This study examines opportunities for renewable resource management when population data are collected by spatial subdivisions. In particular I look at potential applications for the design of management experiments, the distribution of monitoring resources, and the improvement of parameter estimation.
Methods are developed to rank possible groupings of subdivisions for use as experimental units. Factors external to the experiment can cause differences between experimental units. Selecting subdivisions that have reacted similarly in the past to external factors could minimize the risk of external factors creating differences in experimental units.
Methods are developed to identify subdivisions that could provide information about similar subdivisions when monitoring resources are low or when stratified sampling is being used. The use of these subdivisions as "index units" could notify managers of extremely good or bad years in a large number of subdivisions.
Two methods developed by Walters (1986) provide innovative estimation techniques that can be used with subdivided populations. A Bayesian approach allows parameter estimates to be adjusted using a known distribution. Another approach allows similar subdivisions to be estimated jointly more accurately than would be possible individually. Not all renewable resource data sets provide reliable information for use with these applications. Data sets where there is little common variation, high levels of autocorrelation in the noise, or even modest amounts of measurement error are inappropriate for most methods. A series of steps is introduced for managers to test the reliability of the methods on their particular data sets.
Data on Finnish reindeer (Rangifer tarandus tarandus) are used throughout the thesis to illustrate the methods. The reindeer data appear to be appropriate for these methods when tested using the steps developed. Possible experimental units and index units for monitoring are identified. Walters' (1986) methods of parameter estimation are used on the data set as well.
The reindeer data show that subdivisions with similar external effects were located close to one another. This pattern was at least partially caused by the existence of extremely bad years occurring within geographic regions. The reindeer subdivisions are very highly managed and provide little evidence of any kind of density dependence. Managers could potentially benefit by conducting experiments to test the biological limits of the population growth rates and carrying capacities within subdivisions. / Science, Faculty of / Zoology, Department of / Graduate
Identifer | oai:union.ndltd.org:UBC/oai:circle.library.ubc.ca:2429/27799 |
Date | January 1988 |
Creators | Berkson, James Meyer |
Publisher | University of British Columbia |
Source Sets | University of British Columbia |
Language | English |
Detected Language | English |
Type | Text, Thesis/Dissertation |
Rights | For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. |
Page generated in 0.0015 seconds