Approaches to Impact Prediction EIA for Developing Countries - 1 4 . 0 Approaches to Impact Prediction

The scientific and technical credibility of an environmental impact assessment (EIA) relies on the ability of the EIA practitioners to estimate the nature, extent, and magnitude of change in environmental components that may result from project activities. Information about predicted changes is needed for assigning impact significance, prescribing mitigation measures, and designing and developing environmental management plans and monitoring programs. The more accurate the predictions, the more confident the EIA practitioner will be in prescribing specific measures to eliminate or minimize the adverse impacts of development projects. Most of the EIA methods described in Chapter 3 focus on clearly identifying impact pathways. They rely on underlying conceptual models linking project activities to changes in environmental components. In using these methods, predictions of the degree of change may be qualitative or quantitative. Qualitative approaches depend heavily on expert judgement; quantitative approaches rely on mathematical models developed by experts. This chapter discusses some of the technical and scientific methods available to predict environmental changes. It focuses on quantitative models for prediction. 4.1 Predictive Methods In their review of EIA methods, Canter and Sadler (1997) provide a listing of prediction techniques applicable to different aspects of EIA (Table 4-1). Canter (1996) provides an excellent overview, based on American experience, of many of these prediction techniques. In many EIA applications, these basic prediction techniques are actually combined. This is particularly true when using computerized modeling software for specific applications, as the application of a computer model usually requires collection of environmental information to set baseline values for the model's variables and to determine the values for model's parameters. Air 1. emission inventory 2. urban area statistical models 3. receptor monitoring 4. box models 5. single to multiple source dispersion models 6. monitoring from analogs 7. air quality indices Surface Water 1. point and nonpoint waste loads 2. QUAL-IIE and many other quantitative models 3. segment box models 4. waste load allocations 5. water quality indices 6. statistical models for selected parameters 7. water usage studies Ground Water 1. pollution source surveys 2. soil and/or ground water vulnerability indices 3. pollution source indices 4. leachate testing 5. flow and solute transport models 6. relative subsurface transport models Noise 1. individual source propagation models plus additive model 2. statistical model of noise based on population 3. noise impact indices Biological 1. chronic toxicity testing 2. habitat-based methods 3. species population models 4. diversity indices 5. indicators …