Data from: Mapping forest functional type in a forest-shrubland ecotone using SPOT imagery and predictive habitat distribution modelling

This is a portion of the data used in the analysis for the article “Mapping forest functional type in a forest-shrubland ecotone using SPOT imagery and predictive habitat distribution modelling” by T. Assal, P. Anderson, and J. Sibold, Remote Sensing Letters (2015). Five data sets are available for this project: (1) input point data used to derive presence/absence records of deciduous and coniferous forest type (.csv) ; (2) probability of occurrence model for deciduous forest (.img); (3) probability of occurrence model for coniferous forest (.img); (4) synthesis forest cover map derived from the probability of occurrence models (.img) ; (5) evaluation data used to calculate the receiver-operator characteristic area under the curve for each probability of occurrence model (subset of data) and to create a confusion matrix (full data set). All raster files are .img format and coordinate reference system for all data is WGS_1984_UTM_Zone_12North.

For additional information please see:
Assal, T., Anderson, P., & Sibold, J. (2015). Mapping forest functional type in a forest-shrubland ecotone using SPOT imagery and predictive habitat distribution modelling. Remote Sensing Letters, 6(10), 755–764. doi:10.1080/2150704X.2015.1072289

本数据集为T. Assal、P. Anderson与J. Sibold于2015年发表于《遥感快报（Remote Sensing Letters）》的论文《利用SPOT影像与预测性生境分布建模绘制森林-灌丛交错带森林功能型分布图》（Mapping forest functional type in a forest-shrubland ecotone using SPOT imagery and predictive habitat distribution modelling）的分析所用数据集的一部分。本项目共提供5类数据集：(1) 用于提取落叶林与针叶林类型存在/不存在记录的输入点数据（.csv格式）；(2) 落叶林发生概率模型（.img格式）；(3) 针叶林发生概率模型（.img格式）；(4) 基于发生概率模型合成的森林覆盖图（.img格式）；(5) 用于计算各发生概率模型的受试者工作特征曲线下面积（采用数据子集）并构建混淆矩阵（采用完整数据集）的评估数据。所有栅格文件均为.img格式，所有数据的坐标参考系统均为WGS_1984_UTM_Zone_12North。

如需获取更多信息，请参阅：
Assal, T., Anderson, P., & Sibold, J. (2015). Mapping forest functional type in a forest-shrubland ecotone using SPOT imagery and predictive habitat distribution modelling. Remote Sensing Letters, 6(10), 755–764. doi:10.1080/2150704X.2015.1072289