Dissemin is shutting down on January 1st, 2025

Published in

Wiley, Journal of Applied Ecology, 4(59), p. 909-920, 2022

DOI: 10.1111/1365-2664.14114

Links

Tools

Export citation

Search in Google Scholar

Integrated modelling of seabird‐habitat associations from multi‐platform data: A review

This paper was not found in any repository, but could be made available legally by the author.
This paper was not found in any repository, but could be made available legally by the author.

Full text: Unavailable

Green circle
Preprint: archiving allowed
Orange circle
Postprint: archiving restricted
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

AbstractQuantifying current and future overlap between human activities and wildlife is a core and growing aim of ecological study, spurring ever more spatial data collection and diversification of observation techniques (surveys, telemetry, citizen science, etc.).To meet this aim, data collected via multiple platforms, across different geographical and temporal regions, may need to be integrated, yet many ecologists remain unclear about the relationships between data types and therefore how they can be combined.In seabird research, these applied questions can be particularly pressing because many human activities (e.g. tidal and wind renewables, fishing, shipping, etc.) are concentrated in coastal waters, where many seabirds also aggregate, especially while breeding. In addition, seabird coloniality and density dependence present unique analytical challenges.We review the relevant literature on data integration and illustrate it with example models and data (in an accompanying R‐library and supplementary vignette ), to derive methodological and quantitative guidelines for best practice in conducting joint inference for multi‐platform data. We use systematic survey data to motivate the key arguments, but also overview developments in integration with other data (e.g. telemetry tracking, citizen science, mark–recapture).We make recommendations on (a) the use of response and explanatory data, (b) the treatment of survey design and observation errors, (c) exploiting dependencies across space and time, (d) accounting for biological phenomena, such as commuting costs from the colony (i.e. accessibility) and density dependence, and (e) the choice of statistical framework.Synthesis and application. Integrated analysis of multi‐platform data turns many of the seabird‐specific challenges into opportunities for inferring habitat associations and predicting future distributions. Our review proposes practical recommendations for data collection and analysis that will allow seabird conservation to derive maximal benefits from these opportunities.