7910 - Laserfiche WebLink

Home Browse Search

determine, without the encumbrance of personal judgement, locations of study areas which by virtue of representation, can be extrapolated over larger areas. Topographic and geologic maps, stream flow and water quality records, along with species distribution maps and studies, and aerial photographs (where available) collectively provide raw data for the stratification phase. Field tripes over flights are helpful in parti- tioning off rather homogenous stream segments but are often insufficient as the sale means of stratification. A stratified river, segment may be thought of as a series of short reaches with a common morphology. Each of these short reaches is relatively similar to every other short reach within the stratified segment, with the frequency of dissimilar reaches decreasing as a normal distribution function. This means that if several random samples were taken within the population of representative reaches, the probability of obtaining a representative reach "typical" of the stratified segment would be greater than the probability of obtaining one which is "atypical" of the stratified segment. This process is termed "stratified random sampling" and follows accepted statistical sampling methods. Figure 15 illustrates the hierarchy of nomenclature used to focus activities down to a study area. Even though a good job of stratification (identifying the relatively large homogenous segments) has been done, there will always be zones within the stratified reach which are obviously different from the rest of the reach. Consistent with probability theory and the random sampling process, the stratified segment may be simply dissected into a population of small reaches of similar length, and several samples drawn at random from the population. This implicit zonation recognizes that variance between candidate representati- ve reache- reaches does exist, and it is assumed that the different types of candidate reaches will be sampled in the proportion that they occur in the stratified segment. Specifically, it is assumed that the type of reach which occurs the most often is the most likely to be sampled, and the most unusual or atypical reach the least likely to be sampled. The technique of explicit zonation may further be used to ensure that at least one sample is taken from each "different" type of candidate reach. Using the process of explicit zonation, subpopuiations of candidate reaches are generated by a further classification .of reach types within the stratified stream reach. For example, all riffle-pool sequences are clumped together as one discrete population, all meandering reaches as another population, and all braided reaches as yet another population. Explicit zonation essentially means the application of a second stratification process within the larger stratified segment. While this sampling technique ensures that at least one candidate from each reach type will be sampled, the investigator may be faced with the problem of degree of zonation (how different is different?). In the final selection of the representative reach to study, implicit zonation within 415'