been accomplished, decisions concerning the kind of data to collect, the precision in estimates needed to meet objectives, and the actual sampling locations can be made. No attempt is made here to provide a complete review of the literature regarding the design of transuranic field studies. However, the references included should illustrate various designs that have been used in past studies. Our discussion of possible approaches to deciding whether cleanup is needed should be considered only as a guide to what might be done in actual practice. To our knowledge, no cleanup operation has yet been conducted using the probability approach described here. Practical problems in the field will require modifi-~ cation of our suggestions. PRINCIPAL STEPS IN DESIGNING TRANSURANIC FIELD STUDTES Certain factors should be considered in designing transuranic field studies. The description of these which follows has been adapted from Cochran (1963, pp. 5-8). Objectives Identification of objectives is necessary in order to choose the most appropriate field sampling design (see Eberhardt, 1976, for discussion). Since most studies have multiple objectives, a compromise design may be necessary. Some common objectives are: 1. Geographical distribution (spatial pattern) of the contamination over the study site (horizontal and with depth). 2. Inventory (total amount) of contamination present in soil (worldwide fallout or local contamination). 3. Amount of radionuclides present in ecosystem components (soil, vegetation, air, invertebrates, vertebrates). 4, Changes in concentration in ecosystem components over time. 5. Rate of movement of radionuclides from one component to another (soil to vegetation, sediments to water, etc.). 6. Resuspension by wind or from the activities of man and animals. 7. Determination of whether a cleanup operation is necessary or to verify that cleanup has been successful. 8. Evaluation of the hazard to man from a given contamination (this may include all of the above objectives). 576 Population to Be Sampled Cochran (1963) defines "population" to be the aggregate from which the sample is chosen. It is important to distinguish between the "target population" and the "sampled population." In transuranic field studies, the target population is that part of the environment about which information is wanted. The sampled population refers to the population from which samples are actually collected. Inferences always pertain to the sampled population. This is an important point since the sampled population may not correspond to the target population. As an example, consider sampling the sediments of a pond known to be contaminated by plutonium. The objective might be to estimate the average Pu concentration in these sediments. The target population could be the sediment layer in all areas of the pond, whereas the sampled population might be restricted to certain portions of the pond due to practical reasons. Since samples are drawn from only the "sampled population," conclusions from these samples may not apply to the sediments in those areas of the pond that could not be sampled. As another example, consider sampling an area contaminated locally by a nuclear detonation or safety-shot test. The "target population" for soil at these sites is usually surface soil (to some specified depth) over the entire site regardless of terrain, rock content of the soil, vegetation cover, etc. Efforts should be made to insure that the "sampled population" is identical to the “target population” so that no portion of the study area is systematically excluded from sampling. For some kinds of field studies, this goal may be impossible to achieve. For example, suppose that we wish to estimate the total deposition of transuranics from worldwide fallout since the start of atmospheric testing in 1945. The target population could be considered to be the surface area of the entire world, whereas the sampled population is restricted to land sites that have acted as fallout collectors since 1945. Those parts of the target population not included in the sampled population are bodies of water, inaccessible land areas such as mountains, locations where transuranics might have accumulated or washed away, or undisturbed sites that could not be sampled due to practical problems, or deliberate exclusion. Data to Be Collected Careful consideration must be given to insure that all data required to meet the study objectives will be collected. This step requires a clear understanding of objectives so that all essential information is obtained. It may also be necessary to collect certain data in order to properly interpret other data. For example, if im sttu (field) measurement instruments are used, it may be necessary to collect additional soil samples to calibrate the in situ readings with wet chemistry soil concentrations. These special studies may need to be done before the main study begins. Precision Desired Any estimated quantity is subject to error since only a small proportion of potential environmental samples can be collected and analyzed. This error can be reduced in a number of ways such as by taking more samples, using more efficient designs, and statistical analyses, or using more accurate instrumentation and wet chemistry techniques, Since these techniques usually require