Method selection for species distribution modelling: are temporally or spatially independent evaluations necessary?
Method selection for species distribution modelling: are temporally or spatially independent evaluations necessary?
复制标题
物种分布建模的方法选择：时间或空间独立评估是否必要？
DOI：
--
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
A. Hamann
中科院分区：
文献类型：
--
作者：
D. Roberts;A. Hamann
为了评估气候变化背景下栖息地预测的真实性，我们对12种物种分布模型进行了独立评估，其中包括3种新的基于生态系统的建模技术。根据6000年、11000年、14000年、16000年和21000年的931个古生态记录，对北美西部24种树种的生境预测结果进行了验证。此外，我们还评估了基于55000个样地地理分裂的区域外推。接受者工作特征分析表明，交叉验证的预测精度很高（中位AUC为0.90），独立区域和古生态验证的预测精度尚可（0.78和0.75）。令人惊讶的是，我们在任何方法中都没有发现过度参数化的证据。此外，考虑到在非独立评估和独立评估中发现的模型准确性之间的高度相关性，我们得出结论，非独立评估是有效的模型选择工具。基于生态系统的建模方法在模型敏感性方面的表现低于平均水平，但在特定城市统计和对远超出训练数据的外推的稳健性方面表现出色，这表明它们非常适合重建历史生物地理学和冰川避难所。
To assess the realism of habitat projections in the context of climate change, we conduct independent evaluations of twelve species distribution models, including three novel ecosystem-based modelling techniques. Habitat hindcasts for 24 western North American tree species were validated against 931 palaeoecological records from 6000, 11 000, 14 000, 16 000 and 21 000 yr before present. In addition, we evaluate regional extrapolations based on geographic splits of � 55 000 sample plots. Receiver operating characteristic analyses indicated excellent predictive accuracy for cross-validations (median AUC of 0.90) and fair accuracy for independent regional and palaeoecological validations (0.78 and 0.75). Surprisingly, we found little evidence for over-parameterisation in any method. Also, given high correlations found between model accuracies in non-independent and independent evaluations, we conclude that non-independent evaluations are eff ective model selection tools. Ecosystem-based modelling approaches performed below average with respect to model sensitivity but excelled in specifi city statistics and robustness against extrapolations far beyond training data, suggesting that they are well suited to reconstruct historical biogeographies and glacial refugia.