Evolutionary Ecology
 | Demographic estimation methods | |
Adapting mark-recapture statistics for long-lived plants
One of my main research goals is to develop new ways of estimating demographic
and evolutionary parameters in long-lived organisms. After all, fitness is difficult to
estimate if you cannot quantify the lifetime reproductive success of an individual.
This is a significant problem in long-lived organisms, and particularly so in clonal
herbaceous plants.
One of my main research goals is to develop new ways of estimating demographic
and evolutionary parameters in long-lived organisms. After all, fitness is difficult to
estimate if you cannot quantify the lifetime reproductive success of an individual.
This is a significant problem in long-lived organisms, and particularly so in clonal
herbaceous plants.
Some herbaceous plants do not sprout in a given growing season, and may
continue in such a state for many years. This condition, refered to as 'adult
whole-plant dormancy', or 'dormancy' for short, occurs in a fairly large and diverse
group of plants. How can we estimate survival and other important demographic
parameters for such plants?
continue in such a state for many years. This condition, refered to as 'adult
whole-plant dormancy', or 'dormancy' for short, occurs in a fairly large and diverse
group of plants. How can we estimate survival and other important demographic
parameters for such plants?
Caption: An individual orchid monitored over three years. In year 1, the plant is visible, but we
do not find it in year 2. Although we suspect that the plant is dead, when we check its position in
year 3, we find that the plant has sprouted again. If we assume, perhaps for morphological or
density reasons, that the plants we saw in years 1 and 3 are the same, then what happened in
year 2? One possibility is dormancy, and I address how to estimate the probability of dormancy
as well as survival in dormancy-prone plants in Shefferson et al. 2001 (Ecology).
do not find it in year 2. Although we suspect that the plant is dead, when we check its position in
year 3, we find that the plant has sprouted again. If we assume, perhaps for morphological or
density reasons, that the plants we saw in years 1 and 3 are the same, then what happened in
year 2? One possibility is dormancy, and I address how to estimate the probability of dormancy
as well as survival in dormancy-prone plants in Shefferson et al. 2001 (Ecology).
Mark-recapture analysis provides biologists with the means to estimate
demographic parameters even under circumstances when individuals in the
population cannot be perfectly detected. Cormack-Jolly-Seber (CJS) analysis
incorporates an open population framework, in which birth, death, and migration
occur, and uses multinomial probabilistic modeling to estimate survival in the face
of a variable population. No bias occurs due to imperfect detection because these
models are able to use information regarding when individuals were observed to
estimate the probability that an individual that is no longer observed is either alive
(and undetected) or dead. My first contribution into mark-recapture statistics was to
adapt these models for use in dormancy-prone plant populations.
Simply estimating survival is not enough. Transition probabilities are often
necessary in order to piece together the life cycle of an organism, and are crucial in
population viability analysis. Multi-strata (MS) modeling is a form of mark-recapture
analysis in which individuals may not only be born, die, and move in or out of the
population, but may also move from one life stage to another. I have adapted these
analytical methods to dormancy-prone plants, estimating transitions to and from
dormancy, and survival of dormant plants using multi-strata methods.
Demographic parameters may be functions of factors that vary among individuals.
In both CJS and MS modeling, individual covariates provide a means of estimating
how survival, dormancy, and transitions vary with such factors as size and vigor.
Further work is ongoing. Feel free to contact me for more information.
demographic parameters even under circumstances when individuals in the
population cannot be perfectly detected. Cormack-Jolly-Seber (CJS) analysis
incorporates an open population framework, in which birth, death, and migration
occur, and uses multinomial probabilistic modeling to estimate survival in the face
of a variable population. No bias occurs due to imperfect detection because these
models are able to use information regarding when individuals were observed to
estimate the probability that an individual that is no longer observed is either alive
(and undetected) or dead. My first contribution into mark-recapture statistics was to
adapt these models for use in dormancy-prone plant populations.
Simply estimating survival is not enough. Transition probabilities are often
necessary in order to piece together the life cycle of an organism, and are crucial in
population viability analysis. Multi-strata (MS) modeling is a form of mark-recapture
analysis in which individuals may not only be born, die, and move in or out of the
population, but may also move from one life stage to another. I have adapted these
analytical methods to dormancy-prone plants, estimating transitions to and from
dormancy, and survival of dormant plants using multi-strata methods.
Demographic parameters may be functions of factors that vary among individuals.
In both CJS and MS modeling, individual covariates provide a means of estimating
how survival, dormancy, and transitions vary with such factors as size and vigor.
Further work is ongoing. Feel free to contact me for more information.
Copyright 2010 Richard P. Shefferson. All rights reserved.
Research Interests
 |
| |
 |
| |
 |
| |
 |
| |
 |
| |
 |
| |
 |