AWS Public Sector Blog

Leveraging the cloud for rapid climate risk assessments

Four Twenty Seven, an affiliate of Moody’s, is a publisher and provider of data, market intelligence, and analysis related to physical climate and environmental risks. They leverage scientific datasets to provide climate risk scores for real assets and listed securities, informing investors and corporations globally.

As part of the Amazon Sustainability Data Initiative, we invited Colin Gannon, Senior Data Analyst at Four Twenty Seven, to share how the organization is using open data and Amazon Web Services (AWS) to provide insights for climate risk assessments.

Climate Science for Financial Users

Four Twenty Seven builds tools and services that help bring climate data (sourced from government agencies and academic institutions) to public and private organizations so they can better understand their exposure to climate hazards and manage risk in their communities. Four Twenty Seven’s new on-demand scoring application allows users to enter an asset’s location and receive risk scores for each site in real-time.

To get the risk scores, we produce indicators of climate risk derived from the outputs of climate models and other environmental datasets. These values reflect assets’ exposure to impacts from flooding, hurricane-force wind, extreme heat, and drought. Indicators are then aggregated to create scores, which reflect risk from climate change. From manufacturing plants damaged in hurricanes to roads flooding during storms, these hazards impact businesses and investments, with implications on global economies.

Projected increase in cooling degree days (CDD), or the additional number of degrees above 65˚F experienced per year relative to a historical baseline, derived from downscaled global climate models. The lightest areas are those with the least increase in cooling degree days, while the darkest areas are those projected to experience the highest increase in cooling degree days. Source: Four Twenty Seven.

Projected increase in cooling degree days (CDD), or the additional number of degrees above 65˚F experienced per year relative to a historical baseline, derived from downscaled global climate models. The lightest areas are those with the least increase in cooling degree days, while the darkest areas are those projected to experience the highest increase in cooling degree days. Source: Four Twenty Seven.

Overcoming Data Processing Challenges

Providing location-specific risk assessments requires accessing and processing the best climate data available. Climate data poses processing challenges due to the raw file size of climate model outputs, where a single file can be hundreds of megabytes or more, and an entire dataset can be anywhere from tens of terabytes to multiple petabytes.

Likewise, many environmental and scientific datasets impose considerable storage and processing hurdles as a result of their structure and size. To provide on-demand assessments for any location in the world, Four Twenty Seven needs rapid access to these datasets and a cloud infrastructure to process this data and deliver on-demand results. AWS provides the cloud infrastructure needed for the storage and processing of our services.

We also use the AWS Public Dataset Program to access many of the vital datasets we aggregate to create climate risk scores. These public datasets are hosted in Amazon Simple Storage Service (Amazon S3) buckets, eliminating the cost of data acquisition and storage. AWS also supplies the infrastructure needed to build up the APIs that interact with these datasets through Amazon Elastic Compute Cloud (Amazon EC2), Amazon Elastic Container Service (Amazon ECS), and Amazon Simple Queue Service (Amazon SQS).

Using AWS allows us to quickly build and deploy climate data services, at limited cost. By invoking services on-demand, we can develop and test in a minimal environment before scaling to production. Four Twenty Seven’s current deployment uses the managed container service, Amazon ECS, to run Docker containers services in isolated, easily replicable environments. Each climate data service, such as a module for evaluating sea level rise, has its own Amazon ECS cluster, running on a combination of Amazon Fargate and Amazon EC2 launch types, which both offer a high level of scalability and reliability. These, with Amazon SQS, allow us to build an efficient microservices framework, which underpins the delivery of climate data in our services. This flexible framework also allows us to continuously integrate the latest datasets and develop new climate risk metrics as climate science advances.

By using AWS, Four Twenty Seven can process, store, aggregate, and translate climate and environmental datasets into a format that is meaningful, timely, and actionable for users.

Understanding risk to build resilience

Scaling the Solution

Exposure to heat stress. The map shows the risk score of a subset of corporate facilities of publicly listed companies in Four Twenty Seven’s global database colored based on their exposure to extreme heat. Source: Four Twenty Seven.

Exposure to heat stress. The map shows the risk score of a subset of corporate facilities of publicly listed companies in Four Twenty Seven’s global database colored based on their exposure to extreme heat. Source: Four Twenty Seven.

Fortune 500 companies use our data to identify risks in their corporate facilities and supply chains, which can guide their resilience-building efforts internally and with suppliers. Also, real estate investors and municipal bond investors can use this asset-level understanding of climate risk to guide conversations with asset owners and bond issuers around their risks and ways to mitigate that risk.

By providing businesses and investors with granular information on climate risk, this data empowers them to collaborate and make forward-looking decisions that prepare their assets, and the communities they rely upon, for the physical impacts of climate change.

Learn more about the Amazon Sustainability Data Initiative and check out the AWS Public Dataset Program.

Colin Gannon

Colin Gannon

Colin Gannon is a Senior Data Analyst at Four Twenty Seven where he assesses clients’ specific climate risk over space and time, helping to build products which integrate the most recent climate change datasets and research into a user-friendly interface. He has a background in applied climate science, using computational statistics, geospatial analysis, and business intelligence applications to process, analyze and communicate climate data.