Developing effective sampling designs for monitoring natural resources in Alaskan national parks: An example using simulations and vegetation data

Monitoring natural resources in Alaskan national parks is challenging because of their remoteness, limited accessibility, and high sampling costs. We describe an iterative, three-phased process for developing sampling designs based on our efforts to establish a vegetation monitoring program in southwest Alaska. In the first phase, we defined a sampling frame based on land ownership and specific vegetated habitats within the park boundaries and used Path Distance analysis tools to create a GIS layer that delineated portions of each park that could be feasibly accessed for ground sampling. In the second phase, we used simulations based on landcover maps to identify size and configuration of the ground sampling units (single plots or grids of plots) and to refine areas to be potentially sampled. In the third phase, we used a second set of simulations to estimate sample size and sampling frequency required to have a reasonable chance of detecting a minimum trend in vegetation cover for a specified time period and level of statistical confidence. Results of the first set of simulations indicated that a spatially balanced random sample of single plots from the most common landcover types yielded the most efficient sampling scheme. Results of the second set of simulations were compared with field data and indicated that we should be able to detect at least a 25% change in vegetation attributes over 31 years by sampling 8 or more plots per year every five years in focal landcover types. This approach would be especially useful in situations where ground sampling is restricted by access.     

Statistical power for detecting trends with applications to seabird monitoring

Power analysis is helpful in defining goals for ecological monitoring and evaluating the performance of ongoing efforts. I examined detection standards proposed for population monitoring of seabirds using two programs (MONITOR and TRENDS) specially designed for power analysis of trend data. Neither program models within- and among-years components of variance explicitly and independently, thus an error term that incorporates both components is an essential input. Residual variation in seabird counts consisted of day-to-day variation within years and unexplained variation among years in approximately equal parts. The appropriate measure of error for power analysis is the standard error of estimation (S.E._est) from a regression of annual means against year. Replicate counts within years are helpful in minimizing S.E._est but should not be treated as independent samples for estimating power to detect trends. Other issues include a choice of assumptions about variance structure and selection of an exponential or linear model of population change. Seabird count data are characterized by strong correlations between S.D. and mean, thus a constant CV model is appropriate for power calculations. Time series were fit about equally well with exponential or linear models, but log transformation ensures equal variances over time, a basic assumption of regression analysis. Using sample data from seabird monitoring in Alaska, I computed the number of years required (with annual censusing) to detect trends of −1.4% per year (50% decline in 50 years) and −2.7% per year (50% decline in 25 years). At α=0.05 and a desired power of 0.9, estimated study intervals ranged from 11 to 69 years depending on species, trend, software, and study design. Power to detect a negative trend of 6.7% per year (50% decline in 10 years) is suggested as an alternative standard for seabird monitoring that achieves a reasonable match between statistical and biological significance.     

Stochastic variation in reproductive success of a rare frog, Rana sevosa: implications for conservation and for monitoring amphibian populations

Although amphibian populations are thought to be declining in many parts of the world, detailed information on populations in decline are often not available. From 1988 to 2001, we studied temporal variation in the reproductive biology of the only known population of dusky gopher frogs, Rana sevosa Goin and Netting. We found high annual variation in reproductive effort, mortality at the egg and larval stages, and hydroperiod length. No overall trends were apparent in terms of either number of egg masses deposited or in reproductive success, as we found extensive variation among years in the number of egg masses deposited, a high rate of reproductive failure, and no consistent relationship between the number of females present, the number of eggs deposited, and the number of metamorphs emerging. Given the complete isolation of this population from other gopher frogs and the high rate of reproductive failure, the probability of extinction of this population appears to be quite high (0.125-0.316).