Lecture 5 - Estimating animal abundance: Catch-effort Flashcards
What is the catch?
a known number of animals are removed from a habitat
What is effort?
Trapping method
What is the harvest period?
period of time that animals are trapped/caught/removed
Run=time on graph of catch-recovery
What is catch-effort?
a known number of animals are removed (=catch) from a habitat by trapping (=effort) over a period of time (=harvest period)
What is catchability?
The slope of the Leslie catch-effort plot and is the rate at which the trap catches decreases. CPUE decreases as accumulated catch (ki) increases
- slope is exploitation
How do you use the model to estimate the initial size of a population?
the initial size N of an exploited population at the end of the harvest by extrapolating the line to the x-axis
When does this method work?
When the population has been exploited to the point where harvesting at each sample time requires greater and greater catching effort.
ie. # of fish caught in a net declines after each successive fishing trip at a constant rate dictated by the slope.
When is the method used and why?
When a season of harvest is over to calculate the initial population at the beginning of the season
Why is there the possibility of huge bias to be committed?
Related to the size of the net, volunteerism in reporting from the commercial sector, and is bycatch reported at all?
What data is required for the model?
1) catch and effort are used to calculate CPUE
2) accumulated catch ki
What do you do with the data that is collected?
Calculate CPUE and ki and plot CPUE against ki
Calculate the linear regression and extrapolate to the x-axis to determine N (hat)
What happens when CPUE=0?
Represents the exhaustion of the catch
What are the assumptions of the catch-effort model?
1) population is closed
2) individuals have the same probability of being caught throughout the sampling period
3) an animal can be harvested to the point of near extermination
What is meant by a closed population?
N is constant and there are no births/deaths or immigration/emigration
How good is the extrapolation?
It is only as good as the R2 value and if the variability is large the probability model may not be a good fit
What should a linear plot of CPUE(Leslie plot) reveal?
a linear plot should reveal a straight line
What happens if the linear plot regression line is not straight?
Then the basic assumptions of the model are violated
What do you check for the fit of the model?
The R2 (variability of data from the mean/expected?) Parametric assumption/meet statistical assumptions?
What are the parametric assumptions in order of importance?
(No bias)
Homogeneity of variance
Independence
Normality (less important because analysis is robust to variations in normality)
- Brought in on top of model specific assumptions
What can the regression equation be used for?
To determine N (when x=0) but there may be rounding errors (using Krebs this isn’t as much of an issue)
What does Krebs recommend to do in addition to the Leslie plot?
A Ricker Semi-log plot that is used to visually inspect the CPUE data for linearity
What does the Ricker Semi-log plot make possible?
Makes it possible to visually inspect the data to make sure the underlying assumptions are not being violated
The plot may straighten the line and improvement in the line implies assumptions (parametric) are not violated
What is the 10% rule?
A conservative approach to catch-recovery
Allow the population to recover
Only harvest 10% of k
What is a disadvantage of the removal method?
catch-effort has poor precision and a large potential for bias
and it doesn’t work if you don’t meet parametric assumptions
What is the rise in the catch-recovery?
the catch (n, population size)
What do you get if the parametric assumptions are met? Advantage
you can get a population estimate from commercial estimates
If you fit a linear regression model, what ‘baggage’ have you brought?
The fact that parametric assumptions must be met. It becomes a parametric analysis with all the ‘baggage’ that comes with it.
What happens if a population drops below k/2?
approaches extinction/extirpation/ collapse
What is happening if the population is above k/2?
hopefully recovery
What is the MSY?
Maximum Sustainable Yield
- The inflection point at the maximum rise/run
- Stay above for best management
Why do we test for linearity?
So that we can extrapolate the line down to the x-axis and get a good estimate
What kind of slope do we want?
steep slope, can’t use model if slope approaches 0
What is the x-axis where y=0?
where the population total at start of the fishery is and CPUE is 0 (ie no catch)
What do we need to know to get k/2 and how do we get it?
Must know K
- only way to know K is to monitor a lot, volunteerism alone won’t work
What does the Leslie plot give?
The initial population at the end of the catch, but only works if you can draw the population down with a constant slope (linearity)
What happens when CPUE=0
The population is 0
