Data-Smart City Pod: How Cities Can Use Data to Transform Disaster Response

Date: September 17, 2025
Host: Stephen Goldsmith (Professor of Urban Policy, Bloomberg Center for Cities, Harvard University)
Guest: Dr. Andrew Schroeder (VP of Research and Analysis, Direct Relief; Co-Founder, Crisis Ready)

Episode Overview

This episode explores how cities can leverage data and technology to dramatically improve disaster response and preparedness. Dr. Andrew Schroeder shares lessons from his work at Direct Relief and Crisis Ready, touching on practical ways to break down data silos, integrate new tools, and use advanced analytics—like AI and satellite imagery—to direct resources, anticipate crises, and protect vulnerable communities. The discussion is both pragmatic and forward-looking, calling for intentionality in public sector data architecture, cross-sector partnerships, and the elevation of data roles in city government.

Key Discussion Points & Insights

1. The Challenge: Fragmented Data and Undefined Roles

[01:25–04:26]

• Disaster and crisis response is heavily reliant on integrated data from multiple sources, but data is often siloed across agencies (health, environment, emergency management), and no single entity is responsible for “mainstreaming” innovative data approaches.
  • "Often that is nobody's job to introduce new data and methods into these kinds of circumstances." — Dr. Schroeder [02:09]
• Without intentional planning, cities lack the transparency, interoperability, and context needed for effective modeling and response.
  • "We're not designed for interoperability ... not designed for inputs from many different sources." — Dr. Schroeder [04:31]

2. Data Integration: Emerging Solutions

[04:26–06:20]

• The data stewardship movement addresses intentional data integration by creating roles (like Chief Data Officers) and frameworks for both public and private sector data sharing.
• Most crucial data is increasingly produced by private actors, complicating access due to misaligned incentives, legal hurdles, and privacy issues.
  • "The data that's most needed being produced at largest scale is not being produced by the public sector. It's being produced by private actors." — Dr. Schroeder [05:11]

3. Understanding and Modeling Crisis Risk

[06:20–09:53]

• Pre-existing vulnerabilities largely determine who suffers most in crises (e.g., wildfires, pollution spikes).
  • "The places most at risk to the flare up or the crisis ... are the places that were most at risk the day before." — Dr. Schroeder [07:39]
• Effective modeling requires granular and multi-source data (from satellites, sensors, healthcare, etc.) to understand both chronic exposures (like PM2.5, heat) and acute events.
• Health impacts from environmental risks (like particulates) are broader and more complex than previously acknowledged—beyond respiratory illness, affecting brain health, cardiovascular risk, etc.

4. Building Effective City Structures for Data-Driven Response

[09:53–13:32]

• Platforms, data, and people: Mayors must invest in all three for successful disaster response.
  • Use cloud-based solutions (e.g., Snowflake, Google BigQuery, Earth Engine) to integrate data.
  • Recruit data officers and scientists as core public sector staff, competing with private sector for talent.
• Successful strategies require collaboration with nonprofits, universities, and platforms to convene and provide technical and analytic capacity.
  • "Making sure that all of that is functioning is the new job of the mayor." — Dr. Schroeder [13:26]

5. Practical Examples and New Tools

[13:49–15:22]

• Crisis Ready’s "Climateverse" project in India and Mexico City helps cities access, understand, and use climate and environment data, including conversational AI to lower technical barriers.
  • "AI can help with some of that and can help you then understand what data is there to be able to solve that problem." — Dr. Schroeder [14:47]

6. Addressing Health/Environment Data Integration & Legal Hurdles

[15:22–18:15]

• Combining health and environment data for real-time response is possible, but legal frameworks and privacy controls must be tailored to the policy question (e.g., mitigation vs. medical response).
  • "We need to be able to have frameworks for ... data pipelines that are mindful of context and portable depending on the question." — Dr. Schroeder [17:03]
• Start with outcome-focused questions: What do you want to do with the data? This guides appropriate integration and sharing.

7. Case Study: Planning for Extreme Heat

[18:15–23:03]

• Preview of an upcoming Climate Week “tabletop” exercise that will simulate extreme heat response with leaders from multiple cities.
  • Will immerse participants in scenarios to coordinate roles (public health, crisis response, support for the unhoused, hospitals, etc.).
• Cities like Phoenix and Miami are pioneering heat response coordination—efforts include tracking exposures, resource mobilization, and integrating new data sources.
  • "Those coordination efforts are, I think, eye opening ... the joint effort required to mount an effective response." — Dr. Schroeder [21:28]
• Examples of using AI and computer vision: Drones for post-disaster damage assessment (e.g., Texas A&M); large-scale open-data mapping (e.g., Overture Maps in Hurricane Otis aftermath).

8. The Evolving Role of Data in Disaster Management

[23:03–23:54]

• Past emergency management focused on moving physical assets to crisis sites; future-focused management must also prioritize agile data movement for real-time understanding.
  • "They spent a lot of time figuring out how to move physical assets … Not so much time trying to figure out how to move data in order to understand the crisis." — Stephen Goldsmith [23:06]

Notable Quotes & Memorable Moments

• "Often that is nobody's job to introduce new data and methods into these kinds of circumstances."
  – Dr. Andrew Schroeder [02:09]

• "We're not designed for interoperability ... not designed for inputs from many different sources."
  – Dr. Andrew Schroeder [04:31]

• "The places most at risk to the flare up or the crisis ... are the places that were most at risk the day before."
  – Dr. Andrew Schroeder [07:39]

• "Making sure that all of that is functioning is the new job of the mayor."
  – Dr. Andrew Schroeder [13:26]

• "AI can help ... understand what data is there to be able to solve that problem."
  – Dr. Andrew Schroeder [14:47]

• "We need to be able to have frameworks for ... data pipelines that are mindful of context and portable depending on the question."
  – Dr. Andrew Schroeder [17:03]

• "Those coordination efforts are, I think, eye opening ... the joint effort required to mount an effective response."
  – Dr. Andrew Schroeder [21:28]

• "They spent a lot of time figuring out how to move physical assets ... Not so much time trying to figure out how to move data in order to understand the crisis."
  – Stephen Goldsmith [23:06]

Timestamps for Key Segments

• [01:25] - Introduction to Dr. Schroeder’s work in crisis data
• [04:26] - Data fragmentation and stewardship movement
• [07:38] - Granular modeling of crises and health impacts
• [10:33] - Structuring city leadership for data-driven response
• [13:49] - Practical city case: Crisis Ready’s Climateverse
• [15:22] - Integrating health with environmental data—legal and technical considerations
• [18:15] - Extreme heat scenario: Multicity simulation and real-time data use
• [21:45] - Innovations: AI, drones, large-scale map data for disaster assessment
• [23:03] - Shifting emergency management focus from physical to data movement

Summary: Key Takeaways for Listeners

• Effective disaster response requires intentional, cross-sector data integration and clear roles for data stewardship.
• Modern cloud platforms, AI tools, and participatory simulation exercises help cities operationalize data for prevention, mitigation, and real-time crisis management.
• Success hinges on combining platforms, data, and skilled people—supported by strong legal frameworks, outcome-oriented questions, and partnerships with nonprofits, universities, and the private sector.
• Paradigms are shifting from solely moving assets to centering real-time data movement—to save lives and build resilient cities.