Species density models from opportunistic citizen science data

Abstract

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

Spatial Bayesian models project shifts in suitable habitat for Pacific Northwest tree species under climate change

“…Fitting spatial Bayesian hierarchical models can be computationally prohibitive for large data sets. Accordingly, we implemented a variety of strategies from Ver Hoef et al (2021) to speed up estimation, including sparse matrix calculations, simplifying density calculations for Metropolis–Hastings acceptance weights, and using lookup tables for inverse and determinate calculations of $$$ {\mathbf{\sum}}_{\boldsymbol{\uptheta}} $$$ based on different values of $$$ \uprho $$$. When data sets included unobserved response data (i.e., blue oak and white oak), we also increased the sampling rates for $$$ {\boldsymbol{z}}_{\boldsymbol{o}} $$$ and $$$ {\boldsymbol{z}}_{\boldsymbol{u}} $$$ to speed up convergence.…”

“…For computational savings during model fitting, we write $$$ \boldsymbol{z}={\left({{\boldsymbol{z}}_o}^{\prime },{{\boldsymbol{z}}_u}^{\prime}\right)}^{\prime } $$$ where $$$ {\boldsymbol{z}}_o $$$ and $$$ {\boldsymbol{z}}_u $$$ are vectors of spatial random effects associated with observed hexagons and unobserved hexagons, respectively (Ver Hoef et al, 2021). If the response on all FIA plots was observed (i.e., noble fir, coastal Douglas‐fir, and black oak), then $$$ {\boldsymbol{z}}_u $$$ is a zero‐length vector and $$$ \boldsymbol{z}={\boldsymbol{z}}_o $$$.…”

Identifying seasonal distribution patterns of fin whales across the Scotia Sea and the Antarctic Peninsula region using a novel approach combining habitat suitability models and ensemble learning methods

“…In our study, the combination of SDM and a random forest ensemble learning algorithm enabled us to derive information on abundance from a multi-source heterogeneous data set of mainly opportunistic sighting records and to present an additional set of abundance estimates for some of the most frequently visited research areas in the Southern Ocean. While abundance estimates from opportunistic sighting data that include at least some information on search effort have been explored before (Ver Hoef et al, 2021), presence-only (or presence-absence) data that do not contain any information on search effort have not been used to quantify abundances yet. By combining results from both methods based on the same data set, we were able to produce rough estimates of abundance that align with dedicated surveys in select regions and seasons.…”

Modelling the distribution of marine fishery resources: Where are we?

“…Despite this point, when properly used, fishery data become an important and necessary source and, coupled to ENMs/SDMs, could provide a good understanding of the dynamics in distribution of fisheries resources (Pennino et al, 2016). In the terrestrial environment, larger human accessibility, lower cost, and the possibility of nonscientific observations seem to be the main reasons for the discrepancy between the two environments (but see citizen science by Ver Hoef et al, 2021). The growing number of articles is also related to the growing computational power that allows researchers to develop models with some mastery.…”