 Multi-state Occupancy. Multiple Occupancy States Rather than just presence/absence of the species at a sampling unit, ‘occupancy’ could be categorized.

Slides:



Advertisements
Similar presentations
Trustworthy Service Selection and Composition CHUNG-WEI HANG MUNINDAR P. Singh A. Moini.
Advertisements

Analysis of variance and statistical inference.
Supplementation with local, natural-origin broodstock may minimize negative fitness impacts in the wild Initial results of this study were published in.
Structural Equation Modeling Using Mplus Chongming Yang Research Support Center FHSS College.
Change Detection C. Stauffer and W.E.L. Grimson, “Learning patterns of activity using real time tracking,” IEEE Trans. On PAMI, 22(8): , Aug 2000.
Errors in Chemical Analyses: Assessing the Quality of Results
Krishna Pacifici Department of Applied Ecology NCSU January 10, 2014.
Detectability Lab. Outline I.Brief Discussion of Modeling, Sampling, and Inference II.Review and Discussion of Detection Probability and Point Count Methods.
The current status of fisheries stock assessment Mark Maunder Inter-American Tropical Tuna Commission (IATTC) Center for the Advancement of Population.
Paper Discussion: “Simultaneous Localization and Environmental Mapping with a Sensor Network”, Marinakis et. al. ICRA 2011.
Maximum likelihood estimates What are they and why do we care? Relationship to AIC and other model selection criteria.
Species interaction models. Goal Determine whether a site is occupied by two different species and if they affect each others' detection and occupancy.
Common Factor Analysis “World View” of PC vs. CF Choosing between PC and CF PAF -- most common kind of CF Communality & Communality Estimation Common Factor.
458 Fitting models to data – II (The Basics of Maximum Likelihood Estimation) Fish 458, Lecture 9.
Item Response Theory. Shortcomings of Classical True Score Model Sample dependence Limitation to the specific test situation. Dependence on the parallel.
Arizona State University DMML Kernel Methods – Gaussian Processes Presented by Shankar Bhargav.
458 Fitting models to data – III (More on Maximum Likelihood Estimation) Fish 458, Lecture 10.
ECE 8443 – Pattern Recognition LECTURE 06: MAXIMUM LIKELIHOOD AND BAYESIAN ESTIMATION Objectives: Bias in ML Estimates Bayesian Estimation Example Resources:
Detecting trends in dragonfly data - Difficulties and opportunities - Arco van Strien Statistics Netherlands (CBS) Introduction.
Resource Selection Functions and Patch Occupancy Models: Similarities and Differences Lyman L. McDonald Senior Biometrician Western EcoSystems Technology,
Learning Theory Reza Shadmehr logistic regression, iterative re-weighted least squares.
VI. Evaluate Model Fit Basic questions that modelers must address are: How well does the model fit the data? Do changes to a model, such as reparameterization,
CS 782 – Machine Learning Lecture 4 Linear Models for Classification  Probabilistic generative models  Probabilistic discriminative models.
Patch Occupancy: The Problem
Issues concerning the interpretation of statistical significance tests.
Assessing Estimability of Latent Class Models Using a Bayesian Estimation Approach Elizabeth S. Garrett Scott L. Zeger Johns Hopkins University Departments.
Global Analyzing community data with joint species distribution models abundance, traits, phylogeny, co-occurrence and spatio-temporal structures Otso.
Estimating age-specific survival rates from historical ring-recovery data Diana J. Cole and Stephen N. Freeman Mallard Dawn Balmer (BTO) Sandwich Tern.
Workshop on Applied Hierarchical Modeling in BUGS and unmarked Patuxent Wildlife Research Center November 2015.
Introduction to Occupancy Models Key to in-class exercise are in blue
A Synthesis of Annual Estimates of TIR and D for Wild Populations Presenter: Paul Wilson CSS Annual Meeting Apr 2 nd 2010.
Matrix Models for Population Management & Conservation March 2014 Lecture 10 Uncertainty, Process Variance, and Retrospective Perturbation Analysis.
Multiple Season Model Part I. 2 Outline  Data structure  Implicit dynamics  Explicit dynamics  Ecological and conservation applications.
Estimation of Animal Abundance and Density Miscellaneous Observation- Based Estimation Methods 5.2.
Multistate models UF Outline  Description of the model  Data structure and types of analyses  Multistate with 2 and 3 states  Assumptions 
Statistical Methods. 2 Concepts and Notations Sample unit – the basic landscape unit at which we wish to establish the presence/absence of the species.
Estimation of Vital Rates: Use of Index Statistics? Relevance of Detection Probability.
Capture-recapture Models for Open Populations “Single-age Models” 6.13 UF-2015.
Pollock’s Robust Design: Model Extensions. Estimation of Temporary Emigration Temporary Emigration: = individual emigrated from study area, but only temporarily.
 1 Species Richness 5.19 UF Community-level Studies Many community-level studies collect occupancy-type data (species lists). Imperfect detection.
Identify techniques for estimating various populations (quadrats, transects, mark- recapture) Understand the carrying capacity of ecosystems; factors.
Pollock’s Robust Design: Extensions II. Quick overview 1.Separation of Recruitment Components in a single patch context (Source-Sink) 2.Separation of.
Multiple Detection Methods: Single-season Models.
Estimation of State Variables and Rate Parameters Estimation of State Variables and Rate Parameters Overview 5.1 UF UF-2015.
Hypothesis Testing. Statistical Inference – dealing with parameter and model uncertainty  Confidence Intervals (credible intervals)  Hypothesis Tests.
[Part 5] 1/43 Discrete Choice Modeling Ordered Choice Models Discrete Choice Modeling William Greene Stern School of Business New York University 0Introduction.
 Integrated Modelling of Habitat and Species Occurrence Dynamics.
Monitoring and Estimating Species Richness Paul F. Doherty, Jr. Fishery and Wildlife Biology Department Colorado State University Fort Collins, CO.
Additional multistate model applications. Unobservable States single observable state, single unobservable state.
Single Season Model Part I. 2 Basic Field Situation From a population of S sampling units, s are selected and surveyed for the species. Units are closed.
Single Season Occupancy Modeling 5.13 UF Occupancy Modeling State variable is proportion of patches that is occupied by a species of interest.
1 Occupancy models extension: Species Co-occurrence.
 1 Modelling Occurrence of Multiple Species. 2 Motivation Often there may be a desire to model multiple species simultaneously.  Sparse data.  Compare/contrast.
Capture-recapture Models for Open Populations Multiple Ages.
Multiple Season Study Design. 2 Recap All of the issues discussed with respect to single season designs are still pertinent.  why, what and how  how.
Single Season Study Design. 2 Points for consideration Don’t forget; why, what and how. A well designed study will:  highlight gaps in current knowledge.
 Occupancy Model Extensions. Number of Patches or Sample Units Unknown, Single Season So far have assumed the number of sampling units in the population.
Multi-state Occupancy. Multiple Occupancy States Rather than just presence/absence of the species at a sampling unit, ‘occupancy’ could be categorized.
Methods of multivariate analysis Ing. Jozef Palkovič, PhD.
Hierarchical Models. Conceptual: What are we talking about? – What makes a statistical model hierarchical? – How does that fit into population analysis?
Occupancy Models when misclassification occurs. Detection Errors and Occupancy Estimation  Occupancy estimation accounts for issues of detection when.
Chloe Boynton & Kristen Walters February 22, 2017
An introduction to Dynamic Factor Analysis
Combining Species Occupancy Models and Boosted Regression Trees
Extension to the Hybrid Symbolic-Numeric Method for Investigating Identifiability Diana Cole, University of Kent, UK Rémi Choquet, CEFE, CNRS, France.
Parameter Redundancy and Identifiability in Ecological Models
Mathematical Foundations of BME Reza Shadmehr
Multistate models Lecture 10.
Estimating mean abundance from repeated presence-absence surveys
Probabilistic Surrogate Models
Presentation transcript:

 Multi-state Occupancy

Multiple Occupancy States Rather than just presence/absence of the species at a sampling unit, ‘occupancy’ could be categorized into multiple states. breeding/non-breeding/absent disease/no disease/absent index of relative abundance – lots/some/none Key requirements are: highest order state must be observable without error. ambiguity associated with observed lower order states. true state does not change within a season (closure).

Multiple Occupancy States Single Season Study 2 occupancy states 0 = not occupied 1 = non-breeders only 2 = at least some breeders Breeding state cannot be identified with certainty Observed state = 2: detection of breeding, evidence is unambiguous (true state = 2) Observed state = 1: no detection of breeding, evidence is ambiguous (true state = 1 or 2) Observed state = 0: no detection (true state = 0, 1, or 2)

Multiple Occupancy States: Parameters Pr(unit i is in state m ) Pr(observe state l in survey j of site i | true state = m )

Observed State True State Multiple Occupancy States: Parameters

Multiple Occupancy States: Reparameterizations Pr(unit is occupied) Pr(unit is in state 2|unit is occupied) Pr(species is detected|true state = 2) Pr(correctly classify as being in state 2)

Multiple Occupancy States: Reparameterizations 1-  [1] -  [2] not occ  [1] state=1 Occ w/o breeding state=2 occ w/ breeding 1-  not occ  Occupied breeding  [2] 1-R no breeding R

Multiple Occupancy States: Detection History Modeling =  [2] p [1,2] (1-p [1,2] -p [2,2] )p [2,2]

Multiple Occupancy States: Detection History Modeling

A Probabilistic Model Define state-dependent occupancy and detection vectors.

A Probabilistic Model The combination of these statements forms the model likelihood: S L( ,R,p,  | h 1,h 2,…h s )=  Pr(h i ) i =1 Maximum likelihood estimates of parameters can be obtained. However, parameters cannot be site-specific without additional information (covariates).

Multiple Occupancy States: Cal. Spotted Owl Reproduction Eldorado National Forest, 2004 Rocky Gutierrez, Mark Seamans 2 states for each occupied territory: Successfully reproduced Did not successfully reproduce Multiple (5 maximum) visits with “mousing” Definitive evidence of reproduction: Detect young (e.g., moused adult feeds young) Much more likely later in season (last 3 visits) Variation in sampling protocol among 5 Sierra study sites precludes meta-analysis for reproductive rate data

Multiple Occupancy States: Cal. Spotted Owl Reproduction Naïve estimate: Parameter estimates from model

Multiple Occupancy States: Multiple Seasons What about the dynamic processes of change between states over time? Easily accounted for by defining a transition probability matrix.

Multiple Occupancy States: Multiple Seasons

Calculation of model likelihood for observed histories is same as for the multi-season models described previously. Both parameterizations available in PRESENCE.

Example: Cal. Spotted Owl What are the occupancy and reproduction dynamic rates from ? Does the state of a territory last year influence the dynamic rates for either process? Similar parameterization to Nichols et al. (2007) 66 potential nesting territories

Example: Cal. Spotted Owl 24 models fit to the data Top 2 models accounted for 99% AIC model weights 91% 8%

Example: Cal. Spotted Owl (0.04) 0.56 (0.06) 0.89 (0.08) 0.55 (0.06) 0.66 (0.04) 0.55 (0.08) 0.48 (0.09) 0.62 (0.04) 0.91 (0.03) 0.79 (0.04) 0.30 (0.07) 0.87 (0.03) 0.97 (0.03) 0.79 (0.03) 0.74 (0.05) 0.80 (0.04) 0.00 (0.00) 0.93 (0.04) 0.79 (0.06) 0.17 (0.15) 0.83 (0.05) 0.85 (0.08) 0.84 (0.05) 0.26 (0.12) 0.85 (0.05)

Example: Cal. Spotted Owl State (m) in year t (0.11 – 0.26) (0.79 – 0.92) (0.83 – 0.95)

Example: Cal. Spotted Owl

Clear indication that probability of a territory being occupied depends on state in previous year. Reproduction also depends on state in previous year, but nature of relationship varies annually.

Green Frogs, Maryland USA Up to 280 NAAMP listening stations surveyed 3 times from Level of activity scored each survey ‘State’ is maximal call index Number of stations in each state each year

Green Frogs, Maryland USA Multinomial parameterisation used Analysed in WinBUGS flat priors 2 chains of 50,000 iterations

Multiple Occupancy States: Summary Multiple states with uncertain state assignment represent a natural extension of occupancy models. Provides a framework to address interesting ecological questions from landscape-level data. Single- and multiple-season models have been developed to deal with these situations.