Spatial autocorrelation reduces model precision and predictive power in deforestation analyses
Spatial autocorrelation reduces model precision and predictive power in deforestation analyses
复制标题
在森林砍伐分析中，空间自相关会降低模型的精度和预测能力。
DOI：
10.1002/ecs2.1824
复制
发表时间：
2017-05-01
期刊：
影响因子：
2.7
通讯作者：
Davalos, Liliana M.
中科院分区：
文献类型：
--
作者：
Mets, Kristjan D.;Armenteras, Dolors;Davalos, Liliana M.
广义线性模型常用于识别景观过程的协变量和模拟土地利用变化。然而，广义线性模型忽略了土地利用数据的空间组成部分，及其对统计推断的影响。空间自相关可能会人为地减少观测值的方差，并夸大协变量的效应量。为了揭示忽视这一空间组成部分的后果，我们测试了哥伦比亚森林砍伐的空间明确和非空间模型。使用参数估计值、残差空间自相关分析和贝叶斯后验预测检验来比较模型性能。显著的残差相关表明，非空间模型未能充分解释数据的空间结构。后验预测检查显示，空间显式模型具有很强的预测能力的整个范围内的响应变量，只有未能预测离群值，与非空间模型，缺乏预测能力的所有响应值。在远离哥伦比亚中心的地区，非空间模型的预测能力特别低，那里大约有一半的观测结果是聚类的。虽然所有的分析都一致地确定了森林砍伐率的重要协变量的核心，但预测建模需要根据数据的空间结构进行参数估计。为告知日益重要的森林和碳固存政策，土地利用模型必须考虑空间自相关性。
Generalized linear models are often used to identify covariates of landscape processes and to model land-use change. Generalized linear models however, overlook the spatial component of land-use data, and its effects on statistical inference. Spatial autocorrelation may artificially reduce variance in observations, and inflate the effect size of covariates. To uncover the consequences of overlooking this spatial component, we tested both spatially explicit and non-spatial models of deforestation for Colombia. Parameter estimates, analyses of residual spatial autocorrelation, and Bayesian posterior predictive checks were used to compare model performance. Significant residual correlation showed that non-spatial models failed to adequately explain the spatial structure of the data. Posterior predictive checks revealed that spatially explicit models had strong predictive power for the entire range of the response variable and only failed to predict outliers, in contrast with non-spatial models, which lacked predictive power for all response values. The predictive power of non-spatial models was especially low in regions away from Colombia's center, where about half the observations were clustered. While all analyses consistently identified a core of important covariates of deforestation rates, predictive modeling requires parameter estimates informed by the spatial structure of the data. To inform increasingly important forest and carbon sequestration policy, land-use models must account for spatial autocorrelation.