LAST REVISED : 01/21/98, converted to Javascript 7/25/2024
Program JOLLY output contains analyses for as many as 5 different capture-recapture models.
Model A is the standard Jolly-Seber model for open populations (Jolly 1965; Biometrika 52: 225-241).
Model A' is the "Death But No Immigration" model of Darroch, (1959: Biometrika 46: 344-349) and Jolly (1965; Biometrika 52: 241-242).
Model 2 is the capture-resighting model of Brownie and Robson (1983; Biometrics 39:437-453) permitting an effect of initial capture and tagging on first period survival rate.
Model B is the Jolly-Seber model with survival rate assumed constant per unit time and time-specific capture probability.
Model D is the Jolly-Seber model with both survival rate and capture probability assumed constant per unit time.
The following capture-history codes have been added:
3 - not captured in time i, resighted between i and i+1,
4 - captured in time i, resighted between i and i+1.
Definitions and Notation
M(i) = Estimated number of marked animals in the population at time i.
p(i) = Estimated probability that an animal alive at time i is captured in the i-th sample
N(i) = Estimated population size at time i
PHI = Estimated probability that an animal live at time i survives to time i+1
PHI* = Estimated probability that an animal caught for the first time in sample i survives to time i+1 (Model 2 only)
B(i) = Estimated number of new animals recruited during the interval i to i+1 and alive at time i+1 (Recruitment includes birth and immigration)
SE(x) = Standard error of parameter x including non-sampling error terms
SE'(x) = Standard error of parameter x excluding non-sampling error terms
COV(X(i,j)) = Covariance between estimates X(i) and X(j).
r(i) = Number of animals caught in sample i, and recaptured later
z(i) = Number of animals caught before and after sample i, but not caught in sample i
z'(i) = z(i) + animals caught for the first time subsequent to sample i"
Notes:
Estimates of PHI,N,M,and B under models A and A' are computed using bias corrected formulae of Seber (1973:204; The Estimation of Animal Abundance and Related Parameters, Griffin, London)"
except when z[i]=0;{ N(i)= Number caught in sample i.
95% Confidence intervals are computed using the standard error which includes sampling variation (SE(x)).
Estimates which cannot be computed due to poor data (m(i), r(i),z(i),... = zero) are denoted by "Div/0".
(Means do not include these values.)
Example 1- capture-histories, Title=ROBUST DESIGN PAPER: MICROTUS DATA - MONTHS 10-15: