Resources
A curated list of tutorials, courses, and software for working with mobility and geographic data in R and Python. Most of the FloodTraces pipeline and SitReps are built in R, but the challenges are language-agnostic — use whatever your team is most comfortable with.
Mobility data
These resources deal directly with the kind of digital trace and mobility data used in the hackathon.
Using Digital Footprint Data to Measure and Monitor Human Mobility — a hands-on R workshop by Francisco Rowe and colleagues covering Meta–Facebook mobility data: temporal patterns, spatial patterns, and the practicalities of working with the data. The most directly relevant starting point for the hackathon.
mobility— an R package from the COVID-19 Mobility Data Network for parsing travel data, fitting mobility models via Bayesian inference, and simulating mobility patterns. Useful for teams tackling the See It Coming challenge.debiasR— an R package from the DEBIAS project (Francisco Rowe and Carmen Cabrera, University of Liverpool) for assessing and correcting population representation bias in digital trace data. It measures coverage and representativeness bias, adjusts biased origin–destination flows using deterministic and Bayesian multilevel methods, and validates adjusted flows against benchmark data. Directly relevant to the kind of mobility data used in the hackathon.
Cortex Code (CoCo) AI tooling
Explore the project with Cortex Code (CoCo)
Our partners at Snowflake have set up a custom environment for event participants, giving access to their AI coding tool Cortex Code (CoCo). Note that the FloodTraces datasets are not hosted here — this environment provides the CoCo tool itself. It is being made available specially for the hackathon and goes beyond what standard free-trial accounts allow
There are two ways to use it: a command-line tool and the CoCo desktop application — similar to VS Code, but with an AI coding assistant built in.
🔗 Explore the FloodTraces hackathon environment →
This will be introduced on Day 1, after Fawad’s presentation — feel free to start exploring from then.
Geographic Data Science (R & Python)
A course in Geographic Data Science — the ENVS363/563 module from the GDS Lab, with parallel R and Python labs covering spatial data, mapping vector and raster data, points, spatial weights, ESDA, and clustering. A solid foundation if you need to brush up on core methods.
Advanced Geovisualisation — the ENVS456 module (web version), focused on Python-based geovisualisation. Particularly relevant to the Make It Interactive challenge.
Spatial Modelling for Data Scientists — an R-based course from the GDS Lab (Francisco Rowe and Dani Arribas-Bel) covering spatial data wrangling, point pattern analysis, spatial interaction (flow) modelling, spatial econometrics, multilevel modelling, geographically weighted regression, and spatio-temporal analysis. Useful for teams building statistical models of displacement.
Python
- Mapping and Data Visualization with Python — an intermediate full course by Ujaval Gandhi (Spatial Thoughts) covering Matplotlib, GeoPandas, choropleths, basemaps, XArray, interactive maps with Folium and Leafmap, and building dashboards with Streamlit. All notebooks run in Google Colab — no local setup required.
Know a good resource? Add it here.
This page is community-maintained. If you have a tutorial, package, or course that would help other teams, please contribute it: fork the repository, add your resource to this page, and open a pull request. We’ll review and merge useful additions.