Wildfire ignition-distribution modelling: a comparative study in the Huron–Manistee National Forest, Michigan, USA

2013 ◽  
Vol 22 (2) ◽  
pp. 174 ◽  
Author(s):  
Avi Bar Massada ◽  
Alexandra D. Syphard ◽  
Susan I. Stewart ◽  
Volker C. Radeloff
Keyword(s):  
Random Forests ◽  
Model Performance ◽  
Machine Learning Algorithms ◽  
Environmental Data ◽  
Distribution Model ◽  
National Forest ◽  
Similar Species ◽  
Distribution Models ◽  
Distribution Modelling ◽  
Manistee National Forest

Wildfire ignition distribution models are powerful tools for predicting the probability of ignitions across broad areas, and identifying their drivers. Several approaches have been used for ignition-distribution modelling, yet the performance of different model types has not been compared. This is unfortunate, given that conceptually similar species-distribution models exhibit pronounced differences among model types. Therefore, our goal was to compare the predictive performance, variable importance and the spatial patterns of predicted ignition-probabilities of three ignition-distribution model types: one parametric, statistical model (Generalised Linear Models, GLM) and two machine-learning algorithms (Random Forests and Maximum Entropy, Maxent). We parameterised the models using 16 years of ignitions data and environmental data for the Huron–Manistee National Forest in Michigan, USA. Random Forests and Maxent had slightly better prediction accuracies than did GLM, but model fit was similar for all three. Variables related to human population and development were the best predictors of wildfire ignition locations in all models (although variable rankings differed slightly), along with elevation. However, despite similar model performance and variables, the map of ignition probabilities generated by Maxent was markedly different from those of the two other models. We thus suggest that when accurate predictions are desired, the outcomes of different model types should be compared, or alternatively combined, to produce ensemble predictions.

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.

scholarly journals Assessing the effect of fish size on species distribution model performance in southern Chilean rivers

2018 ◽  
Author(s):  
Daniel Zamorano ◽  
Fabio Labra ◽  
Marcelo Villarroel ◽  
Luca Mao ◽  
Shaw Lucy ◽  
...  
Keyword(s):  
Body Size ◽  
Species Distribution ◽  
River Flow ◽  
Species Distribution Model ◽  
Anthropogenic Activities ◽  
Model Performance ◽  
Distribution Model ◽  
Distribution Models ◽  
Fish Size ◽  
River Fish

Despite its theoretical relationship, the effect of body size on the performance of species distribution models (SDM) has only been assessed in a few studies of terrestrial taxa. We aim to assess the effect of body size on the performance of SDM in river fish. We study seven Chilean freshwater fish, using models trained with three different sets of predictor variables: ecological (Eco), anthropogenic (Antr) and both (Eco+Antr). Our results indicate that the performance of the Eco+Antr models improves with fish size. These results highlight the importance of two novel predictive layers: the source of river flow and the overproduction of biotopes by anthropogenic activities. We compare our work with previous studies that modeled river fish, and observe a similar relationship in most cases. We discuss the current challenges of the modeling of riverine species, and how our work helps suggest possible solutions.

scholarly journals SDMtoolbox 2.0: the next generation Python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses

PeerJ ◽  
2017 ◽  
Vol 5 ◽  
pp. e4095 ◽  
Author(s):  
Jason L. Brown ◽  
Joseph R. Bennett ◽  
Connor M. French
Keyword(s):  
Species Distribution ◽  
Species Distribution Model ◽  
Species Distribution Modeling ◽  
Environmental Data ◽  
Distribution Model ◽  
Data Types ◽  
Distribution Models ◽  
Distribution Modeling ◽  
Modeling Software ◽  
Cost Paths

SDMtoolbox 2.0 is a software package for spatial studies of ecology, evolution, and genetics. The release of SDMtoolbox 2.0 allows researchers to use the most current ArcGIS software and MaxEnt software, and reduces the amount of time that would be spent developing common solutions. The central aim of this software is to automate complicated and repetitive spatial analyses in an intuitive graphical user interface. One core tenant facilitates careful parameterization of species distribution models (SDMs) to maximize each model’s discriminatory ability and minimize overfitting. This includes carefully processing of occurrence data, environmental data, and model parameterization. This program directly interfaces with MaxEnt, one of the most powerful and widely used species distribution modeling software programs, although SDMtoolbox 2.0 is not limited to species distribution modeling or restricted to modeling in MaxEnt. Many of the SDM pre- and post-processing tools have ‘universal’ analogs for use with any modeling software. The current version contains a total of 79 scripts that harness the power of ArcGIS for macroecology, landscape genetics, and evolutionary studies. For example, these tools allow for biodiversity quantification (such as species richness or corrected weighted endemism), generation of least-cost paths and corridors among shared haplotypes, assessment of the significance of spatial randomizations, and enforcement of dispersal limitations of SDMs projected into future climates—to only name a few functions contained in SDMtoolbox 2.0. Lastly, dozens of generalized tools exists for batch processing and conversion of GIS data types or formats, which are broadly useful to any ArcMap user.

scholarly journals Species versus within-species niches: a multi-modelling approach to assess range size of a spring-dwelling amphibian

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Forough Goudarzi ◽  
Mahmoud-Reza Hemami ◽  
Mansoureh Malekian ◽  
Sima Fakheran ◽  
Fernando Martínez-Freiría
Keyword(s):  
Species Distribution ◽  
Conservation Status ◽  
Species Conservation ◽  
Landscape Scale ◽  
Species Level ◽  
Environmental Data ◽  
Similar Species ◽  
Fine Scale ◽  
Distribution Models ◽  
Area Of Occupancy

AbstractSpecies Distribution Models (SDMs) can be used to estimate potential geographic ranges and derive indices to assess species conservation status. However, habitat-specialist species require fine-scale range estimates that reflect resource dependency. Furthermore, local adaptation of intraspecific lineages to distinct environmental conditions across ranges have frequently been neglected in SDMs. Here, we propose a multi-stage SDM approach to estimate the distributional range and potential area of occupancy (pAOO) of Neurergus kaiseri, a spring-dwelling amphibian with two climatically-divergent evolutionary lineages. We integrate both broad-scale climatic variables and fine-resolution environmental data to predict the species distribution while examining the performance of lineage-level versus species-level modelling on the estimated pAOO. Predictions of habitat suitability at the landscape scale differed considerably between evolutionary level models. At the landscape scale, spatial predictions derived from lineage-level models showed low overlap and recognised a larger amount of suitable habitats than species-level model. The variable dependency of lineages was different at the landscape scale, but similar at the local scale. Our results highlight the importance of considering fine-scale resolution approaches, as well as intraspecific genetic structure of taxa to estimate pAOO. The flexible procedure presented here can be used as a guideline for estimating pAOO of other similar species.

scholarly journals Assessing the effect of fish size on species distribution model performance in southern Chilean rivers

2018 ◽  
Author(s):  
Daniel Zamorano ◽  
Fabio Labra ◽  
Marcelo Villarroel ◽  
Luca Mao ◽  
Shaw Lucy ◽  
...  
Keyword(s):  
Body Size ◽  
Species Distribution ◽  
River Flow ◽  
Species Distribution Model ◽  
Anthropogenic Activities ◽  
Model Performance ◽  
Distribution Model ◽  
Distribution Models ◽  
Fish Size ◽  
River Fish

Despite its theoretical relationship, the effect of body size on the performance of species distribution models (SDM) has only been assessed in a few studies of terrestrial taxa. We aim to assess the effect of body size on the performance of SDM in river fish. We study seven Chilean freshwater fish, using models trained with three different sets of predictor variables: ecological (Eco), anthropogenic (Antr) and both (Eco+Antr). Our results indicate that the performance of the Eco+Antr models improves with fish size. These results highlight the importance of two novel predictive layers: the source of river flow and the overproduction of biotopes by anthropogenic activities. We compare our work with previous studies that modeled river fish, and observe a similar relationship in most cases. We discuss the current challenges of the modeling of riverine species, and how our work helps suggest possible solutions.

scholarly journals Evaluating the influence of spatially varying catchability on multispecies distribution modelling

2020 ◽  
Vol 77 (5) ◽  
pp. 1841-1853
Author(s):  
Chongliang Zhang ◽  
Yong Chen ◽  
Binduo Xu ◽  
Ying Xue ◽  
Yiping Ren
Keyword(s):  
Species Distribution ◽  
Predictive Power ◽  
Species Distribution Model ◽  
Distribution Model ◽  
Distribution Models ◽  
Distribution Modelling ◽  
Species Response ◽  
Challenges And Opportunities ◽  
Empirical Tests ◽  
Spatially Varying

Abstract Varying catchability is a common feature in fisheries and has great impacts on fisheries assessments and species distribution models. However, spatial variations in catchability have been rarely evaluated, especially in the multispecies context. We advocate that the need for multispecies models stands for both challenges and opportunities to handle spatial catchability. This study evaluated the influence of spatially varying catchability on the performance of a novel joint species distribution model, namely Hierarchical Modelling of Species Communities (HMSC). We implemented the model under nine simulation scenarios to account for diverse spatial patterns of catchability and conducted empirical tests using survey data from Yellow Sea, China. Our results showed that ignoring variability in catchability could lead to substantial errors in the inferences of species response to environment. Meanwhile, the models’ predictive power was less impacted, yielding proper predictions of relative abundance. Incorporating a spatially autocorrelated structure substantially improved the predictability of HMSC in both simulation and empirical tests. Nevertheless, combined sources of spatial catchabilities could largely diminish the advantage of HMSC in inference and prediction. We highlight situations where catchability needs to be explicitly accounted for in modelling fish distributions, and suggest directions for future applications and development of JSDMs.

scholarly journals Data quantity is more important than its spatial bias for predictive species distribution modelling

2020 ◽  
Author(s):  
Willson Gaul ◽  
Dinara Sadykova ◽  
Hannah J. White ◽  
Lupe León-Sánchez ◽  
Paul Caplat ◽  
...  
Keyword(s):  
Species Distribution ◽  
Species Distribution Model ◽  
Species Distribution Modelling ◽  
Prediction Performance ◽  
Training Data ◽  
Distribution Model ◽  
Spatial Bias ◽  
Distribution Models ◽  
Distribution Modelling ◽  
Modelling Method

ABSTRACTBiological records are often the data of choice for training predictive species distribution models (SDMs), but spatial sampling bias is pervasive in biological records data at multiple spatial scales and is thought to impair the performance of SDMs. We simulated presences and absences of virtual species as well as the process of recording these species to evaluate the effect on species distribution model prediction performance of 1) spatial bias in training data, 2) sample size (the average number of observations per species), and 3) the choice of species distribution modelling method. Our approach is novel in quantifying and applying real-world spatial sampling biases to simulated data. Spatial bias in training data decreased species distribution model prediction performance, but only when the bias was relatively strong. Sample size and the choice of modelling method were more important than spatial bias in determining the prediction performance of species distribution models.

scholarly journals Incorporating uncertainty in predictive species distribution modelling

2012 ◽  
Vol 367 (1586) ◽  
pp. 247-258 ◽  
Author(s):  
Colin M. Beale ◽  
Jack J. Lennon
Keyword(s):  
Species Distribution ◽  
Goodness Of Fit ◽  
Spatial Scales ◽  
Predictive Performance ◽  
Species Distribution Modelling ◽  
Control Measures ◽  
Distribution Model ◽  
Distribution Models ◽  
Distribution Modelling ◽  
Demographic Models

Motivated by the need to solve ecological problems (climate change, habitat fragmentation and biological invasions), there has been increasing interest in species distribution models (SDMs). Predictions from these models inform conservation policy, invasive species management and disease-control measures. However, predictions are subject to uncertainty, the degree and source of which is often unrecognized. Here, we review the SDM literature in the context of uncertainty, focusing on three main classes of SDM: niche-based models, demographic models and process-based models. We identify sources of uncertainty for each class and discuss how uncertainty can be minimized or included in the modelling process to give realistic measures of confidence around predictions. Because this has typically not been performed, we conclude that uncertainty in SDMs has often been underestimated and a false precision assigned to predictions of geographical distribution. We identify areas where development of new statistical tools will improve predictions from distribution models, notably the development of hierarchical models that link different types of distribution model and their attendant uncertainties across spatial scales. Finally, we discuss the need to develop more defensible methods for assessing predictive performance, quantifying model goodness-of-fit and for assessing the significance of model covariates.

Species distribution modelling of the Southern Ocean benthos: a review on methods, cautions and solutions

2021 ◽  
pp. 1-24
Author(s):  
Charlène Guillaumot ◽  
Bruno Danis ◽  
Thomas Saucède
Keyword(s):  
Southern Ocean ◽  
Species Distribution ◽  
Model Performance ◽  
Species Distribution Modelling ◽  
Projection Area ◽  
Distribution Models ◽  
Distribution Modelling ◽  
Influence Model ◽  
Modelling Studies

Abstract Species distribution modelling studies the relationship between species occurrence records and their environmental setting, providing a valuable approach to predicting species distribution in the Southern Ocean (SO), a challenging region to investigate due to its remoteness and extreme weather and sea conditions. The specificity of SO studies, including restricted field access and sampling, the paucity of observations and difficulties in conducting biological experiments, limit the performance of species distribution models. In this review, we discuss some issues that may influence model performance when preparing datasets and calibrating models, namely the selection and quality of environmental descriptors, the spatial and temporal biases that may affect the quality of occurrence data, the choice of modelling algorithms and the spatial scale and limits of the projection area. We stress the importance of evaluating and communicating model uncertainties, and the most common evaluation metrics are reviewed and discussed accordingly. Based on a selection of case studies on SO benthic invertebrates, we highlight important cautions to take and pitfalls to avoid when modelling the distribution of SO species, and we provide some guidelines along with potential methods and original solutions that can be used for improving model performance.

Comparing predictions of fisheries bycatch using multiple spatiotemporal species distribution model frameworks

2020 ◽  
Vol 77 (1) ◽  
pp. 146-163 ◽  
Author(s):  
Brian C. Stock ◽  
Eric J. Ward ◽  
Tomoharu Eguchi ◽  
Jason E. Jannot ◽  
James T. Thorson ◽  
...  
Keyword(s):  
Random Forests ◽  
Species Distribution ◽  
Markov Random Fields ◽  
Fishery Management ◽  
Species Distribution Model ◽  
Additive Models ◽  
Distribution Model ◽  
Range Of Movement ◽  
Distribution Models ◽  
Model Framework

Spatiotemporal predictions of bycatch (i.e., catch of nontargeted species) have shown promise as dynamic ocean management tools for reducing bycatch. However, which spatiotemporal model framework to use for generating these predictions is unclear. We evaluated a relatively new method, Gaussian Markov random fields (GMRFs), with two other frameworks, generalized additive models (GAMs) and random forests. We fit geostatistical delta-models to fisheries observer bycatch data for six species with a broad range of movement patterns (e.g., highly migratory sea turtles versus sedentary rockfish) and bycatch rates (percentage of observations with nonzero catch, 0.3%–96.2%). Random forests had better interpolation performance than the GMRF and GAM models for all six species, but random forests performance was more sensitive when predicting data at the edge of the fishery (i.e., spatial extrapolation). Using random forests to identify and remove the 5% highest bycatch risk fishing events reduced the bycatch-to-target species catch ratio by 34% on average. All models considerably reduced the bycatch-to-target ratio, demonstrating the clear potential of species distribution models to support spatial fishery management.

Sign in / Sign up

Export Citation Format

Share Document