scholarly journals The effect of local species composition on the distribution of an avian invader

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Tali Magory Cohen ◽  
Roi Dor
Keyword(s):  
Species Composition ◽  
Species Distribution ◽  
Species Interactions ◽  
Species Distribution Models ◽  
Predictive Ability ◽  
Distribution Models ◽  
Common Myna ◽  
Improve Model ◽  
Local Species ◽  
Occurrence Patterns

Abstract Estimating the potential distribution of invasive species has been primarily achieved by employing species distribution models (SDM). Recently introduced joint species distribution models (JSDM) that include species interactions are expected to improve model output. Here we compare the predictive ability of SDM and JSDM by modelling the distribution of one of the most prolific avian invaders in the world, the common myna (Acridotheres tristis), in a recent introduction in Israel. Our results indicate that including information on the local species composition did not improve model accuracy, possibly because of the unique characteristics of this species that include broad environmental tolerance and behavior flexibility. However, the JSDM provided insights into co-occurrence patterns of common mynas and their local heterospecifics, suggesting that at this time point, there is no evidence of species exclusion by common mynas. Our findings suggest that the invasion potential of common mynas depends greatly on urbanization and less so on the local species composition and reflect the major role of anthropogenic impact in increasing the distribution of avian invaders.

scholarly journals Deep Multi-species Embedding

2017 ◽  
Author(s):  
Di Chen ◽  
Yexiang Xue ◽  
Daniel Fink ◽  
Shuo Chen ◽  
Carla P. Gomes
Keyword(s):  
Neural Network ◽  
Species Distribution ◽  
Species Interactions ◽  
Deep Neural Network ◽  
Species Distribution Models ◽  
Habitat Preferences ◽  
Bird Species ◽  
Single Species ◽  
Additional Contribution ◽  
Distribution Models

Understanding how species are distributed across landscapes over time is a fundamental question in biodiversity research. Unfortunately, most species distribution models only target a single species at a time, despite strong ecological evidence that species are not independently distributed. We propose Deep Multi-Species Embedding (DMSE), which jointly embeds vectors corresponding to multiple species as well as vectors representing environmental covariates into a common high-dimensional feature space via a deep neural network. Applied to bird observational data from the citizen science project eBird, we demonstrate how the DMSE model discovers inter-species relationships to outperform single-species distribution models (random forests and SVMs) as well as competing multi-label models. Additionally, we demonstrate the benefit of using a deep neural network to extract features within the embedding and show how they improve the predictive performance of species distribution modelling. An important domain contribution of the DMSE model is the ability to discover and describe species interactions while simultaneously learning the shared habitat preferences among species. As an additional contribution, we provide a graphical embedding of hundreds of bird species in the Northeast US.

scholarly journals Incorporating Context Dependency of Species Interactions in Species Distribution Models

2017 ◽  
Vol 57 (1) ◽  
pp. 159-167 ◽  
Author(s):  
Nina K. Lany ◽  
Phoebe L. Zarnetske ◽  
Tarik C. Gouhier ◽  
Bruce A. Menge
Keyword(s):  
Species Distribution ◽  
Species Interactions ◽  
Species Distribution Models ◽  
Context Dependency ◽  
Distribution Models

Dealing with overprediction in species distribution models: How adding distance constraints can improve model accuracy

2020 ◽  
Vol 431 ◽  
pp. 109180 ◽  
Author(s):  
Poliana Mendes ◽  
Santiago José Elías Velazco ◽  
André Felipe Alves de Andrade ◽  
Paulo De Marco
Keyword(s):  
Species Distribution ◽  
Species Distribution Models ◽  
Model Accuracy ◽  
Distribution Models ◽  
Distance Constraints ◽  
Improve Model

Improved understanding and prediction of freshwater fish communities through the use of joint species distribution models

2020 ◽  
Vol 77 (9) ◽  
pp. 1540-1551 ◽  
Author(s):  
Tyler Wagner ◽  
Gretchen J.A. Hansen ◽  
Erin M. Schliep ◽  
Bethany J. Bethke ◽  
Andrew E. Honsey ◽  
...  
Keyword(s):  
Species Distribution ◽  
Species Interactions ◽  
Species Distribution Models ◽  
Fish Communities ◽  
Freshwater Ecosystems ◽  
Habitat Associations ◽  
Distribution Models ◽  
Fisheries Science ◽  
Flexible Framework ◽  
Abiotic Habitat

Two primary goals in fisheries research are to (i) understand how habitat and environmental conditions influence the distribution of fishes across the landscape and (ii) make predictions about how fish communities will respond to environmental and anthropogenic change. In inland, freshwater ecosystems, quantitative approaches traditionally used to accomplish these goals largely ignore the effects of species interactions (competition, predation, mutualism) on shaping community structure, potentially leading to erroneous conclusions regarding habitat associations and unrealistic predictions about species distributions. Using two contrasting case studies, we highlight how joint species distribution models (JSDMs) can address the aforementioned deficiencies by simultaneously quantifying the effects of abiotic habitat variables and species dependencies. In particular, we show that conditional predictions of species occurrence from JSDMs can better predict species presence or absence compared with predictions that ignore species dependencies. JSDMs also allow for the estimation of site-specific probabilities of species co-occurrence, which can be informative for generating hypotheses about species interactions. JSDMs provide a flexible framework that can be used to address a variety of questions in fisheries science and management.

scholarly journals A Robust Prediction Model for Species Distribution Using Bagging Ensembles with Deep Neural Networks

2021 ◽  
Vol 13 (8) ◽  
pp. 1495
Author(s):  
Jehyeok Rew ◽  
Yongjang Cho ◽  
Eenjun Hwang
Keyword(s):  
Neural Networks ◽  
Species Distribution ◽  
Best Practice ◽  
Deep Neural Networks ◽  
Species Distribution Models ◽  
Species Distribution Model ◽  
Environmental Data ◽  
Distribution Model ◽  
Distribution Models ◽  
Robust Prediction

Species distribution models have been used for various purposes, such as conserving species, discovering potential habitats, and obtaining evolutionary insights by predicting species occurrence. Many statistical and machine-learning-based approaches have been proposed to construct effective species distribution models, but with limited success due to spatial biases in presences and imbalanced presence-absences. We propose a novel species distribution model to address these problems based on bootstrap aggregating (bagging) ensembles of deep neural networks (DNNs). We first generate bootstraps considering presence-absence data on spatial balance to alleviate the bias problem. Then we construct DNNs using environmental data from presence and absence locations, and finally combine these into an ensemble model using three voting methods to improve prediction accuracy. Extensive experiments verified the proposed model’s effectiveness for species in South Korea using crowdsourced observations that have spatial biases. The proposed model achieved more accurate and robust prediction results than the current best practice models.

Sign in / Sign up

Export Citation Format

Share Document