scholarly journals Improving species distribution models for invasive non‐native species with biologically informed pseudo‐absence selection

2019 ◽  
Vol 46 (5) ◽  
pp. 1029-1040 ◽  
Author(s):  
Daniel Chapman ◽  
Oliver L. Pescott ◽  
Helen E. Roy ◽  
Rob Tanner
Keyword(s):  
Species Distribution ◽  
Native Species ◽  
Species Distribution Models ◽  
Distribution Models

scholarly journals Constructing ecological indices for urban environments using species distribution models

2021 ◽  
Author(s):  
Ariel Levi Simons ◽  
Stevie Caldwell ◽  
Michelle Fu ◽  
Jose Gallegos ◽  
Michael Gatheru ◽  
...  
Keyword(s):  
Los Angeles ◽  
Species Distribution ◽  
Indicator Species ◽  
Environmental Conditions ◽  
Native Species ◽  
Species Distribution Models ◽  
Urban Environments ◽  
Ecological Indices ◽  
Distribution Models ◽  
The Mean

Abstract In an increasingly urbanized world, there is the need for a framework to assess ecological conditions in these anthropogenically dominated environments. Using species observations from the Global Biodiversity Information Facility (GBIF), along with remotely sensed environmental layers, we used MaxEnt to construct species distribution models (SDMs) of native and non-native species in Los Angeles. 25 native and non-native Indicator species were selected based on the sensitivities of their SDM, as measured by the Symmetric Extremal Dependence Index (SEDI), to environmental gradients. These SDMs were summarized to produce ecological indices of native and non-native biodiversity in Los Angeles. We found native indicator species to have a greater sensitivity to environmental conditions than their non-native counterparts, with the mean SEDI score of native and non-native species MaxEnt models being 0.72 and 0.71 respectively. While both sets of species were sensitive to land use categories and housing density, native species were more sensitive to natural landscape variables while non-native ones were more sensitive to measures of water and soil contamination. Using random forest modeling we also found our native index could be more reliably predicted, given environmental conditions, than its non-native counterpart. The mean Pearson correlation between actual and predicted index values were 0.86 and 0.84 for native and non-native species. From these results we conclude that using SDMs to predict the biodiversity of environmental species is a suitable approach towards evaluating ecological conditions in urban environments, with the environmental sensitivity of native SDMs outperforming non-native ones.

scholarly journals Conservation biogeography of lotic fishes in the Missouri and Colorado River basins

2019 ◽  
Author(s):  
◽  
Landon Lee Pierce
Keyword(s):  
Species Distribution ◽  
Fish Assemblage ◽  
Native Species ◽  
Species Distribution Models ◽  
Colorado River ◽  
River Basins ◽  
Assemblage Structure ◽  
Native Fish ◽  
Fish Assemblage Structure ◽  
Distribution Models

To improve our understanding of lotic fish ecology and improve conservation efforts, I 1) identified potentially ecologically important tributaries (PEITs) and evaluated their effects on fish assemble structure, 2) evaluated factors affecting spatial transferability of species distribution models (SDMs), and 3) evaluated the drivers of non-native fish establishment in the Missouri and Colorado River basins (MRB and CRB). The effects of PEIT likely vary among rivers as all Missouri River PEITs affected fish assemblage structure, but only half of upper Colorado River basin PEITs affected fish assemblage structure. Species distribution models transferred from the MRB to the CRB for 15 of 25 species, but transferability was not predictable based on species characteristics, re-enforcing the hypothesis that transferability is species-and contextspecific. Support for Human Activity, Biotic Resistance and Biotic Acceptance hypotheses as the drivers of non-native fish establishment varied by family, but these hypotheses rarely explained significant variability in the probability of non-native Salmonidae, Catostomidae, and Cyprinidae occurrence. These results may suggest that other factors (e.g., natural factors) drive non-native species distributions at the spatial (i.e., grain-stream segment; extents-physiographic divisions, and MRB and CRB combined) and taxonomic (i.e., family) scales considered in this study. This study aids conservations efforts by providing an efficient approach for identifying ecologically important tributaries and improving predictions of non-native species establishment.

scholarly journals An invasion risk map for non-native aquatic macrophytes of the Iberian Peninsula

2017 ◽  
Vol 74 (1) ◽  
pp. 055 ◽  
Author(s):  
Argantonio Rodríguez-Merino ◽  
Rocío Fernández-Zamudio ◽  
Pablo García-Murillo
Keyword(s):  
Iberian Peninsula ◽  
Species Distribution ◽  
Native Species ◽  
Aquatic Macrophytes ◽  
Species Distribution Models ◽  
Green Revolution ◽  
High Sensitivity ◽  
Risk Map ◽  
Invasion Risk ◽  
Distribution Models

Freshwater systems are particularly susceptible to non-native organisms, owing to their high sensitivity to the impacts that are caused by these organisms. Species distribution models, which are based on both environmental and socio-economic variables, facilitate the identification of the most vulnerable areas for the spread of non-native species. We used MaxEnt to predict the potential distribution of 20 non-native aquatic macrophytes in the Iberian Peninsula. Some selected variables, such as the temperature seasonality and the precipitation in the driest quarter, highlight the importance of the climate on their distribution. Notably, the human influence in the territory appears as a key variable in the distribution of studied species. The model discriminated between favorable and unfavorable areas with high accuracy. We used the model to build an invasion risk map of aquatic macrophytes for the Iberian Peninsula that included results from 20 individual models. It showed that the most vulnerable areas are located near to the sea, the major rivers basins, and the high population density areas. These facts suggest the importance of the human impact on the colonization and distribution of non-native aquatic macrophytes in the Iberian Peninsula, and more precisely agricultural development during the Green Revolution at the end of the 70’s. Our work also emphasizes the utility of species distribution models for the prevention and management of biological invasions.

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