In any particular situation, dealing with environmental risk involves two steps:

Transcription

1 Presented by: Gargi Bhatt 04/27/2005 ENVIRONMENTAL RISK ASSESSMENT Introduction: The field of Environmental statistics is one of the rapid growths at the moment. Environmental decision making is prevalent in much of the world, and politicians and other decision makers are requesting new tools to for understanding the state of the environment. A successful environmental risk assessment requires the talents of many academic disciplines: statistics, chemistry, physics, biology, ecology, geology, hydrology and engineering. Because of their skills in analyzing data and computations, statisticians can play a role in each one of the steps above. Additionally, a statistician can play a key role in the process because of the complexity of a risk assessment. Data or modeling uncertainty in any one of the steps may have a significant effect on the results and meaning of the assessment. Therefore, the description of uncertainty in the assessment, which is a skill area of statisticians, becomes a crucial part of communicating the results of a complex assessment. Key words: environmental monitoring, design, compositional data, risk assessment, spatial covariance, visualization. In any particular situation, dealing with environmental risk involves two steps: 1 Risk assessment: Assessment of risk associated with the environment involves the use of factual information to define the health effects of exposure of humans and/or the environment to hazardous material and situations. For example, this involves quantification of the extent to which a chemical moves from the location of a spill via one or more exposure pathways to create an exposure scenario for a nearby resident, and a determination of the deleterious consequences, if any, of that exposure scenario. It includes the following: Environmental assessments and environmental impact analyses Ecological and human health risk assessments for programs, projects, chemicals,

2 Products and natural resources Stressor identification Toxicology and exposure assessment Microbial risk assessments Hazard identification Custom model and software development Statistical (multivariate, time series, regression, probabilistic) approaches Uncertainty and variability analyses 2. Risk management: Management of the risk associated with the environment considers the process of weighing policy and remediation alternatives, integrating the results of risk assessment with engineering data and with social economic and political concerns to reach a decision involving the management of risk. It includes: Risk-based decision analysis models, databases, and software Comparative risk assessments Qualitative evaluations Adaptive environmental management Cost-benefit analysis, net present resource valuation, and habitat equivalency estimation Bayesian decision analysis Product stewardship Issue tracking Regulatory process implementation (chemical registration, NPDES permitting, TMDLs, Water Quality Criteria, etc.) Environmental policy analysis and negotiation Industrial ecology, energy efficiency, and life-cycle analyses Expert opinions Stakeholder involvement, community outreach, public relations, and education Risk communication Estimation of risk: Assessment of risks can be done by various methods. Mathematical models to quantify the exposure scenarios are important elements of the assessment, probability distributions and statistical tools as information are utilized as a part of the mathematical modeling effort. Numerous model classification systems exist, including: The selection of a model is influenced by at least three conditions: 1. The judgment of the investigator in the specific application of the various modeling techniques. 2. The extent of the database

3 .3.The physical system being modeled. The next level of concern is whether the various elements of the system combine to create significant potential exposure risk. The result is there are four methods to estimate the risk. The steps in the risk assessment process are: 1. Hazard identification is the analysis of an environmental situation to ascertain if there is the potential for an exposure of an organism (including a human) or ecosystem to an environmental stressor that may cause harm. 2. Dose-response assessment is the process of characterizing the relation between the dose of an agent received by a receptor (organism or ecosystem) and the incidence of an adverse effect on that receptor. 3. Exposure assessment is the process of measuring or estimating the intensity, frequency, or duration of a human or ecological exposure to agents that are currently in the environment, or may be present in the future. 4. Risk characterization is the process of estimating the incidence of an adverse effect under the conditions of exposure described in the exposure assessment. It also includes the narrative description of the meaning of the assessment, including uncertainties in the preceding steps. Recently much emphasis has been put on uncertainty analysis of these risk assessments. Primarily it has been noted that the values of the parameters in the model are subjected to uncertainty, which then propagates through the whole assessment and results in uncertainty about the final risk. The method of probabilistic risk analysis is used to assess the risk. Each one of the steps requires the collection of data (such as the collection of lead concentrations in soil or water) and/or the use of mathematical models (such as ones that describe the movement of contaminants in the environment or define the cancer incidence from exposure to levels of uranium). Risk assessments have now become part of the analysis that a proposed new chemical or other product must undergo before it may be placed on the market. Risk assessment methods (Uncertainty analyses method): Probabilistic risk assessment techniques are widely used which consider the variability and uncertainty in parameter values and models. The best known probabilistic techniques to incorporate uncertainty are: 1 Monte Carlo analysis 2 Latin-Hypercube procedures.

4 Monte Carlo analysis: In Monte Carlo simulation, distributions are specified for each parameter that account for both variability and uncertainty. Uncertainty plays a major role in all risk assessment related to environmental quality. It allows sensitivity analyses.monte Carlo method requires few assumptions. What is Monte Carlo analysis? Monte Carlo analysis is a method initially developed in the 1940s that uses statistical sampling techniques to obtain a probabilistic approximation to the solution of a mathematical equation or model. As such, it is a tool that can be used for conducting probabilistic risk assessments. A probabilistic risk assessment is an assessment that estimates the probability or likelihood that particular risk values would result from exposure to contaminants at a site. In the context of an ERA, a probabilistic risk analysis involves the use of methods for estimating the probability that ecological receptors will be harmed by environmental contaminants. Although there are some ERAs where probabilistic analyses have been included, all ERAs are required to calculate a single value or point estimate for evaluating whether a contaminant will harm a particular ecological receptor (e.g., a hazard quotient). When single values are used to describe risks in an ERA, it is known as a deterministic risk analysis. In a deterministic risk analysis, it is implied that the determined risk values adequately represent the risks to organisms and little or no information is provided about the probability that a particular value will result from exposure to a site s contamination. Monte Carlo analysis is used to determine the probability of occurrence for the point estimates of a deterministic risk Assessment and, in this way, deal with the uncertainty associated with these assessments. Some examples of Monte Carlo analysis: Example 1: A Deterministic Model for Compound Interest Monte Carlo simulation is a method for iteratively evaluating a deterministic model using sets of random numbers as inputs. This method is often used when the model is complex, nonlinear, or involves more than just a couple uncertain parameters. A simulation can typically involve over 10,000 evaluations of the model, a task which in the past was only practical using super computers. Example 2: Stochastic model By using random inputs, you are essentially turning the deterministic model into a stochastic model. Example 2 demonstrates this concept with a very simple problem.

5 Example 2: A Stochastic Model for a Hinge Assembly In Example 2, we used simple uniform random numbers as the inputs to the model. However, a uniform distribution is not the only way to represent uncertainty. Before describing the steps of the general MC simulation in detail, a little word about uncertainty propagation: The Monte Carlo method is just one of many methods for analyzing uncertainty propagation, where the goal is to determine how random variation, lack of knowledge, or error affects the sensitivity, performance, or reliability of the system that is being modeled. Monte Carlo simulation is categorized as a sampling method because the inputs are randomly generated from probability distributions to simulate the process of sampling from an actual population. So, we try to choose a distribution for the inputs that most closely matches data we already have, or best represents our current state of knowledge. The data generated from the simulation can be represented as probability distributions (or histograms) or converted to error bars, reliability predictions, tolerance zones, and confidence intervals. (See Figure 2). Uncertainty Propagation Figure 2: Schematic showing the principal of stochastic uncertainty propagation. (The basic principle behind Monte Carlo simulation.) The steps in Monte Carlo simulation corresponding to the uncertainty propagation shown in Figure 2 are fairly simple, and can be easily implemented in Excel for simple models. All we need to do is follow the five simple steps listed below: Step1: Create a parametric model, y = f(x 1, x 2,..., x q ).

6 Step2: Generate a set of random inputs, x i1, x i2,..., x iq. Step 3: Evaluate the model and store the results as y i. Step 4: Repeat steps 2 and 3 for i = 1 to n. Step 5: Analyze the results using histograms, summary statistics, confidence intervals, etc. Latin-Hypercube procedure: It is an alternative to the Monte Carlo procedure, but it is a more efficient method of parameter sampling. The procedure requires the division of the input parameter distribution into N segments of equal probability,1/n where N is the number of realizations. An equal number of samples are drawn from each segment to provide a representation of the full range of possible input parameter values. This distribution method ensures that the tail of the distribution are efficiently accounted for, a feature that is not ensured when using the Monte Carlo technique. This sampling technique procedure is performed for each input variable that has been assigned a level of uncertainty. Examples on Latin Hypercube design: A method of comparison of leakage rates through leach ate collection system at the bottom of a landfill uses 100 Latin hypercube simulations in order to represent the range of possible values for the input parameters assigned to characterize the level of uncertainty. A major component for the design of a landfill involves the leach ate collection and low permeable linear system placed at the bottom of a landfill. The system includes features such as high permeability (high transmissivity) drainage layer and drainage tiles and low permeability liner or barrier underlying the drainage layer and tiles. The low permeability liner is constructed of a single material of natural soils or FML s. Other applications of environmental risk assessment: 1 Identifying, evaluating, and managing persistent, bioaccumulative, toxic (PBT) substances and other chemicals 2 Modeling environmental pathways and assessing exposures for a wide variety of pollutants 3 Developing statistical methods, models, and guidance 4 Conducting statistical risk-based ranking, planning, and evaluation 5 Providing regulatory impact assessment and support 6 Assessing site-specific human health and ecological risks, in support of developing cleanup standards and selecting remedies 7 Assessing risks at currently operating and planned industrial facilities 8 Conducting statistical evaluations of site and background data 9 Biological sciences applications. 10 Computational statistics.

7 REFERENCES: Book: Statistical Procedures for analysis of environmental monitoring data and risk assessment by Edward A.McBean and Frank A. Rovers. 4