The Climate Intelligence Ecosystem’s "as a Service" features support communities of data stewards and users alike and are flexible in meeting the common data challenges across weather, environmental, energy, and earth science domains. The Ecosystem enables climate data-driven policy decisions, accelerating insights using cloud-based Artificial Intelligence/Machine Learning (AI/ML) and advanced analytics. The Ecosystem’s flexible open-source and cloud-based services are built on open architecture with the ability to integrate across multiple cloud platforms. The Ecosystem provides an on-demand data science workbench capability for analytics and modeling with no environment or infrastructure setup required for users. The Ecosystem’s data catalog can be loaded with high-quality environmental and population data from authoritative sources like NOAA, USGS, FEMA, IPCC, and others. A secure API layer offers the ability to feed external capabilities, such as Digital Twins or commercial visualization tools like Esri, Tableau, and Qlik. The Ecosystem not only optimizes the data value chain, but also enables actionable information for a variety of climate adaptation and resilience services.
A few examples of the application of the Climate Intelligence Ecosystem include 1) creation of a sub-surface water intelligence projection pilot; 2) use of EPA and other data, and Digital Twin technology to showcase air quality heatmaps and scenario planning by introduction of Electric Vehicles (EVs), and more. The same design principles are used for other projects including a recent capability to visually explore data via geospatial services serving over 100,000 unique Earth science users with over 50 million data requests in a year.
For the water intelligence pilot in particular, the project ingested, transformed, and stored USGS groundwater data, Coupled Model Intercomparison Project Phase 6 (CMIP6), and other datasets in the cloud using a common schema. Data science techniques, such as linear regression, Support Vector Machine (SVM), random forest, and gradient boosting models, were performed to merge the datasets and train ML models to ultimately project future subsurface water depths useful to water managers and long-term community master planners.
Communities, individuals, and businesses are in a race against time to make better climate-informed decisions as climate changes continue to accelerate. Data and technologies are often disconnected from the mission owner or decision maker. With the Climate Intelligence Ecosystem’s cloud-based and AI/ML-fueled platform, Booz Allen has created the conduit, from data management to climate analytics, to more rapidly link data and technology to a variety of mission outcomes.
In this presentation, we will discuss the framework and approaches behind the Climate Intelligence Ecosystem and share insights and lessons learned from recent research and prototypes stemming from our work and how they are beginning to accelerate the Climate Services value chain.

