Spatially explicit Bayesian hierarchical models improve estimates of avian population status and trends

被引:1
|
作者
Smith, Adam C. [1 ]
Binley, Allison D. [2 ]
Daly, Lindsay [1 ]
Edwards, Brandon P. M. [1 ,2 ]
Ethier, Danielle [2 ,3 ]
Frei, Barbara [4 ]
Iles, David [1 ]
Meehan, Timothy D. [5 ]
Michel, Nicole L. [5 ]
Smith, Paul A. [6 ]
机构
[1] Environm Climate Change Canada, Canadian Wildlife Serv, Ottawa, ON, Canada
[2] Carleton Univ, Dept Biol, Ottawa, ON, Canada
[3] Birds Canada, Port Rowan, ON, Canada
[4] Environm Climate Change Canada, Wildlife Res Div, Montreal, PQ, Canada
[5] Natl Audubon Soc, New York, NY USA
[6] Environm Climate Change Canada, Wildlife Res Div, Ottawa, ON, Canada
关键词
abundance; Bayesian; biological monitoring; GAM; hierarchical; iCAR; spatially explicit; abondance; bayesien; suivi biologique; hierarchique; spatialement explicite; BREEDING BIRD SURVEY; AUTOREGRESSIVE MODELS; SHOREBIRDS; DECLINES; TEMPERATURE; STRATEGIES; RESPONSES; DIPPER;
D O I
10.1093/ornithapp/duad056
中图分类号
Q95 [动物学];
学科分类号
071002 ;
摘要
Population trend estimates form the core of avian conservation assessments in North America and indicate important changes in the state of the natural world. The models used to estimate these trends would be more efficient and informative for conservation if they explicitly considered the spatial locations of the monitoring data. We created spatially explicit versions of some standard status and trend models applied to long-term monitoring data for birds across North America. We compared the spatial models to simpler non-spatial versions of the same models, fitting them to simulated data and real data from 3 broad-scale monitoring programs: the North American Breeding Bird Survey (BBS), the Christmas Bird Count, and a collection of programs we refer to as Migrating Shorebird Surveys. All the models generally reproduced the simulated trends and population trajectories when there were many data, and the spatial models performed better when there were fewer data and in locations where the local trends differed from the range-wide means. When fit to real data, the spatial models revealed interesting spatial patterns in trend, such as recent population increases along the Appalachian Mountains for the Eastern Whip-poor-will (Antrostomus vociferus), that were much less apparent in results from the non-spatial versions. The spatial models also had higher out-of-sample predictive accuracy than the non-spatial models for a selection of species using BBS data. The spatially explicit sharing of information allows fitting the models with much smaller strata, allowing for finer-grained patterns in trends. Spatially informed trends will facilitate more locally relevant conservation, highlight areas of conservation successes and challenges, and help generate and test hypotheses about the spatially dependent drivers of population change. center dot We created population trend models that share information on species abundance and trend among neighboring regions and compared those to models without spatial information.center dot Trends and population trajectories from the spatial models were more accurate, precise, and better predicted future observations than the non-spatial models.center dot Spatial information allows for finer-scale stratifications, producing more detailed spatial patterns and highlighting areas for local conservation.center dot The predictions from these models (trends, trajectories, maps of patterns in trends, etc.) will improve our understanding of the status of bird populations in time and space, and guide future research into the drivers of ongoing avian population declines. Les estimations des tendances des populations constituent le c oe ur des evaluations pour la conservation aviaire en Amerique du Nord et indiquent des changements importants de l'etat de l'environnement naturel. Les modeles utilises pour estimer ces tendances seraient plus efficaces et plus instructifs pour la conservation s'ils prenaient explicitement en compte les emplacements spatiaux des donnees de suivi. Nous avons cree des versions spatialement explicites de certains modeles standards de statut et de tendance appliques aux donnees de suivi a long terme des oiseaux en Amerique du Nord. Nous avons compare les modeles spatiaux a des versions non spatiales plus simples des memes modeles, en les adaptant a des donnees simulees et a des donnees reelles provenant de trois programmes de suivi a grande echelle : le Releve des oiseaux nicheurs de l'Amerique du Nord (BBS), le Recensement des oiseaux de Noel (CBC) et un ensemble de programmes que nous appelons Releves d'oiseaux de rivage migrateurs (MSS). Tous les modeles ont generalement reproduit les tendances simulees et les trajectoires demographiques lorsqu'il y avait beaucoup de donnees, et les modeles spatiaux ont donne de meilleurs resultats lorsqu'il y avait moins de donnees et pour des endroits ou les tendances locales differaient des moyennes a l'echelle de l'aire de repartition. Lorsqu'ils ont ete ajustes aux donnees reelles, les modeles spatiaux ont revele des patrons spatiaux interessants dans les tendances, telles que les recentes augmentations de population le long des Appalaches pour Antrostomus vociferus, qui etaient beaucoup moins apparents dans les resultats des versions non spatiales. Les modeles spatiaux avaient aussi une plus grande precision predictive hors echantillon que les modeles non spatiaux pour une selection d'especes utilisant les donnees du BBS. Le partage spatialement explicite des informations permet d'ajuster les modeles avec des strates beaucoup plus petites, ce qui permet d'obtenir des tendances plus fines. Les tendances tenant compte des donnees spatiales faciliteront une conservation plus pertinente au niveau local, mettront en evidence les succes et les defis dans les aires de conservation, et aideront a generer et a tester des hypotheses sur les facteurs de changement demographique spatialement dependants.
引用
收藏
页数:16
相关论文
共 50 条
  • [1] Spatially explicit Bayesian clustering models in population genetics
    Francois, Olivier
    Durand, Eric
    [J]. MOLECULAR ECOLOGY RESOURCES, 2010, 10 (05) : 773 - 784
  • [2] Tropical deforestation in Madagascar: analysis using hierarchical spatially explicit, Bayesian regression models
    Agarwal, DK
    Silander, JA
    Gelfand, AE
    Dewar, RE
    Mickelson, JG
    [J]. ECOLOGICAL MODELLING, 2005, 185 (01) : 105 - 131
  • [3] SPATIALLY EXPLICIT POPULATION-MODELS
    PULLIAM, HR
    DUNNING, JB
    [J]. ECOLOGICAL APPLICATIONS, 1995, 5 (01) : 2 - 2
  • [4] Uncertainty in spatially explicit population models
    Minor, E. S.
    McDonald, R. I.
    Treml, E. A.
    Urban, D. L.
    [J]. BIOLOGICAL CONSERVATION, 2008, 141 (04) : 956 - 970
  • [5] Spatially explicit integrated population models
    Chandler, Richard B.
    Clark, Joseph D.
    [J]. METHODS IN ECOLOGY AND EVOLUTION, 2014, 5 (12): : 1351 - 1360
  • [6] Dispersal in spatially explicit population models
    South, A
    [J]. CONSERVATION BIOLOGY, 1999, 13 (05) : 1039 - 1046
  • [7] A spatially explicit hierarchical model to characterize population viability
    Campbell, Steven P.
    Zylstra, Erin R.
    Darst, Catherine R.
    Averill-Murray, Roy C.
    Steidl, Robert J.
    [J]. ECOLOGICAL APPLICATIONS, 2018, 28 (08) : 2055 - 2065
  • [8] Spatially Explicit Trends in the Global Conservation Status of Vertebrates
    Rodrigues, Ana S. L.
    Brooks, Thomas M.
    Butchart, Stuart H. M.
    Chanson, Janice
    Cox, Neil
    Hoffmann, Michael
    Stuart, Simon N.
    [J]. PLOS ONE, 2014, 9 (11):
  • [9] Dealing with Uncertainty in Spatially Explicit Population Models
    Thorsten Wiegand
    Eloy Revilla
    Felix Knauer
    [J]. Biodiversity & Conservation, 2004, 13 : 53 - 78
  • [10] Dealing with uncertainty in spatially explicit population models
    Wiegand, T
    Revilla, E
    Knauer, F
    [J]. BIODIVERSITY AND CONSERVATION, 2004, 13 (01) : 53 - 78