The secret assumption of transfer functions: problems with spatial autocorrelation in evaluating model performance
The secret assumption of transfer functions: problems with spatial autocorrelation in evaluating model performance
复制标题
DOI：
10.1016/j.quascirev.2005.05.001
复制
发表时间：
2005-11
影响因子：
4
通讯作者：
R. Telford;H. Birks
中科院分区：
文献类型：
--
作者：
R. Telford;H. Birks
传递函数的预测能力的估计假设测试站点独立于建模站点。存在空间自相关的情况下的交叉验证严重违反了这一假设。这一假设及其违反的后果之前尚未讨论过。我们通过模拟表明，自相关环境中传递函数模型的预期 r2 可能很高，并且不像通常假设的那样接近于零。我们研究了北大西洋有孔虫海面温度训练集，通过交叉验证，现代模拟技术（MAT）和人工神经网络（ANN）优于基于单峰物种-环境响应模型的传递函数方法。然而，当使用空间独立的测试集（南大西洋）时，所有模型都具有相似的预测能力。我们表明，即使考虑了温度，有孔虫组合也存在空间结构，这可能是由于其他环境变量的自相关所致。由于 MAT 的残差几乎没有显示空间结构，与单峰响应模型的残差相比，我们认为 MAT 不恰当地内化了非温度空间结构以提高其性能。我们认为，迄今为止发布的 MAT 和 ANN 模型对海面温度的预测能力的大多数（如果不是全部）估计都过于乐观且具有误导性。
The estimation of the predictive power of transfer functions assumes that the test sites are independent of the modelling sites. Cross-validation in the presence of spatial autocorrelation seriously violates this assumption. This assumption and the consequences of its violation have not been discussed before. We show, by simulation, that the expected r2of a transfer function model from an autocorrelated environment can be high, and is not near zero as commonly assumed. We investigate a foraminiferal sea surface temperature training set for the North Atlantic, for which, with cross-validation, the modern analogue technique (MAT) and artificial neural networks (ANN) outperform transfer function methods based on a unimodal species-environment response model. However, when a spatially independent test set, the South Atlantic, is used, all models have a similar predictive power. We show that there is a spatial structure in the foraminiferal assemblages even after accounting for temperature, presumably due to autocorrelations in other environmental variables. Since the residuals from MAT show little spatial structure, in contrast to the residuals of unimodal response models, we contend that MAT has inappropriately internalized the non-temperature spatial structure to improve its performance. We argue that most, if not all, estimates of the predictive power of MAT and ANN models for sea surface temperatures hitherto published are over-optimistic and misleading.