Laserfiche WebLink
In concert with systematic sampling of habitat parameters over the entire study area, systematic <br />monitoring of species use will also occur throughout the entire study area. The monitoring <br />protocols pertaining to species use do not involve definitions of habitat types ("suitable habitat") <br />within which sampling is concentrated or restricted. Protocols for monitoring use locations of <br />the target species are designed to allocate known search effort throughout a defined study area, <br />regardless of habitat suitability. Thus, habitat characteristics at use locations documented by the <br />monitoring will be contrasted to habitat characteristics throughout the study area. In some <br />protocols, the study area has been defined through the use of habitat (fish monitoring will only <br />occur in water) but this has only occurred when the habitat can be defined by the Program. It is <br />recognized that the species monitoring protocols collect data only on individuals using the <br />central Platte River (the lower Platte River for pallid sturgeon), not on the entire whooping crane, <br />least tern, and piping plover populations. Therefore, these results are only applicable to the <br />populations' use of this area and are biased for inference to the entire population. <br />V.E.2. Research Design <br />The hypothesized relationships among species and habitat associations and species and habitat <br />response to Program land and water manageinent (treatments) will be evaluated with research. <br />Proper research designs will produce accurate and precise results with an efficient use of <br />resources. Research designs will include both experimental and observational studies. <br />Inferences to the cause and effect relationship will be possible with experimental research while <br />inferences with observational studies will be limited to associations (Keuhl 1994). <br />There are many components of the statistical design of experimental research. Each IMRP <br />research project will be designed for a specific research question and will ideally contain the <br />following components: controls, randoinization, and replication. The use of control areas will <br />enable efficient estimation of treatment effects (Keuhl 1994). Without controls, there will be no <br />benchmark estimate of changes that would have occurred in the treatment areas regardless of the <br />treatment (Keuhl 1994). Randomizatiou is a critical component of experimental design. <br />Randomization applies to the selection of experimental units from the population and forms the <br />basis of the applicability of research results to the population. Randomization also applies to the <br />application of the treatments to experimental units enabling the experiment to account for <br />confounding factors (Neter et al. 1996). Replication refers to the duplication of the study design <br />to multiple experimental units. Replication provides an estimate of experimental error and <br />increases the precision for tests of treatment effects (Neter et al. 1996). Analyses of <br />experimental research with controls, randomization, and replication are described in many <br />statistical texts (Box et al. 1978, Keuhl 1992, Neter et al. 1996). <br />In the cases where IMRP research proj ect designs are not able to incorporate controls, <br />randomization, and replication, constrained study designs will be developed (Skalski and Robson <br />1992). It is anticipated that system wide Program effectiveness research will not always be able <br />to incorporate each of the statistical design components because of the lack of replication at the <br />treatment level (the Platte River) and lack of control areas (water treatments effecting the entire <br />reach). Instead, small scale manipulative studies will be conducted on Program lands and <br />inferences will be restricted to the project area with system wide conclusions left to professional <br />judgment. <br />September l, 2006 Adaptive Management Plan 36