forecast calibration techniques.
Frontiers
Climate Services (2009–2013)
There is widespread recognition that the time is ripe to develop a national effort in the provision of climate services. Such services should provide information to the nation and to the world to assist in understanding, anticipating, and responding to climate, climate change, and climate variability and their impacts and implications.
Stakeholders in the areas of water resources, environmental quality, surface transportation, air transportation, manufacturing, public health, disaster–preparedness and relief, national security, and construction would benefit greatly from information about the next season's weather anomalies and about longer–range climate trends. This Climate Services Frontier proposed here for NCAR involves the development of knowledge and modeling techniques that will allow information about seasonal and longer climate trends to be customized for, and made available to, a wide range of stakeholders.
Three areas of endeavor are currently being considered in beginning this large effort. The first work area described below focuses on the health impacts of climate and weather, and how we can better serve that community. The second area recognizes that one of the most important horizons for local planning for weather and climate impacts is seasonal, and identifies and applies practical seasonal–prediction downscaling methods. The third work area aims to develop and provide community tools that will make it more feasible for universities and other organizations to provide climate services to state and local constituents by developing and offering a software toolkit and best–practices recommendations for generating and interpreting model guidance.
Climate, Weather, and Health
Human health is inextricably connected to weather and climate through complex interactions among natural and social systems. Climate change has been recognized by the WHO as a significant threat to public health, and it will have major impacts through natural disasters (heat waves, droughts, and other extreme events), transmission of infectious diseases such as dengue fever and Lyme disease, and the increased adverse affects of air pollution. Initial activities in RAL are motivated by the fact that current and recent–history information on the near–surface micro–environment are needed for defining regions that are suitable for the development of infectious diseases or for the proliferation of disease vectors. The WRF–based Real–Time Four–Dimensional Data Assimilation System and the High–Resolution Land Data Assimilation System are ideal tools for defining these micro–environments. Together with GIS, these models can provide mappings about imminent climate and weather–related disease threats. Proposals have been submitted, related to application of these methods to human and agricultural diseases. Existing work on other climate–related health threats, such as heat waves, will be expanded, and coordinated with the urban–weather activities at NCAR.
Actions:
2009–2013 – We are submitting proposals to a broad set of potential sponsors, and are also developing demonstration and proof–of–concept cases.
Targeted Sponsors: We were recently successful in obtaining sponsorship from Google. We have submitted proposals to NASA and NSF for exploring coupled infectious–disease modeling. We have support from the Inter–America Institute to explore urban vulnerability to the health impacts of heat/cold waves and air pollution, and are in the process of submitting to an environmental RFP from the Centers for Disease Control and Prevention to further investigate heat vulnerability.
Anticipated Collaborators: We anticipate working closely with groups such as the CDC, Colorado State University, University of Texas, Ecuadorian public health staff and epidemiologists, Mexican medical entomologists, the U.S. Department of Agriculture, Florida Departments of Health and Agriculture, the University of Ghana, Columbia University, and others.
Specific Measurements of Success: The accuracy of these models in depicting future patterns of heat–related vulnerability and infectious–disease transmission, and the value of the results to stakeholder communities, will be used to measure the success of this initiative.
Climate Services Using Practical Seasonal–Forecast Downscaling Methods
Most stakeholders in the U.S. are presently limited to the use of coarse–resolution, publicly available anomaly maps, in spite of the fact that they need the guidance for planning on the county and city level. A wide variety of statistical–downscaling methods is available for use, and the methods are scientifically relatively mature, but their application to meet specific needs (e.g., estimating river discharge) often requires customization. Two parallel activities will advance the use of seasonal–forecast downscaling to meet societal needs. One is to review the available statistical methods, and determine which ones are most adaptable for meeting the requirements of a variety of stakeholders. The other is to make progress on the development, testing, and evaluation of dynamical–downscaling methods, where limited–area models are run with boundary conditions provided by global seasonal forecasts. Direct interaction with stakeholders at each stage of the development and deployment cycle will be crucial.
Actions:
Targeted Sponsors: Organizations working in the areas of water resources, environmental quality, surface transportation, air transportation, manufacturing, public health, disaster–preparedness and relief, and construction.
Anticipated Collaborators: ISSE staff, and climate modelers in the Climate and Global Dynamics Division (CGD).
Specific Measurements of Success: Success will be measured by our ability to meet the needs of the collaborating stakeholders, and by objective verification of the downscaled seasonal forecasts.
Software Toolkit for Regional Climate Projections
There is great interest by domestic and foreign governments, industries, and humanitarian organizations in obtaining regionally specific information on the impact of climate change on various infrastructures and the environment. One initial such effort in RAL has been in collaboration with the Atmospheric Chemistry Division (ACD), where WRF regional–climate simulations for Houston have been performed to evaluate the impact of climate change on local air quality. The previously described efforts at regional downscaling applied to current climates have been an essential exercise for verifying the ability of the modeling tools to replicate current climates worldwide – a first step before applying a model for future–climate downscaling in a new location.
Actions:
Targeted Sponsors: Many UCAR–Member universities have interests or ongoing activities in modeling the regional impacts of climate change. These tools will provide a means for NCAR and the university community to more effectively meet the needs of local, National and global stakeholder groups that require actionable information.
Anticipated Collaborators: This effort will provide opportunities for collaboration with other groups at NCAR such as ACD and CGD.
Specific Measurements of Success: Success will be measured by the ease with which users can employ the software and data to address practical problems.