Search the Community
Showing results for tags 'detection'.
Found 1 result
Assessing the Probability of Species Occurrence and Detection with Occupancy Models Tuesday, Feb. 24, 2015; 8AM-5PM Cost: $60 USD Maximum Enrollment: 25 Presented by the Society for Northwestern Vertebrate Biology http://thesnvb.org/annual-meeting-2015/ Instructors: Mike Adams and Tara Chestnut, USGS Amphibian Research and Monitoring Initiative (ARMI Program) In ecological research, it is often of interest to ask questions about the probability of species occurrence when detection is imperfect. For example, how does the probability that a pond is occupied by red-legged frogs relate to the occurrence of non-native fish?; or the occurrence of a pathogen relate to water chemistry? Such analyses rely on survey data that result from efforts to detect the species of interest and measure site characteristics. These analyses are hampered by false negatives: the failure to detect a species at a site where it is present. Occupancy models can be used to estimate the probability that a site is occupied (and other parameters related to changes in occupancy) despite false negatives. They use repeated observations to estimate and account for the probability that a single observation will detect a species that is present. Since their initial development in 2002, there have been several important expansions of the occupancy modeling framework including multi-season models that estimate local extinction and colonization rates, two-species models that deal with false negatives in two species simultaneously, multi-state models that allow response categories beyond just presence and absence, and integrated habitat models that deal with sites that may or may not exist in a given year (e.g., ephemeral ponds). We have also come to better understand some of the pitfalls of occupancy models. The main goals of this workshop are to: 1) Offer a primer to the basic occupancy model and introduction to the free software program PRESENCE; 2) Overview of the range of occupancy models currently available; 3) Offer hands on experience to help get you started with occupancy studies and one-on-one consultations to address questions regarding your specific project. The target audience: Participants are beginners to occupancy analyses but have a basic understanding of regression analysis. To register: http://thesnvb.org/event/2015-meeting-registration/ Please bring your own PC laptop with the latest version of PRESENCE installed (http://www.mbr-pwrc.usgs.gov/software). Otherwise you should be able to pair with someone with a laptop. Contact Michael_Adams@usgs.gov with any questions.