2020
DOI: 10.1101/2020.05.27.118901
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

bbsBayes: An R Package for Hierarchical Bayesian Analysis of North American Breeding Bird Survey Data

Abstract: The North American Breeding Bird Survey (BBS) is the primary ecological monitoring program used to assess the population, status, and trend of North American birds. As such, accessible analysis of BBS data is crucial to wildlife conservation/management and ecological science in North America. The R package bbsBayes was developed as a wrapper for the analysis of BBS data using hierarchical Bayesian models, including the models currently used by the Canadian Wildlife Service and the United States Geological Surv… Show more

Help me understand this report
View published versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
12
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
4
1

Relationship

2
3

Authors

Journals

citations
Cited by 6 publications
(12 citation statements)
references
References 16 publications
0
12
0
Order By: Relevance
“…Population trends and trajectories were estimated using a Bayesian hierarchical Generalized Additive Model with Year Effects (GAMYE) 32 in R 51 . The survey-wide analyses were run during the annual analysis of the BBS data conducted by the Canadian Wildlife Service 52 , and additional summaries and maps were created with the R-package bbsBayes 53 . Data were stratified both by geopolitical regions (states and provinces), as well as internationally ratified Bird Conservation Regions (BCRs) 54 .…”
Section: Methodsmentioning
confidence: 99%
“…Population trends and trajectories were estimated using a Bayesian hierarchical Generalized Additive Model with Year Effects (GAMYE) 32 in R 51 . The survey-wide analyses were run during the annual analysis of the BBS data conducted by the Canadian Wildlife Service 52 , and additional summaries and maps were created with the R-package bbsBayes 53 . Data were stratified both by geopolitical regions (states and provinces), as well as internationally ratified Bird Conservation Regions (BCRs) 54 .…”
Section: Methodsmentioning
confidence: 99%
“…We used a general additive model with a random year effect, the current model used by the Canadian Wildlife Service (Edwards and Smith 2021). Otherwise, we used the default model parameters (e.g., burn‐in = 10,000, chains = 3, iterations = 20,000, thinning = 10) with a heavy‐tailed t‐distribution for the extra‐Poisson error distribution, as suggested by Edwards and Smith (2021). Only black duck conservation regions where enough survey routes were available could be modeled (i.e., ≥3 routes with black duck observations; ≥3 years with non‐zero observations on ≥1 route).…”
Section: Methodsmentioning
confidence: 99%
“…We determined regional indices of black duck abundance using the North American BBS data (1966–2019; Pardieck et al 2020) analyzed using hierarchical Bayesian analysis within the bbsBayes package (version 2.3.8.2020; Edwards and Smith 2021). We used a general additive model with a random year effect, the current model used by the Canadian Wildlife Service (Edwards and Smith 2021).…”
Section: Methodsmentioning
confidence: 99%
“…We aggregated route‐level count data into population indices within 2° latitude x 2° longitude grid cells (mean area 36,267 km 2 · cell −1 , range 29,192−43,596 km 2 ) with a Bayesian hierarchical model (heavy‐tailed GAMYE) (Edwards & Smith 2020). The model, which accounts for observer effects among other sources of bias, has the following structure: logfalse(λs,j,tfalse)=θs+ωj+ηIj,t+εs,j,t+normalΔstwhere (λ s,j,t ) is the mean count per grid cell ( s ), observer‐route combination ( j ), and year ( t ); θ s is a grid‐cell‐specific intercept; ω j is an observer‐route effect; ηI [ j,t ] is a first‐time observer effect; ε s,j,t is a random effect to account for overdispersion; and Δ s ( t ) is a generalized additive model based on year (for details see Smith & Edwards [2020]).…”
Section: Methodsmentioning
confidence: 99%