ProWis: A Visual Approach for Building, Managing, and Analyzing Weather Simulation Ensembles at Runtime

TL;DR

ProWis tackles the complexity of building, running, and analyzing weather simulation ensembles by providing a provenance-aware visual analytics platform that orchestrates WRF workflows at runtime. The backend uses a MonetDB-based provenance store and Apache Airflow to manage directed-acyclic graph workflows, while the frontend offers Setup, Simulation, and Ensemble views with domain brushing, PCA-based clustering, sunburst time aggregations, and heat matrices for ensemble analysis. Two Brazilian rainfall case studies demonstrate the system's ability to rapidly assemble ensembles, monitor ongoing runs, and compare scenarios in space and time, revealing underpredictions and ICBC/parameterization sensitivities. The work shows that integrating provenance, automated run management, and rich visual analytics can significantly reduce manual effort, improve reproducibility, and enhance decision-support in operational weather forecasting.

Abstract

Weather forecasting is essential for decision-making and is usually performed using numerical modeling. Numerical weather models, in turn, are complex tools that require specialized training and laborious setup and are challenging even for weather experts. Moreover, weather simulations are data-intensive computations and may take hours to days to complete. When the simulation is finished, the experts face challenges analyzing its outputs, a large mass of spatiotemporal and multivariate data. From the simulation setup to the analysis of results, working with weather simulations involves several manual and error-prone steps. The complexity of the problem increases exponentially when the experts must deal with ensembles of simulations, a frequent task in their daily duties. To tackle these challenges, we propose ProWis: an interactive and provenance-oriented system to help weather experts build, manage, and analyze simulation ensembles at runtime. Our system follows a human-in-the-loop approach to enable the exploration of multiple atmospheric variables and weather scenarios. ProWis was built in close collaboration with weather experts, and we demonstrate its effectiveness by presenting two case studies of rainfall events in Brazil.

Figures (8)

• Figure 1: WPS and PRC sub-workflows. Geogrid, Ungrib, and Metgrid process terrain, ICBC, and meteorological data. Real and WRF consume the WPS output and perform the simulation.
• Figure 2: Overview of the ProWis components.
• Figure 3: ProWis's DAGs and tasks.
• Figure 4: (a) The Setup view lets users configure WRF simulations interactively. (b--d) Interactions used to set up domains.
• Figure 5: (a) Simulation view. (b) Example of the runs overview graph (middle) from the São Paulo case study (Section \ref{['subsec:case2']}). It shows six runs, 5 completed and 1 in progress. By hovering node 2, the run's information is shown. The user can interact with a run by clicking on its node. (c) Scatter plot from the same case study colored according to the ICBC source. Runs with the same ICBC generated similar precipitation results; (d) Scatter plot of the Maricá case study (Section \ref{['subsec:case1']}) colored by the cumulus physical process and showing no precipitation pattern.