Submerse: Visualizing Storm Surge Flooding Simulations in Immersive Display Ecologies

TL;DR

Submerse addresses the need for effective 3D, immersive visualization of storm surge flooding by delivering an end-to-end pipeline that converts flood simulations into a to-scale 3D GIS-enabled scene on immersive displays. It combines an adaptive quadtree-based height-field reconstruction with Gerstner wave synthesis to depict flood progression and direction, and introduces an AR auxiliary display for focused, personalized analysis. The system includes an automatic camera view-finding and path-generation component for multi-POI exploration, and a novel AR focus+context auxiliary display to support collaborative interpretation. Evaluations with domain experts in NYC using the Stony Brook Reality Deck demonstrate improved understanding and decision support, highlighting the practical impact of immersive flood visualization for disaster planning and response.

Abstract

We present Submerse, an end-to-end framework for visualizing flooding scenarios on large and immersive display ecologies. Specifically, we reconstruct a surface mesh from input flood simulation data and generate a to-scale 3D virtual scene by incorporating geographical data such as terrain, textures, buildings, and additional scene objects. To optimize computation and memory performance for large simulation datasets, we discretize the data on an adaptive grid using dynamic quadtrees and support level-of-detail based rendering. Moreover, to provide a perception of flooding direction for a time instance, we animate the surface mesh by synthesizing water waves. As interaction is key for effective decision-making and analysis, we introduce two novel techniques for flood visualization in immersive systems: (1) an automatic scene-navigation method using optimal camera viewpoints generated for marked points-of-interest based on the display layout, and (2) an AR-based focus+context technique using an auxiliary display system. Submerse is developed in collaboration between computer scientists and atmospheric scientists. We evaluate the effectiveness of our system and application by conducting workshops with emergency managers, domain experts, and concerned stakeholders in the Stony Brook Reality Deck, an immersive gigapixel facility, to visualize a superstorm flooding scenario in New York City.

Figures (10)

• Figure 1: An example of constructing a water surface mesh using an adaptive grid. For flood datapoints defined on a geographic location (a), we discretize the data on a quadtree (b) and interpolate the values to construct a smooth and continuous field as shown in (c) and (d). A cell without a datapoint (e) is defined by averaging the values of its neighboring defined cells (f). To interpolate a value for the red point in (g), we apply bilinear interpolation on the cells that bound the point. However, since one of the sampling cells, marked in red, has a lower cell resolution, we divide the cell (h) and assign a value by upsampling. The upsampling is done by applying bilinear interpolation on the neighboring same-resolution cells (i). A lower resolution value is calculated by averaging the value of all its child higher resolution cells (downsampling). Finally, a value for the point is constructed by applying bilinear interpolation on the same-resolution cells (j).
• Figure 2: (a) shows vertex motion following a Gerstner wave, and (b) illustrates the difference in calculating analytical and mesh normals.
• Figure 3: Two examples (a) and (b) of our dynamic quadtree update based on camera motion. Red represents the deleted quadrants and the newly added nodes are shaded in green.
• Figure 4: System protocol of our AR-based auxiliary (aux) display. (a) Localization is initialized by registering an image marker, placed at the physical location of the defined head. (b) The tracked position and orientation of the device are transformed from the physical to the virtual space, and data interaction is supported by casting a ray from the physical device position to the virtual scene. Selection and depth are performed by tapping and swiping gestures. (c) Complementary visualizations are rendered on a temporary render texture to accurately augment additional virtual objects on the aux display, as shown in (d).
• Figure 5: Examples of Submerse auxiliary (aux) display visualizations.