DOC PREVIEW
UI WLF 448 - Sightability Models

This preview shows page 1-2-3-4 out of 13 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 13 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 13 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 13 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 13 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 13 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

1 Estimating Abundance: Sightability Models Visibility Bias  Virtually all counts from the air or ground are undercounts because can’t see all the animals due to vegetation cover or topographic irregularity  Solutions utilize mark-resight methods, distance estimation (line transects), a correction factor or a sightability model2 Elk in Brushfield How many? Elk in Light Timber3 Medium Timber Heavy Timber4 Sightability Model • Attempts to remove visibility bias by estimating a correction factor for each group of animals seen. • Adaptable to a variety of conditions. • Cost efficient, especially once model built • Only works if model is applicable and if visibility averages at least 33%. Developing a Sightability (or Visibility Bias) Model • Mark elk (deer, sheep, etc.) groups with radio-collars or have observers on ground keep track of individual groups when helicopter/plane passes over. • Fly aerial survey over the geographic area where the marked groups occur. • Determine which individual groups were seen and which groups were missed.5 Developing Sightability Models • Identify which factors such as group size, tree and shrub cover, snow cover, weather, observers, type of helicopter, etc. influenced whether a group was seen or missed. • Important: factors must be ones that will have the same effect each time a survey is conducted Developing Sightability Models  Keep some factors constant such as type of helicopter or fixed-wing, experience of observers, speed of flight, height above ground, etc.  Estimate the effects of the other important factors we can’t control such as group size, vegetation cover, etc. using logistic regression.6 So how many seen of known total for each variable of interest? Samuels et al (1987)7 Sightability Model:Analysis  Logistic regression is one of a number of statistical models that can be used to analyze the observations of groups seen and groups missed.  Where pi is the probability of seeing a group  e.g. X1 = group size, X2 = veg. cover 0 1 1 2 2logit( ) log ...1iiipp X Xp      Samuels et al (1987)8 Probability of Seeing Elk How does sightability of elk change with group size and veg cover? Probability of Seeing Elk How does sightability of elk change with group size and veg cover?9 Factors Affecting Elk Sightability  Size of group  Percent vegetation cover around group  Percent snow cover  Secondary factors also statistically signif.: – Activity (moving vs. still) – Observer experience – Composition (Bull groups vs. others) – Type of helicopter or fixed-wing Sightability Model  Use the logistic regression model to calculate the probability that each group is seen.10 Simple Application  Suppose we see a group of 3 elk in an open forest with 40% cover of obscuring vegetation.  If our logistic regression model estimates that only ½ of groups of 3 in 40% cover are seen (p=0.5), then if we saw this one group of 3 animals, there was probably another group of 3 that we missed. Simple Application  So if we saw 3 there were actually 6 in the area.  How? Probability of detection = 0.5  True N = Nobs /Prob. of det. = 3 / 0.5 = 611 Simple Application  If the next group we saw was a group of 2 animals in 80% cover and the model said that we only have a 20% chance of detecting such a group (p=0.2)  We would correct this group of 2 to represent 2/0.2 or 10 animals in the population. Lochsa River Elk Herd  This sightability model was applied to the elk herd wintering on the Lochsa River in 1985.  Half of the winter range was flown obtaining a raw count of 2718 elk.  When the sightability model corrections were applied to the counts the corrected estimate was 4775 with 90% bound of 458.12 Lochsa Elk Herd 0 1000 2000 3000 4000 5000 6000 Comp Cnt Sightability Can be applied to similar areas/ conditions, or new sightability models created  More recent applications use model selection approach to choose best model: Gilbert & Moeller (2008) – elk in central Cascades (WA)13 Results w parameters for 2 best models Describe how sightability changes in each model Raw and adjusted counts by various classes. Does using sightability model to adjust make a


View Full Document

UI WLF 448 - Sightability Models

Download Sightability Models
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Sightability Models and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Sightability Models 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?