Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regions
Large branchiopods are one of the most iconic species groups in temporary pools. Due to their specific biology and massive habitat degradation, they are one of the most threatened freshwater species groups. In this study, we used a unique dataset on the distribution of large branchiopods on a nation...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2020-09-01
|
Series: | Global Ecology and Conservation |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2351989420300068 |
id |
doaj-6b251084b23444a3ad9644885762a078 |
---|---|
record_format |
Article |
spelling |
doaj-6b251084b23444a3ad9644885762a0782020-11-25T02:09:52ZengElsevierGlobal Ecology and Conservation2351-98942020-09-0123Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regionsTobias Schernhammer0Johannes Wessely1Erich Eder2Ulrich Straka3Franz Essl4Division of Conservation Biology, Vegetation Ecology and Landscape Ecology, Department of Botany and Biodiversity Research, Rennweg 14, 1030, Vienna, Austria; Corresponding author.Division of Conservation Biology, Vegetation Ecology and Landscape Ecology, Department of Botany and Biodiversity Research, Rennweg 14, 1030, Vienna, AustriaSigmund Freud University Vienna, Medical School, Freudplatz 3, A-1020, Vienna, AustriaInstitute of Zoology, Department of Integrative Biology and Biodiversity Research, University of Natural Resources and Life Sciences, Gregor Mendel-Straße 33, A-1180, Vienna, AustriaDivision of Conservation Biology, Vegetation Ecology and Landscape Ecology, Department of Botany and Biodiversity Research, Rennweg 14, 1030, Vienna, AustriaLarge branchiopods are one of the most iconic species groups in temporary pools. Due to their specific biology and massive habitat degradation, they are one of the most threatened freshwater species groups. In this study, we used a unique dataset on the distribution of large branchiopods on a national scale (Austria), data on the distribution of temporary pools, and environmental variables to model the potential distribution of seven large branchiopod species. For species distribution modeling, we used six predictors which were assumed to be potentially relevant for explaining large branchiopod distribution. We used three modeling techniques (Generalized Linear Models, General Boosted Models, Random Forests) for distribution modeling. While some predictors (e.g. mean annual temperature, mean annual precipitation and the proportion of arable land at the temporary pool) were important for all target species, the importance of the other variables (size of the temporary pool, connectivity, flooding probability) was variable among species, and partly also between the three model techniques used. The resulting ensemble distribution maps, which are based on ensemble models of the above-mentioned modeling techniques, show the distribution of occurrence probabilities for the modeled species. These maps can serve as a baseline for future surveying regions that have been under-sampled so far. Furthermore, as these maps identify regions where large branchiopods likely do occur, conservation measures can be prioritized towards these regions. Finally, we discuss conservation measures which should improve land use by farmers to appropriately protect large branchiopods and their habitats, temporary pools.http://www.sciencedirect.com/science/article/pii/S2351989420300068Agricultural landscapeConservationLand useLarge branchiopodsModelingSpecies distribution |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Tobias Schernhammer Johannes Wessely Erich Eder Ulrich Straka Franz Essl |
spellingShingle |
Tobias Schernhammer Johannes Wessely Erich Eder Ulrich Straka Franz Essl Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regions Global Ecology and Conservation Agricultural landscape Conservation Land use Large branchiopods Modeling Species distribution |
author_facet |
Tobias Schernhammer Johannes Wessely Erich Eder Ulrich Straka Franz Essl |
author_sort |
Tobias Schernhammer |
title |
Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regions |
title_short |
Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regions |
title_full |
Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regions |
title_fullStr |
Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regions |
title_full_unstemmed |
Modelling the distribution of large branchiopods (Crustacea: Notostraca, Anostraca & Spinicaudata) for predicting occurrences in poorly sampled regions |
title_sort |
modelling the distribution of large branchiopods (crustacea: notostraca, anostraca & spinicaudata) for predicting occurrences in poorly sampled regions |
publisher |
Elsevier |
series |
Global Ecology and Conservation |
issn |
2351-9894 |
publishDate |
2020-09-01 |
description |
Large branchiopods are one of the most iconic species groups in temporary pools. Due to their specific biology and massive habitat degradation, they are one of the most threatened freshwater species groups. In this study, we used a unique dataset on the distribution of large branchiopods on a national scale (Austria), data on the distribution of temporary pools, and environmental variables to model the potential distribution of seven large branchiopod species. For species distribution modeling, we used six predictors which were assumed to be potentially relevant for explaining large branchiopod distribution. We used three modeling techniques (Generalized Linear Models, General Boosted Models, Random Forests) for distribution modeling. While some predictors (e.g. mean annual temperature, mean annual precipitation and the proportion of arable land at the temporary pool) were important for all target species, the importance of the other variables (size of the temporary pool, connectivity, flooding probability) was variable among species, and partly also between the three model techniques used. The resulting ensemble distribution maps, which are based on ensemble models of the above-mentioned modeling techniques, show the distribution of occurrence probabilities for the modeled species. These maps can serve as a baseline for future surveying regions that have been under-sampled so far. Furthermore, as these maps identify regions where large branchiopods likely do occur, conservation measures can be prioritized towards these regions. Finally, we discuss conservation measures which should improve land use by farmers to appropriately protect large branchiopods and their habitats, temporary pools. |
topic |
Agricultural landscape Conservation Land use Large branchiopods Modeling Species distribution |
url |
http://www.sciencedirect.com/science/article/pii/S2351989420300068 |
work_keys_str_mv |
AT tobiasschernhammer modellingthedistributionoflargebranchiopodscrustaceanotostracaanostracaampspinicaudataforpredictingoccurrencesinpoorlysampledregions AT johanneswessely modellingthedistributionoflargebranchiopodscrustaceanotostracaanostracaampspinicaudataforpredictingoccurrencesinpoorlysampledregions AT ericheder modellingthedistributionoflargebranchiopodscrustaceanotostracaanostracaampspinicaudataforpredictingoccurrencesinpoorlysampledregions AT ulrichstraka modellingthedistributionoflargebranchiopodscrustaceanotostracaanostracaampspinicaudataforpredictingoccurrencesinpoorlysampledregions AT franzessl modellingthedistributionoflargebranchiopodscrustaceanotostracaanostracaampspinicaudataforpredictingoccurrencesinpoorlysampledregions |
_version_ |
1724922020887527424 |