Search Results

You are looking at 1 - 8 of 8 items for

  • Author or Editor: Stephen F. Mueller x
  • All content x
Clear All Modify Search
Stephen F. Mueller

Abstract

Data on atmospheric levels of sulfur dioxide (SO2) and sulfate were examined to quantify changes since 1989. Changes in sulfur species were adjusted to account for meteorological variability. Adjustments were made using meteorological variables expressed in terms of their principal components that were used as predictors in statistical models. Several models were tested. A generalized additive model (GAM)—based in part on nonparametric, locally smoothed predictor functions—computed the greatest association between sulfate and the meteorological predictors. Sulfate trends estimated after a GAM-based adjustment for weather-related influences were found to be primarily downward across the eastern United States by as much as 6.7% per year (average of −2.6% per year), but large spatial variability was noted. The most conspicuous characteristic in the trends was over portions of the Appalachian Mountains where very small (average = −1.6% per year) and often insignificant sulfate changes were found. The Appalachian region also experienced a tendency, after removing meteorological influences, for increases in the ratio RS of sulfate sulfur to total sulfur. Before 1991, this ratio averaged 0.33 across all sites. Appalachian increases in RS were equivalent to 0.07 during 1989–2001 (significant for most sites at the 0.05 level), or nearly 2 times the average change at the other sites. This suggests that conditions over the Appalachians became notably more efficient at oxidizing SO2 into sulfate. Alternatively, subtle changes in local deposition patterns occurred, preferentially in and near mountainous monitoring sites, that changed the SO2–sulfate balance.

Full access
Stephen F. Mueller

Abstract

Ambient ozone data collected at two sites in the Great Smoky Mountains National Park (GSMNP) are summarized and compared with data from an urban and a low-elevation rural site. The ozone climatology in the park is found to be similar to that of other remote sites in the southern Appalachian Mountain region. As expected, terrain elevation is identified as a major factor influencing local ozone levels. Episodes of high ozone concentrations (≥90 ppb) in the park are shown to be primarily attributable to the transport of ozone into the park from outside. Backward air trajectories computed for high-ozone episodes in the GSMNP reveal that no preferred source regions exist, although some episodes appear to be associated with transport from urban areas.

Full access
Stephen F. Mueller

Abstract

Daily (24 h) and hourly air quality data at several sites are used to examine the performance of the fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5)–Community Multiscale Air Quality Model (CMAQ) system over a 3-month period in 2003. A coarse (36 km) model grid was expected to provide relatively poor performance for ozone and comparatively better performance for fine particles, especially the more regional sulfate and carbonaceous aerosols. However, results were different from this expectation. Modeling showed significant skill for ozone at several locations but very little skill for particulate species. Modeling did poorly identifying surface wind directions associated with the highest and lowest pollutant exposures at most sites, although results varied widely by location. Model skill appeared to be better for ozone when spatial–temporal (S–T) patterns were examined, due in part to the ability of the model to reproduce much of the temporal variance associated with the diurnal photochemical cycle. At some sites the modeling even performed well in replicating the directional variability of hourly ozone despite relatively low spatial resolution. MM5–CMAQ spatial (directional) representation of 24-h-average particulate data was not good in most cases, but model skill improved somewhat when hourly data were examined. Modeling exhibited skill for sulfate at only one of nine sites using 24-h data averaged by daily resultant wind direction, at two of six sites when hourly data were averaged by direction, and at four of six sites when the combined spatial and temporal variance of sulfate was examined. Results were generally poorer for total carbon aerosol mass and total mass of particulate matter with diameter of less than 2.5 μm (PM2.5). The primary result of this study is that an S–T analysis of pollutant patterns reveals model performance insights that cannot be realized by only examining model error statistics as is typically done for regulatory applications. Use of this S–T analysis technique is recommended for better understanding model performance during longer simulation periods, especially when using grids of finer spatial resolution for applications supporting local air quality management studies. Of course, using this approach will require measuring semicontinuous fine particle data at more sites and for longer periods.

Full access
Lawrence M. Reisinger and Stephen F. Mueller

Abstract

Forty-five tetroon flights made during the summer of 1980 for the PEPE/NEROS regions pollution studies, sponsored by the U.S. Environmental Protection Agency, were compared to computed trajectories based on National Weather Service rawinsonde wind fields. Most tetroon data were obtained for travel times of less than 10 h and travel distances of less than 150 km.

Two trajectory computation algorithms were used. No significant differences were found between the two comparisons. Results of the comparisons indicate that the median measure of direction difference—the angle, determined in a clockwise sense, between tetroon and computed forward trajectory position vectors—has a bias of 11°. The average standard deviation of the direction difference is ±28°. About 10% of the total direction difference variance could be due to random tetroon motion; the remaining 90% is probably the result of error in the trajectory algorithms and/or the input data. Other significant results of the trajectory comparisons are also described.

Full access
Qi Mao, Stephen F. Mueller, and Hann-Ming Henry Juang

Abstract

A limited-area spectral model—the Regional Spectral Model—developed at the National Centers for Environmental Prediction is used to prepare daily quantitative precipitation forecasts out to 48 h for the Tennessee and Cumberland River basins in the southeastern United States. One year of these forecasts is evaluated against data from a network of 243 rain gauges and against traditional man–machine forecasts provided under contract to Tennessee Valley Authority river system managers. The intent of this study was to determine whether the model forecasts, made at greater spatial resolution than those typically available from other sources, offered any advantages to water resource managers responsible for making critical day-to-day decisions affecting flood control, navigation, and hydropower production. The model’s performance, determined using a variety of statistical measures, was found to be more accurate than the traditional forecasts. In particular, the model had less bias and lower root-mean-square error, and was more accurate in the timing of precipitation events. The model’s advantage was especially evident in 24–48-h forecasts and for heavy precipitation events. Three specific case studies of model performance are described to illustrate the model’s abilities under conditions that could significantly influence river management decisions.

Full access
Stephen F. Mueller, Jonathan W. Mallard, Qi Mao, and Stephanie L. Shaw

Abstract

A study of fugitive dust emissions from a pile of crushed coal revealed that, in addition to dust being emitted into the atmosphere during periods of pile-management (human) activity, it is also emitted during periods without human activity. This “natural” emission in itself is not surprising given past work on wind erosion of particulate matter from aggregate piles. However, hourly downwind measurements of fine particle (PM10) mass concentrations at two sites revealed that excessive dust was present in the air even when wind speeds were below the erosion threshold estimated from nearby wind speed measurements and regulatory guidance on coal pile aerodynamic characteristics. During periods of natural emissions with higher wind speeds, downwind concentrations were strongly associated with µ 2—the squared excess of 1-min maximum wind speed above the erosion threshold—consistent with previous work on wind erosion potential. However, 88% of hourly concentrations coincided with lower winds for which wind speed was not a good predictor of airborne dust levels. Evidence was found that natural low-wind PM10 concentrations varied significantly with relative humidity, air temperature, and turbulence parameters (σ u and σ w). Smoke from coal combustion was ruled out as a significant factor in PM10 levels, but statistical evidence along with visual observation suggests that microscale turbulent airflows, including dust devils, were a significant source of PM10 during low wind speeds over the pile. The localized behavior of the turbulence makes it very difficult to develop a strong statistical model of natural downwind concentrations on the basis of off-pile meteorological measurements.

Full access
Stephen F. Mueller, Aaron Song, William B. Noms, Shekar Gupta, and Richard T. McNider

Abstract

Airflow patterns and pollution transport in the southern Appalachian Mountains region of the southeastern United States are examined using mesoscale meteorological models and a Lagrangian particle dispersion model (LPDM). The two primary goals of this work are 1) to identify a meteorological modeling methodology that can be used in regional photochemical modeling, and 2) to identify large regional ozone precursor sources that may impact the southern Appalachians during periods having high ozone levels. Four episodes characterized by measured high levels of ozone (1-h average concentrations greater than 90 ppb) at remote monitoring sites are the focus of the modeling efforts. To address the first goal, several methods of airflow modeling involving varying degrees of complexity are examined to find one that reliably simulates the complex wind patterns that occur. A hydrostatic model with homogeneous initialization, a nonhydrostatic model with homogeneous initialization, and a nonhydrostatic model with nonhomogeneous initialization and four-dimensional data assimilation (FDDA) are evaluated against available wind observations. The method using nonhomogeneous initialization and FDDA is found to best reproduce observed wind patterns. Results of a test of model sensitivity to the strength of the FDDA are described.

In addressing the second project goal, a LPDM driven by computed meteorological fields is used to simulate the potential for ozone precursor emissions (in the form of NOx) to be transported from nearby major sources toward the mountains. LPDM simulations indicate that one of the urban areas was the most likely source to influence the monitoring sites experiencing high ozone levels during three of the four episodes. However, none of the plumes are computed to be over the monitoring sites for the length of time that the high ozone concentrations were actually observed. Detailed air quality data for one episode suggest the presence of a large urban plume passing over the mountains and originating from outside the modeling domain. This implies that a larger domain is needed for photochemical modeling.

Full access
Qi Mao, Richard T. McNider, Stephen F. Mueller, and Hann-Ming Henry Juang

Abstract

An optimal model output calibration (MOC) algorithm suitable for surface air temperature forecasts is proposed and tested with the National Centers for Environmental Prediction Regional Spectral Model (RSM). Differing from existing methodologies and the traditional model output statistics (MOS) technique, the MOC algorithm uses forecasts and observations of the most recent 2–4 weeks to objectively estimate and adjust the current model forecast errors and make refined predictions. The MOC equation, a multivariate linear regression equation with forecast error being the predictand, objectively screens as many as 30 candidates of predictors and optimally selects no more than 6. The equation varies from day to day and from site to site. Since it does not rely on long-term statistics of stable model runs, the MOC minimizes the influence of changes in model physics and spatial resolution on the forecast refinement process.

Forecast experiments were conducted for six major urban centers in the Tennessee Valley over the period of 27 June to 30 July 1997. Surface air temperature forecasts out to 72 h were produced based upon RSM runs initialized from 0000 UTC observations. Performance of the MOC for minimum and maximum temperature forecasts was assessed by determining mean forecast error (BIAS), mean absolute error (MAE), and root-mean-square errors (rmse) for both the MOC-adjusted and nonadjusted RSM output. The same statistical measures for Nested Grid Model–MOS forecasts over the experiment period were also provided for instruction. A skill score was calculated to demonstrate the improvement of refined forecasts with the MOC over the RSM. On average for the six sites, reduction of forecast errors by the MOC ranged from 58% to 98% in BIAS, 40% to 52% in MAE, and 33% to 46% in rmse. It also showed that the error frequencies of the refined forecasts had Gaussian distributions with the peak centered around zero. The error bands were narrower using the MOC and there were decreases in large forecast errors, especially during the first 48 h. The Wilcoxon signed-rank test was performed to verify that populations of the forecast errors before and after the MOC adjustment were statistically far enough apart to be distinct at a high significance level. Forecast experiments were also conducted to address the issue of sensitivity of the MOC by varying the length of the time series used in deriving the MOC equation. It was found that the mean biases of the refined forecasts slightly increased and the MAE and rmse slightly decreased with increasing length of the time series from 2 to 4 weeks. The study demonstrated the usefulness of the MOC for objective temperature forecasting, especially in an operational environment in which changes in model physics and configurations are continually implemented and no long-term stable model runs are readily available to generate MOS forecast guidance. The MOC is also informative for diagnosing and tuning model physics. The study suggests that the MOC may be applied to other weather forecast elements, such as precipitation, cloud cover, visibility, and wind speed and direction.

Full access