European Landscape Attractiveness Indicator

European Landscape Attractiveness Raster Dataset This raster layer represents a machine learning-based prediction of perceived attractiveness across Europe, extending a German model based on surveys. Values indicate how people are likely to rate the attractiveness of landscape based on visible features like land cover, infrastructure, and elevation. Predictions are derived from a model trained on the output of a German reference model (Roth et al., 2021), which was based on a survey of German residents rating landscape images. The extrapolation was done by replicating and retraining a model using pan-European features.欧洲景观吸引力栅格数据集 本栅格图层基于机器学习方法，对欧洲全域的主观感知景观吸引力开展预测，其在一项基于调研数据构建的德国本土模型基础上进行拓展。栅格数值可反映人们基于土地覆盖、基础设施与高程等可见景观特征，对该区域景观吸引力的主观评分倾向。 该预测模型以德国参考模型（Roth et al., 2021）的输出结果作为训练数据构建而成；该德国参考模型的构建依据为德国居民对景观图像进行评分的调研数据集。 本次空间外推工作通过复制并重新训练模型完成，训练过程采用了全欧洲尺度的景观特征变量。