IMPROVE Gray Literature

This presentation reviews the current (as of 2013) methods used to correct for the positive organic carbon artifact in the IMPROVE and CSN programs; analyzes the measured carbon artifacts on backup and field blank filters and makes recommendations on how best to correct of the positive organic artifacts and harmonize the procedures in the IMPROVE and CSN programs.

This report examines the sensitivity of estimated dv values to the choice of a strategy for handling missing values. The report also examines, via statistical simulation studies, the closeness of the estimated dv values to the “true values” under different scenarios for the occurrences of missing values. It is found that, when missing values occur infrequently, there doesn’t appear to be a practical difference in the estimated dv regardless of the method used. However, as the frequency of occurrence of missing values increases, differences due to estimation method becomes noticeable.

On October 6-8, 2008, the U.S. EPA’s Office of Air Quality Planning and Standards (OAQPS) held an expert workshop on urban visibility preferences and valuation to identify and discuss methods and materials that could be used in “next step” projects to develop additional information about people’s preferences for reducing existing impairment of urban visibility, and about the value of improving urban visibility. This document summarizes findings from this workshop and contains some of the presentations.

An extended abstract describing the method and results for calculating the natural haze levels using the revised IMPROVE aerosol reconstructed light extinction algorithm and a revised natural haze level algorithm. This was presented at the AWMA Aerosol & Atmospheric Optics: Visual Air Quality and Radiation specialty conference at Moab, Utah, April 2008.

In September 2007, several member of the IMPROVE steering committee and others were given a tour of the Navajo Generating Station (NGS), a coal fired power plant producing 2250 megawatts of electricity. The tour included the wet scrubbers which were installed in the 1990s to reduce the emissions of the SO2 and other pollutants. Following is a brochure provided to the participants which shows a schematic of NGS and the flow of material to generate the electricity.

This manuscript outlines and demonstrates a set of procedures that could be used in a detailed FLAG level II and III visibility assessment. The procedure uses advanced visibility modeling to simulate the instantaneous degradation of visual air quality indexes along idealized sight paths under various ambient lighting conditions. This work was presented at the AWMA Specialty Conference - Guideline on Air Quality Models: Applications and FLAG Developments, Denver, CO, 2006.

The Natural Haze Levels II Committee was established in the Spring of 2006 to review and refine, as appropriate, a methodology developed by Roger Ames (CIRA) for applying the new IMPROVE algorithm for estimating light extinction from aerosol species concentrations to natural species concentration estimates. The ultimate purpose of this committee was to determine natural haze estimates for the 20% best and 20% worst day for each of the visibility-protected class I areas. This is the final report of the committee describing the procedures and final results for each of the visibility protected area.

The elemental composition of particulate matter collected on several IMPROVE filters was examined by analyzing the filters hundreds of times using the non-destructive XRF analysis. The hundreds of results allowed the estimation of the distributions of potential analytical outcomes which were examined using counting statistics. POSTER : XRFposter.ppt

A presentation comparing data from collocated IMPROVE and STN monitors. This presented at the AAAR Specialty Conference in Atlanta, GA, February 2005

In straight-line regression, the least squares estimator of the slope is sensitive to outliers and the associated confidence interval is affected by non-normality of the dependent variable. A simple and robust alternative to least squares regression is Theil regression, first proposed by H. Theil (1950). Theil’s method actually yields an estimate of the slope of the regression line. Several approaches exist for obtaining a nonparametric estimate of the intercept. In this paper, we describe a SAS macro for implementing Theil regression where the estimation of the intercept is based on Graybill and Iyer (1994).

In November 2002 a number of questions arose regarding the accuracy of data obtained from the IMPROVE B Module, especially for nitrate. In particular, wintertime nitrate concentrations at many sites were below historical levels for about four years, from 1996-97 to 1999-2000. This paper reports on a number of sampler tests conducted to investigate these questions.

IMPROVE adjusts for positive sampling artifacts by subtracting nylon filter field blank values and quartz after filter values from measured ion and carbon concentrations, respectively. Artifacts have been observed to vary substantially among filter manufacturing lots, especially for nylon filters. Sulfate and nitrate artifact corrections are currently less than 10% of the measured values, but organic carbon artifacts can exceed 50% of the measured values.

The purpose of this report is to determine if we wish to revise the current equation for the soil composite. This report examines four aspects of the soil equation. The first is what to do when aluminum is missed because of x-ray interference. The second is how to treat soil potassium. The third is how to treat soil elements not included in the equation. The fourth is a suggested change in the equation for the uncertainty in the soil composite concentrations.

An NPS brochure describing the Rocky Mountain Atmospheric Nitrogen and Sulfur Study (ROMANS) study including the problem of increasing nitrogen deposition and its effects on the ecosystem, the measurement campaign conducted as part of ROMANS and the study objectives.

In August, 2006, ARS performed additional QA on the IMPROVE Optec LPV-2 transmissometer data set in preparation for re-submittal to CIRA. The goals of this additional QA were as follows: 1) Check for completeness and continuity of data 2) Verify that reported RH and AT are within instrument specifications 3) Verify lamp and calibration information. QA checks prompted some changes in the raw (Level-A) files and in the transmissometer lamp files. ARS reprocessed all historical transmissometer data to include these changes. A total of 1151 files were reprocessed for the years 1987–2004 and are included on the accompanying CD.

This report assess the potential impact of a proposed source in the Four Corners basin on haze on the Colorado Plateau. This is done by first examining past monitoring and modeling studies to determine the relevant atmospheric processes that would cause the source to have it largest contribution to haze. Then diagnostic modeling is conducted using a particle dispersion model to reproduce these processes to simulate the source contribution to sulfate concentrations at the Grand Canyon NP and other class I areas. Radiative transfer modeling is then conducted to simulate the impact of sulfate concentrations on the scene.

This study investigates the potential impacts of a proposed power plant located in the Four Corners basin on class I areas during January 2001 with the “Comprehensive Air Quality Model with Extensions” (CAMx) photochemical dispersion model. The MM5 mesoscale model was used to generate 4 km horizontal resolution meteorological fields used as input for CAMx.

The proposed new algorithm described in this report was adopted for use by the IMPROVE Steering Committee in December 2005.

This document is a summary report by a subcommittee established by the IMPROVE Steering Committee at their July 2005 meeting to recommend a refined algorithm that would replace or be made available as an alternative to the current approach. The subcommittee included those who worked on the internal review as well as scientists who have been critical of the original IMPROVE algorithm. The primary purpose of this document is to describe the subcommittee’s recommended revised algorithm, characterize its performance, and summarize the rationale for each of the changes from the currently used algorithm.

In October 2005 UC Davis resubmitted all of the IMPROVE aerosol data for the five year Regional Haze Rule (RHR) baseline period, 2000-2004. The data were resubmitted to correct several errors and discrepancies in the data in order to provide the RHR analysts with the best available data set. There were four systematic changes that affected large blocks of data and this document provides a summary of these changes.

The Interagency Monitoring of Protected Visual Environments (IMPROVE) protocols for reconstructing ambient light extinction from measured aerosol species is the basis for evaluating compliance under the Regional Haze Rule. We review the assumptions involved in computing reconstructed light extinction using the IMPROVE protocol. This review includes examining the biases in the measurements of aerosol composition, the assumed chemical forms of aerosol species, particle hygroscopicity, and assumed mass scattering efficiencies. We present a thorough survey of estimates of mass scattering efficiencies from recent peer-reviewed literature. Furthermore, we use IMPROVE nephelometry and composition data to estimate mass scattering efficiencies using a variety of methods. The current mass scattering efficiencies applied in the IMPROVE equation are then interpreted in the context of this survey and results derived from the IMPROVE data analyses. Finally, a summary of provisional recommendations for refinements to the IMPROVE equation and a discussion of important uncertainties to consider in the assumptions is presented. Although tentative recommendations of refinements to the IMPROVE equation are presented, final refinements to the IMPROVE equation await future discussions of the results presented here.

ABSTRACT: In this study, a statistical method known as Positive Matrix Factorization (PMF) was used to analyze IMPROVE monitoring data collected at two west coast Class I areas over two time periods, 1991-1995 and 2000-2003. These Class I areas were Mt. Rainier National Park in Washington, and Yosemite National Park in California. PMF generated source profiles associated with each source of fine particulates, and generated a time-dependent series of fine particulate concentrations from each source in these two Class I areas. The light extinction of each source was determined by summing the light extinction of all the light absorbing chemical species in each source. The average light extinction of each source for 1991-1995 and 2000-2003 were summed to determine the total light extinction in each Class I area for these time periods. The total light extinction in each Class I area in 2002 was also determined for the worst 20% visibility days, which is one of the parameters identified in EPA’s Regional Haze Rule to determine progress towards improving visibility in Class I areas. The composition of the biomass source in Yosemite was also examined to determine the relative contribution of fine particulates from biomass combustion and biogenic emissions to visibility impairment on the 20% worst visibility days in 2002. The results show that PMF can be used as a tool to help determine which sources have the most significant impact on visibility in Class I areas, and how the visibility impairment from each source varies over time.

This paper uses Lorenz-Mie theory, reasonable estimates of the variation of ambient aerosol properties, and the optical characteristics of currently available nephelometers and light scattering photometers to investigate the theoretical limits of the accuracy and precision of PM2.5 mass measurements estimated by nephelometry.

The integrating nephelometer directly measures the scattering coefficient of atmospheric aerosols. The Optec NGN integrating nephelometer was designed to operate in low power, low maintenance, and ambient field conditions. This required two main compromises in the optical design:

· The use of a photodiode rather than a photomultiplier tube (PMT) detector, and

· A wide band filter to compensate for the reduced sensitivity of the photodiode compared to a PMT.

The following is a theoretical examination of the effect of these design requirements on the measurement accuracy of the Optec NGN integrating nephelometer and an experimental determination of the measurement precision of the instrument. Included in the theoretical analysis are comparisons to three other currently available integrating nephelometers:

· TSI 3563: TSI, Inc., St. Paul, Minnesota

· Belfort 1590 : Belfort Instrument Company, Baltimore, Maryland

· Radiance Research M903: Radiance Research Inc., Seattle, Washington

This report summarizes Air Resources Specialists' (ARS) operational experience with the Optec Next Generation Nephelometer (NGN) from 1992 to 1994. It describes several design, manufacturing and operating deficiencies that were discovered and instrument upgrades to address these deficiencies.

The DRI/OGC analyzers used for IMPROVE carbon analysis since 1987 are obsolete and spare parts for repairs are no longer manufactured and increasingly difficult to obtain. The Model 2001 carbon analyzer has been designed and tested as a replacement for future IMPROVE sample analysis. This report summarizes a number of analyses comparing the DRI/OGC and Model 2001 analyzers under different configurations. The similarities and differences are reported and recommendations are made for a smooth transition from the current DRI/OGC analyzer to the Model 2001.

A power point presentation on the assessment of the comparability between the current DRI/OGC carbon analyzer used in IMPROVE and the new Model 2001 carbon analyzer proposed as the DRI/OGC replacement.

Positive Matrix Factorization (PMF) was used to model IMPROVE monitoring data collected at the Mt. Zion and Wishram monitoring sites in the Columbia River Gorge National Scenic Area. PMF was able to identify source categories that contribute to the fine particulates (PM2.5) measured at each site, the average percent PM2.5 mass allocation from each source category, and the time-dependent allocation from each source category. PMF identified 9 source categories for the Mt. Zion site based on 1996-98 data, and, 9 source categories for the Wishram site based on 1993-95 data.

A presentation describing the IMPROVE particulate monitoring network and the spatial and seasonal patterns of the composite aerosol components from all data collected in 2001. Some results from IMPROVE special studies are also included. The was presented at the Science Colloquium on Measuring and modeling aerosols, June 14-16 in Baltimore, MD. Many of the figures of the particulate patterns can be found in the paper Malm et al., 2004, J. of Geo. Res. Vol 109, D03306

Describes some of the limitations and issue concerning the transmissometer data and problems with using the long-term transmissometer records for trend analysis.

Reconstructed mass excluding nitrate is compared with gravimetric mass for each year of the IMPROVE network. Two parameters were used to investigate the trends. The first is the slope of the regression line through the origin, assuming gravimetric mass is independent. The second is the ratio of the two means. The slope emphasizes points with large mass, while the ratio reflects the average samples. No significant historical trends in the difference between the measured and reconstructed fine mass was found.

An Explanation as to why Na is multiplied by a factor of 2.5 to reconstructed sea salt.

In 1999 the IMPROVE network began to expand to monitor visibility in all Class I areas that visibility was deemed an important attribute of the Area. This document describes the procedure used to select the new IMPROVE monitoring sites.

A 1995 report providing an overview of the IMPROVE network. This document describes the IMPROVE monitors and filter analysis, estimation of uncertainty in the measures species, calculation of the major aerosol types and their uncertainty, as well as the effect of changing the analysis from PIXE to XRF for elements between Fe and Pb on the minimum detection limits and precision.

A document containing images representing the cleanest and the haziest clear sky, non-weather, days found in the IMPROVE 35mm slide data base. Extinction coefficient estimates are also provided based upon the slide contrast measurements, the average species concentrations of the 20% lowest and highest fine mass days, and annual estimates of the f(rh).

Examination and estimation of uncertainties in light extinction derived from apparent contrast measurements from images of natural targets.

This document summarize the limitations and value of light extinction and visual range measurements from slide based images.

This report analyzes the sampling completeness for the IMPROVE aerosol monitoring network in 2001. It was found that the median site had a 92% completion rate, which is excellent. Ninety percent of the sites had a completion rate greater than 75%, which is the annual criterion for acceptance. If we consider only the 103 sites were installed before January 2001, then 12 failed the annual criterion and 3 others failed the quarter or 10-consecutive criteria.

The IMPROVE denuder has a collection efficiency for SO ₂ of >99% when freshly coated. There is some question about whether there is breakthrough of SO ₂ and HNO ₃ after being used for a year, at least at sites with high SO ₂ concentrations. This report examines the effect of breakthrough of SO ₂ on the sulfate measurements and of HNO ₃ on the nitrate measurements.

The primary purpose of flow control is to keep the volume flow rate through the cyclone or PM ₁₀ inlet as close as possible to the nominal rate for the 2.5 or 10 µm cut. There are several factors producing a variation in the flow rate for a critical orifice: the diurnal and seasonal variation of ambient temperature, the change in pressure drop across the filter associated with variation in clean filters, and the change in pressure drop across the filter associated with mass loading. The estimated total variation in inlet flow rate with a critical orifice controller from all three factors is generally less than 3%. This is similar to the standard deviation in flow rates observed in the network. The conclusion is that a critical orifice system is reasonable if a variation in flow rate of approximately 3% is acceptable.

Long term trends in the EC/OC ratios in the IMPROVE network have been found. This report examines the question as to whether these trends are real or caused by analytical changes in the IMPROVE network. In addition, the appendix examines the seasonal variation and long term trends in the organic and elemental carbon artifacts.

Significant changes were made in the IMPROVE cassettes in the new version of the sampler installed in the year 2000. The Version 2 cassette sampler are made of an inert plastic, Delrin®, and the filters sit on stainless steel screens. The Version 1 cassettes were polycarbonate and sat on cellulose drain discs over polycarbonate support grids. This report examines the change in the organic and elemental carbon artifact after the Version 2 cassettes were installed.

Warren White noted a trend in the ratio of nylon sulfate / Teflon sulfur over the course of the IMPROVE network. This is extremely critical in determining sulfate trends. This report concludes that there is no trend in the ratio of analytical calibrations, but there is a trend in the relationship at low concentrations. This is seen in the intercept and in distributions. The low end concentrations are primarily affected by the artifact that is subtracted from the samples. This change in low end concentrations can account for the trend in the ratios. No mechanism for producing such a change has been identified.

This report details the changes in the sampling and filter analysis in the IMPROVE program that could affect the fine particle sulfur or sulfate measurements. It was found that three issues could produce invalid data and the possible impact of these issue is examined using eastern US data.

The IMPROVE (Interagency Monitoring of Protected Visual Environments) network collects PM _2.5 particles on Teflon, nylon, and quartz filters using a modular, cyclone-based sampler with critical orifice flow control. Understanding the protocols of IMPROVE is important because IMPROVE will continue to provide background monitoring at Class I areas, and because the protocols for measuring chemical species in the Federal Reference Method network at urban and suburban sites will probably be similar. The IMPROVE filters are analyzed by seven methods to obtain concentrations of mass, elements H and Na-Pb, ions NO ₃ ^- and SO ₄ ⁼ , organic and elemental carbon, and the coefficient of absorption. The replicate precision for most analytical methods is approximately 4%. The artifacts for all methods except carbon are small or zero. The artifact problem for organic carbon is minimized by allowing each filter to reach saturation prior to collection. Redundant measurements by the three filters permit improved quality control. Comparisons of sulfur with sulfate and organic carbon with organic hydrogen permit assurance that (1) the calculated uncertainties are reasonable, (2) the analytical methods are correctly calibrated, (3) the flow rates of the various modules are in agreement, and (4) the individual concentrations are valid.

This paper examines the annual and seasonal spatial patterns of the major aerosol components (sulfate, nitrate, organics, elemental carbon (EC), and soil) from the IMPROVE monitoring network. The data were from 70 monitoring sites with one year of data from March 1996 through February 1998.

The Model LPV-2 Long Path Visibility Transmissometer consists of a constant output light source transmitter and a computer controlled photometer receiver. The irradiance at 550nm wavelength from the transmitter can be measured to a high degree of accuracy both day and night, and over a path length of up to 15 km depending on expected extinction values. Both the receiver and transmitter operate from a 12 volt battery source and use 5 and 34 watts respectively. The signal is processed by an internal CMOS 8-bit computer and the output voltage can made proportional to extinction or visual range. The units can operate for long periods of time unattended in a continuous or timed cycle mode. The self resetting and battery backup systems ensure continued operation even after power blackouts and computer lockup due to local static electric discharges. Both units can be synchronized, programmed and calibrated at the home station and installed in the field ready for operation.

The Model NGN-2 (Next Generation Nephelometer) uses a unique integrating open-air design that allows accurate measurement of the scattering extinction coefficient of ambient air. Because of the open-air design, relative humidity and temperature of the air sample are essentially unchanged, thus the aerosol is negligibly modified when brought into the optical measuring chamber. With improvements made to the prototype NGN-1, the production Model NGN-2 nephelometer alters the temperature of the air sample by less than 0.5° C. Extinction due to scatter can accurately be measured from Rayleigh to 100% saturated fog conditions. Using principals and techniques developed for the popular and successful Optec LPV transmissometer, the NGN-2 features low power consumption, solid compact design, and easy operation - all necessary parameters when installing and operating a measuring instrument in remote locations. Housed in a single environmentally sealed enclosure, the nephelometer contains all the necessary components for ambient standalone operation. The NGN-2 consists of a large inlet with motorized door, measuring chamber, clean air pump for calibration, blower, solenoid activated inlet for a span gas, temperature sensor, real time clock and rain detector for closing the inlet during wet conditions. See Figure 1-2 for a cross-sectional view of the NGN-2. A CMOS computer controls all operating functions and outputs data either in digital or analog format. External to the NGN-2 unit, a 13.8V DC, 5 amp power supply, compressed span gas and a data logger or printer are required.