Detecting and Visualizing Observation Hot-Spots in Massive Volunteer-Contributed Geographic Data across Spatial Scales Using GPU-Accelerated Kernel Density Estimation

Volunteer-contributed geographic data (VGI) is an important source of geospatial big that support research and applications. A major concern on VGI quality the underlying observation processes are inherently biased. Detecting hot-spots thus helps better understand bias. Enabled by parallel kernel density estimation (KDE) computational tool can run multiple GPUs (graphics processing units), this study conducted point pattern analyses tens millions iNaturalist observations to detect visualize volunteers’ across spatial scales. It was achieved setting varying KDE bandwidths in accordance with scales at which be detected. The succession estimated surfaces were then rendered a sequence map for visual detection hot-spots. This offers effective geovisualization scheme hierarchically detecting massive datasets, useful understanding pattern-shaping drivers operate exemplifies supported high-performance computing capable efficiently visualizing multi-scale contributes expanding toolbox analytics.