Evaluation of wind fields used in Grand Canyon Visibility Transport Commission analyses

The Grand Canyon Visibility Transport Commission (GCVTC) was established by the U.S. Congress to assess the potential impacts of projected growth on atmospheric visibility at Grand Canyon National Park and to make recommendations to the U.S. Environmental Protection Agency on what measures could be taken to avoid such adverse impacts. A critical input to the assessment tool used by the commission was three-dimensional model-derived wind fields used to transport the emissions. This paper describes the evaluation of the wind fields used at various stages in the assessment. Wind fields evaluated included those obtained from the Colorado State University Regional Atmospheric Modeling System (RAMS), the National Meteorological Center's Nested Grid Model (NGM), and the National Oceanic and Atmospheric Administration's Atmospheric Transport and Dispersion (ATAD) trajectory model. The model-derived wind fields were evaluated at multiple vertical levels at several locations in the southwestern United States by determining differences between model predicted winds and winds that were measured using radiosonde and radar wind profiler data. Model-derived winds were also evaluated by determining the percent of time that they were within acceptable differences from measured winds. All models had difficulties, generally meeting the acceptable criteria for less than 50% of the predictions. The RAMS model had a persistent bias toward southwesterly winds at the expense of other directions, especially failing to represent channeling by north-south mountain ranges in the lower levels. The NGM model exhibited a substantial bias in the summer months by extending northwesterly winds in the eastern Pacific Ocean well inland, in contrast to the observed southwesterlies at inland locations. The simpler ATAD trajectory model performed somewhat better than the other models, probably because of its use of more upper air sites. The results of the evaluation indicated that these wind fields could not be used to reliably predict source-receptor impacts on a particular day; thus, seasonally averaged impacts were used in the GCVTC assessment.     

Clean Air Corridors: A Geographic and

Meteorological factors, pollutant emissions, and geographic regions related to transport of low optical extinction coefficient air to Grand Canyon National Park were examined. Back trajectories were generated by two models, the Atmospheric Transport and Dispersion Model (ATAD) and an approach using the Nested Grid Model output for a Lagrangian particle transport model (NGM/ CAPITA). Meteorological information along the trajectories was analyzed for its relationship to visibility at the Grand Canyon. Case studies considered days with anomalously clean air from the southwest and dirty air from the northwest. Clean air was most frequently from the north and northwest, rarely from the south. Low emissions, high ventilation and washout by precipitation was associated with clean air. All clean days with transport from the Los Angeles area had upper-level low pressure over the region with high ventilation and usually abundant precipitation. The dirtiest days with transport from the northwest were affected by forest fires.     

Assessment of the nested grid model estimates for driving regional visibility models in the southwestern United States

The Nested Grid Model (NGM) is a primitive-equation meteorological model that is routinely exercised over North America for forecasting purposes by the National Meteorological Center. While prognostic meteorological models are being increasingly used to drive air quality models, their use in conducting annual simulations requires significant resources. NGM estimates of wind fields and other meteorological variables provide an attractive alternative since they are typically archived and readily available for an entire year. Preliminary evaluation of NGM winds during the summer of 1992 for application to the region surrounding the Grand Canyon National Park showed serious shortcomings. The NGM winds along the borders between California, Arizona and Mexico tend to be northwesterly with a speed of about 6 m/sec, while the observed flow is predominantly southerly at about 2-5 m/sec. The mesoscale effect of a thermal low pressure area over the highly heated Southern California and western Arizona deserts does not appear to be represented by the NGM because of its coarse resolution and the use of sparse observations in that region. Tracer simulations and statistical evaluation against special high resolution observations of winds in the southwest United States clearly demonstrate the northwest bias in NGM winds and its adverse effect on predictions of an air quality model. The "enhanced" NGM winds, in which selected wind observations are incorporated in the NGM winds using a diagnostic meteorological model provide additional confirmation on the primary cause of the northwest bias. This study has demonstrated that in situations where limited resources prevent the use of prognostic meteorological models, previously archived coarse resolution wind fields in which additional observations are incorporated to correct known biases provide an attractive option.     

Assessment of the Nested Grid Model Estimates for Driving Regional Visibility Models in the Southwestern United States

ABSTRACT

The Nested Grid Model (NGM) is a primitive-equation meteorological model that is routinely exercised over North America for forecasting purposes by the National Meteorological Center. While prognostic meteorological models are being increasingly used to drive air quality models, their use in conducting annual simulations requires significant resources. NGM estimates of wind fields and other meteorological variables provide an attractive alternative since they are typically archived and readily available for an entire year. Preliminary evaluation of NGM winds during the summer of 1992 for application to the region surrounding the Grand Canyon National Park showed serious shortcomings. The NGM winds along the borders between California, Arizona and Mexico tend to be northwesterly with a speed of about 6 m/sec, while the observed flow is predominantly southerly at about 2-5 m/sec. The mesoscale effect of a thermal low pressure area over the highly heated Southern California and western Arizona deserts does not appear to be represented by the NGM because of its coarse resolution and the use of sparse observations in that region. Tracer simulations and statistical evaluation against special high resolution observations of winds in the southwest United States clearly demonstrate the northwest bias in NGM winds and its adverse effect on predictions of an air quality model. The “enhanced” NGM winds, in which selected wind observations are incorporated in the NGM winds using a diagnostic meteorological model provide additional confirmation on the primary cause of the northwest bias. This study has demonstrated that in situations where limited resources prevent the use of prognostic meteorological models, previously archived coarse resolution wind fields in which additional observations are incorporated to correct known biases provide an attractive option.     

Comparison of the ARL/ATAD constant level and the NCAR isentropic trajectory analyses for selected case studies

The following study is a comparison of the National Center for Atmospheric Research (NCAR) isentropic trajectory model and the Air Resources Laboratories Atmospheric Transport and Dispersion (ARL/ATAD) variable height trajectory model. Back-trajectories from each model were computed within the boundary layer and at two levels above the boundary layer over periods in July and December of 1977 and February of 1978. All trajectories terminated in Charlottesville, Virginia.Assessment of the models was achieved through the study of potential vorticity measurements computed along NCAR trajectories, wind shear values computed along the ARL/ATAD trajectories, as well as through consideration of synoptic patterns available during the case study periods. A root mean square (r.m.s.) analysis was performed to quantify model differences.Results of this study show that in the absence of frontal movement the ARL/ATAD model is a better choice within the boundary layer, but only because the model is less expensive to run. NCAR trajectories are probably more realistic in the vicinity of fronts or other large sources of vertical movement. Above the boundary layer, both models produce similar trajectories when the atmosphere is barotropic; NCAR trajectories appear more accurate in baroclinic atmospheres because of better treatment of vertical motion, r.m.s. studies show that NCAR and ARL/ATAD trajectories differ more during winter than summer, especially after 48 h of trajectory calculation r.m.s. trajectory differences remain similar for different levels for a given season and period of time.     

Modeling dispersion from toxic gas released after a train collision in Graniteville, SC

The Savannah River National Laboratory (SRNL) Weather Information and Display System was used to provide meteorological and atmospheric modeling/consequence assessment support to state and local agencies after the collision of two Norfolk Southern freight trains on the morning of January 6, 2005. This collision resulted in the release of several toxic chemicals to the environment, including chlorine. The dense and highly toxic cloud of chlorine gas that formed in the vicinity of the accident was responsible for 9 fatalities and caused injuries to more than 500 others. Transport model results depicting the forecast path of the ongoing release were made available to emergency managers in the county's Unified Command Center shortly after SRNL received a request for assistance. Support continued over the ensuing 2 days of the active response. The SRNL also provided weather briefings and transport/consequence assessment model results to responders from the South Carolina Department of Health and Environmental Control, the Savannah River Site (SRS) Emergency Operations Center, Department of Energy headquarters, and hazard material teams dispatched from the SRS. Operational model-generated forecast winds used in consequence assessments conducted during the incident were provided at 2-km horizontal grid spacing during the accident response. High-resolution Regional Atmospheric Modeling System (RAMS, version 4.3.0) simulation was later performed to examine potential influences of local topography on plume migration in greater detail. The detailed RAMS simulation was used to determine meteorology using multiple grids with an innermost grid spacing of 125 m. Results from the two simulations are shown to generally agree with meteorological observations at the time; consequently, local topography did not significantly affect wind in the area. Use of a dense gas dispersion model to simulate localized plume behavior using the higher-resolution winds indicated agreement with fatalities in the immediate area and visible damage to vegetation.     

The Sixty-Sixth Annual Meeting of the Air Pollution Control Association

Abstract

The U.S. Clean Air Act, amended in 1990, mandated the establishment of the Grand Canyon Visibility Transport Commission (GCVTC). The commission is required to submit a report to the U.S. Environmental Protection Agency addressing visibility issues in the region, including "the establishment of clean air corridors, in which additional restrictions on increases in emissions may be appropriate to protect visibility in affected Class I areas." This paper presents a methodology to identify candidate geographic areas for consideration for Clean Air Corridor (CAC) status for Colorado Plateau Class I areas. The methodology uses thousands of model determined trajectories over a five year period (1988 to 1992) to indicate the paths taken by air that arrives during clean air conditions at Class I areas. These clean air back-trajectories identify upwind areas where pollution emissions could jeopardize currently pristine visibility. Using this methodology, six candidate areas are identified, ranging in size from 75,000 to 506,000 square miles, and permitting varying levels of visibility protection for clean air days at Grand Canyon, Canyonlands, and Petrified Forest National Parks. Assuming effective emissions management of the CAC, the larger the CAC, the greater the visibility protection during clean air conditions.     

Comparison of multi-receptor and single-receptor trajectory source apportionment (TSA) methods using artificial sources

Two trajectory source apportionment methods were tested using an artificially generated data set to determine their ability to detect the known sources. The residence time or conditional probability method developed by Ashbaugh et al. [1985. A residence time probability analysis of sulfur concentrations at Grand Canyon National Park. Atmospheric Environment 19(8), 1263–1270] uses a single receptor at a time, whereas the new multi-receptor (MURA) method developed here uses several receptors at once in an attempt to detect the sources with higher accuracy. The methods were first tested using a simulation with a single source and then with another simulation using four sources. The ability of the methods to detect the sources was quantified for each simulation. The MURA trajectory method proved to be superior at identifying sources for these simulations.     

Numerical study of atmospheric dispersion of a substance released from an industrial complex in the southern coast of Korea

Diurnal variations of wind field and pollutant dispersion in a complex terrain with a shoreline were investigated under the insolation conditions of summer and winter. The area is located in the south of the Korean Peninsula and includes a large petrochemical industrial complex. The Regional Atmospheric Modeling System (RAMS) was used in the simulation study. Initially, horizontally homogeneous wind fields were assumed on the basis of sounding data at the nearby upper-air station for days with morning wind speeds below 2 m s⁻¹. On these days, the sea breeze prevailed in summer while the land breeze lasted for a few hours in the morning; the effect of synoptic winds was strong in winter with some inclusion of wind variations owing to the interaction between sea and land. The predicted wind direction at the location of the weather station captured an important change of the sea/land breeze of the observed one. In the morning, both in summer and winter, complicated wind fields with low wind speeds resulted in high pollutant concentrations almost all over the area. On the other hand, in the afternoon, the wind field was rather uniform and the terrain effects were not significant even in the mountainous area with the development of a mixing layer.     

Comparison of the MURA and an improved single-receptor (SIRA) trajectory source apportionment (TSA) method using artificial sources

Two trajectory source apportionment methods were tested using an artificially generated data set to determine their ability to detect the known sources. The forward-looking step from the multi-receptor trajectory analysis (MURA) method was added to the conditional probability (CP)method of Ashbaugh et al. [1985. A residence time probability analysis of sulfur concentrations at Grand Canyon National Park. Atmospheric Environment 19(8), 1263–1270] to develop the single-receptor forward CP (SIRA) method. The multi-receptor (MURA) and the SIRA methods were tested with three simulations using artificially generated sources. The ability of the methods to detect the sources was quantified for each simulation. The first simulation showed that the SIRA method is an improvement over the original CP method. The MURA trajectory method proved to be superior at identifying sources for the simulation located in the west and comparable to the SIRA method for the two simulations located in the east.     

Application of back-trajectory techniques to the delimitation of urban clean air zones

Following a request by environmental authorities in New Zealand, atmospheric modelling has been applied to delimitation of clean air zones for urban areas. This approach involved the integration of a kinematic trajectory model with an atmospheric mesoscale model to identify the spatial extent of the catchment of air affecting air pollution concentrations in Christchurch on nights of high air pollution. The Regional Atmospheric Modelling System (RAMS) was configured for the region and idealised simulations performed to obtain predicted wind fields for synoptic situations typical of winter smog events. The predicted surface wind fields on two grids, with horizontal resolutions of 1.5 and 0.5 km, respectively, were then used to derive back-trajectories from the late evening peak of pollution over the central city (around 2200 NZST) to the time at which people tend to first light their domestic fires (around 1800 NZST). The results indicate that although winds are often light, the air tends to travel from a significant distance outside the city boundary over this time period. In particular, cold air typically travels up to 20 km from the Canterbury Plains to the west into the city during these air pollution events, as well as from small valleys in the Port Hills to the south of the city. This research illustrates the significance of upstream sources of air for providing relatively clean air to the city, and acting as buffer zones. It is, therefore, possible to identify the area around the city to which urban air quality is particularly sensitive. This area could either be designated as a buffer zone, or included within the clean air zone of the city. This technique also provides a useful tool for identifying the role of different components of the local wind field responsible for air pollution dispersion and transport in different parts of the area.     

The project MOHAVE tracer study: study design,data quality,and overview of results

In the winter and summer of 1992, atmospheric tracer studies were conducted in support of project MOHAVE, a visibility study in the southwestern United States. The primary goal of project MOHAVE is to determine the effects of the Mohave power plant and other sources upon visibility at Grand Canyon National Park. Perfluorocarbon tracers (PFTs) were released from the Mohave power plant and other locations and monitored at about 30 sites. The tracer data are being used for source attribution analysis and for evaluation of transport and dispersion models and receptor models. Collocated measurements showed the tracer data to be of high quality and suitable for source attribution analysis and model evaluation. The results showed strong influences of channeling by the Colorado River canyon during both winter and summer. Flow from the Mohave power plant was usually to the south, away from the Grand Canyon in winter and to the northeast, toward the Grand Canyon in summer. Tracer released at Lake Powell in winter was found to often travel downstream through the entire length of the Grand Canyon. Data from summer tracer releases in southern California demonstrated the existence of a convergence zone in the western Mohave Desert.     

The Relationship Between Atmospheric Transport and the Particle Scattering Coefficient at the Grand Canyon

Air trajectory and particle scattering data (bsp) for the period 1984-1989 are used to determine the relationship between atmospheric transport and visual air quality at the Grand Canyon National Park. Using cluster analysis, 72-hour back-trajectories arriving four times per day were grouped into distinct transport patterns. Northwesterly and southerly/southwesterly flow dominate in the winter and summer seasons, respectively. Comparisons of bsp values accompanying different transport patterns showed a clear relationship between air flow pathway and light scattering due to small particles during the non-summer months only. An index is defined which describes the percentage of annual trajectories belonging to specific transport routes delivering predominantly clear air to the GCNP.     

The effects of scrubber installation at the Navajo Generating Station on particulate sulfur and visibility levels in the Grand Canyon

Grand Canyon National Park (GCNP) is a mandatory Class I federal area that is afforded visibility protection under the Federal Clean Air Act. In this paper, we have examined the effects on visibility and particulate sulfur (Sp) at GCNP as a result of reducing sulfur dioxide (SO2) emissions by 90% from the Navajo Generating Station (NGS). Scrubbers were retrofitted to each of the three units at NGS during 1997, 1998, and 1999. The Inter-agency Monitoring of Protected Visual Environments aerosol network database affords us an opportunity to examine trends in Sp and extinction both prescrubber and postscrubber. The NGS impacts GCNP primarily during the winter (December to February). During winter, at times, there are fogs, stratus, and high-relative humidity in the Grand Canyon. When the NGS plume interacts with these fogs and stratus, rapid conversion of SO2 to Sp can occur. A variety of analytical techniques were used, including cumulative frequency plots of Sp and extinction, and chemical mass balance and tracer source apportionment analysis. We also deployed P value statistical analysis of "extreme" Sp values. Before scrubbers were installed, values of Sp approaching 2 microg/m3 were occasionally observed. Because scrubbers have been installed, high levels of Sp have been markedly reduced. Statistical P value analysis suggests that these reductions were significant. Furthermore, we have also observed that Sp has been reduced throughout the cumulative frequency curve during winter by approximately 33% since scrubbers were installed. By contrast, during summer when the NGS impact on the Canyon is minimal, there has been only a relatively small decrease in Sp.     

Influence of sea-land breezes on the tempospatial distribution of atmospheric aerosols over coastal region

The influence of sea-land breezes (SLBs) on the spatial distribution and temporal variation of particulate matter (PM) in the atmosphere was investigated over coastal Taiwan. PM was simultaneously sampled at inland and offshore locations during three intensive sampling periods. The intensive PM sampling protocol was continuously conducted over a 48-hr period. During this time, PM2.5 and PM(2.5-10) (PM with aerodynamic diameters < 2.5 microm and between 2.5 and 10 microm, respectively) were simultaneously measured with dichotomous samplers at four sites (two inland and two offshore sites) and PM10 (PM with aerodynamic diameters < or =10 microm) was measured with beta-ray monitors at these same 4 sites and at 10 sites of the Taiwan Air Quality Monitoring Network. PM sampling on a mobile air quality monitoring boat was further conducted along the coastline to collect offshore PM using a beta-ray monitor and a dichotomous sampler. Data obtained from the inland sites (n=12) and offshore sites (n=2) were applied to plot the PM10 concentration contour using Surfer software. This study also used a three-dimensional meteorological model (Pennsylvania State University/National Center for Atmospheric Research Meteorological Model 5) and the Comprehensive Air Quality Model with Extensions to simulate surface wind fields and spatial distribution of PM10 over the coastal region during the intensive sampling periods. Spatial distribution of PM10 concentration was further used in investigating the influence of SLBs on the transport of PM10 over the coastal region. Field measurement and model simulation results showed that PM10 was transported back and forth across the coastline. In particular, a high PM10 concentration was observed at the inland sites during the day because of sea breezes, whereas a high PM10 concentration was detected offshore at night because of land breezes. This study revealed that the accumulation of PM in the near-ocean region because of SLBs influenced the tempospatial distribution of PM10 over the coastal region.     

Stochastic simulation of meteorological variability for long-range atmospheric transport—II. Long-term statistical models

Methods are developed for simulating meteorological inputs for statistical atmospheric transport models. Stochastic models are presented for the annual mean wind vector, the horizontal diffusion coefficient, the fraction of time it is raining, and the quantity of precipitation. The models are derived by aggregating the statistical models for short-term meteorological processes given in the companion paper of Small and Samson (Atmospheric Environment, 1989, 23, 2813–2824). Normal distributions are used for each variable. The normal distribution is derived directly for the mean wind vector based on the normality of short-term winds, and serves as an acceptable approximation for the other variables given their low interannual coefficient of variation. Correlation between variables is considered, including correlation between winds and precipitation, between the fraction of time it is raining and the total quantity of precipitation at a site, and between the annual precipitation quantity at multiple sites. Development and confirmation of the interannual distributions is supported by long-term meteorological data, including a nine-year backward trajectory record computed for a site in western Pennsylvania. Application of the method is demonstrated with a statistical atmospheric LRT model used to predict distributions of annual precipitation sulfates in eastern North America. Predicted interannual coefficients of variation of wet SO₄²⁻ concentrations range from about 10 to 15 per cent, similar to the level of interannual variation inferred from long-term data sets.     

Impact of using prognostic and objective wind fields on the photochemical modeling of Athens, Greece

A three-dimensional Eulerian photochemical model is used to follow the dynamics of ozone, NO_x, and CO over the Athens area, for 25 May 1990, the day considered in the APSIS project. A unique aspect of this work lies in the study of the impacts of the wind field preparation methods on the concentrations predicted by the model. Three sets of wind fields are developed. The first one used is derived from a prognostic meteorological model. The second one is calculated from available wind observations using objective: methods. For these two cases, a previous day is simulated, using the same conditions, to develop preconditioned initial conditions for the following day. For the third simulation, again two days are simulated, this time using the observed winds for each of the two days modeled. The predictions using the prognostically derived and the objective analysis wind fields are significantly different, particularly for the primary pollutants. Comparing predictions to the observations did not favor any particular method of wind field preparation. In this case, when using the prognostically derived field, the simulations are very sensitive to boundary conditions. In contrast, when using the wind fields constructed by objective methods, the simulations became most sensitive to emissions and initial conditions. This comes directly from the different residence times in the domain, which are governed by the wind speed.     

Validation studies of turbulence closure schemes for high resolutions in mesoscale meteorological models – A case of gas dispersion at the local scale

This work is a contribution to a large project, aimed at the development of an advanced environmental assessment modelling system to be used in Japan. The modelling system here considered consisted of the RAMS and HYPACT coupled models. The RAMS code was modified to properly simulate local scale phenomena using a fine mesh size of 250 m. In this direction, the main aim here was to investigate the effect of the choice of the turbulence closure scheme on the dispersion of pollutants. Our modified version of the RAMS/HYPACT model chain was validated using field experiments which were carried out by the Japan Atomic Energy Research Institute (JAERI) in the area of Mt. Tsukuba (Japan). The mean flow, turbulence and concentration fields obtained using two alternative turbulence closure schemes are compared. A discussion on the different performances of the turbulence closures is presented and the influence of the closure schemes over the plume dispersion is investigated.     

A modeling study of visibility in the Grand Canyon

Using a backward version of the Monte Carlo radiative transfer model, radiance values in the Grand Canyon were simulated to within the precision of 10%. The contiguous spectral contrast was introduced to distinguish between adjacent areas of the same target and compared to the apparent spectral contrast. The contiguous spectral contrast is a valuable tool in evaluating visibility because in some cases, the target became more distinguishable when viewed against the sky while the target features became less distinguishable. Average equilibrium radiance values were calculated and incorporated into the Koschmieder estimate; the Koschmieder estimate was deemed to be of limited value in the Grand Canyon because of the violation of the assumptions used to derive the estimate.It was shown that apparent spectral contrasts determined from first order scattering results were within 10% of these contrasts determined from multiple order scattering results. However, higher order scattering must be considered when radiance values are calculated for the Grand Canyon scene.     

Aerosol light scattering measurements as a function of relative humidity: a comparison between measurements made at three different sites

The water uptake by fine aerosol particles in the atmosphere has been investigated at three rural National Parks in the United States (Great Smoky Mountains, Grand Canyon and Big Bend National Parks). The relative humidity (RH) of sample aerosols was varied from less than 20% to greater than 90% using Perma Pure drying tubes as the scattering coefficient of the aerosol was measured with a Radiance Research M903 nephelometer. Data from these studies show that growth curves at all the three sites are similar in shape but the magnitude of growth can vary considerably from day to day. The growth curves from Great Smoky Mountains show smooth continuous growth over the entire range of RH, while the growth curves from the Grand Canyon and Big Bend show smooth and continuous growth on some days and deliquescence on other days. Comparing 12-h filter samples of chemical composition data with the aerosol growth curves, we find that higher fractions of soluble inorganic compounds (sulfate and nitrate) produce growth curves of greater magnitude than do higher concentrations of either organic carbon or soil material.