Geoscience Reference
In-Depth Information
Live Recaptures
Possible
Live Recaptures
Not Possible
General Mark-
Recapture
Models
Models for
Returns of
Dead Animals
Two Sample Closed
Populations
Several Samples
Combination of
Open-and Closed-
Population Models
Closed-Population
Models
Open-Population
Models
Jolly-Seber Model
Simpli cations and
Generalizations of
Jolly-Seber Model
Comparison of
Survival Rates in
Di‚erent Populations
Age-Dependent
Model
FIGURE 8.1
Relationships between models for mark-recapture data. (Based on figures provided by Pollock,
K.H.
Journal of the American Statistical Association
86: 225-238, 1991.)
recognized; and (4) all samples are effectively instantaneous and releases are
made immediately after sampling.
The JS estimators were derived using the principle of maximum likeli-
hood. However, they can also be justified intuitively in the manner described
next. First, consider the estimation of the number
M
j
of marked animals in
the population just before the
j
th sample. This consists of the
m
j
marked
animals seen in the
j
th sample plus those that were in the population but
not captured. To estimate the latter, it can be argued that the probability
of an animal in the population at the time of the
j
th sample being recap-
tured in a later sample can be estimated by
z
j
/(
M
j
-
m
j
), where
z
j
is the
number of animals seen before and after the
j
th sample but not in the
j
th
sample, and also by
r
j
/
R
j
, where
R
j
is the number of animals released from
