Visual Analysis of Weather Ensembles

Source: DALL-E

Theoretical (Analytical):

Practical (Implementation):

Literature Work:

Overview

A weather ensemble is a set of computer simulations to estimate and predict future weather conditions.
Instead of simply running a single forecast, multiple simluations are carried out with slightly different inital starting conditions, producing a wide range of possible outcome scenarios.
The visualization and analysis of these ensembles allow meterologists to gain insights on weather uncertainty, and more accurately predict weather patterns and extreme weather events.

Problem Statement

Due to the large, multidimensional spatiotemporal datasets generated by these simulations, the visual exploration and analysis of these ensembles remains a challenging task.

Traditional methods of data analysis often fall short when dealing these large and complex datasets, which constitutes the need for further analysis within this field.

The goal of this project is to develop novel visualization techniques that enables domain experts to gain insights that were previously unobtainable.

Tasks

The following list comprises exemplary tasks that could be solved within the scope of this project. The development of own ideas aside from this list is also encouraged. 

  • Visualization Techniques: Which known visualizations can be employed to visualize spatiotemporal trends in ensemble visualizations?
  • Uncertainty Assessment: How can we quantify and visually communicate the uncertainty contained inside an ensemble simulation?
  • Input Parameter Sensitivity Analysis: What effects do small changes in the input parameters have on the outcome of the simulation analysis?
  • Pattern Extraction: How can patterns contained accross different ensemble members effectively identified and categorized?

Requirements

  • Interest in spatiotemporal data
  • Basic knowledge about information visualization, data mining and geographic information systems
  • Good programming skills in: JavaScript/TypeScript, D3.js, Python, PostGIS

 

Scope/Duration/Start

  • Scope: Bachelor/Master

Contact

References

  • Q. Shu, H. Guo, J. Liang, L. Che, J. Liu and X. Yuan, "EnsembleGraph: Interactive visual analysis of spatiotemporal behaviors in ensemble simulation data", IEEE PacificVis, 2016, doi: 10.1109/PACIFICVIS.2016.7465251
  • J. Wang, S. Hazarika, C. Li and H. -W. Shen, "Visualization and Visual Analysis of Ensemble Data: A Survey", IEEE TVCG, 2019, doi: 10.1109/TVCG.2018.2853721
  • M. Zhang, L. Chen, Q. Li, X. Yuan and J. Yong, "Uncertainty-Oriented Ensemble Data Visualization and Exploration using Variable Spatial Spreading", IEEE TVCG, 2021, doi: 10.1109/TVCG.2020.3030377