Science and Judgment in Risk Assessment (1994)

TABLE L-1 Potential Data Needs for Calculation of Emissions

Process Vents

1.

Volumetric flow rate of vent gas

2.

Vent-gas discharge temperature

3.

Concentration of individual or aggregate HAP

4.

Operating hours per year of unit operation

5.

Molecular weight of gas

6.

Efficiency of control device

7.

Production rate during measurement

Fugitive Emission

1.

Numbers of pumps, valves, flanges, pressure-relief valves, open-ended lines, and compressors

2.

Screening level

3.

Weight % of HAPS in stream

4.

Percent leaking equipment

5.

Other HAPS characterization

6.

Frequency of leak checking

Loading Emission

1.

Type of cargo carrier

2.

Mode of operation

3.

Annual volume of liquid loaded

4.

Temperature of liquid loaded

5.

Weight in percent of HAP in loaded material

6.

True vapor pressure of HAP loaded

7.

Molecular weight of HAP

8.

Efficiency of control device

Storage-Tank Emissions

1.

Material stored

2.

Diameter of tank

3.

Rim seal type

4.

Tank, roof, and shell color

5.

Ambient temperature

6.

Wind speed

7.

Density and partial pressure of chemical

8.

Molecular weight

9.

Vapor pressure

10.

Efficiency of control device

11.

Type of storage tank

12.

Annual throughput

13.

Number and diameter of columns

Emission Factors

1.

Magnitude of input into the process

2.

Production level

Wastewater Sources

1.

Volumetric flow rate of wastewater

2.

Concentration

3.

Production rate during flow determination

4.

Production rate during concentration determination

Source: EPA, 1991c.

•

Data on emissions of pollutants that result from production, storage, use, and disposal (discussed in previous section).

•

Data on physical and chemical properties of pollutants (see Table L-2). For example, the vapor pressure of a chemical pollutant plays a major role in determining exchange of the chemical between the atmosphere and other environmental media. The vapor pressures of chemicals vary widely from those of gases (such as CO, CO2, and SO2), with vapor pressures of more than 1 atm, to those of aromatic compounds, organophosphates, dioxins, and other non-criteria pollutants, which are often in the range of 10-8-10-3 atm. VOCs generally have vapor pressures of greater than 10-3 atm and semivolatile compounds vapor pressures of 10-8-10-3 atm. Lead and other inorganic species are volatile as well. Water solubilityis important, because, with vapor pressure, it determines the distribution of a pollutant in the atmosphere. Water-soluble vapors, for example, might be efficiently scrubbed from air by rainfall or fog deposition—processes that can minimize human exposure, at least by inhalation. Suspended dust or aerosol particles can adsorb vapors of the pollutant and may also play a major role in determining the rate of exchange of chemicals between the atmosphere and other environmental media.

•

Data on transformation, degradation, and sequestration of pollutants in the environment (Table L-2), including chemical, biologic, and physical data:

—

Chemical data (e.g., for atmospheric oxidation and photochemical reactions). Chemical breakdown depends on molecular structure, and for some substances breakdown is rapid. If the chemical is susceptible to nucleophilic attack, oxidation, or hydroxylation, alterations can occur rapidly and change the potential exposure dramatically.

—

Biologic data (e.g., on degradation by metabolic action of microorganisms). Alterations by biologically mediated reactions are enormously variable, and data are needed on products of alteration; for example, do emissions tend to become more toxic or less toxic?

—

Physical data (e.g., on solubility and gravitational settlement). For particles, gravitational settlement or sedimentation increases with the aerodynamic diameter of the particle. Physical processes that occur in the atmosphere can affect particle-removal efficiency. Hydroscopic particles can increase in size because of the accumulation of water from the vapor phase in the atmosphere; this growth can help in their removal by sedimentation and washout.

•

Data on rate of removal of pollutants by various routes. For example, the rate of catalytic oxidation of SO2 decreases if the water concentration in the atmosphere falls below that necessary to maintain catalyst droplets. The critical point seems to be the percent relative humidity; above this, rates of catalytic oxidation increase dramatically. In clean air, SO2 emissions are only very slowly oxidized via homogeneous reactions of the gas phase to SO2 vapor. The development of the kind of information described here is important for the pre-

TABLE L-2 Physicochemical Properties of Chemical and Its Atmospheric Environment Important in Transport-Fate Calculations

Properties of Cbemical

Properties of Environment

Physical properties:

Particulate load:

Molecular weight

For dust, other solid particulate matter

Density

For liquid aerosols

Vapor pressure (or boiling point)

Oxidant level

Water solubility

Temperature

Henry's constant (air-water distribution coefficient)

Relative humidity

Lipid solubility (or octanol-water distribution coefficient)

Amount and intensity of sunlight

Amount and frequency of precipitation

Soil sorption constant

Meteorologic characteristics:

Chemical properties:

Ventilation

Rate constants for

Inversion

Oxidation

Surface cover:

Hydrolysis

Water

Photolysis

Vegetation

Microbial decomposition

Soil type

Other modes of decomposition

Particle properties:

Size

Surface area

Chemical composition

Solubility

diction of risk associated with environmental pollutants. Such data could be used to identify the most probable routes through the environment and provide clues to the rate of degradation (alteration) from source to receptor. Knowing the probable routes and sinks, one can identify populations that should have special attention in an evaluation of potential health effects. More refined approaches might include selecting or developing models to estimate transport and fate of pollutants.

•

Data on types of models to predict the persistence, transport, and fate of pollutants, including their input requirements, degree of accuracy and precision, and method of validation. Several models of aerial dissipation have been reported.

•

Contaminants (e.g., types, in which media, at what concentrations, and for what durations).

•

Exposed population (e.g., who is at risk, where, and under what circumstances; how long they are exposed and to what degree; and their intake of the contaminant from air, food, water, or through other relevant routes).

•

all data should be collected with validated methods under strict quality-assurance and quality-control standards.

•

A clear statement of uncertainty is fundamental to all analytic reports (Keith et al., 1983). Errors are likely to be greater with airborne trace-amount toxicants than with ''criteria pollutants," which tend to occur at much higher concentrations. This is because the relative accuracy of instruments often decreases at low concentrations.

•

A contaminant might be present but below the detection limit of the equipment. In this case, the concentration of the contaiminant should not be assumed to be zero. Rather, the detection limit (or some agreed-on fraction of it) should be used in the processing of data.

•

Vapors must be discriminated from particle-bound residues in air monitoring, especially for toxicants of low to intermediate vapor pressure.

•

Data on trace toxicants should be confirmed by mass spectrometry or other confirmatory method to increase confidence in the results.

•

Clinical. Outcome and disease data are reported for members of the general population, including, if known:

—

A description of the outcome(s).

—

The diagnostic criteria used.

—

A description of individual characteristics that might affect outcomes (age, pre-existing illness, etc.).

—

Exposure history, including dose and time frames.

•

Toxicologic. Outcome and disease data are reported for persons (usually volunteers, not members of the general population) after exposure under controlled experimental conditions, including:

—

Description of the hypotheses tested.

—

The criteria used to select the study groups.

—

The relevance of the outcomes to the general population or specified subpopulations (e.g., potential high-risk groups).

—

The diagnostic and detection methods.

—

The experimental conditions.

—

Personal characteristics that might affect exposure and outcome (e.g., age, sex, and pre-existing conditions).

•

Epidemiologic. Outcome and disease data are collected on groups of people in real-world settings. These data should be accompanied by:

—

A description of the hypotheses tested.

—

Criteria applied to select groups observed.

—

Study methods and target-group participation rates.

—

Diagnostic criteria for clearly defined outcomes.

—

Exposure history and characteristics, including period and doses relevant to outcome studied.

—

Evaluation of characteristics that might affect exposure and outcome (e.g., age, employment, activity patterns, and pre-existing health conditions).

—

Appropriate statistical analyses of comprehensive outcome measures (e.g., point and range estimates, dose-response data, time-response analysis, and measures of association and significance)

—

Interpretation of the findings, including analysis of generalizability, bias, and other confounding issues.