Figure 1: Tagger System Architecture TaggerVR: Immersive Data Analytics for Geoscience Oculus Rift and Leap Motion interfaces for Interactive Visual Analytics of Large Geoscientific Datasets in Cloud Repositories Peter Morse (Author) School of Physical Sciences (Earth Sciences) University of Tasmania Hobart, Australia peter.morse@utas.edu.au Anya Reading School of Physical Sciences (Earth Sciences) University of Tasmania Hobart, Australia anya.reading@utas.edu.au Christopher Lueg School of Engineering and ICT University of Tasmania Hobart, Australia Sarah Kenderdine iGLAMS, NIEA, Art & Design University of New South Wales Sydney, Australia Abstract—‘TaggerVR’ is a work-in-progress immersive Virtual Reality implementation of the ‘Tagger’ software application developed by the author (Morse). Tagger is an interactive software tool designed to visualize, characterize, sample and tag large geoscientific datasets hosted in local and cloud-based repositories using a THREDDS Data Server and OPeNDAP. TaggerVR implements a VR GUI using Human Interface Devices (HID) such as the Oculus Rift headset and Leap Motion Controller via Quartz Composer, Syphon and the Unity3D Game Engine. Keywords—Virtual Reality; Visual Analytics; Geoscience Data; Cloud Repository; Human-Computer Interaction; Game Engines; Visual Programming I. TAGGER SOFTWARE AND WORKFLOW Tagger is a software application and workflow developed in the Quartz Composer (QC) Visual Programming Language for the display of, and interaction with, large geoscientific datasets [1]. It implements an intuitive animated graphing interface for time-variant 1-dimensional data, with optional parametrically- driven animated overlays. Multiple 1D value streams can be concurrently graphed for comparison purposes. Animated graphs can be displayed in a variety of standard formats with user-controlled animation parameters (such as speed of animation, color and scale), levels of detail (zooming and subsampling) as well as unconventional visual cues such as particle systems and parametric geometry (e.g. running averages over user-defined selection ranges) driven by the underlying values of incoming data. QC draws highly-performant OpenGL graphics to screen, via the Mac OSX graphics stack[2], significantly out- performing conventional plotting libraries such as MatPlotLib[3]. It also enables the development of unconventional GUIs and HID interaction, including deployment upon unconventional screen devices such as VR headsets (Oculus Rift via QC or Unity3D), immersive dome systems and arbitrary projection geometries, in concert with novel interaction modalities beyond mouse and keyboard (e.g. Leap Motion Controller 1 [4]). II. THREDDS, OPENDAP, NETCDF Tagger was designed to enable the rapid preliminary visualization of large geoscientific datasets hosted in cloud repositories. A THREDDS data server (TDS 2 ) was deployed within an Ubuntu Linux Virtual Machine (VM) hosted upon the Australian NCRIS NeCTAR Research Cloud 3 . The TDS was developed by NCAR Unidata specifically for the hosting and 1 https://www.leapmotion.com accessed 6/4/2015 2 http://www.unidata.ucar.edu/software/thredds/current/tds/ accessed 6/4/2015 3 https://www.nectar.org.au accessed 6/4/2015