Environmental Engineering Reference
In-Depth Information
Box 2.9
Estimating survival rate from population estimates.
In the simplest case, in which numbers of young of the year,
N
0
, and older,
N
1
,
individuals are estimated at two or more points in time (usually annually), time-
specific but age-independent survival rate,
S
t
, is given by:
N
1
,
t
1
N
t
S
t
However, in long-lived species, demographic rates may vary greatly with age.
Age-specific variation in survival rate can be quantified by generalising the above
equation for any age class,
a
:
N
a
1,
t
1
N
a
,
t
S
a
,
t
Because older age classes are generally scarce, and survival rate often reaches a
roughly stable value after a period of maturation, it will usually be both sensible
and reasonable to define a maximum age class,
A
, beyond which survival is
assumed to be constant. In this case, the survival rate for individuals aged between
A
and the maximum recorded age,
A
max
, is:
A
max
N
a
,
t
1
a
A
1
S
A
,
t
A
max
1
N
a
,
t
a
A
The problem with this approach is that it relies on very accurate and precise age-
specific population estimates, which will rarely be achievable. A more practical
approach in some cases might be to quantify only total population size, and carry
out independent surveys of age-structure based on a sampling method that takes
account of any differences in detectability between age classes. However, this adds
further sampling error to the calculation. Relatively small amounts of error in the
constituent numbers propagate into considerable error in survival rate, which has
a strong influence on the assessment of sustainability. Given the low degree of
precision yielded by most methods for estimating abundance and age structure,
estimates of survival rate based on this approach should be treated with caution.
Static life tables
Given the difficulty of obtaining precise age-structured population estimates,
count-based calculation of age-specific survival rate more commonly uses a sample
of the population to define age structure. Samples may be taken visually if age
classes are sufficiently obvious in the field, but otherwise depend on the retrieval of
dead individuals, usually through harvest. Methods for determining age depend
