Laserfiche WebLink
<br />" <br /> <br />Concepts and Definitions <br />Statistical concepts <br /> <br />Statistical errors in trend detection, given a <br />specific sampling design and estimation <br />procedure: <br />- Type I: What Is the probabifity that we infer a <br />significant trend in population size when there is <br />none? <br />- Type II: Given a specific trend in population size, <br />what is the probability that we detect it? <br /> <br />Panel Assessment of Data, <br />Estimation Methods, Sampling <br />Designs, and Constraints <br />Upper Basin <br /> <br />4. Relatively short term ( 4 years) data set <br />5. Fall sampling is less problematic and <br />produces larger capture probabilities <br />6. Available recapture data among years not <br />currently utilized <br /> <br />Panel Assessment of Data, <br />Estimation Methods, Sampling <br />Designs, and Constraints <br />Grand Canyon <br /> <br />6. Long term data set (15 years) <br />7. Appropriate to use age-specific open models <br />to take advantage at among year <br />information, but biases in estimates of N are <br />possible due to unaccounted variation in <br />capture probabilities <br />8. ASMR model based on similar foundation to <br />well-established Jolly-Seber Age models <br />9. No formal model selection in ASMR <br /> <br />Panel Assessment of Data, <br />Estimation Methods, Sampling <br />Designs, and Constraints <br />Upper Basin <br /> <br />1. Number of sampling occasions and capture <br />probabilities are low <br />2. Geographic and demographic closure <br />assumptions are probably OK (if Black <br />Rocks and Westwater combined) <br />3. Reasonable population estimates are <br />possible for Black Rocks and Westwater <br />populations, but are problematic for Vamps, <br />Desolation, and Cataract <br /> <br />Panel Assessment of Data, <br />Estimation Methods, Sampling <br />Designs, and Constraints <br />Grand Canyon <br /> <br />1. Capture probabilities are generally higher <br />than UB and higher in spring (LCR) than in <br />fall (LCR and mainstem) <br />2. Large spring population size in LCR <br />3. Only 2 sampling occasions within a season <br />4. Geographic closure during spring sampling <br />of LCA a potential problem due to timing of <br />spawning migration <br />5. Demographic closure within a year OK <br /> <br />Recommendations <br /> <br />1. Continue fall sampling in the Upper Basin <br />and spring sampling in LCA only <br />2. Increase number of sampling occasions <br />within a year to 4 in the Upper Basin <br />3. Increase number of sampling occasions to 4 <br />in the LCA or conduct telemetry to improve <br />assessment of potential bias in population <br />estimates and geographic closure <br />4. Conduct cross-validation of ASMR and Jolly- <br />Seber~Age method <br /> <br />02301 <br /> <br />3 <br />