Evaluating species distribution model predictions through time against paleozoological records

IF 2.3 2区 生物学 Q2 ECOLOGY
Ignacio A. Lazagabaster, Chris D. Thomas, Juliet V. Spedding, Salima Ikram, Irene Solano-Regadera, Steven Snape, Jakob Bro-Jørgensen
{"title":"Evaluating species distribution model predictions through time against paleozoological records","authors":"Ignacio A. Lazagabaster,&nbsp;Chris D. Thomas,&nbsp;Juliet V. Spedding,&nbsp;Salima Ikram,&nbsp;Irene Solano-Regadera,&nbsp;Steven Snape,&nbsp;Jakob Bro-Jørgensen","doi":"10.1002/ece3.70288","DOIUrl":null,"url":null,"abstract":"<p>Species distribution models (SDMs) are widely used to project how species distributions may vary over time, particularly in response climate change. Although the fit of such models to current distributions is regularly enumerated, SDMs are rarely tested across longer time spans to gauge their actual performance under environmental change. Here, we utilise paleozoological presence/absence records to independently assess the predictive accuracy of SDMs through time. To illustrate the approach, we focused on modelling the Holocene distribution of the hartebeest, <i>Alcelaphus buselaphus</i>, a widespread savannah-adapted African antelope. We applied various modelling algorithms to three occurrence datasets, including a point dataset from online repositories and two range maps representing current and ‘natural’ (i.e. hypothetical assuming no human impact) distributions. We compared conventional model evaluation metrics which assess fit to current distributions (i.e. True Skill Statistic, TSS<sub>c</sub>, and Area Under the Curve, AUC<sub>c</sub>) to analogous ‘paleometrics’ for past distributions (i.e. TSS<sub>p</sub>, AUC<sub>p</sub>, and in addition Boyce<sub>p</sub>, F2-score<sub>p</sub> and Sorensen<sub>p</sub>). Our findings reveal only a weak correlation between the ranking of conventional metrics and paleometrics, suggesting that the models most effectively capturing present-day distributions may not be the most reliable to hindcast historical distributions, and that the choice of input data and modelling algorithm both significantly influences environmental suitability predictions and SDM performance. We thus advocate assessment of model performance using paleometrics, particularly those capturing the correct prediction of presences, such as F2-score<sub>p</sub> or Sorensen<sub>p</sub>, due to the potential unreliability of absence data in paleozoological records. By integrating archaeological and paleontological records into the assessment of alternative models' ability to project shifts in species distributions over time, we are likely to enhance our understanding of environmental constraints on species distributions.</p>","PeriodicalId":11467,"journal":{"name":"Ecology and Evolution","volume":"14 10","pages":""},"PeriodicalIF":2.3000,"publicationDate":"2024-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11496045/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ecology and Evolution","FirstCategoryId":"99","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/ece3.70288","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ECOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Species distribution models (SDMs) are widely used to project how species distributions may vary over time, particularly in response climate change. Although the fit of such models to current distributions is regularly enumerated, SDMs are rarely tested across longer time spans to gauge their actual performance under environmental change. Here, we utilise paleozoological presence/absence records to independently assess the predictive accuracy of SDMs through time. To illustrate the approach, we focused on modelling the Holocene distribution of the hartebeest, Alcelaphus buselaphus, a widespread savannah-adapted African antelope. We applied various modelling algorithms to three occurrence datasets, including a point dataset from online repositories and two range maps representing current and ‘natural’ (i.e. hypothetical assuming no human impact) distributions. We compared conventional model evaluation metrics which assess fit to current distributions (i.e. True Skill Statistic, TSSc, and Area Under the Curve, AUCc) to analogous ‘paleometrics’ for past distributions (i.e. TSSp, AUCp, and in addition Boycep, F2-scorep and Sorensenp). Our findings reveal only a weak correlation between the ranking of conventional metrics and paleometrics, suggesting that the models most effectively capturing present-day distributions may not be the most reliable to hindcast historical distributions, and that the choice of input data and modelling algorithm both significantly influences environmental suitability predictions and SDM performance. We thus advocate assessment of model performance using paleometrics, particularly those capturing the correct prediction of presences, such as F2-scorep or Sorensenp, due to the potential unreliability of absence data in paleozoological records. By integrating archaeological and paleontological records into the assessment of alternative models' ability to project shifts in species distributions over time, we are likely to enhance our understanding of environmental constraints on species distributions.

Abstract Image

根据古生物学记录评估物种分布模型的时间预测。
物种分布模型(SDMs)被广泛用于预测物种分布如何随时间变化,特别是在应对气候变化时。虽然此类模型与当前分布的拟合度经常被列举出来,但却很少在更长的时间跨度内对 SDMs 进行测试,以衡量它们在环境变化下的实际表现。在这里,我们利用古生物学的存在/消失记录来独立评估SDM在不同时期的预测准确性。为了说明这种方法,我们重点研究了全新世疣鼻动物(Alcelaphus buselaphus)的分布建模,这是一种广泛分布于热带稀树草原的非洲羚羊。我们将各种建模算法应用于三个分布数据集,包括一个来自在线资料库的点数据集和两个代表当前分布和 "自然 "分布(即假设没有人类影响)的分布范围图。我们将评估当前分布拟合度的传统模型评估指标(即真实技能统计量(TSSc)和曲线下面积(AUCc))与过去分布的类似 "古计量学 "指标(即 TSSp、AUCp 以及 Boycep、F2-scorep 和 Sorensenp)进行了比较。我们的研究结果表明,传统指标的排名与古计量学指标的排名之间只有微弱的相关性,这表明最有效地捕捉现今分布的模型可能不是最可靠的后向预测历史分布的模型,输入数据和建模算法的选择都会对环境适宜性预测和 SDM 性能产生重大影响。因此,由于古生物记录中的缺失数据可能不可靠,我们主张使用古计量学方法评估模型性能,特别是那些能够正确预测存在的方法,如 F2-scorep 或 Sorensenp。通过将考古学和古生物学记录整合到评估替代模型预测物种分布随时间变化的能力中,我们有可能加深对物种分布的环境制约因素的理解。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
4.40
自引率
3.80%
发文量
1027
审稿时长
3-6 weeks
期刊介绍: Ecology and Evolution is the peer reviewed journal for rapid dissemination of research in all areas of ecology, evolution and conservation science. The journal gives priority to quality research reports, theoretical or empirical, that develop our understanding of organisms and their diversity, interactions between them, and the natural environment. Ecology and Evolution gives prompt and equal consideration to papers reporting theoretical, experimental, applied and descriptive work in terrestrial and aquatic environments. The journal will consider submissions across taxa in areas including but not limited to micro and macro ecological and evolutionary processes, characteristics of and interactions between individuals, populations, communities and the environment, physiological responses to environmental change, population genetics and phylogenetics, relatedness and kin selection, life histories, systematics and taxonomy, conservation genetics, extinction, speciation, adaption, behaviour, biodiversity, species abundance, macroecology, population and ecosystem dynamics, and conservation policy.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信