scholarly journals Balancing transferability and complexity of species distribution models for rare species conservation

2020 ◽  
Vol 27 (1) ◽  
pp. 95-108
Author(s):  
Nolan A. Helmstetter ◽  
Courtney J. Conway ◽  
Bryan S. Stevens ◽  
Amanda R. Goldberg
Keyword(s):  
Species Distribution ◽  
Rare Species ◽  
Species Distribution Models ◽  
Species Conservation ◽  
Distribution Models

scholarly journals Trends and gaps in the use of citizen science derived data as input for species distribution models: A quantitative review

PLoS ONE ◽  
2021 ◽  
Vol 16 (3) ◽  
pp. e0234587
Author(s):  
Mariano J. Feldman ◽  
Louis Imbeau ◽  
Philippe Marchand ◽  
Marc J. Mazerolle ◽  
Marcel Darveau ◽  
...  
Keyword(s):  
Citizen Science ◽  
Species Distribution ◽  
Species Distribution Models ◽  
Spatial Scales ◽  
Species Conservation ◽  
Innovative Technology ◽  
Data Types ◽  
Species Occurrence ◽  
Distribution Models ◽  
Digital Platforms

Citizen science (CS) currently refers to the participation of non-scientist volunteers in any discipline of conventional scientific research. Over the last two decades, nature-based CS has flourished due to innovative technology, novel devices, and widespread digital platforms used to collect and classify species occurrence data. For scientists, CS offers a low-cost approach of collecting species occurrence information at large spatial scales that otherwise would be prohibitively expensive. We examined the trends and gaps linked to the use of CS as a source of data for species distribution models (SDMs), in order to propose guidelines and highlight solutions. We conducted a quantitative literature review of 207 peer-reviewed articles to measure how the representation of different taxa, regions, and data types have changed in SDM publications since the 2010s. Our review shows that the number of papers using CS for SDMs has increased at approximately double the rate of the overall number of SDM papers. However, disparities in taxonomic and geographic coverage remain in studies using CS. Western Europe and North America were the regions with the most coverage (73%). Papers on birds (49%) and mammals (19.3%) outnumbered other taxa. Among invertebrates, flying insects including Lepidoptera, Odonata and Hymenoptera received the most attention. Discrepancies between research interest and availability of data were as especially important for amphibians, reptiles and fishes. Compared to studies on animal taxa, papers on plants using CS data remain rare. Although the aims and scope of papers are diverse, species conservation remained the central theme of SDM using CS data. We present examples of the use of CS and highlight recommendations to motivate further research, such as combining multiple data sources and promoting local and traditional knowledge. We hope our findings will strengthen citizen-researchers partnerships to better inform SDMs, especially for less-studied taxa and regions. Researchers stand to benefit from the large quantity of data available from CS sources to improve global predictions of species distributions.

scholarly journals Magnolia mercedesiarum (subsect. Talauma, Magnoliaceae): a new Andean species from northern Ecuador, with insights into its potential distribution

Phytotaxa ◽  
2018 ◽  
Vol 348 (4) ◽  
pp. 254 ◽  
Author(s):  
J.-ANTONIO VÁZQUEZ-GARCÍA ◽  
DAVID A. NEILL ◽  
VIACHESLAV SHALISKO ◽  
FRANK ARROYO ◽  
R. EFRÉN MERINO-SANTI
Keyword(s):  
New Species ◽  
Species Distribution ◽  
Rare Species ◽  
Potential Distribution ◽  
Species Distribution Models ◽  
Distribution Area ◽  
Distribution Models ◽  
The Andes ◽  
Vegetation Sampling ◽  
A New Species

Magnolia mercedesiarum, a new species from the eastern slopes of the Andes in northern Ecuador, is described and illustrated, and a key to Ecuadorian Magnolia (subsect. Talauma) is provided. This species differs from M. vargasiana in having broadly elliptic leaves that have an obtuse base vs. suborbicular and subcordate to cordate, glabrous stipular scars, more numerous lateral veins per side and fewer stamens. It also differs from M. llanganatensis in having leaf blades broadly elliptic vs. elliptic, longer petioles, less numerous lateral leaf veins per side, larger fruits and more numerous petals and carpels. Using MaxEnt species distribution models and IUCN threat criteria, M. mercedesiarum has a potential distribution area of less than 3307 km² and is assessed as Endangered (EN): B1 ab (i, ii, iii). The relevance of systematic vegetation sampling in the discovery of rare species is highlighted.

scholarly journals Modelling the distribution of rare invertebrates by correcting class imbalance and spatial bias

2022 ◽  
Author(s):  
Willson B Gaul ◽  
Dinara Sadykova ◽  
Hannah J White ◽  
Lupe León-Sánchez ◽  
Paul Caplat ◽  
...  
Keyword(s):  
Species Distribution ◽  
Rare Species ◽  
Species Distribution Models ◽  
Class Imbalance ◽  
Prediction Performance ◽  
Good Prediction ◽  
Soil Arthropods ◽  
Spatial Bias ◽  
Distribution Models ◽  
Under Sampling

Aim: Soil arthropods are important decomposers and nutrient cyclers, but are poorly represented on national and international conservation Red Lists. Opportunistic biological records for soil invertebrates are often sparse, and contain few observations of rare species but a relatively large number of non-detection observations (a problem known as class imbalance). Robinson et al. (2018) proposed a method for sub-sampling non-detection data using a spatial grid to improve class balance and spatial bias in bird data. For taxa that are less intensively sampled, datasets are smaller, which poses a challenge because under-sampling data removes information. We tested whether spatial under-sampling improved prediction performance of species distribution models for millipedes, for which large datasets are not available. We also tested whether using environmental predictor variables provided additional information beyond what is captured by spatial position for predicting species distributions. Location: Island of Ireland. Methods: We tested the spatial under-sampling method of Robinson et al. (2018) by using biological records to train species distribution models of rare millipedes. Results: Using spatially under-sampled training data improved species distribution model sensitivity (true positive rate) but decreased model specificity (true negative rate). The decrease in specificity was minimal for rarer species and was accompanied by substantial increases in sensitivity. For common species, specificity decreased more, and sensitivity increased less, making spatial under-sampling most useful for rare species. Geographic coordinates were as good as or better than environmental variables for predicting distributions of two out of six species. Main Conclusions: Spatial under-sampling improved prediction performance of species distribution models for rare soil arthropod species. Spatial under-sampling was most effective for rarer species. The good prediction performance of models using geographic coordinates is promising for modeling distributions of poorly studied species for which little is known about ecological or physiological determinants of occurrence.

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