Term
| What's another term for an open model? |
|
Definition
|
|
Term
| What does a Jolly-Seber model allow us to do? |
|
Definition
| Estimate pop size when there are births/deaths and/or immigration/emigration |
|
|
Term
| Who thought of Jolly-Seber first? |
|
Definition
| The one with the stupid glasses! No not really. Published simultaneously in 1965. |
|
|
Term
| In the Jolly-Seber model, what does COVPHI(i) mean? |
|
Definition
| Covariance between survival estimates for times i-1, i |
|
|
Term
| What are 3 assumtions of the Jolly-Seber model? |
|
Definition
| 1) Every animal present in the pop at the time of the ith sample has the same prob of capture (pi) |
|
|
Term
| What's the general approach to running a study for Jolly-Seber analysis? |
|
Definition
Capture/recapture scenario
Each individual is marked and recorded with unique code/name/whatever
Capture history is represented by 1, 0 (but sometimes 2, 3, 4, he didn't say what they mean)
0 = no capture, 1 = capture |
|
|
Term
| What does program Jolly do? |
|
Definition
1) Computes survival rate, pop size, and immigration rate estimates
2) Provides a Goodness-of-Fit test to determine if the model is appropriate
3) Provides other models which allow relaxations of the genreal Jolly-Seber model |
|
|
Term
| What are the other models available for Jolly program? |
|
Definition
|
|
Term
| What's model A good for in Jolly? |
|
Definition
| allows for death, but no immigration |
|
|
Term
| What's model B good for in Jolly? |
|
Definition
| constant survival rate with time-specific capture probability |
|
|
Term
| What's model 2 good for in Jolly? |
|
Definition
| assumes survival rates are different for animals which are captured for the first time than unmarked or previously marked animals |
|
|
Term
| What's model D good for in Jolly? |
|
Definition
| Constant survival rate and constant capture probability |
|
|
Term
| Program Jolly allows the user to test different models for the best fit using what test? |
|
Definition
|
|
Term
| True or False, Jolly-Seber is least useful with long-term studies |
|
Definition
|
|
Term
| How do you avoid problems with closure in longer term studies? |
|
Definition
1) Shorten duration of trapping
2) pick times of decreased movement/births/deaths |
|
|
Term
| 2 reasons Jolly-Seber pop estimates are not as reliable as closed-model estimates |
|
Definition
1) more information is gleaned from data instead of assumed, this means larger variances
2) models are based on Mt and do not allow individual heterogeneity or behavioral responses to trapping. Don't ask me what Mt is. |
|
|
Term
| 3 problems with Jolly-Seber |
|
Definition
1) sensitive to capture heterogeneity
2) capture hetero.. causes large underestimation of population size. This is not a problem if capture probability >.5
3) also sensitive to changes in animal behavior |
|
|
Term
| What is a finite sampling approach? |
|
Definition
| Given a fixed area (trapping grids), how many are in it? |
|
|
Term
| What is a distance sampling approach? |
|
Definition
| Given the detection of N objects, how many objects are estimated to be within the area? |
|
|
Term
| What are the 2 types of point count surveys? |
|
Definition
1) Extensive
2) Intensive |
|
|
Term
| What is an extensive point count survey? |
|
Definition
| Intended for a series of points, minimum 250 m apart, usually on roads/trails over and entire region/large area |
|
|
Term
| What is an intensive point count survey? |
|
Definition
| used on a maller scale than extensive, use with mist nets or search plots |
|
|
Term
| What do point counts allow us to study? |
|
Definition
| Yearly changes of bird pops at fixed pts, differences in spp compositions among habitats, abundance of spp, diversity of sites |
|
|
Term
| How is a point count usually completed? |
|
Definition
1) and obs stands at a point and records all spp seen.heard at a fixed or sometimes unlimited distance
2) A particular area is surveyed by placing multiple points around and recording all spp deteted at each point
3) The survey is generally repeated after some time interval that is relevant to the study (1x/mo or 1x/yr) |
|
|
Term
| When do you use point count surveys? |
|
Definition
-Preferred method for forests and difficult terrain
-Some birds need to "acclimate to your movement and presence before they can be detected"
-DON'T use for: water fowl and landbirds that are very quiet/loud, nocturnal, or flocking |
|
|
Term
| Skills needed for point count surveys? |
|
Definition
-Sight/sound ID all spp in area
-Judge distance well
May take 2-8 weeks to learn 80-90% of spp present; may want a pre-survey period to learn all birds |
|
|
Term
| Required equipment for point count surveys? |
|
Definition
-Timing device and binocs/scope
-Mark points with map/GPS/flagging
-Time to conduct one point count should be <4 hrs - vehicle may be needed |
|
|
Term
| Wha are 3 points to consider when designing an extensive point count? |
|
Definition
1) route should encompass all habitats of a region or site and any mist netting or nest searching plots
2) Use a systematic design to place points, not random ("points should be placed evenly throughout an area w/o regard to habitat configurations unless separate estimations for a habitat are required")
3) Points must be at least 250 m apart in forested habitats (ea. point must be statistically independent, may have to increase distance between points in open habitats) |
|
|
Term
| What are 4 points to consider when designing an extensive point count? |
|
Definition
1) Conduct w/in a study plot for mist netting/nest searches
2) 9-16 points in a grid (3x3, 3x4, 4x4)
3) Birds counted from these points are combined into a single mean so distance between points is less critical
4) Don't include areas from outside the plot |
|
|
Term
| Explain data collection for intensive point counts. |
|
Definition
1) approach plots w/ minimal disturbance
2) begin count asap when you arrive
3) spend 5 mins at point before you begin if travel time is < 15 mins
4) 10 mins at point before you begin if travel time is > 15 mins |
|
|
Term
| What do you record at each point in intensive point count surveys? |
|
Definition
1) Point # (yes, he did give these instructions in the lecture)
2) Date/time
3) spp in the order of detection
4) distance to each bird (where first observed - birds fleeing recorded at point of takeoff) |
|
|
Term
| What are some rules to follow when repeating point counts? |
|
Definition
1) Timing of survey should be constant year to year (should not differ by more than one week from the date of first count)
2) Start of count should not differ by more than 30 mins
3) Same observer from year to year (if possible) |
|
|
Term
| How to you move from data to estimates of abundace in point counts? |
|
Definition
1) Total Bird Abundance = mean # birds detected per survey
2) Ind Spp adundance = mean # individuals of a spp detected per survey
3) Spp richness is mean # spp detected per survey |
|
|
Term
| How do you compare between sites in point count surveys? |
|
Definition
| "Standard statistical techniques." No, really. The man has rocks for brains. |
|
|
Term
| True or false, point counts are readily compared to other studies. |
|
Definition
|
|
Term
| What is one major assumption that is a problem in point counts? |
|
Definition
| Use of abunance data in pt counts assume the data are positively related to density in a consistent manner (bird detectability remains constant) |
|
|
Term
| What are 3 TYPES of factors that affect bird detectability? |
|
Definition
1) Observer ability to detect and ID birds correctly
2) Environmental factors that affect bird behavior and frequency
3) Physical and behavioral attributes of birds |
|
|
Term
| What are some factors linked to observer in bird surveys? |
|
Definition
1) Observer performance (vision, age, experience, fatigue, motivation, hearing...)
2) Observer behavior and clothing color |
|
|
Term
| What are some environmental factors that affect bird behavior and observer efficiency? |
|
Definition
| Wind, precip, temp, cloud cover, light intensity, topography, vegetation characteristics |
|
|
Term
| Physical attributes of birds that make them more/less conspicuous: |
|
Definition
| body size, coloration, vocalizations (loudness, rate, sonic range), flight behavior, physiological traits, flock size, density, age, sex |
|
|
Term
| True or false- bird detectability can change from survey to survey |
|
Definition
|
|
Term
| Standard error, coefficient variable, and confidence interval are all a meadure of what? |
|
Definition
|
|
Term
| What is a basic field procedure for point transect? |
|
Definition
Similar to point count, but you measure the radial distance of the bird to the point count.
Transects have 0 length
Must meet the same assumptions as in line transects
All distance est originate from the point |
|
|
Term
| How to you ensure correct distance measurements in point transects? |
|
Definition
Range finder
Training
With calls, identify an interval (10-20 m) rather than exact distance |
|
|
Term
| How do you set up distance intervals for point transects? |
|
Definition
The length of interval should increase with disance from central point
Use 4-8 intervals (too few=low precision, too many=error prone and time consuming)
Final interval should have no upper bound (i.e., >200) |
|
|
Term
| Why do point transects require more samples than line transects? |
|
Definition
Required to get the same level of precision
Repeated sampling can be used to increase N
Independence not violated as long as Ind are not detected more than one time from the same line or point in a given survey |
|
|
Term
| How much do you truncate data in point transects? Line transects? |
|
Definition
| Truncate 10% of the right tail in point transects and 5% in line transects. |
|
|
Term
| True or false. Objects around the point should be located at an instant in time. |
|
Definition
|
|
Term
| In practice, how long does a point count last? |
|
Definition
| Minutes (should be instant). You can manage this by only counting individuals with a known position at the start of survey; define a "snapshot" moment; mobile objects can lead to overestimation of density; this process leads to lower sample size, but lower bias as well. |
|
|
