All:
The announcement below was just sent to my by James Gibbs. I have used
versions of the program below for over 10 years now and this latest
version is even more useful and user friendly. Its been a big help in our
work setting sample sizes for all sorts of monitoring programs.
Queries should go directly to James Gibbs [email protected]
sam
Sam Droege
w 301-497-5840 h 301-390-7759 fax 301-497-5624
USGS Patuxent Wildlife Research Center
BARC-EAST, BLDG 308, RM 124 10300 Balt. Ave., Beltsville, MD 20705
Http://www.pwrc.usgs.gov
Excuse my wandering.
How can one be orderly with this?
It's like counting leaves in a garden,
along with the sound notes of partridges,
and crows.
Sometimes organization
and computation become absurd.
Rumi
-----------------------------------------------------------------------------------------
A Windows version of Program MONITOR is now available at:
http://www.esf.edu/efb/gibbs/monitor/
Program MONITOR uses simulation procedures to estimate the statistical
power of ecological monitoring programs. Because many design variables
interact to determine the capacity of a monitoring program to detect
trends, intuitive answers are elusive for such basic questions as: "How
many plots should I monitor?" or "How often should I conduct surveys?" or
"For how long should I monitor?"
The specific components of sampling design that program MONITOR integrates
are:
? Number of plots monitored
? Frequency of plot re-measurement each season
? Occasions of plot surveys through time
? Magnitude of measurement values on each plot
? Variation in measurements on each plot (all sources of variation pooled
together, or within- ("sampling") and between-season ("process") variation
partitioned separately)
? Magnitude of ongoing population trends
? Significance level associated with trend detection
? Number of tails considered in statistical tests of trend significance
? Data type: Normally distributed and log-normally distributed or
presence/absence
? Pattern of variation in the relationship of magnitude of measurements
versus their variance over time (coefficient of variation constant versus
proportional to the mean measurement)
? Whole versus fractional measures (rounding)
? Measurement truncation
? Data type: Presence/absence
Most users will apply Program MONITOR to population monitoring but the
program is applicable to any designing monitoring programs for any index
measured over time and space. Program MONITOR can be used to find an
optimal sampling design once you have declared desired power levels and
maximum sampling effort you are willing to undertake. The program also is
a useful tool for teaching students about trend detection and sampling
design.
