Predictive Systems for Disaster Resilience

Setup

A lot of the data that could warn us about disasters already exists. Population movement shows up in call detail records, environmental stress in satellite imagery, vulnerability in ground surveys. The catch is that it lives in different formats, resolutions, and institutions, and so do the researchers working on it.

Cornell’s “Data Science to Build Resilience and Improve Humanitarian Response” Thought Summit brings that research into one room, along with the funders and field partners who could act on it. The output is a white paper, and I’m one of the people writing it.

The work

Reading the research submitted to the summit (CDR feature construction, satellite imagery pipelines, multi-modal data fusion) and synthesizing it into one argument about where predictive disaster-resilience systems stand and what they need next. I also engage the summit’s funders and partners directly.