Item: ADVANCING ACCESSIBILITY IN AVALANCHE FORECASTING STRATEGIES FOR INTERPRETING GEOSPATIAL DATA
-
-
Title: ADVANCING ACCESSIBILITY IN AVALANCHE FORECASTING STRATEGIES FOR INTERPRETING GEOSPATIAL DATA
Proceedings: Proceedings, International Snow Science Workshop, Whistler, BC, Canada, 2026
Authors:
- Tyson Rettie [ Avalanche Canada, Revelstoke, BC, Canada ]
Date: 2026-09-28
Abstract: Building on prior work presented in Visually Impaired Forecasting and Accessible Software (Rettie, 2023), this paper examines advances in accessibility within avalanche forecasting tools. Public forecasts in Canada and Colorado are produced using the AvlD 2.0 platform, which introduced flexible forecast boundaries that can be redrawn as conditions evolve. An enhanced accessibility mode allowed a visually impaired forecaster to use all the features with a screen reader instead of the map-based interface used by sighted forecasters. Over the past three years, the techniques developed for the enhanced accessibility mode in AvlD 2.0 have been extended to tools supporting the diverse data sources used in avalanche hazard assessment, including weather stations, avalanche observations, metegram weather forecasts, and snowpack model dashboards. Each of these tools presents highly spatial or graphical data that are traditionally inaccessible to screen reader users. The primary strategy is to systematically convert spatial data into structured, navigable table formats that retain the relationships and context needed for hazard analysis. While this approach has proven effective for some complex datasets, its applicability is limited, and certain data is still not practical to interpret in this format. Key examples include highly visual data streams such as map-based synoptic weather products and avalanche photos, neither of which can be effectively represented in tabular form. This paper presents the current strategies for making geospatial data accessible to screen reader users and identifies key forecasting resources that remain difficult to navigate. It then explores potential solutions to address these accessibility gaps, including the use of AI to interpret and summarize primarily visual data in a text-based format.
Object ID: ISSW2026_P2.37.pdf
DOI: https://doi.org/10.15788/1790098986
Language of Article: English
Presenter(s): Tyson Rettie
Keywords: Screen Reader, Accessibility, Snowpack Models, Metteograms, Forecasting
Page Number(s): 1383 - 1390
-