On the Web at http://dust.ess.uci.edu/prp/prp_itr/prp_itr.pdf NSF Information Technology Research (ITR) Proposal Submitted: February 26, 2004 Last modified: Tuesday 13th April, 200417:50 Next Round Due: December 2004?

ITR-(ASE+NHS)-(dmc+sim): Interactive Mesoscale Forecasts, Visualization, and Environmental Planning
Dr. Charles S. Zender Department of Earth System Science University of California at Irvine Dr. Renato B. Pajarola Department of Computer Science University of California at Irvine

1. Finalize Hardware/Support Budget (Pajarola, Rehbaum, Ross, Wessel) (done) (a) Visualization Cluster (VC) (Davison, Evans, Pajarola, Wessel) (done) (b) Forecast Cluster (FC) (Evans, Wessel) (done) (c) Network infrastructure (Davison, Hildebrand, Wessel) (done) (budget is missing $2535 needed for fiber-pulling labor) (d) Visualization Gizmos (Pajarola) (done) (e) Text on NACS MPC support and sysadmin (Wessel) (done) 2. Cal-(IT)2 Coordination (a) Ask Yee for visualization/student space (Pajarola) (done) (b) Pursue OptIPuter node status (budget implications?) (Zender) (done) (c) Include paragraph on applicability to HiPerWall (Zender) (not done) 3. Letters of Support/Collaboration/Senior Personnel Inquiries (Zender) (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l) Abdibekov (IoM, Aral Sea) (received) Chavez (USGS, Mojave) (received) Dabdub (UCI MAE, Mojave) (will participate next round) Frost (SDSU, Kazakhstan, Mojave) (may participate next round) Glantz (NCAR, Kazakhstan) (declined) Goulden (ESS, Mojave) (no response) Kuester (Cal-(IT)2 /HiPerWall) (misplaced) Mehrotra (Cal-(IT)2 /RESCUE) (received) Neff (CU/USGS, Mojave) (received) Purvis (Claremont Colleges, Aral Sea) (received) Randerson (ESS, Mojave) (no response) Reynolds (USGS, Mojave) (received)

4. Proposal submitted Feb. 26, 2004 (done)


The next round of ITR proposals is due around December 2004. This proposal is so strong already that I plan to revise and re-submit (again!) if we are declined this round. Here are my initial thoughts on how to make a stronger proposal: 1. Strenghthen Mojave Desert (MD) ROI. Strengthen or remove Aral Sea (AS). (a) (b) (c) (d) (e) (f) Drop Aral Sea ROI if Reviewers' comments too hard to address Ramp-up AS without de-emphasizing MD in years 3­5 How to entrain Glantz for AS? Solicit School of Social Ecology collaborator for AS Entrain Israeli group (Rudich?) Kazakhstani visualization through OptIPuter (Frost?)

2. Describe specific regional projects and environmental planning in text and/or letters of support. (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (l) (m) CDHS (respiratory studies, valley fever) CHP (likely dust/fog road closure locations/seasons) DHS (NBC WMD plume scenarios) DWP (Airborne contaminants) FAA (Santa Ana dust/smoke plume morphology) JPL (complement coastal ocean forecasts) NPS (Mojave stresses/drought, sources of visibility reduction) RESCUE (prototype dust/smoke response plan?) SC4 (standardize scenarios for future MD simulations) SCCOOS (cause of blooms) UCI MAE (SCAQMD model eastern boundary eddy dust flux) UNCCD (land use/soil treatments to reduce desertification) USGS (deposition characterization, valley fever habitat)

3. Strengthen forecast/visualization/application links (a) Joerg Meyer (UCI EECS) will Co-PI (b) Donald Dabdub (UCI MAE, Mojave) will Co-PI (c) Provide reviewer web access to prototype dust storm rendering 4. Strengthen project management (a) 0.5 FTE project scientist/web evangelist

ii


Project Summary. We propose to improve forecasts and real and accelerated-time visualizations of mesoscale environmental change on multiple timescales to provide infrastructure for scientific, civil, and security-related decisions involving weather and aerosol dispersal. To accomplish this, we will converge existing programs of scientific research on boundary layer aerosol prediction with research on environmental visualization and immersive forecast interaction. Our approach is twofold: 1. Create turn-key forecasting capability for particle entrainment, dispersal, and deposition using existing aerosol and mesoscale forecast models (ASE, NHS). 2. Use and improve visualization technology to realistically render the boundary layer forecasts guided by decision-makers' assumptions of natural and anthropogenic land-use disturbance, climate change, and point source emissions (dmc,sim). We call our integration of interactive mesoscale forecasts with visualization capabilities critical for decision-makers the Laboratory for Environmental Planning (LEP). LEP links three sophisticated information technology layers into a flexible environmental planning and decision-making system. LEP's core is the (existing) Weather Research and Forecast (WRF) model with aerosol module enhancements. WRF will simulate Regions of Interest (ROIs), areas of high scientific, civil, and security interest that have significant and variable particulate concentrations on multiple timescales. The second layer is the high-performance visualization front-end for rendering the real and accelerated-time simulations from user specified projections, vantage points, and realistic illumination (ASE). The third layer controls the forecast boundary conditions with interactive devices for decision-makers to feed-back into the simulation and modify the forecast in real time (dmc,sim). LEP initially targets arid regions for their relative visual simplicity, single aerosol type (soil mineral dust), and impacts on downwind populations (NHS). The LEP team has the requisite expertise and data to simulate and evaluate two complementary ROIs that span the spectrum of landscape disturbance from natural to heavily disturbed. The Mojave Desert ROI is a relatively undisturbed, periodically active particulate source upwind of densely populated areas (Los Angeles, Las Vegas). The other ROI is the Aral Sea, a heavily disturbed, persistent source upwind of sparsely populated areas. Both ROIs are susceptible to rapid activation and, potentially, mitigation. Moreover, LEP will visualize any user-prescribed passive particulate emissions, e.g., wildfire smoke (during Santa Anas) and radiological, biological, chemical explosions (i.e., NBC WMD) (NHS) and will be useful in emergency response planning. Intellectual Merit: 1. Inert tracer and desert aerosol dispersal by mesoscale processes (e.g., gust fronts, dry convection, topography) is little studied and will improve our understanding of fast time-scale visibility and biogeochemical fluxes, and security hazards (ASE, NHS, dmc, sim). 2. Comparison of satellite and in situ observations of dust aerosol to rendered forecast data will improve our understanding and model representation of the relative roles of topography, geomorphology, and disturbance in aerosol emission, transport, and deposition (ASE, sim). 3. Applying advanced rendering techniques to mesoscale forecasts will lead to new and more efficient environmental visualization algorithms (ASE, sim). 4. The ROIs to be hindcast/forecast and monitored are conducive to science-guided planning and remediation decisions (NHS, dmc, sim). Broader Impacts: LEP 1. Enhances research and teaching infrastructure with its convergent integration of physically based-forecasting and visualization (ASE). 2. Benefits society by enabling interactive exploration of the outcomes of land use change and disturbance scenarios on wind erosion and visibility, and the prediction of newly vulnerable landscapes in arid environments undergoing climate change (ASE). 3. Increases regional security by improving simulation of (and potentially response to) dispersal of specified inert particulate tracers (e.g., NBC WMD) (NHS).


ITR-(ASE+NHS)-(dmc+sim): Interactive Mesoscale Forecasts, Visualization, and Environmental Planning 1 Preamble

The south west United States (SW US) experiences adverse impacts from fast timescale changes in an arid environment. For example, a multi-day dust storm from the Mojave Desert blanketed Las Vegas, Nevada, April 15­17, 2002. It kept inhabitants indoors and inactive, closed construction sites, and shut down air travel. In Fall 2003, metropolitan Southern California was downwind of the dry desert regions during an intense and prolonged Santa Ana wind event (Raphael, 2003). Numerous natural and anthropogenic fires sent clouds of smoke, ash, dust, and pollutants over much of greater Los Angeles and San Diego. These dramatic events illustrate how high variability, mesoscale weather patterns over desert regions affect millions of American citizens each year (Pappagianis and Einstein, 1978; Prospero, 1999; Kolivras et al., 2001; Zender and Talamantes, 2004). Recently planners are more concerned that domestic or international terrorists release nuclear, biological, and chemical (NBC) weapons of mass destruction (WMD) in the US. Planning for atmospheric natural hazards (dust storms, wildfire smoke) at first seems unrelated to preparing for security and WMD hazards. However, both natural and anthropogenic hazards are isomorphic computer modeling and visualization problems insofar as they comprise following the emission, transport, and deposition of passive aerosols (i.e., dust, soot, radionuclides) or gases (Sehmel, 1980; Guelle et al., 1998). Concomitant with the increasing awareness of the agricultural, climatic, economic, and respiratory impact of dust storms (e.g., Glantz, 1994; Youlin et al., 2002) has been a dramatic increase in our understanding and ability to detect and to predict arid region aerosol emission and transport in recent years (e.g., Shao and Leslie, 1997; Marticorena et al., 1997; Wald et al., 1998; Zender et al., 2003a). This forecast technology has been developed for scientific (e.g., Collins et al., 2001), civil (e.g., Claiborn et al., 1998), and military (e.g., Hogan and Brody, 1993; van Donk et al., 2003) applications. Use of these mesoscale aerosol models in SW US civil planning and homeland security has not kept pace. To help remedy this, we will develop the Laboratory for Environmental Planning (LEP). LEP will produce fast-timescale forecasts of SW US dust and other aerosols, similar to efforts underway to predict intense Gobi and African dust storms (e.g., Shao et al., 2003). Moreover, LEP will integrate scientific forecasting with advanced visualization/rendering and decisionmaking capabilities. The time seems right to integrate recent advances in weather and aerosol forecast technology with advances in interactive visualization and rendering environments so that planners may take full advantage of fast yet quantitative environmental simulations. Let us begin to make civil planners, park rangers, epidemiologists, firefighters, aviation officials, and homeland security personnel familiar with the potential of mesoscale weather and aerosol forecasts to help them understand past events and help them plan for future events. We will help close the forecast-technology-planning gap with technology friendlier to use, easier to understand, and interactively configurable for environmental planning. The decision-makers will control these quantitatively rigorous forecasts and hindcasts in a highly realistic visualization space. Our Information Technology Research (ITR) for National Priorities project would develop a physically based, scientifically evaluated, mesoscale environmental virtual reality and visualization facility. 1


2

Introduction

We propose to improve forecasts and real and accelerated-time visualizations of mesoscale environmental change on multiple timescales to provide infrastructure for scientific, civil, and securityrelated decisions involving weather and aerosol dispersal. To accomplish this, we will converge existing programs of scientific research on boundary layer aerosol prediction with research on environmental visualization and immersive forecast interaction. Our approach is twofold: 1. Create turn-key forecasting capability for particle entrainment, dispersal, and deposition using existing aerosol and mesoscale forecast models. 2. Use and improve visualization technology to realistically render the boundary layer forecasts guided by decision-makers' assumptions of natural and anthropogenic land-use disturbance, climate change, and point source emissions. We call our integration of interactive mesoscale forecasts with visualization capabilities critical for decisionmakers the Laboratory for Environmental Planning (LEP). LEP links three sophisticated information technology layers into a flexible environmental planning and decision-making system. LEP's core is the (existing) Weather Research and Forecast (WRF) model with aerosol module enhancements. WRF will simulate Regions of Interest (ROIs), areas of high scientific, civil, and security interest that have significant and variable particulate concentrations on multiple timescales. The second layer is the high-performance visualization front-end for rendering the real and accelerated-time simulations from user specified projections, vantage points, and realistic illumination. The third layer controls the forecast boundary conditions with interactive devices for decision-makers to feed-back into the simulation and modify the forecast in real time. LEP initially targets arid regions for their relative visual simplicity, single aerosol type (soil mineral dust), and impacts on downwind populations. 2.1 Personnel Directly Involved with LEP Table 1 lists the primary researchers affiliated with LEP (see supplementary letters of collaboration or biographical sketches). This non-exhaustive list highlights unique and complementary interests of the research groups. The participants are "matrixed" into overlapping fields of interest. The LEP team has the requisite expertise and data to simulate and evaluate two complementary ROIs that span the spectrum of landscape disturbance from natural to heavily disturbed. The Mojave Desert ROI is a relatively undisturbed, periodically active particulate source upwind of densely populated areas (Los Angeles, Las Vegas) (Bach et al., 1996). The other ROI is the Aral Sea, a heavily disturbed, persistent source upwind of sparsely populated areas (Youlin et al., 2002). Both ROIs are susceptible to rapid activation and, potentially, mitigation (Glantz, 1994). Moreover, LEP will visualize any user-prescribed passive particulate emissions, e.g., wildfire smoke (during Santa Anas) and radiological, biological, chemical explosions (i.e., NBC WMD) and will be useful in emergency response planning. There is no institution in the US better qualified than UCI to integrate research on arid region erosion, hydrology, and climate change can with research on real-time environmental rendering and simulation interaction. The science arising from our studies of arid region behavior can be leveraged into useful decision-making tools when combined with effective, intuitive visualization and interaction methods. In terms of computing power, the LEP requires two relatively inexpensive, of the shelf clusters of about a dozen PCs each. It is the interdisciplinary collaboration of personnel and entrainment of outside decision-makers that make LEP unique. LEP fills a niche, allowing faculty, post-graduates and students in Earth System Science and Electrical Engineering and Computer Science to integrate their disciplines while learning to apply their scientific spe2


cialties with environmental planners working on real-world problems. The first proof-of-concept ROI (the Mojave Desert) is practically in our backyard and we are highly motivated to produce accurate forecasts and stunning visualizations of its behavior now and under forcing from natural (e.g., drought) and anthropogenic (e.g., greenhouse warming) scenarios. Although primarily a simulation and planning tool for arid and semi-arid regions, LEP will produce the mesoscale hindcasts and forecasts necessary to conduct original scientific research on which mesoscale processes control fast-timescale biogeochemical aeolian fluxes. Over the next five years, LEP will allow UCI Earth System Science (ESS) researchers to bridge the scale of their studies from global change to regional and mesoscale change for the first time. Thus LEP will serve as a focus point for ESS's increasing interest in modeling and measuring Southern Californian climate variability and change.

3

Background

3.1 Results from Prior NSF Funding Zender is a Co-PI on ATM-0214430, "Collaborative Proposal: Using Measurements from the Columbia Plateau Eolian System to Improve Global-Scale Models of Mineral-Dust Aerosols", 8/1/2002­7/30/2005. This project has resulted so far in four national meeting presentations with manuscripts in preparation (Sweeney et al., 2002, 2003b,a; Zender et al., 2003b). Our manuscript studies the range of uncertainty in LGM dust mass and radiative budgets to uncertainty in vegetation reconstruction. We show that a significant fraction of the observed LGM increase in Pacific Ocean dust deposition is attributable to vegetation change. Our paper in press (Grini and Zender, 2004) explains how the twin processes of saltation and sandblasting (SS) relate to loess formation. These SS physics were implemented in DEAD and will be used in the LEP proposal for accurate simulation of small (dust) to large (loess and sand) particle entrainment which is especially important to mesoscale visibility. Zender is PI on ATM-0321380 "Acquisition of an Earth System Modeling Facility (ESMF) for Coupled Climate, Chemistry, and Biogeochemistry Studies". After negotiating the best price supercomputer through an open bid competition in summer 2003, we awarded IBM the ESMF contract in October 2003. The ESMF opened to early users in early February 2004. It is currently undergoing final acceptance testing by UCI and configuration by IBM prior to being fully devoted to coupled climate studies. As discussed in the Facilities section, this ITR may use the ESMF as a source of real-time global (as opposed to mesoscale) coupled climate model simulation data should entrepreneurial members of the visualization team become interested or entrained in the problem of global visualizations. Global visualizations, however, are beyond the scope of this proposal and no funding is presently available or sought to pursue them. Pajarola has not received NSF funding for projects related to this ITR proposal. 3.2 Overview Mineral dust aerosol plays many roles in the environment, including visibility impairment and air quality (Prospero, 1999; Bian and Zender, 2003), radiative forcing of climate (Tegen et al., 1996; Sokolik et al., 2001), and transport of biogeochemical nutrients to remote ecosystems (Martin and Fitzwater, 1988; Moore et al., 2002). Forecasting dust movement is difficult because of the great spatial and temporal variations in its emissions mechanisms and constraints (Balkanski et al., 1996; Gillette, 1999; Prospero et al., 2002). Recent research has greatly improved our ability to explain the global cycle of mineral dust distribution on climate timescales (e.g., Tegen and Fung, 1994; 3


Ginoux et al., 2001; Zender et al., 2003a). However, our ability to accurately forecast dust activity at the meso- and synoptic scales ( 10­1000 km, timescale of hours to days) remains relatively weak due to inadequate boundary data sets (e.g., Marticorena et al., 1997), and incomplete understanding of the complex interactions between surface meteorology and soil structure which result in dust emissions (e.g., Marticorena and Bergametti, 1995; Shao et al., 1996; Alfaro and Gomes, 2001). We will investigate the interaction of mesoscale weather systems with dust entrainment and dust storm structure by using and improving our Dust Entrainment And Deposition (DEAD) model in the mesoscale Weather Research and Forecasting (WRF) model. The resulting mesoscale forecasting system offers novel opportunities to use high speed realistic environmental visualization technologies in scientifically and societally useful ways. The computationally intensive nature of weather forecasting (Hogan and Brody, 1993) and dust entrainment processes (Shao et al., 1996; Marticorena et al., 1997) makes high-speed computers necessary to simulate mesoscale events in real time. We will adapt a state-of-the-art environmental visualization system to the back end of the mesoscale weather and dust forecast model. We will use the forecast fields in real time to interactively visualize the boundary layer from user-selected perspectives and anthropogenic emissions sources. In arid and semi-arid environments, mineral dust aerosol is usually the dominant aerosol in terms of mass, visibility impairment, and temporal variability (e.g., Mbourou et al., 1997; Chin et al., 2002). Thus our single system will simultaneously forecast and render the most significant features of dynamically changing arid environments. Our goals are to 1. Enhance predictive skill and understanding of mesoscale natural and anthropogenic dust erodibility, emissions, and transport 2. Develop efficient physically based visualization technologies and systems for interactive arid environment simulation 3. Entrain environmental planners to advise development and use of LEP to maximize societal preparedness for mesoscale environmental change LEP's three components (scientific, visualization, and planning) are synergistic. To facilitate environmental planning by outside experts, about half of LEP is devoted to an interactive visualization and control system. We will render predicted aerosol (not only dust) plumes above synthetic surfaces constructed from geomorphic terrain processes such as upstream area, local slope, USGS land-use category, and MODIS surface imagery (Schaaf et al., 2002). We anticipate learning more about complex linkages between land-use and erodibility by visualizing different combinations of geophysical variables together. The scientific inference of land-use-turbulence-erosion linkages will drive the visualization development in part. Also driving the visualization research is the need to extract and visualize the scientifically meaningful data from the vast data stream generated in real time. LEP's primary non-scientific purpose, to train and assist in environmental planning, depends strongly on the accuracy, realism, and flexibility of the forecasts and visualization system. We expect that planners will exhort the modeling and visualization teams to produce their finest work possible because of the societal importance of their decisions. Planners, in turn, will be able to devise smarter strategies for urban growth/desert disturbance and weather response due to the flexibility and accuracy of LEP. We do not intend to develop an operational, 24 × 7, global dust forecast system responsive to aviation, civil, and/or military needs (such as the NOGAPS system at the Naval Research Lab) 4


(Hogan and Brody, 1993). LEP is a system where research forecasts and visualization inform and improve each other. When LEP forecasts a potentially hazardous aerosol event, we will consult with the UC Irvine Responding to Crises and Unexpected Events (RESCUE) Project on whether and how best to disseminate the information (see attached letter of support from Sharad Mehrotra). The central scientific areas that we will address are 1. Dust storm formation and structure; 2. Predictability; and 3. Response to environmental change. The visualization problems that we will address are 1. Multi-resolution visualization of dynamically changing environments; 2. Physically realistic rendering of arid terrain and vegetation in changing light conditions; 3. Optimal communication interfaces between forecast and rendering components. Our scientific and technical research plans to address these questions is given in Section 4. Appendices contain the Budget Justification, letters of support from collaborators, and a list of Acronyms and Abbreviations. 3.3 Scientific Visualization A real-time scientific visualization system powerful enough to allow interactive exploration of mesoscale environmental forecasts depends on efficient solutions and system integration of various components including: large-scale terrain rendering, accurate surface light backscattering, atmospheric attenuation and cloud simulation. In particular, multi-resolution level-of-detail (LOD) approaches as well as a parallel cluster-based visualization system are required to provide the necessary rendering power and scalability to high-resolution mesoscale atmospheric data sets. Below we briefly summarize state-of-the-art rendering algorithms and visualization systems in this context to pinpoint current limitations that will be addressed by research in this project. Efficient real-time terrain visualization algorithms and systems for displaying very large-scale grid-digital elevation models have mainly concentrated on bin-tree or quadtree multi-resolution triangulation methods such as (Lindstrom et al., 1996; Duchaineau et al., 1997; Pajarola, 1998b; Balmelli et al., 1998; Gerstner, 1999; Evans et al., 2001; Lindstrom and Pascucci, 2001; Pajarola, 2002) and patch-based rendering (Levenberg, 2002; Cignoni et al., 2003a,b). The common focus is to define an effective hierarchical multi-resolution data structure that allows quick, feature-adaptive and continuous LOD terrain surface extraction. This minimal-complexity terrain surface then allows fast polygonal rendering. The main focus of research in this area has concentrated on rendering large terrains locally on a single computer (node). The few proposed parallel terrain rendering approaches are aimed at shared-memory Symmetric Multi Processing (SMP) or Massively Parallel Processing (MPP) systems (Vezina and Robertson, 1991; Li et al., 1996; Cohen-Or et al., 1996) and do not take advantage of hardware accelerated polygonal rendering (Agranov and Gotsman, 1995; Li et al., 1996; Cohen-Or et al., 1996). Hence these previous approaches do directly adapt and scale well to cluster-based systems. Atmospheric attenuation is closely related to the low-albedo case in direct volume rendering (DVR) of color-and-opacity fields defined over a regular 3D grid of volume elements (voxels) (Kajiya and Von Herzen, 1984). Popular DVR approaches include raycasting (Levoy, 1990), shearwarping (Lacroute and Levoy, 1994), 3D texture mapping (Cabral et al., 1994), cell projection (Shirley and Tuchman, 1990) and splatting (Westover, 1990). Due to the excellent combination of rendering quality and performance (see also comparison in (Meissner et al., 2000)), and because of the direct analogy to particles1 we focus on splatting-based DVR in this project. In this context, much work has been invested into improving performance of footprint rasterization (e.g., Mueller et al., 1998, 1999a; Huang et al., 2000; Zwicker et al., 2001), occlusion culling (Mueller et al.,
1

which mainly determine atmospheric attenuation

5


¨ 1999b) and sparse volume traversal (Orchard and Moller, 2001). Common splatting methods suffer from a software rasterization of splat-footprints that limits better rendering performance. On the other hand, splatting is among the fastest parallel volume rendering methods (Li et al., 1997). However, development of cluster-parallel splatting algorithms is largely missing (Wittenbrink, 1998). Furthermore, real-time rendering of physically realistic atmospheric effects such as light attenuation and scattering, or clouds is an extremely difficult volume rendering problem (e.g., Nishita ¨ et al., 1996; Elinas and Sturzlinger, 2000; Harris and Lastra, 2001; Nishita and Dobashi, 2001) and has not been addressed in a larger scale nor extended to cluster-rendering. A major effort of this project is system development of the environmental visualization component. To date no similar visualization systems exist that link an interactive mesoscale weather forecast module with a real-time 3D rendering system. Commercial or semi-commercial systems as Vis5d (Hibbard and Santek, 1990) or Open Visualization Data Explorer (IBM Research, ) provide high-level API toolkits. However, they are not aimed at realistic rendering, real-time performance on large data sets (e.g. no LOD support) or integration with an interactive mesoscale weather forecast model. Recent developments such as (Riley et al., 2003, 2004) provide isolated technical solutions to individual tasks (i.e. fast realistic cloud rendering) but do not address overall system development and integration. Major system issues previously left out are: (1) The direct coupling of efficient multi-resolution terrain elevation models with ground surface bi-directional radiance distribution function (BRDF) estimates and down-welling flux information from the WRF model that allows for real-time physically realistic ground illumination. (2) Integration of simulated light scattering and attenuation data into multi-resolution volume rendering for physically realistic boundary layer visualization. (3) User input interaction that allows the specification and spatial placement of tracer emissions as well as land-use change as interactive feedback-loop to the WRF model. (4) Design and implementation of a client-server based environmental forecast simulation-visualization system based on cluster-parallel simulation and rendering engines.

4

Research Plan

We will accomplish our first goal of providing accurate mineral aerosol forecasts and hindcasts for the visualization system by iterative refinements to WRF-DEAD based on evaluations in the ROIs (Zender), informed by powerful real time visualization technology which allows us to assess and explore model sensitivities and biases (Pajarola). The second goal of our project objectives is to develop a powerful visualization system capable of interactive visualization of the large-scale terrain surface and atmospheric volume data from the WRF model. High-speed network connectivity between the forecast simulation and the visualization engine provides the bandwidth for fast update rates of mesoscale atmospheric data between the two systems. Cluster-based computing on both sides is exploited to achieve accelerated-time weather forecasting and interactive visualization. In the following we outline the research activities in terrain and atmospheric visualization (Section 4.2) as well as in system development (Section 4.3) that are necessary to achieve our goals for LEP. The third goal, entraining environmental planners to use LEP and guide its development to be applicable to real world problems in the ROIs, is largely a problem of human factors, project coordination, and effective communication. Our methods of addressing these project management issues are discussed in Section 6 below.

6


4.1 Macrophysics and Microphysics of Dust Entrainment and Transport The Dust Entrainment And Deposition (DEAD) model (Zender, 2003) is a widely used and extensively tested mineral dust prediction module. DEAD has been used and evaluated in global and regional studies of dust emissions (Zender et al., 2003a,c; Luo et al., 2003; Mahowald et al., 2002, 2003), radiative forcing (Collins et al., 2001, 2002), chemistry (Bian and Zender, 2003), and microphysics (Grini et al., 2002; Grini and Zender, 2004). Current state-of-the-art dust entrainment models include DEAD (Zender et al., 2003a), GOCART (Ginoux et al., 2001), CARMA (Colarco et al., 2002), and NOGAPS (Hogan and Brody, 1993). All of these models and others have been used in global forecasts by various institutions. A description of the relative strengths and weaknesses of these models is beyond the scope of this proposal, but the available model evaluations show (in our opinion) that no dust model performs better than DEAD. The physical processes in DEAD are summarized in Zender et al. (2003a), Zender et al. (2003c), and Grini and Zender (2004). In particular, DEAD suits LEP well because it accurately mobilizes large silt and sand-sized particles (10 < D > 200 µm) which settle too quickly to be important in global scale models, but which are extremely important in mesoscale events. A rich variety of physical processes and scientific questions that are neglected in current global scale models become relatively more important at the mesoscale and will be the focus of our scientific investigations. Mesoscale dust forecasts are sensitive to the relative roles of particular dynamic processes in determining total dust mobilization and structure. These mesoscale processes include gust fronts, mountain winds, surface turbulence, and dry and wet convection. Other processes best addressed at the mesoscale rather than the global scale include the role plant phenotype in determining surface roughness lengths and drag partitioning (Raupach et al., 1993), particle asphericity effects on sedimentation (Ginoux, 2003), fine scale Geo-morphologic and topographic contributions to soil erodibility (Zender et al., 2003c), and entrainment of large particles in anthropogenically disturbed environments (Gillette, 1988; Batt and Peabody II, 1999; Saxton et al., 2000). Recent evidence suggests direct suspension may be more important than saltation-sandblasting in anthropogenic environments (David Chandler, personal communication, 2002; Dale Gillette, personal communication, 2002). As these processes are represented and better understood at the mesoscale, we will transfer (and parameterize) this knowledge for application in our global models. The central scientific areas we will address with this research are 1. Formation and Structure: (a) What are the relative contributions to dust loading by gust fronts, surface turbulence, mean winds, and dry and wet convection? (b) How does the spatial and temporal structure of a dust storm depend on the presence or absence of these elements? (c) How are dust storm mass budgets and visibility reductions partitioned between large and small particles under a variety of conditions? 2. Predictability: (a) To what extent are simulations of mesoscale dust events consistent with synoptic and in-situ observations in the ROIs? (b) Which bioclimatic mobilization constraints (e.g., wind speed, soil moisture, vegetation, soil texture, disturbance) most affect dust predictability in the ROIs? Which of these 7


constraints alter most under climate change and thus present environmental planners with the problem of adapting to a dustier climate or exploiting a less turbid climate? 3. Response to environmental change: (a) How is dustiness expected to change in ROIs as a result of natural and/or anthropogenic climate and land-use change? (b) Which regions are most vulnerable to climatological forecast rainfall changes? This research requires extending the physics in DEAD to account for processes which are relatively more important at the mesoscale than at the global scale. The first such process is the entrainment and transport of large particles. However, large particles are thought to contribute significantly to aerosol mass transport on global scales, although the mechanisms by which large particles remain in suspension are not well understood and consequently are poorly modeled (Zender et al., 2003a; Maring et al., 2003; Ginoux, 2003; Colarco et al., 2002). The role of large dust particles in visibility reduction increases with proximity to dust source regions, which in the case of military operations may often be local. We are ready to assess the importance of large silt and sand in mesoscale regions using our saltation-sandblasting physics (Grini et al., 2002; Grini and Zender, 2004). We will extend DEAD to allow direct entrainment of very large sand and gravel sizes using the threshold speed measurements of Batt and Peabody II (1999). This extension will allow realistic deflation to result from extreme events and even helicopter-generated winds. Besides extension to higher wind speeds, DEAD will allow dust production by user-specified anthropogenic sources, such as land use disturbance, vehicle movement, or explosion. Vehiculargenerated dust will be parameterized from existing observations based on soil texture, moisture and vehicle type (Saxton et al., 2000). Specification of anthropogenic disturbance and source will be done through offline files initially, and eventually by user-controlled real time input using external controls such as joysticks. LEP will allow the user to visualize natural and anthropogenic dust separately or together, and thus to evaluate the relative strength of anthropogenic disturbance against the natural background. 4.2 Boundary Layer Visualization The visualization engine has to address realistic ground illumination and rendering of high-resolution mesoscale terrain elevation models as well as visualization of atmospheric attenuation and cloud formation. As indicated in Section 3.3 the visualization engine exploits cluster-parallelism for fast rendering. It also features a high-resolution tiled display system capable of 10 mega-pixel (Mp) image resolutions (see also Figure 1 and Section 4.3). In this context, the main problems are efficient data distribution techniques and algorithms for cluster-parallel terrain and atmosphere visualization. The visualization problems that we will confront and address in our project on interactive rendering of forecasts are: 1. Ground (surface) rendering: (a) To what extent must efficient multi-resolution terrain models be modified to support efficient load-balancing on distributed cluster-based and tiled-display systems? (b) What are the main effects for ground illumination from the diffuse sky-/sunlight transfer through the boundary layer atmosphere? How can we efficiently model these for real-time terrain rendering? 8


2. Atmosphere (volume) visualization: (a) Efficiently coupling the dynamic large-scale atmospheric aerosol distribution and radiance data with the visualization system, (b) Optimizing approximation algorithms for physically correct light scattering and extinction to obtain throughput required for interactive rendering (c) Algorithms and data structures for efficient cluster-parallel visualization 4.2.1 Ground Rendering Terrain surface rendering will be based on our own extensive work on multi-resolution terrain rendering frameworks (Pajarola, 1998a,b; Pajarola et al., 1998; Pajarola and Widmayer, 2001; Pajarola, 2002; Pajarola et al., 2002; Bao and Pajarola, 2003; Lario et al., 2003). This project requires extending such efficient terrain rendering approaches to cluster-parallel tiled-display rendering. In this context, the efficient parallelization and distribution of workload of the rendering task is the major problem (see overviews (Molnar et al., 1994; Crockett, 1997)). Existing work on parallel rendering exhibits limiting factors such as requiring all data to reside in main memory (Samanta et al., 2000, 2001; Humphreys et al., 2001) or poor load-balancing (Correa et al., 2002) for unevenly distributed geometry among display tiles. This work will address the development of a distributed multi-resolution terrain surface representation that supports dynamic load balancing between the rendering nodes. While maximally interleaved and distributed data partitioning supports efficient rendering on each individual rendering node resulting in partial full-resolution images, it imposes a large compositing cost for combining large images from all nodes for each rendered frame. Therefore, sort-first (data partitioning) and sort-last (image compositing) strategies must carefully be combined in this context (see also (Samanta et al., 2000)). Additionally, this terrain multi-resolution representation will be designed to maintain very large-scale terrain data out-of-core on external (disk) memory and to provide seamless access from all rendering nodes. Ground illumination is mainly dependent on the sky- and sunlight absorption and transport through the boundary layer atmosphere, and reflection properties of the surface. The WRF module computes direct and diffuse down-welling radiance and thus provides the critical terrain surface irradiance information needed for accurate ground illumination. Second, the necessary surface reflection properties will be modeled in form of realistic BRDFs derived from the NASA MODIS and MISR instruments (Schaaf et al., 2002; Tsvetsinskaya et al., 2002). Similar to color texturing from satellite images or aerial photographs, sampled BRDF data is interpolated and mapped onto the ground surface. 4.2.2 Atmospheric Visualization Boundary layer visibility attenuation and scattering are modeled by a volume rendering approach of sky- and sunlight absorption and transport through the atmosphere. WRF uses atmospheric and particle scattering and absorption properties to calculate radiative fluxes between cells in the atmosphere and thus provides the visualization system with partial results of the scattering equation for direct volume rendering (Kajiya and Von Herzen, 1984). Hence as scattering is provided by the WRF model, the remaining main rendering task is numerical integration and attenuation of radiative fluxes along the view-direction of the observer similar to the low-albedo case of volume 9


rendering (Blinn, 1982). The required physical light attenuation and scattering properties can be derived from DEAD and WRF by extending the calculation of upwelling and downwelling irradiance to incorporate scattering in multiple directions for increased visual fidelity. Zender knows the appropriate modifications to make to the WRF radiation code in order to expose more detail to the visualization system where desired. The WRF solar radiation code is based on the accurate, efficient, atmospheric solar radiative transfer code (Briegleb, 1992) used in other climate/weather models such as the NCAR CCM3, CAM, and RegCM2 These models all employ versions of the CCM3 CRM radiation code which PI Zender maintains on behalf of NCAR. For some visualizations we might want to employ more than the standard two-stream adding-doubling approach used for visible radiation in order to obtain realistic radiances at more polar angles (Stamnes et al., 1988; Zender et al., 1997; Zender, 1999). An alternative to increasing the radiation code in WRF is to explore anisotropic scattering phase functions for given particle mixtures that can be computed off-line and accessed via look-up tables. Mesoscale atmospheric models easily constitute of millions or even billions of samples (voxels). Without efficient multi-resolution techniques that can approximate rendering at different levels-of-detail (LODs) and without hardware accelerated rendering, such large volumes cannot be visualized at interactive frame rates. Thus interactive visualization will be achieved by a multiresolution octree volume representation (e.g., Samet, 1989; Laur and Hanrahan, 1991; Grosso and Greiner, 2000) and splatting-based direct volume rendering techniques (e.g., Westover, 1990; Laur and Hanrahan, 1991; Crawfis and Max, 1993; Mueller et al., 1998, 1999a; Huang et al., 2000; Zwicker et al., 2001). Similar to (Crawfis and Max, 1993), our approach will incorporate perspectively projected splatting footprints as texture maps and exploit hardware acceleration by rendering splats as textured sprites. This task will draw on our experience in large-scale volume rendering (Chopra and Meyer, 2002, 2003; Meyer et al., 2003) and result in a combined view-dependent rendering of both the terrain and atmosphere using hybrid rendering approaches (Lengen et al., 1998; Meyer et al., 2003). 4.3 LEP System LEP system design and implementation must address the following issues: 1. How does the time-stepping forecast model (WRF) send 2D and 3D output (tracer concentrations, RGB radiance fields) to the visualization engine's multiresolution terrain and atmosphere rendering model? 2. What hardware, software and networking infrastructure components and configurations efficiently support real-time forecast rendering? 3. What user-interaction, input, and visualization features best support environmental planning? The basic interaction and visualization mode of LEP will be interactive walk- and fly-through exploration of the simulated ROI, similar to Figure 1 a. This image is from our preliminary terrain rendering system ViRGIS (Pajarola et al., 1998; Pajarola and Widmayer, 2001), and does not include the physically realistic simulated ground illumination or atmospheric effects to be incorporated in LEP. Fundamental rendering features will include enabling and disabling various illumination, light scattering and attenuation effects via a mouse- or wand-controlled user-interface (e.g., menus, dialog windows, parameter panels etc.). Further interaction functionality will include specification of boundary condition changes (erodibility, land-use and vegetation type) and anthro10


pogenic disturbances (type, location, duration and magnitude of tracer emissions fluxes), see also Figure 2. Input modes are tabular data, and, ultimately, interactive manipulation using external controls such as mouse/joystick or wireless hand-held point-and-click devices. One user-interaction scenario the user (i.e., planner or decision-maker) sketches out a region on the screen using the mouse that is converted by the system to a geo-spatial terrain region. Then the user may change the land-use properties of the selected region. Altered properties are fed-back as time-varying boudary conditions to the WRF simulation module and will result in a modified forecast. Moreover, as shown in Figure 1 b, LEP will provide high-resolution images through the use of a tiled multi-screen display system that incorporates in its initial configuration a 3 × 3 arrangement of nine LCD panels capable of rendering 10 Mp. Large screen real-estate not only supports large

a)

b)

Figure 1: (a) Example interactive environmental exploration. (b) Synchronized distributed rendering on tiled-display wall. field-of-view and hi-res imagery but also allows for simultaneous display of various atmospheric multi-field data attributes such as convective fluxes, different species (e.g., dust, smoke) and current and time-integrated aerosol deposition. Ultimately, the large display may be partitioned to display different forecast simulations. For increased user immersion, the visualization system will support configurations for stereo-rendering2 . To cope with the computational cost of accelerated-time forecasts and interactive visualization of very large environmental data, LEP requires cluster-parallel computing for both tasks as illustrated in Figure 2. Hence the WRF/DEAD simulation engine is driven by a 10 node (20 CPU) Beowulf cluster, and also includes a large high-performance RAID storage system to archive simulated test-scenarios and to serve as asynchronous data-transfer cache for the visualization system. The forecast simulation software is based on extensions of the Beowulf/MPI based WRF and DEAD modules. The visualization engine is powered by a parallel-rendering cluster consisting of 10 nodes with high-performance graphics cards, each driving one of the LCD displays. It also includes a medium RAID storage system. The rendering software for the tasks outlined in Sections 4.2.1 and 4.2.2 will be based on the MPI and Chromium (Humphreys et al., 2002) concurrent computing and rendering libraries. As the forecast system will produce an anticipated data rate of 1­2 Gb s-1 every few seconds, even with only few field-attributes provided to the visualization
2

via use of stereo shutter-glasses and rendering of separate images for the left and right eye

11


system, LEP incorporates its own multi Gb s-1 network infrastructure.3
Forecast System Archived Observations Forecast Cluster Network Network Router/Switch Forecast Model Network Router/Switch Visualization System Visualization Cluster
9 8 7 6 5 4 3 2 1 0
RAID

3D Display Wall
9 6 3 8 5 2 7 4 1

Console
0

Gridded Meteorology

High-speed

User Navigation Forecast Modification: Tracer emissions, land-use change

WRF/DEAD Model RAID Storage System

Real-Time Terrain, Atmosphere and Object Rendering

Figure 2: System organization of the LEP. In hindcast mode, LEP is driven by meteorological inputs from observations such as National Centers for Environmental Prediction (NCEP) reanalyses (e.g., Kalnay, 1996) or from archived model data. In forecast mode, the atmospheric model at the core of LEP prognoses the state variables (e.g., wind) required to predict dust emissions and cloud formation. The visualization system of LEP obtains updated aerosol and radiance flux data from the predictive forecast model every few seconds (downlink data transfer). Upon transfer, this data is organized and processed to be easily accessible to the visualization engine's multiresolution terrain and atmosphere rendering. A high-speed RAID system caches the incoming atmospheric field data and provides a transparent single-image disk to all rendering nodes. This allows all nodes to concurrently access the data for rendering from external memory (i.e. via memory-mappedfiles). Uplink communication consists of low-bandwidth parameter changes to the WRF/DEAD model and will cause the forecast to temporarily interrupt its calculation to adjust for the modified boundary layer and atmosphere conditions.

5

Simulation and Forecast Evaluation

DEAD has a long and ongoing history of evaluation against measurements, including the station and satellite measurements and field campaigns (Rasch et al., 2001; Collins et al., 2001, 2002; Zender et al., 2003a; Luo et al., 2003; Mahowald et al., 2002, 2003). DEAD has also been evaluated against wind-tunnel measurements in the field and laboratory (Iversen and White, 1982; Gillette ´ et al., 1998; Fecan et al., 1999; Alfaro and Gomes, 2001). DEAD has been used operationally by P. Rasch and W. Collins of NCAR in 72-hour forecasts of global dust aerosol distributions in the MATCH Aerosol assimilation framework since 2001, as well as in other GCMs. We are currently evaluating DEAD against near real time station measurements of PM10 and saltation in the Columbia Plateau (data from Washington State University and the Columbia Plateau Air Quality Project). With LEP we will initiate new evaluations against saltation and deposition at multiple stations in the Mojave and against cameras which automatically photograph key playas
3

required bandwidth exceeds current production-network backbone available on site

12


when wind speeds exceed 6 m s-1 (data from Rich Reynolds, Pat Chavez, and Marith Reheis, USGS) (Reheis, 1997). However, we recognize that the ongoing global and station evaluations are not particularly appropriate for assessing predictions of mesoscale dust storms. To be certain that we improve representation of dust storm processes in LEP, we must also identify and use new sources of satellite data. We will use GOES/TOMS/MODIS imagery/aerosol index/optical depth (Wald et al., 1998; King et al., 1999; Torres et al., 2002). By comparison of model to observations, we will obtain fundamental metrics to assess the forecasting skill of LEP, including: · Probability of Detection: How often are dust storms observed but not predicted? What commonalities do missed forecasts share? · False Alarm Rate: How often dust storms predicted but not observed? · Forecast bias: What is the difference between observed and predicted dust levels? These three metrics are appropriate to mesoscale forecasts and may be diagnosed from instrumentation available in the ROIs and from satellite. Because LEP allows interactive specification of anthropogenic sources, simulations may also be evaluated against vehicle-influenced visibility (van Donk et al., 2003). Drought is a primary natural cause of increased dust in semi-arid regions (Prospero and Nees, 1986; Nicholson et al., 1998). Agriculture is a primary anthropogenic source of dust in semi-arid regions (Glantz, 1994; Saxton et al., 2000; Chandler et al., 2002; Youlin et al., 2002). Livestock grazing in both ROIs may be crucial to explaining observed patterns of deflation as much as 30 years after the last anthropogenic activity (Neff et al., 2004). Collaborator Pat Chavez conducts ongoing photogrammetric measurements and GOES infrared dust detection over the Mojave to assess source region activity. Adequate expenses are requested for travel to the Mojave and to the IoM and Aral Sea in Kazakhstan to train Abdibekov in these methods. In addition, we have assembled a climatology of all detectable dust source regions and their emission efficiency from TOMS satellite observations (Herman et al., 1997; Prospero et al., 2002), and MODIS aerosol classifications are now a standard product. we are able to identify regions of interest worldwide where dust, both natural and anthropogenic, poses a potential visibility problem. 5.1 Broader Impacts LEP will combine and enhance existing weather models and display technologies into a coordinated forecast and interactive visualization system of weather and atmospheric information at mesoscale dimensions. This inter-disciplinary research project will advance the penetration of information technology into areas with beneficial application to society, including environmental forecasting, land-use planning and erosion mitigation. Utilizing LEP technology aviation authorities in arid regions could forecast visibility impairment and engine risk due to dust storms. Urban planners downwind of dusty regions could prioritize landscape vulnerability to land use change and forecast traffic warnings. Conservationists could develop "what if " visualizations of arid landscapes based on a range of climate, conservation, and land use change scenarios. Health and security sectors could simulate transport and deposition patterns of passive particulates in mesoscale regions, from specified point sources (e.g., dirty bombs). The DEAD model and visualization system are based on and will continue to be made available as open-source software. Finally, LEP provides an exceptional educational opportunity to rapidly demonstrate and experience the dynamic visual impacts of environmental change on daily to inter-annual timescales 13


(e.g., from dust storms to desertification). Graduate students participating in LEP will benefit from a unique learning experience working at the interface of weather and aerosol forecasting and information technology and display. These students will be advantaged by exposure to scientific questions and applications beyond their own dissertation work. Zender will also display and discuss the mesoscale visualization results to his 350-student lower division course on the Atmosphere. 5.2 Synergy with Ongoing Work This ITR project will enhance adoption and support for DEAD in solving wind-erosion problems in countries with limited scientific infrastructure. This proposal is synergistic with ISTC-funded proposal K-424 by Abdibekov to simulate Aral Sea dust fluxes. Zender is their US project advisor, and they already use DEAD. This proposal is synergistic with NSF ATM-0214430 (discussed in Section 3.1), where Zender requested and was denied funding for mesoscale erosion simulations of the Columbia Plateau in eastern Washington State. This proposal synergizes with a pending proposal to DOE ARM where Dr. Steven Ghan, PI, and others including Zender seek to prototype a next generation aerosol scheme in WRF and then move it to CCSM. PI Zender is an affiliate scientist with NCAR and collaborates extensively with scientists there responsible for aerosol implementations in internationally used models including MATCH, CCSM, and WRF. He will ensure that extensions and improvements to the aerosol capabilities of DEAD and WRF are made available to the WRF Chemistry and Aerosol working group, led by Dr. Peter Hess, and to the NCAR CCSM Atmospheric, Land, and Biogeochemistry Working Groups, led by Drs. Bill Collins, Gordon Bonan, and Natalie Mahowald. The software technology to expose the key prognostic variables of the running forecast model to interactive modification will be an important contribution useful to many similar community forecast models. 5.3 Extensibility and Technology Transfer Many other ROIs which might benefit from such an integrated mesoscale forecasting and decisionmaking system. For example, climate change and anthropogenic tracers are both problems confronting the Arctic National Wildlife Refuge (ANWR). In that sense, our initial ROIs are a specific examples of a generic class of problems which could benefit from LEP's capabilities. The UCI ESS Department has national strengths in arid region disturbances, hydrology, and climate change. Once we debug the forecast-visualization-decision process loop, we are committed to expand LEP or transfer our technology to other institutions interested in spearheading similar projects on other regions.

6

Project Coordination

PI Zender takes overall responsibility for project coordination. Interactions among scientific, visualization, decision-making and field representatives of the LEP team will occur at a variety of pre-planned and informal bi-lateral and multi-lateral, project-wide meetings. Zender directs two other multi-investigator projects, the Earth System Modeling Facility, similar in scale to the LEP, and the netCDF Operators, a smaller scale, unfunded, OpenSource software project. He makes efficient use of project coordination software such as Mailman (for project mailing lists), wreq (a work-request tracking system for prioritizing tasks), and extensive documentation on project Home Pages. These techniques maximize project transparency and minimize confusion that arises through misunderstood responsibilities, requests, and goals. All LEP software design, construction, and modification will employ Concurrent Versioning System (CVS) to facilitate distributed 14


development. To facilitate collaboration, all model software and data will be made immediately available via the LEP homepage to any interested outside investigator. We believe strongly in unfettered exchange of software and data. Sub-groups of the immediately participating personnel (Table 1) will naturally form along disciplinary and ROI boundaries. These sub-groups will communicate outside of any pre-planned mechanism, though within a medium (e.g., mail-lists) viewable by all. Phone conferences will be held when immediate "all hands" meetings are required. The specific scientific/technical responsibilities of PIs, senior personnel, and collaborators receiving funds are: 1. PI Zender will direct environmental forecasts and evaluation. Zender will direct the summer workshops and outreach efforts to entrain local educational, civil, and security organizations including CDHS, CHP, DHS, FAA, NPS, and RESCUE. Zender's group will embed DEAD into WRF, study dust storm structure in ROIs, and make and evaluate improvements in dust storm forecasting. The post-doc will conduct original research applying mesoscale forecasts to environmental decisions in the ROIs. The ESS graduate student will work with Zender on micro-to-mesoscale processes, dynamics, prediction, and consequences of Aeolian erosion. 2. Co-PI Pajarola is in charge of environmental visualization and rendering and will direct two UCI graduate students. Pajarola's group will design and implement real time visualization algorithms for WRF/DEAD, along with protocols for efficient interactive and offline specification of anthropogenic disturbances. Pajarola's group will contribute extensively to international synergies by making subsets of LEP forecasts efficiently available through the Web. 3. Senior Personnel Joerg Meyer has primary charge of volume rendering and optimizing a multi-resolution data representation to work in a hybrid visualization with Pajarola's viewdependent terrain visualization. The hybrid algorithm will be implemented by a graduate student under Meyer's and Pajarola's joint supervision. 4. Senior Personnel Rich Reynolds will provide in situ measurements of Mojave erosion, advise on pilot environmental planning questions, and coordinate LEP with ongoing USGS projects to predict SW US response to natural and anthropogenic forcing. Reynolds is in charge of assessing the visibility impact of the planned doubling of Fort Irwin in the next few years. Fort Irwin is (anecdotally) the largest current dust source in the Mojave (van Donk et al., 2003). 5. Collaborator Ualikhan Abdibekov, Institute of Mathematics, Almaty, Kazakhstan (see attach letter of support) is our primary pre-committed International Collaboration and will lead the Aral Sea ROI group. The Aral Sea provides a wonderful end-member for arid region disturbance studies because it is an anthropogenic environmental disaster from which we hope to learn and to which we hope to apply erosion causation and mitigation strategies. Zender is an advisor to Abdibekov on ISTC-funded proposal K-424 to simulate Aral-Sea dust fluxes. Abdibekov will gather and supply in situ Aral Sea data and liaison between US and international researchers interested in applying LEP in Kazakhstan. Abdibekov will train on LEP in Year 2 and host our field visit to the Aral Sea in Year 3. This will permit Abdibekov's group to contribute to LEP's design for the complex Aral Sea project, and to obtain skills transfer in environmental planning for the Aral Sea based on our Mojave results. 6. Collaborator Katie Purvis will use LEP heavily to forecast, understand, and plan for health effects of particulate plumes downwind of the Aral Sea. She will advise Zender's group on 15


how to get science done in Kazakhstan. 7. Collaborator Pat Chavez will provide satellite and photogrammetric evaluations of erosion in the Mojave, and use LEP to better understand the roles of bioclimatic controls on erosion. 8. Zender, Pajarola, and Meyer will all perform studies of passive tracer releases (e.g., explosions) in the Mojave ROI. Passive tracer studies with prescribed magnitudes and source locations require the least geophysical knowledge and are excellent forecast diagnostics (flow markers). At the same time, these studies are extremely important for igniting interest in LEP from the Civil, Security, and DHS communities. The long-range planning directions that a facility like LEP will take are difficult to predict and subject to pre-emption by emerging threats. For this reason, the above tasks mainly represent the scientific and technical responsibilities which must be carried through to make environmental planning for real-world problems possible. Four distinct types of pre-planned multi-investigator meetings will ensure project cohesion and steady progress to important milestones. Field Site visits (FV), National Conferences (NC), Summer Workshops (SW), and User Visits to LEP (UV). The summer workshops in years three and five bring together students, researchers, and decision-makers to learn and improve the capabilities of LEP in helping to solve real world problems. Invited attendees will represent local educational, civil, and security organizations including California Department of Health Services (CDHS), California Highway Patrol (CHP), Department of Homeland Security (DHS), Federal Aviation Administration (FAA), National Park Service (NPS), and Responding to Crises and Unexpected Events (RESCUE). The project time-line shows how the requested travel funds support intra- and extra-project communication and progress toward milestones. This time-line only details travel involving multiple project investigators. Funds used solely to present investigator-specific results at national and international conferences are requested in the proposal, and are not detailed here: Year 1. Milestones: 1a. Integrate DEAD with WRF; 1b. Render first passive tracers; 1c. Hindcast and render April 15, 2002 dust storm. Travel: 1. FV: Three-day visit to Mojave to familiarize modelers with measurement locations, source regions, and terrain (Pajarola, Zender, Reynolds, Chavez, ESS graduate student, post-doc). Year 2. Milestones: 2a. Mesoscale dust storm mass budgets by process; 2b. Real-time forecast and rendering; 2c. First forecast interactivity. Travel: 1. UV: Rich Reynolds pilots study on Fort Irwin impacts expansion. 2. UV: Pat Chavez evaluates modeled Mojave source regions against his satellite and in situ measurements 3. UV: Ualikhan Abdibekov trains on LEP for one week, exchanges information required to commence Aral Sea ROI studies. 4. UV: Katie Purvis trains on LEP in Mojave and helps guide development of Aral Sea ROI. 5. NC: PI Zender and ESS Post-doc attend AMS Natural Hazards meeting to present science and to provide hands-on demonstration of LEP forecast/decision technology Year 3. Milestones: 3a. Finish Fort Irwin study; 3b Hyper-real surface tiling with MISR BRDFs; 3c Fully interactive forecast boundary conditions. Travel: 16


1. FV: Zender visits Abdibekov in Almaty, Kazakhstan and tours the Aral Sea to familiarize modelers with ROI terrain and available erosion data. 2. UV: Katie Purvis pilots study on particulate exposure downwind of Aral Sea ROI. 3. NC: Zender and ESS grad student attend AGU Fall AGU meeting to present science and to provide hands-on demonstration of LEP forecast/decision technology to other attendees 4. UV: 1­2 unspecified Southern California civil/security planners train on and use LEP 5. SW: Summer Workshop I brings together and train all US LEP participants with civil and security planners from CDHS, CHP, DHS, FAA, NPS, and RESCUE. Year 4. Milestones: 4a. Mojave climate change simulations; 4b. Initial Aral Sea forecasts; 4c. Fully hybrid volume rendering/surface rendering. Travel: 1. NC: Zender and ESS Post-doc attend AMS Natural Hazards meeting to present science and to provide hands-on demonstration of LEP forecast/decision technology to other attendees 2. UV: Katie Purvis pilots study on particulate exposure downwind of Mojave 3. UV: 1­2 unspecified Southern California civil/security "decision-makers" train on and use LEP 4. UV: Zender and Rich Reynolds conduct pilot study on Valley Fever habitat and dispersal changes with climate (Kolivras et al., 2001; Zender and Talamantes, 2004) Year 5. Milestones: 5a. Aral Sea dust storm mitigation studies; 5b. Visualize LEP forecasts from Cal-(IT)2 /UCSD via OptIPuter; 5c. Visualization-derived improvements to ROI erodibility Travel: 1. NC: Zender and ESS grad student attend AGU Fall AGU meeting to present science and to provide hands-on demonstration of LEP forecast/decision technology to other attendees 2. UV: Katie Purvis finishes particulate exposure downwind of Aral Sea and Mojave ROIs 3. UV: 1­2 unspecified Southern California civil/security "decision-makers" visit to train on and use LEP 4. SW: Summer Workshop II. All US LEP participants interact with, re-acquaint, and train civil and security planners with new LEP features. Due to the complexity of realistically rendering and flying through forecasts, we assume that the visualization facility will be in a rough but usable state by the end of Year 2. Early User Visits, i.e., before Year 3, are reserved for bona fide collaborators and colleagues who will provide feedback on how to improve the system while it is under development. The two all-hands workshops are in years three and five. By that time we expect the interactive visualization system to be much more user friendly, robust, efficient, and versatile. Note that the vast majority of our pre-overhead budget is spent on salaries, benefits and tuition ( 84%), and travel and meetings for scientists, students, planners, and decision-makers ( 6%), rather than on equipment ( 7%). The greatest challenges faced by this project are not computational or networking speed because we can always scale our simulation resolution as needed to obtained the desired visualization throughput. Constructing a system that performs stunning visualizations of mesoscale forecasts will certainly be difficult and will require cutting edge visualization development. Nevertheless, our greatest challenge is to build the facility in such a user-friendly way that students, researcher, and decision-makers see using it as an easy opportunity to approach real-world environmental problems quantitatively yet intuitively. This will require 17


careful leveraging of our modest travel and outreach funding to integrate our talented and interdisciplinary team members. Hence we put more effort up front to outline the human rather than the technical dimensions of our project coordination.

18


References Bibliography
Agranov, G. and C. Gotsman, 1995: Algorithms for rendering realistic terrain image sequences and their parallel implementation. The Visual Computer, 11(9), 455­464. 3.3 Alfaro, S. C. and L. Gomes, 2001: Modeling mineral aerosol production by wind erosion: Emission intensities and aerosol size distributions in source areas. J. Geophys. Res., 106(D16), 18075­18084. 3.2, 5 Bach, A. J., A. J. Brazel and N. Lancaster, 1996: Temporal and spatial aspects of blowing dust in the Mojave and Colorado deserts of southern California, 1993­1994. Phys. Geography, 17(4), 329­353. 2.1 Balkanski, Y., M. Schulz, B. Marticorena, G. Bergametti, W. Guelle, F. Dulac, C. Moulin and C. E. Lambert, 1996: Importance of the source term and of the size distribution to model the mineral dust cycle. in S. Guerzoni and R. Chester, editors, The Impact of Desert Dust Across the Mediterranean, pp. 69­76, Boston, MA. Kluwer Academic Pub. 3.2 Balmelli, L., S. Ayer and M. Vetterli, 1998: Efficient algorithms for embedded rendering of terrain models. in Proceedings IEEE International Conference on Image Processing ICIP 98, pp. 914­ 918. 3.3 Bao, X. and R. Pajarola, 2003: LOD-based clustering techniques for optimizing large-scale terrain storage and visualization. in Proceedings SPIE Conference on Visualization and Data Analysis, pp. 225­235. 4.2.1 Batt, R. G. and S. A. Peabody II, 1999: Threshold friction velocities for large pebble gravel beds. J. Geophys. Res., 104(D20), 24273­24279. 4.1, 4.1 Bian, H. and C. S. Zender, 2003: Mineral dust and global tropospheric chemistry: Relative roles of photolysis and heterogeneous uptake. J. Geophys. Res., 108(D21), 4672, doi:10.1029/2002JD003143. 3.2, 4.1 Blinn, J., 1982: Light reflection functions for simulation of clouds and dusty surfaces. in Proceedings ACM SIGGRAPH 82, pp. 21­29. ACM Press. 4.2.2 Briegleb, B. P., 1992: Delta-Eddington approximation for solar radiation in the NCAR community climate model. J. Geophys. Res., 97(D7), 7603­7612. 4.2.2 Cabral, B., N. Cam and J. Foran, 1994: Accelerated volume rendering and tomographic reconstruction using texture mapping hardware. in Proceedings Symposium on Volume Visualization, pp. 91­98. 3.3 Chandler, D. G., K. E. Saxton, J. Kjelgaard and A. J. Busacca, 2002: A technique to measure fine-dust emission potentials during wind erosion. Soil Sci. Soc. Am. J., 66, 1127­1133. 5 Chin, M., P. Ginoux, S. Kinne, O. Torres, B. N. Holben, B. N. Duncan, R. V. Martin, J. A. Logan, A. Higurashi and T. Nakajima, 2002: Tropospheric aerosol optical thickness from the GOCART model and comparisons with satellite and sun photometer measurements. J. Atmos. Sci., 59(3), 461­483. 3.2 Chopra, P. and J. Meyer, 2002: Tetfusion: An algorithm for rapid tetrahedral mesh simplification. in Proceedings IEEE Visualization 2002, pp. 133­144. Computer Society Press. 4.2.2 Chopra, P. and J. Meyer, 2003: Topology sensitive volume mesh simplification with planar quadric error metrics. in IASTED International Conference on Visualization, Imaging, and Image Processing (VIIP 2003), pp. 908­913. 4.2.2 Cignoni, P., F. Ganovelli, E. Gobbetti, F. Marton, F. Ponchio and R. Scopigno, 2003a: BDAM - batched dynamic adaptive meshes for high performance terrain visualization. in Proceedings EUROGRAPHICS 2003, pp. 505­514. also in Computer Graphics Forum 22(3). 3.3 Cignoni, P., F. Ganovelli, E. Gobbetti, F. Marton, F. Ponchio and R. Scopigno, 2003b: Planetsized batched dynamic adaptive meshes (P-BDAM). in Proceedings IEEE Visualization 2003,


pp. 147­155. Computer Society Press. 3.3 Claiborn, C., B. Lamb, A. Miller, J. Beseda, B. Clode, J. Vaughan, L. Kang and C. Newvine, 1998: Regional measurements and modeling of windblown agricultural dust: The Columbia Plateau PM10 Program. J. Geophys. Res., 103(D16), 19753­19767. 1 Cohen-Or, D., E. Rich, U. Lerner and V. Shenkar, 1996: A real-time photo-realistic visual flythrough. IEEE Transactions on Visualization and Computer Graphics, 2(3), 255­265. 3.3 Colarco, P., O. Toon and B. Holben, 2002: Saharan dust transport to the caribbean during PRIDE: Part 1. Influence of dust sources and removal mechanisms on the timing and magnitude of downwind AOD events from simulations of and remote sensing observations. J. Geophys. Res., 108(D19), 8589, doi:10.1029/2002JD002658. 4.1, 4.1 Collins, W. D., P. J. Rasch, B. E. Eaton, D. W. Fillmore, J. T. Kiehl, C. T. Beck and C. S. Zender, 2002: Simulation of aerosol distributions and radiative forcing for INDOEX: Regional climate impacts. J. Geophys. Res., 107(D19), 8028, doi:10.1029/2000JD000032. 4.1, 5 Collins, W. D., P. J. Rasch, B. E. Eaton, B. Khattatov, J.-F. Lamarque and C. S. Zender, 2001: Forecasting aerosols using a chemical transport model with assimilation of satellite aerosol retrievals: Methodology for INDOEX. J. Geophys. Res., 106(D7), 7313­7336. 1, 4.1, 5 Correa, W. T., J. T. Klosowski and C. T. Silva, 2002: Out-of-core sort-first parallel rendering for cluster-based tiled displays. in Proceedings Eurographics Workshop on Parallel Graphics and Visualization, pp. 89­96. 4.2.1 Crawfis, R. and N. Max, 1993: Texture splats for 3D scalar and vector field visualization. in Proceedings IEEE Visualization 93, pp. 261­266. Computer Society Press. 4.2.2 Crockett, T. W., 1997: An introduction to parallel rendering. Parallel Computing, 23, 819­843. 4.2.1 Duchaineau, M., M. Wolinsky, D. E. Sigeti, M. C. Miller, C. Aldrich and M. B. Mineev-Weinstein, 1997: Roaming terrain: Real-time optimally adapting meshes. in Proceedings IEEE Visualization 97, pp. 81­88. 3.3 ¨ Elinas, P. and W. Sturzlinger, 2000: Real-time rendering of 3D clouds. Journal of Graphics Tools, 5(4), 33­45. 3.3 Evans, W., D. Kirkpatrick and G. Townsend, 2001: Right-triangulated irregular networks. Algorithmica, 30(2), 264­286. 3.3 ´ Fecan, F., B. Marticorena and G. Bergametti, 1999: Parametrization of the increase of the aeolian erosion threshold wind friction velocity due to soil moisture for arid and semi-arid areas. Annales Geophysicae, 17, 149­157. 5 Gerstner, T., 1999: Multiresolution compression and visualization of global topographic data. ¨ ¨ Tech. Rep. 29, Institut fur Angewandte Mathematik, Universitat Bonn, to appear in Geoinformatica 2001. 3.3 Gillette, D. A., 1988: Threshold friction velocities for dust production for agricultural soils. J. Geophys. Res., 93(D10), 12645­12662. 4.1 Gillette, D. A., 1999: A qualitative geophysical explanation for "hot spot" dust emitting source regions. Contr. Atmos. Phys., 72(1), 67­77. 3.2 Gillette, D. A., B. Marticorena and G. Bergametti, 1998: Change in the aerodynamic roughness height by saltating grains: Experimental assessment, test of theory, and operational parameterization. J. Geophys. Res., 103(D6), 6203­6209. 5 Ginoux, P., 2003: Effects of non-sphericity on mineral dust modeling. J. Geophys. Res., 108(D2), 4052, doi:10.1029/2002JD002516. 4.1, 4.1 Ginoux, P., M. Chin, I. Tegen, J. Prospero, B. Holben, O. Dubovik and S.-J. Lin, 2001: Sources and distributions of dust aerosols simulated with the GOCART model. J. Geophys. Res., 106(D17), 20555­20273. 3.2, 4.1 2


Glantz, M. H., 1994: Drought Follows the Plow: Cultivating Marginal Areas. Cambridge Univ. Press, New York, NY. 1, 2.1, 5 Grini, A. and C. S. Zender, 2004: Roles of saltation, sandblasting, and wind speed variability on mineral dust aerosol size distribution during the Puerto Rican Dust Experiment (PRIDE). In Press in J. Geophys. Res. 3.1, 4.1, 4.1 Grini, A., C. S. Zender and P. Colarco, 2002: Saltation sandblasting behavior during mineral dust aerosol production. Geophys. Res. Lett., 29(18), 1868, doi:10.1029/2002GL015248. 4.1, 4.1 Grosso, R. and G. Greiner, 2000: Hierarchical tetrahedral-octahedral subdivision for volume visualization. The Visual Computer, 16(6), 357­369. 4.2.2 Guelle, W., Y. J. Balkanski, M. Schulz, F. Dulac and P. Monfray, 1998: Wet deposition in a global size-dependent aerosol transport model 1. Comparison of a 1 year 210 Pb simulation with ground measurements. J. Geophys. Res., 103(D10), 11429­11445. 1 Harris, M. J. and A. Lastra, 2001: Real-time cloud rendering. in Proceedings EUROGRAPHICS 2001, pp. 76­84. also in Computer Graphics Forum 20(3). 3.3 Herman, J. R., P. K. Bhartia, O. Torres, C. Hsu, C. Seftor and E. Celarier, 1997: Global distribution of UV-absorbing aerosols from Nimbus 7/TOMS data. J. Geophys. Res., 102(D14), 16911­ 16922. 5 Hibbard, W. and D. Santek, 1990: The VIS-5D system for easy interactive visualization. in Proceedings IEEE Visualization 90, pp. 129­134. Computer Society Press. 3.3 Hogan, T. F. and L. R. Brody, 1993: Sensitivity studies of the Navy's global forecast model parameterizations and evaluation of improvements to NOGAPS. Mon. Weather Rev., 121, 2373­2395. 1, 3.2, 3.2, 4.1 Huang, J., K. Mueller, N. Shareef and R. Crawfis, 2000: Fastsplats: Optimized splatting on rectilinear grids. in Proceedings IEEE Visualization 2000, pp. 219­226. Computer Society Press. 3.3, 4.2.2 Humphreys, G., M. Eldridge, I. Buck, G. Stoll, M. Everett and P. Hanrahan, 2001: WireGL: A scalable graphics system for clusters. in Proceedings ACM SIGGRAPH 2001, pp. 129­140. ACM Press. 4.2.1 Humphreys, G., M. Houston, R. Ng, R. Frank, S. Ahern, P. D. Kirchner and J. T. Klosowski, 2002: Chromium: A stream-processing framework for interactive rendering on clusters. in Proceedings ACM SIGGRAPH 2002, pp. 693­702. ACM Press. 4.3 IBM Research, : Open visualization data explorer. http://www.research.ibm.com/dx/. 3.3 Iversen, J. D. and B. R. White, 1982: Saltation threshold on Earth, Mars, and Venus. Sedimentology, 29, 111­119. 5 Kajiya, J. T. and B. P. Von Herzen, 1984: Ray tracing volume densities. in Proceedings ACM SIGGRAPH 84, pp. 165­174. ACM Press. 3.3, 4.2.2 Kalnay, E., 1996: The NCEP/NCAR 40-year reanalysis project. Bull. Am. Meteorol. Soc., 77(3), 437­471. 4.3 ´ King, M. D., Y. J. Kaufman, D. Tanre and T. Nakajima, 1999: Remote sensing of tropospheric aerosols from space: Past, present, and future. Bull. Am. Meteorol. Soc., 80(11), 2229­2259. 5 Kolivras, K. N., P. S. Johnson, A. C. Comrie and S. R. Yool, 2001: Environmental variability and coccidioidomycisis (valley fever). Aerobiologia, 17(1), 31­42. 1, 4 Lacroute, P. and M. Levoy, 1994: Fast volume rendering using a shear-warp factorization of the viewing transformation. in Proceedings ACM SIGGRAPH 94, pp. 451­458. ACM Press. 3.3 Lario, R., R. Pajarola and F. Tirado, 2003: Hyperblock-QuadTIN: Hyper-block quadtree based triangulated irregular networks. in Proceedings IASTED Invernational Conference on Visualization, Imaging and Image Processing (VIIP 2003), pp. 733­738. 4.2.1 3


Laur, D. and P. Hanrahan, 1991: Hierarchical splatting: A progressive refinement algorithm for volume rendering. in Proceedings ACM SIGGRAPH 91, pp. 285­288. ACM Press. 4.2.2 Lengen, R. v., J. Meyer, M. Matzat and H. Hagen, 1998: Shadow generation for volumetric data sets. The Visual Computer, 14(2), 83­94. 4.2.2 Levenberg, J., 2002: Fast view-dependent level-of-detail rendering using cached geometry. in Proceedings IEEE Visualization 2002, pp. 259­266. Computer Society Press. 3.3 Levoy, M., 1990: Efficient ray tracing of volume data. ACM Transactions on Graphics, 9(3), 245­ 261. 3.3 Li, P. P., W. H. Duquette and D. W. Curkendall, 1996: RIVA: A versatile parallel rendering system for interactive scientific visualization. IEEE Transactions on Visualization and Computer Graphics, 2(3), 186­201. 3.3 Li, P. P., S. Whitman, R. Mendoza and J. Tsiao, 1997: ParVox: A parallel splatting volume rendering system for distributed visualization. in Proceedings IEEE Parallel Rendering Symposium, pp. 7­14. 3.3 Lindstrom, P., D. Koller, W. Ribarsky, L. F. Hodges, N. Faust and G. A. Turner, 1996: Real-time, continuous level of detail rendering of height fields. in Proceedings SIGGRAPH 96, pp. 109­ 118. ACM SIGGRAPH. 3.3 Lindstrom, P. and V. Pascucci, 2001: Visualization of large terrains made easy. in Proceedings IEEE Visualization 2001, pp. 363­370. Computer Society Press. 3.3 Luo, C., N. M. Mahowald, C. S. Zender and J. del Corral, 2003: A 22-year climatology of mineral aerosols. Submitted to J. Geophys. Res. 4.1, 5 Mahowald, N. M., C. Luo, J. del Corral and C. S. Zender, 2003: Interannual variability in atmospheric mineral aerosols from a 22-year model simulation and observational data. J. Geophys. Res., 108(D12), 4352, doi:10.1029/2002JD002821. 4.1, 5 Mahowald, N. M., C. S. Zender, C. Luo, D. Savoie, O. Torres and J. del Corral, 2002: Understanding the 30 year Barbados desert dust record. J. Geophys. Res., 107(D21), 4561, doi:10.1029/2002JD002097. 4.1, 5 Maring, H., D. L. Savoie, M. A. Izaguirre, L. Custals and J. S. Reid, 2003: Mineral dust aerosol size distribution change during atmospheric transport. In Press in J. Geophys. Res. 4.1 Marticorena, B. and G. Bergametti, 1995: Modeling the atmospheric dust cycle: 1. Design of a soil-derived dust emission scheme. J. Geophys. Res., 100(D8), 16415­16430. 3.2 ´ Marticorena, B., G. Bergametti, B. Aumont, Y. Callot, C. N'Doume and M. Legrand, 1997: Modeling the atmospheric dust cycle: 2. Simulation of Saharan dust sources. J. Geophys. Res., 102(D4), 4387­4404. 1, 3.2 Martin, J. H. and S. E. Fitzwater, 1988: Iron-deficiency limits phytoplankton growth in the northeast Pacific subarctic. Nature, 331(6154), 341­343. 3.2 Mbourou, G. N., J. J. Bertrand and S. E. Nicholson, 1997: The diurnal and seasonal cycles of wind-borne dust over Africa north of the equator. J. Appl. Meteorol., 36, 868­882. 3.2 Meissner, M., J. Huang, D. Bartz, K. Mueller and R. Crawfis, 2000: A practical evaluation of popular volume rendering algorithms. in Proceedings ACM Symposium on Volume Visualization 2000, pp. 81­90. 3.3 Meyer, J., R. Borg, I. Takanashi, E. Lum and B. Hamann, 2003: Segmentation and texture-based hierarchical rendering techniques for large-scale real-color biomedical image data. in Data Visualization - The State of the Art, pp. 169­182. Kluwer Academic Publishers. 4.2.2 Molnar, S., M. Cox, D. Ellsworth and H. Fuchs, 1994: A sorting classification of parallel rendering. IEEE Computer Graphics and Applications, 14(4), 23­32. 4.2.1 Moore, J. K., S. C. Doney, D. M. Glover and I. Y. Fung, 2002: Iron cycling and nutrient-limitation 4


patterns in surface waters of the World Ocean. Deep-Sea Res. II, 49, 463­507. 3.2 ¨ Mueller, K., T. Moller and R. Crawfis, 1999a: Splatting without the blur. in Proceedings IEEE Visualization 99, pp. 363­370. Computer Society Press. 3.3, 4.2.2 ¨ Mueller, K., T. Moller, J. E. Swan II, R. Crawfis, N. Shareef and R. Yagel, 1998: Splatting errors and antialiasing. IEEE Transactions on Visualization and Computer Graphics, 4(2), 178­191. 3.3, 4.2.2 Mueller, K., N. Shareef, J. Huang and R. Crawfis, 1999b: High-quality splatting on rectilinear grids with efficient culling of occluded voxels. IEEE Transactions on Visualization and Computer Graphics, 5(2), 116­134. 3.3 Neff, J., R. Reynolds, J. Belnap and P. Lamothe, 2004: Multi-decadal impacts of grazing on soil and physical biogeochemical properties in Southeast Utah. Submitted to Ecological Applications. 5 Nicholson, S. E., C. J. Tucker and M. B. Ba, 1998: Desertification, drought, and surface vegetation: An example from the West African Sahel. Bull. Am. Meteorol. Soc., 79(5), 815­829. 5 Nishita, T. and Y. Dobashi, 2001: Modeling and rendering of various natural phenomena consisting of particles. in Proceedings Computer Graphics International 2001, pp. 149­156. 3.3 Nishita, T., E. Nakamae and Y. Dobashi, 1996: Display of clouds and snow taking into account multiple anisotropic scattering and sky light. in Proceedings ACM SIGGRAPH 96, pp. 379­386. ACM Press. 3.3 ¨ Orchard, J. and T. Moller, 2001: Accelerated splatting using a 3D adjacency data structure. in Proceedings Graphics Interface 2001, pp. 191­200. 3.3 Pajarola, R., 1998a: Access to large scale Terrain and Image Databases in Geoinformation Sys¨ tems. Ph.D. thesis, Dept. of Computer Science, ETH Zurich. 4.2.1 Pajarola, R., 1998b: Large scale terrain visualization using the restricted quadtree triangulation. in Proceedings IEEE Visualization 98, pp. 19­26,515. 3.3, 4.2.1 Pajarola, R., 2002: Overview of quadtree-based terrain triangulation and visualization. Tech. Rep. UCI-ICS-02-01, Information & Computer Science, University of California Irvine. 3.3, 4.2.1 Pajarola, R., M. Antonijuan and R. Lario, 2002: QuadTIN: Quadtree based triangulated irregular networks. in Proceedings IEEE Visualization 2002, pp. 395­402. Computer Society Press. 4.2.1 Pajarola, R., T. Ohler, P. Stucki, K. Szabo and P. Widmayer, 1998: The alps at your fingertips: Virtual reality and geoinformation systems. in Proceedings 14th International Conference on Data Engineering, ICDE '98, pp. 550­557. IEEE. 4.2.1, 4.3 Pajarola, R. and P. Widmayer, 2001: Virtual geoexploration: Concepts and design choices. International Journal of Computational Geometry and Applications, 11(1), 1­14. 4.2.1, 4.3 Pappagianis, D. and H. Einstein, 1978: Tempest from Tehachapi takes toll or Coccidioides conveyed aloft and afar. West. J. Med., 129(6), 527­530. 1 Prospero, J. M., 1999: Long-term measurements of the transport of African mineral dust to the southeastern United States: Implications for regional air quality. J. Geophys. Res., 104(D13), 15917­15928. 1, 3.2 Prospero, J. M., P. Ginoux, O. Torres, S. E. Nicholson and T. E. Gill, 2002: Environmental characterization of global sources of atmospheric soil dust derived from the NIMBUS7 TOMS absorbing aerosol product. Rev. Geophys., 40(1), 2­1­2­31. 3.2, 5 ~ Prospero, J. M. and R. T. Nees, 1986: Impact of the North African drought and El Nino on mineral dust in the Barbados trade winds. Nature, 320, 735­738. 5 Raphael, M. N., 2003: The Santa Ana winds of California. Earth Interactions, 7(8), 15. 1 Rasch, P. J., W. D. Collins and B. E. Eaton, 2001: Understanding the Indian Ocean Experiment INDOEX aerosol distributions with an aerosol assimilation. J. Geophys. Res., 106(D7), 7337­ 5


7355. 5 Raupach, M. R., D. A. Gillette and J. F. Leys, 1993: The effect of roughness elements on wind erosion threshold. J. Geophys. Res., 98(D2), 3023­3029. 4.1 Reheis, M. C., 1997: Dust deposition downwind of Owens (dry) Lake, 1991­1994: Preliminary findings. J. Geophys. Res., 102(D22), 25999­26008. 5 Riley, K., D. Ebert, C. Hansen and J. Levit, 2003: Visually accurate multi-field weather visualization. in Proceedings IEEE Visualization 2003, pp. 279­286. Computer Society Press. 3.3 Riley, K., D. Ebert, C. Hansen and J. Levit, 2004: A sytstem for realistic weather rendering. in Proceedings of the 84th American Meteorological Society Annual Meeting, p. forthcoming. 3.3 Samanta, R., T. Funkhouser and K. Li, 2001: Parallel rendering with K-way replication. in Proceedings IEEE Symposium on Parallel and Large-Data Visualization and Graphics. Computer Society Press. 4.2.1 Samanta, R., T. Funkhouser, K. Li and J. P. Singh, 2000: Hybrid sort-first and sort-last parallel rendering with a cluster of PCs. in Proceedings Eurographics Workshop on on Graphics Hardware, pp. 97­108. 4.2.1 Samet, H., 1989: The Design and Analysis of Spatial Data Structures. Addison Wesley, Reading, Massachusetts. 4.2.2 Saxton, K., D. Chandler, L. Stetler, B. Lamb, C. Claiborn and B.-H. Lee, 2000: Wind erosion and fugitive dust fluxes on agricultural lands in the Pacific Northwest. Trans. of the ASAE, 43(3), 623­630. 4.1, 4.1, 5 Schaaf, C. B., F. Gao, A. H. Strahler, W. Lucht, X. Li, T. Tsang, N. C. Strugnell, X. Zhang, Y. Jin, J.-P. Muller, P. Lewis, M. Barnsley, P. Hobson, M. Disney, G. Roberts, M. Dunderdale, C. Doll, R. P. d'Entremont, B. Hu, S. Liang, J. L. Privette and D. Roy, 2002: First operational BRDF, albedo nadir reflectance products from MODIS. Remote Sensing of Environment, 83, 135­148. 3.2, 4.2.1 Sehmel, G. A., 1980: Particle and gas dry deposition: A review. Atmos. Environ., 14(9), 983­1011. 1 Shao, Y. and L. M. Leslie, 1997: Wind erosion prediction over the Australian continent. J. Geophys. Res., 102(D25), 30091­30105. 1 Shao, Y., M. R. Raupach and J. F. Leys, 1996: A model for predicting aeolian sand drift and dust entrainment on scales from paddock to region. Aust. J. Soil Res., 34(D25), 309­342. 3.2 Shao, Y., Y. Yang, J. Wang, Z. Song, L. M. Leslie, C. Dong, Z. Zhang, Z. Lin, Y. Kanai, S. Yabuki and Y. Chun, 2003: Northeast Asian dust storms: Real-time numerical prediction and validation. J. Geophys. Res., 108(D22), 4691, doi:10.1029/JD003667. 1 Shirley, P. and A. Tuchman, 1990: A polygonal approximation to direct scalar volume rendering. Computer Graphics (Workshop on Volume Visualization), 24(5), 63­63. 3.3 Sokolik, I. N., D. M. Winker, G. Bergametti, D. A. Gillette, G. Carmichael, Y. J. Kaufman, L. Gomes, L. Schuetz and J. E. Penner, 2001: Introduction to special section: Outstanding problems in quantifying the radiative impacts of mineral dust. J. Geophys. Res., 106(D16), 18015­ 18027. 3.2 Stamnes, K., S.-C. Tsay, W. Wiscombe and K. Jayaweera, 1988: Numerically stable algorithm for discrete-ordinate-method radiative transfer in multiple scattering and emitting layered media. Appl. Opt., 27(12), 2502­2509. 4.2.2 Sweeney, M. R., A. J. Busacca and D. Gaylord, 2003a: High accumulations rates and the generations of thick Palouse loess via topographic traps, Juniper Canyon. Proc. Geol. Soc. Amer. 2003 Meeting. 3.1 Sweeney, M. R., A. J. Busacca, D. R. Gaylord and C. Zender, 2002: Provenance of Palouse loess related to late quaternary glacial outburst flooding in the Pacific Northwest. Eos, Trans. Amer. 6


Geophys. Union, 83(47), H22B­0899. 3.1 Sweeney, M. R., A. J. Busacca, C. A. Richardson, M. Blinnikov and E. McDonald, 2003b: The Columbia Plateau dust engine during the last glacial maximum: Trouble with cold starts. Proc. XVI International Quaternary Association (INQUA) Congress. 3.1 Tegen, I. and I. Fung, 1994: Modeling of mineral dust in the atmosphere: Sources, transport, and optical thickness. J. Geophys. Res., 99(D11), 22897­22914. 3.2 Tegen, I., A. A. Lacis and I. Fung, 1996: The influence on climate forcing of mineral aerosols from disturbed soils. Nature, 380, 419­422. 3.2 Torres, O., P. K. Bhartia, J. R. Herman, A. Sinyuk, P. Ginoux and B. Holben, 2002: A longterm record of aerosol optical depth from TOMS observations and comparison to AERONET measurements. J. Atmos. Sci., 59(3), 398­413. 5 Tsvetsinskaya, E. A., C. B. Schaaf, F. Gao, A. H. Strahler, R. E. Dickinson, X. Zeng and W. Lucht, 2002: Relating MODIS-derived surface albedo to soils and rock types over Northern Africa and the Arabian peninsula. Geophys. Res. Lett., 29(9), doi:10.1029/2001GL014096. 4.2.1 van Donk, S. J., X. Huang, E. L. Skidmore, A. B. Anderson, D. L. Gebhart, V. E. Prehoda, and E. M. Kellogg, 2003: Wind erosion from military training lands in the Mojave Desert, California, U.S.A. J. Arid Env., 54, 687­703. 1, 5, 4 Vezina, G. and P. K. Robertson, 1991: Terrain perspectives on a massively parallel SIMD computer. in Proceedings CG International 91, pp. 163­188. 3.3 ´ Wald, A. E., Y. J. Kaufman, D. Tanre and B.-C. Gao, 1998: Daytime and nighttime detection of mineral dust over desert using infrared spectral contrast. J. Geophys. Res., 103(D24), 32307­ 32313. 1, 5 Westover, L., 1990: Footprint evaluation for volume rendering. in Proceedings SIGGRAPH 90, pp. 367­376. ACM SIGGRAPH. 3.3, 4.2.2 Wittenbrink, C. M., 1998: Survey of parallel volume rendering algorithms. in Proceedings Parallel and Distributed Processing Techniques and Applications, pp. 1329­1336. 3.3 Youlin, Y., V. Squires and L. Qi, 2002: Global alarm: Dust and sandstorms from the World's drylands. Tech. rep., United Nations Convention to Combat Desertification (UNCCD), Bangkok, Thailand. 1, 2.1, 5 Zender, C. S., 1999: Global climatology of abundance and solar absorption of oxygen collision complexes. J. Geophys. Res., 104(D20), 24471­24484. 4.2.2 Zender, C. S., 2003: Dust Entrainment and Deposition (DEAD) model. "http://dust.ess. uci.edu/dead". 4.1 Zender, C. S., H. Bian and D. Newman, 2003a: Mineral Dust Entrainment And Deposition (DEAD) model: Description and 1990s dust climatology. J. Geophys. Res., 108(D14), 4416, doi:10.1029/2002JD002775. 1, 3.2, 4.1, 4.1, 5 Zender, C. S., B. Bush, S. K. Pope, A. Bucholtz, W. D. Collins, J. T. Kiehl, F. P. J. Valero and J. Vitko, Jr., 1997: Atmospheric absorption during the Atmospheric Radiation Measurement (ARM) Enhanced Shortwave Experiment (ARESE). J. Geophys. Res., 102(D25), 29901­29915. 4.2.2 Zender, C. S., M. Flanner and J. Adams, 2003b: LGM dust distribution and radiative forcing: Sensitivity to vegetation reconstruction. Proc. XVI International Quaternary Association (INQUA) Congress. 3.1 Zender, C. S., D. J. Newman and O. Torres, 2003c: Spatial heterogeneity in aeolian erodibility: Uniform, topographic, geomorphic, and hydrologic hypotheses. J. Geophys. Res., 108(D17), 4543, doi:10.1029/2002JD003039. 4.1 Zender, C. S. and J. Talamantes, 2004: Environmental factors controlling Valley Fever incidence in Kern County, CA. Manuscript in Preparation, "http://dust.ess.uci.edu/ppr/ppr_ 7


ZeT04.pdf". 1, 4 Zwicker, M., H. Pfister, J. van Baar and M. Gross, 2001: EWA volume splatting. in Proceedings IEEE Visualization 2001, pp. 29­36. Computer Society Press. 3.3, 4.2.2

8


6.1 Budget Justification % NB: Do not use LaTeX formatting in Budget Justification since must % upload into Liz's Word document PERSONNEL Dr. Zender is the lead PI and will oversee the planning and coordination of the project. He will have primary responsibility for the forecasting work and will lead interactions with decision-makers and environmental planners. One month of summer salary each year is requested. Dr. Pajarola is the Co-PI and will have lead responsibilities for the work on visualization. One month of summer salary each year is requested. Dr. Meyer is a Senior Personnel who will have particular responsibility for work on volume rendering. One month of summer salary each year is requested. Rich Reynolds is with the USGS in Denver and is an expert on the Mojave Desert ROI. Pat Chavez is with the USGS in Flagstaff and is an expert on using satellite imagery to study wind erosion, with particular reference to the Mojave Desert. Ualikhan Abdibekov is with the Institute of Mathematics, Almaty, Kazakhstan, and leads their nationally recognized modeling studies on the dessication, deflation, and salinization of the Aral Sea, our other ROI. Katie Purvis is in the Joint Science Department of the Claremont Colleges; her group will use the LEP to understand the distribution of wind-eroded and radionuclide-rich sediment originating near the Aral Sea and affecting Kazakhstani health. These senior personnel will provide expertise and assistance in the field but will draw no salary from the project. To Be Named---Postdoctoral Scholar. 100\% effort will be contributed to the project. Salary is based on a published scale of \$50,472 annually. The post-doc will conduct original research applying mesoscale forecasts to environmental decisions in the ROIs. They will help train the graduate students in this endeavor. To be Named---Graduate Student Researchers~III. Funds are requested to support three non-resident graduate students each year of the project. Salary is estimated using the published scale of \$34,956 annually for GSR III, the level for students who are past the Masters degree but not yet advanced to candidacy. The students will be employed 49\% during the AY and 100\% during the summer, the maximum permissible per UC policy. One student will work with Zender in understanding/improving arid region forecasts and two will assist Pajarola in the visualization work. Student support for Meyer will be sought from other sources.


7

Facilities, Equipment, and Other Resources

7.1 Computer and Networking LEP is well-situated to take advantage of UCI's fastest network connections. The UCI Network Infrastructure provides researchers with 1.0 Gb s-1 access to the high-performance network of Cal(IT)2 and to the Gb-backbone of UCINet. UCI will upgrade this link to 10 Gb s-1 in the near future. This will remove one potential bottleneck to the Forecast Cluster (FC) which was designed to deliver about 1.5 Gb s-1 in a typical configuration. The Cal-(IT)2 building is designed with redundant Gb Ethernet links to the UC Irvine backbone and will support one and ten Gb s-1 links to other research facilities at UCI. A conceptual network called CalREN-XD (Experimental Development) that will leverage the visualization capabilities of the LEP is under development. An example of this is the "OptIPuter", which establishes a private, direct link between UCSD and UCI over the CalREN-HPR or a CalREN-XD circuit. The main impact of the OptIPuter is to operate a computer that has geographically distributed components. The visualization and compute clusters of this ITR will also be interconnected to the OptIPuter. Theoretically, this will allow the Visualization cluster to render on geographically remote displays such as at UCSD. This capability would prove useful in times of natural (e.g., dust storm) or anthropogenic emergency (e.g., radiologic plumes). PI Zender is director of the Earth System Modeling Facility (ESMF), an NSF-supported MRI facility dedicated to coupled global climate, chemistry, and biogeochemistry simulations. The ESMF is an 88-CPU Power4+ IBM supercomputer with 192 GB RAM and 32 TB of RAID storage. Although funding was not requested in this proposal to perform real and accelerated time visualizations of coupled global model simulations, the ESMF is available for groups who are funded for such visualizations. The ESMF will be made available for LEP forecasting if ESMF nodes would otherwise be unused, and if short forecasting benchmarks are needed on machines faster than the forecasting cluster. Should LEP entrain visualization and decision experts interested in global-scale applications, the ESMF will gladly make available accounts and data. Funding for bi-directional 2 Gb s-1 connections between the ESMF and UCI's Campus portal is requested as part of PI Zender's pending SEI proposal. Co-PI Pajarola is the founder of the Computer Graphics Lab (CGL) in the School of Infor mation and Computer Sciences (ICS) and is part of the Visualization Trust at Cal-(IT)2 Senior personnel Meyer leads the Creative Interactive Visualization Laboratory (CIVL) research group. The two research labs CGL and CIVL offer various mid-range to high-end graphics workstations, display systems, video editing equipment and printing to support research and software development in scientific visualization. CGL hosts a basic 10-node 3 × 3 tiled-display rendering cluster infrastructure that is at the core of the proposed visualization system. As part of this proposal this cluster will be upgraded to satisfy the computational infrastructure needs of LEP. 7.2 Maintenance and Technical Support Network and Academic Computing Services NACS is the largest IT organization at UCI. Dr. Frank Wessel manages the NACS Research Computing Support Group (RCS). RCS provides customized support and facilitates user access to high-performance computing (HPC) resources, software, training, and development of the UCI research infrastructure. NACS RCS staff led by Dr. Wessel will facilitate the design, set-up, acquisition, and networking of the Forecast and Visualization clusters.


NACS provides free system administration, co-location in the NACS Machine Room, and software sharing to researchers who partner with NACS on Medium Performance Computing (MPC) Beowulf Clusters. In exchange, researchers contribute 25% of their compute cycles back to the campus. The principal benefit to researchers in joining the MPC partnership is to obtain a robust computing environment at low-recurring cost. The Forecast Cluster (FC) will be situated in the MPC cluster, co-located in the NACS machine room in a NACS-provided rack. The Visualization Cluster (VC) will be remotely located from the Forecast Cluster in Prof. Pajarola's laboratory, immediately adjacent to the visualization system. The VC will be administered by IT staff in the School of Information and Computer Science. To implement the required network connection speed, NACS will upgrade network facilities with additional switches and interconnects provided for in the budget.

2


8

Acronyms and Abbreviations
Table 2: Acronyms and Abbreviations Abbreviation ACE AMWG ANWR ASE BRDF Cal-(IT)2 CAM CARRE SDSU CCSM CDHS CE CGL CHRS CIVL CLM CVS DEAD DWP EECS ESIG ESMF ESS FAA FC FOI FSU FV GB GCM GOES Gb IC ICS IOM ISTC JPL Description Army Corps of Engineers (CCSM) Atmospheric Model Working Group Arctic National Wildlife Refuge Advanced Science and Engineering Bi-directional Reflectance Distribution Function California Institute for Telecommunications and Information Technology Community Atmosphere Model Central Asia Research and Remediation Exchange San Diego State University Community Climate System Model California Department of Health Services Civil Engineering Computer Graphics Lab Center for Hydrometeorology and Remote Sensing Creative Interactive Visualization Laboratory Common Land Model Concurrent Versions System Dust Entrainment And Deposition Model Department of Water and Power Electrical Engineering and Computer Science Environmental and Societal Impacts Group Earth System Modeling Facility Earth System Science (Department) Federal Aviation Administration Forecast Cluster Field Of Interest Former Soviet Union Field site Visits Gigabyte General Circulation Model Geostationary Earth Orbiting Satellite Gigabit International Conferences Information and Computer Sciences Institute of Mathematics (Almaty, Kazakhstan) International Science and Technology Center Jet Propulsion Laboratory


Table 2: (continued) Abbreviation Description

LEP Laboratory for Environmental Planning LOD Level-of-detail MAE Mechanical and Aerospace Engineering MISR Multi-angle Imaging Spectro-Radiometer (satellite instrument) MNP Mojave National Preserve MODIS Moderate Resolution Imaging Spectroradiometer (satellite instrument) MPC Medium Performance Computing NACS Network and Computing Services NASA National Aeronautic and Space Administration NBC Nuclear, Biological, and Chemical weapons NC National Conferences NCAR National Center for Atmospheric Research NCEP National Center for Environmental Prediction NHS National and Homeland Security NPS National Park Service PI Principle Investigator RAID Redundant Array of Independent Disks RCS Research Computing Services RESCUE Responding to Crises and Unexpected Events RGB Red Green Blue ROI Region Of Interest RT Radiative Transfer SC4 = SC4 Southern California Climate Change Consortium SCAQMD South Coast Air Quality Management District SCCOOS Southern California Coastal Ocean Observing System SDSC San Diego Supercomputer Center SEI Science and Engineering Informatics SP Senior Personnel SS Saltation-Sandblasting SW Southwest SW Summer Workshops TB Terabyte UNCCD United Nations Convention to Combat Desertification UV User Visits to LEP VC Visualization Cluster WMD Weapons of Mass Destruction WRF Weather Research and Forecasting (model)

2


9

List of All Personnel Associated with Proposal, Collaborators and CoEditors of Project Senior Personnel, their Post-docs, and their Thesis Advisors
Adam, David, USGS, retired Agrawal, Divyakant, Computer Science, University of California Santa Barbara Ammann, C. A. (NCAR) Artemyev, O National Nuclear Center, Kurchatov, Kazakhstan Ayuso, Robert USGS, Reston Belnap, Jayne USGS, Moab Bernasek, S.L. Princeton University Bian, H. (NASA/UMBC) Bielak, Jacobo (Carnegie Mellon University, CEE) Bocarsly, Andrew Princeton University Bonan, G. B. (NCAR) Bradbury, J.Platt, USGS, Denver Busacca, A. (WSU) Callender, Edward USGS, RI Canagaratna, M Aerodyne Research, Inc., Billerica, MA Carlsen, Tina M. Lawrence Livermore National Laboratory Chambers, Douglas B. USGS, Charleston, WV Chavez, Pat, Jr., USGS, Flagstaff Clow, Gary USGS, Denver Colarco, P. (GSFC) Collins, W. D. (NCAR) Colman, Steven USGS, Woods Hole Cooper, W. A. (NCAR) Cullen, Alison University of Washington Dean, Walter USGS, Denver Dillner, Ann Arizona State University ElZarki, Magda, Computer Science, University of California Irvine El Abbadi, Amr, Computer Science, University of California Santa Barbara Famiglietti, J. (UCI) Fenves, Gregory L. (UC Berkeley, CEE) Forester, Richard USGS, Denver Fulton, Robert California Desert Consortium, Cal State Fullerton Gawalt, Ellen University of Chicago Gaylord, D. (WSU) Gerstner, Thomas, Applied Mathematics, University of Bonn Ghertner, Asher University of California, Berkeley Gill, Tom Texas Tech Univ. Goldhaber, Martin USGS, Denver Goldstein, Harland USGS, Denver Grini, A. (U. Oslo) Gross, Markus, Computer Graphics Lab, ETH Zurich


Guidotti, Patrick, Mathematics, University of California Irvine Hagen, Hans (University of Kaiserslautern, Germany, CS) Hamann, Bernd (UC Davis, Center for Image Processing and Integrated Computing) Harriss, Robert C. National Center for Atmospheric Research, Boulder, CO Herndon, Scott Aerodyne Research, Inc., Billerica, MA Hinkley, Todd USGS, Denver Ibraev, Nurlan State Agency for Health Care in East-Kazakhstan Oblast Jayne, John Aerodyne Research, Inc., Billerica, MA Jimenez, Jose University of Colorado, Boulder Jones, Edward G. (UC Davis, Center for Neuroscience) Joy, Kenneth I. (UC Davis, CS) Jumba, Isaac O. University of Nairobi, Kenya Kammen, Daniel M. University of California, Berkeley Kerwin, Micahel Univ. of Denver Kiehl, J. T. (NCAR) Kiehl, J. T. (NCAR) Kolb, Chuck Aerodyne Research, Inc., Billerica, MA Kuester, F. (UCI) Lamothe, Paul USGS, Denver Lancaster, Nick Desert Research Inst (Reno) and USGS, Reston Larson, Edwin Univ. of Colorado Lu, Gang Millennium Chemicals, Baltimore, MD Luiszer, Fred Univ. of Colorado MacKinnon, David USGS, Flagstaff Mahowald, N. M. (NCAR) Meeker, Greg USGS, Denver Meenakshisundaram, Gopi, Computer Science, University of California Irvine Miller, Douglas Naval Postgraduate School, Monterey, CA Miller, Mark E. USGS, Moab Moore, J. K. (UCI) Moscato, Silvina (most recent: USGS, Denver) Neff, Jason Univ. of Colorado Okin, G. (U. Virginia) Okin, Greg Univ. of Virginia Olson, Arthur J. (The Scripps Research Institute, La Jolla) Orsini, Douglas Metrohm Peak Pajarola, R. (UCI) Peterson, Leif E. Baylor College of Medicine Phillips, Susan USGS, Moab Rapp, Joshua (most recent: Univ. Vermont) Rasch, P. J. (NCAR) Reheis, Marith USGS, Denver Robert C. Harriss National Center for Atmospheric Research Roberts, Helen Univ. of Wales Rosenbaum, Joseph USGS, Denver 2


Rossignac, Jarek, Graphics Visualization & Usability Center, Georgia Tech Sainz, Miguel, Computer Science, University of California Irvine Sanford, Robert Univ. of Denver Schwartz, J.S. Princeton University Sharber, Anna C. National Research Council Sides, Stuart USGS, Flagstaff Silva, Phil Utah State University Soltesz, Deborah USGS, Flagstaff Steven L. Bernasek Princeton University Stojadinovic, Bozidar (UC Berkeley, CEE) Stucki, Peter, Multimedia Laboratory, University of Zurich Susin, Antonio, Mathematics, Universita Politecnica de Catalunya Barcelona Thomas, G. T. (CU) Tigges, Richard USGS, retired Tirado, Francisco, Computer Science, Complutense University Madrid Torres, O. (NASA GSFC) Ulsh, Brant A. McMaster University, Ontario Canada Urban, Frank USGS, Denver Valero, F. P. J. (Scripps) Velasco, Miguel USGS, Flagstaff Wandiga, Shem University of Nairobi, Kenya Weber, Rodney Georgia Institute of Technology Werner, Cynthia A. Texas A&M University, College Station Widmayer, Peter, Theoretical Computer Science, ETH Zurich Wilhelmi, Olga V. National Center for Atmospheric Research, Boulder, CO Wischgoll, Thomas (UC Irvine, EECS) Worsnop, Douglas Aerodyne Research, Inc., Billerica, MA Yount, James USGS, Denver Yu, S. (Duke) Zender, Charlie, Earth System Sciences, University of California Irvine Zhang, Junfeng Rutgers University, Piscataway, NJ

3


List is Alphabetical by Surname. Collaborators of Zender: Ammann, C. A. (NCAR) Bian, H. (NASA/UMBC) Bonan, G. B. (NCAR) Busacca, A. (WSU) Colarco, P. (GSFC) Collins, W. D. (NCAR) Famiglietti, J. (UCI) Gaylord, D. (WSU) Grini, A. (U. Oslo) Kiehl, J. T. (NCAR) Kuester, F. (UCI) Mahowald, N. M. (NCAR) Moore, J. K. (UCI) Okin, G. (U. Virginia) Pajarola, R. (UCI) Rasch, P. J. (NCAR) Valero, F. P. J. (Scripps) Yu, S. (Duke) Torres, O. (NASA GSFC) Thomas, G. T. (CU) Kiehl, J. T. (NCAR) Cooper, W. A. (NCAR) Collaborators of Pajarola: El Abbadi, Amr, Computer Science, University of California Santa Barbara Agrawal, Divyakant, Computer Science, University of California Santa Barbara ElZarki, Magda, Computer Science, University of California Irvine Gerstner, Thomas, Applied Mathematics, University of Bonn Guidotti, Patrick, Mathematics, University of California Irvine Meenakshisundaram, Gopi, Computer Science, University of California Irvine Susin, Antonio, Mathematics, Universita Politecnica de Catalunya Barcelona Tirado, Francisco, Computer Science, Complutense University Madrid Zender, Charlie, Earth System Sciences, University of California Irvine Rossignac, Jarek, Graphics Visualization & Usability Center, Georgia Tech Widmayer, Peter, Theoretical Computer Science, ETH Zurich Gross, Markus, Computer Graphics Lab, ETH Zurich Stucki, Peter, Multimedia Laboratory, University of Zurich Sainz, Miguel, Computer Science, University of California Irvine (postdoc) Collaborators of Meyer: Bielak, Jacobo (Carnegie Mellon University, CEE) Fenves, Gregory L. (UC Berkeley, CEE) Jones, Edward G. (UC Davis, Center for Neuroscience) 4


Joy, Kenneth I. (UC Davis, CS) Olson, Arthur J. (The Scripps Research Institute, La Jolla) Stojadinovic, Bozidar (UC Berkeley, CEE) Hamann, Bernd (UC Davis, Center for Image Processing and Integrated Computing), post-doc. adv. Hagen, Hans (University of Kaiserslautern, Germany, CS), graduate advisor Wischgoll, Thomas (UC Irvine, EECS), postgraduate scholar (total number: 1) Collaborators of Purvis: Artemyev, O National Nuclear Center, Kurchatov, Kazakhstan Bernasek, S.L. Princeton University Bocarsly, Andrew Princeton University Canagaratna, M Aerodyne Research, Inc., Billerica, MA Carlsen, Tina M. Lawrence Livermore National Laboratory Cullen, Alison University of Washington Dillner, Ann Arizona State University Gawalt, Ellen University of Chicago Ghertner, Asher University of California, Berkeley Harriss, Robert C. National Center for Atmospheric Research, Boulder, CO Herndon, Scott Aerodyne Research, Inc., Billerica, MA Ibraev, Nurlan State Agency for Health Care in East-Kazakhstan Oblast Jayne, John Aerodyne Research, Inc., Billerica, MA Jumba, Isaac O. University of Nairobi, Kenya Jimenez, Jose University of Colorado, Boulder Kammen, Daniel M. University of California, Berkeley Kolb, Chuck Aerodyne Research, Inc., Billerica, MA Lu, Gang Millennium Chemicals, Baltimore, MD Orsini, Douglas Metrohm Peak Peterson, Leif E. Baylor College of Medicine Schwartz, J.S. Princeton University Sharber, Anna C. National Research Council Silva, Phil Utah State University Ulsh, Brant A. McMaster University, Ontario Canada Wandiga, Shem University of Nairobi, Kenya Weber, Rodney Georgia Institute of Technology Werner, Cynthia A. Texas A&M University, College Station Wilhelmi, Olga V. National Center for Atmospheric Research, Boulder, CO Worsnop, Douglas Aerodyne Research, Inc., Billerica, MA Zhang, Junfeng Rutgers University, Piscataway, NJ Steven L. Bernasek Princeton University (Graduate advisor) Robert C. Harriss National Center for Atmospheric Research (Post-Doctoral advisor) Collaborators of Reynolds: Adam, David, USGS, retired Ayuso, Robert USGS, Reston 5


Belnap, Jayne USGS, Moab Bradbury, J.Platt, USGS, Denver Callender, Edward USGS, RI Chambers, Douglas B. USGS, Charleston, WV Chavez, Pat, Jr., USGS, Flagstaff Clow, Gary USGS, Denver Colman, Steven USGS, Woods Hole Dean, Walter USGS, Denver Forester, Richard USGS, Denver Fulton, Robert California Desert Consortium, Cal State Fullerton Gill, Tom Texas Tech Univ. Goldhaber, Martin USGS, Denver Goldstein, Harland USGS, Denver Hinkley, Todd USGS, Denver Kerwin, Micahel Univ. of Denver Lamothe, Paul USGS, Denver Lancaster, Nick Desert Research Inst (Reno) and USGS, Reston Luiszer, Fred Univ. of Colorado MacKinnon, David USGS, Flagstaff Meeker, Greg USGS, Denver Miller, Douglas Naval Postgraduate School, Monterey, CA Miller, Mark E. USGS, Moab Moscato, Silvina (most recent: USGS, Denver) Neff, Jason Univ. of Colorado Okin, Greg Univ. of Virginia Phillips, Susan USGS, Moab Rapp, Joshua (most recent: Univ. Vermont) Reheis, Marith USGS, Denver Rosenbaum, Joseph USGS, Denver Roberts, Helen Univ. of Wales Sanford, Robert Univ. of Denver Tigges, Richard USGS, retired Sides, Stuart USGS, Flagstaff Soltesz, Deborah USGS, Flagstaff Urban, Frank USGS, Denver Velasco, Miguel USGS, Flagstaff Yount, James USGS, Denver Larson, Edwin Univ. of Colorado

6


9.1 1. 2. 3. 4. 5. 6. 7. 8. 9.

Supplementary Documents ${DATA}/prp/prp_itr/prp_itr_ltr_nsf.tex ${DATA}/prp/prp_itr/prp_itr_ltr_abdibekov.pdf ${DATA}/prp/prp_itr/prp_itr_ltr_purvis.pdf ${DATA}/prp/prp_itr/prp_itr_ltr_chavez.doc ${DATA}/prp/prp_itr/prp_itr_ltr_sorooshian.pdf ${DATA}/prp/prp_itr/prp_itr_ltr_neff.pdf ${DATA}/prp/prp_itr/prp_itr_clb.pdf ${DATA}/prp/prp_itr/prp_itr_ltr_mehrotra.pdf ${DATA}/prp/prp_itr/prp_itr_abb.pdf


Table 1: Researchers Affiliated with LEP ITR Project Institutionb UCI ESS Researcher Charlie Zender LoIc PI ROI/FOIa MD AS SVd × × EPe × Relevant Interests Aeolian erosion, transport, deposition, composition, valley fever Scientific visualization, terrain rendering Aral Sea dust forecasting, amelioration, salinization SW US erosion, satellite, photographic monitoring of source regions Regional air quality/atmospheric chemistry/aerosol modeling of LA air-shed Aral Sea monitoring, assessment, policy Scientific visualization, interactive systems, VR Crisis/Hazard Response and Management Large-scale visualization, volume rendering, VR SW US land use change, aeolian erosion and biogeochemistry Radionuclide dispersion/exposure via dust; health Climate change and land use in SW US, station obs. of saltation, wind Hydrology, sustainability of semi-arid landscapes

UCI ICS IOMf USGS

Renato Pajarola Ualikhan Abdibekov Pat Chavez

Co-PI SP Coll. × ×

×

UCI MAE

Donald Dabdub

Co-PI

×

SDSU Geology UCI EECS UCI ICS UCI EECS CU/USGS

Eric Frost Falko Kuester Sharad Mehrotra Joerg Meyer Jason Neff

Coll. User User Co-PI Supp.

×

× × × ×

×

Claremont

Katie Purvis

Coll.

×

×

USGS

Richard Reynolds Soroosh Sorooshian

SP

×

UCI CE/ESS
a

User

×

×

Regions and Fields of Interest and Expertise: MD = Mojave Desert; AS = Aral Sea; SV = Scientific Visualization; EP = Environmental Planning b Primary institutional and departmental affiliation c Level of Involvement: PI = Principle Investigator; SP = Senior Personnel, integral to accomplishing project goals; Coll.= Collaborator, providing or using information/expertise important to project success; Supp.= Supporter, directs ~ ~ projects that may benefit from LEP; User, will use LEP facility or data at no cost to project. d Interests/expertise in scientific visualization, interactive systems, and virtual reality (VR). e Interests/expertise in environmental planning, weather-society interaction, emergency response f Institute of Mathematics, Kazakhstan

2