7143

• UCREFRP Catalog Number: 7143
• Author: Inskip, P. D.
• Title: Habitat Suitability Index Models
• USFW Year: 1982.
• USFW - Doc Type: Northern Pike.
• Copyright Material: NO

water at different times in the future. If it is known that the fluctuation
mostly occurs within a range over which the suitability index does not change,
then no further consideration of the variability is necessary. Problems
arise, however,- when stochastic variation in a model variable can result in
drastically different HSI's for different years. Ideally, sufficient long
term data to estimate probabilities of occurrence of different levels of the
habitat variable would be available. The average length of the frost-free
season (average number of days between the last spring occurrence and first
fall occurrence of an air temperature of 0° C} between 1941 and 1970 is known
for many areas in the United States (National Oceanographic and Atmospheric
Administration 1978). Average values for selected areas in southern Canada
were compiled by Chapman and Brown (19b6). Long term water temperature and
water level data are also available in some cases. It is recommended that
suitability indices used for determination of HSI's be averages of suitability
indices corresponding to measured, estimated, or projected conditions over a
series of years. Even if long term data are not available, the degree to
which water temperature or water level measurements for a single year are
"representative" can be evaluated by comparing air temperature, precipitation,
andior stream discharge data for the year in which the measurements were made
with corresponding values for preceding years. Climatological summaries and
records for U.S. Geological Survey stream gauging stations are possible sources
of such information.
Several reviewers had reservations about including V6 (average length of
frost-free season) in the model. The variable is not useful for distinguishing
between water bodies that are close to one another and at the same elevation.
It is useful when considered over a wider geographic area. If the model is to
be applied over a limited area and comparisons of HSI's are to be made only
for water bodies within that area, then V6 can be deleted.
Daily variations in water temperature can be considerable. Daily means,
rather than extremes, should be used in this model, because northern pike can
tolerate brief exposures to moderately high temperatures. It is recommended
that V, (maximum weekly average water temperature) be determined by averaging
the means of daily minimum and maximum water temperatures for the warmest
(water temperature) week of the year. The minimum daily temperature usually
occurs sometimes near dawn, and the maximum generally occurs in the early to
middle afternoon. Day-to-day monitoring of water temperature is not always
possible. Whatever procedure is used to estimate V, should account for the
daily temperature cycle.
A measurement depth of 1 to 2 m was specified for water temperature (V,).
In cases where curve B is to be used (e.g., unstratified lakes or stratified
lakes with < 1.5 ppm dissolved oxygen in the meta]imnion), measurements should
be made in the littoral zone from 1 m below the surface to the bottom. The
most favorable temperature for northern pike, based on the SI curve for V,,
should be used to determine V,.

26