CASCADE Research Projects

This research advances Engineering With Nature approaches by integrating natural systems into traditional built infrastructure to improve resilience of water systems across the western United States. The project broadens the definition of natural infrastructure to include diverse natural assets such as wetlands, soils, and forest ecosystems, and develops tools to analyze how combinations of natural and built assets perform under stress. It aims to produce modeling and visualization platforms for scenario-based analysis that can help stakeholders plan portfolios of interoperable natural-built systems. The work aligns hydrologic science, data-driven analysis, and stakeholder engagement to support resilient and equitable water infrastructure planning.

The understanding and containment of epidemics involves many factors, such as how people behave and interact, how they move around, and the decisions made by governments and organizations to control the spread. Because of this complexity, building accurate and timely models to predict and manage outbreaks in near real time is challenging and time consuming. It also requires copious data, which are often unavailable when it is urgently needed. To address these challenges, this project develops PanAX, a new computational system to improve how we prepare for and respond to evolving epidemics. The core idea is to use existing data and models in smarter ways and based on situational awareness, so we do not have to start from scratch every time.

CausalBench develops a transparent, fair, and easy-to-use causal benchmarking platform, aiming to (a) enable the advancement of research in causal learning by facilitating scientific collaboration in novel algorithms, datasets, and metrics and (b) promote scientific objectivity, reproducibility, fairness, and awareness of bias in causal learning research. CausalBench provides services for downloading and exploring benchmarking data, algorithms, models, metrics, and benchmark results.

This project develops novel database system architectures that embed strong privacy protections into machine learning workflows. The research focuses on creating systems where sensitive information can be processed and analyzed without exposing raw data. By integrating privacy mechanisms at the database level, the project aims to support secure analytics while maintaining performance. This effort bridges database design, security, and machine-learning requirements to protect data in real-world applications.

APPEX is a multi-disciplinary NSF center focused on understanding how localized outbreaks evolve into global pandemics through socio-biological and environmental data integration. The center develops predictive models that include biological, social, and information systems to forecast pandemic growth and support decision-making. APPEX engages with public health and policy communities to translate research insights into actionable strategies for anticipation and mitigation. This center also emphasizes education and collaboration across fields to strengthen pandemic preparedness infrastructure.

This international project studies AI-enabled control and optimization of district heat pump systems to accelerate equitable and resilient building decarbonization. The project investigates the use of artificial intelligence to optimize district-scale heat pump systems, which are key to reducing carbon emissions in building energy systems. The research combines systems modeling with real-world deployments to improve energy efficiency and decarbonize heating infrastructure. By embedding AI control and optimization, the project aims to enhance operational performance under varying demand and environmental conditions. The work contributes to scalable decarbonization strategies across urban and community energy systems.

This NSF CAREER award supports research into redesigning analytics database systems to better support the lifecycle of machine learning model serving. Traditional database systems are optimized for queries, but not for serving ML models at scale; this project seeks to bridge that gap by developing architectural extensions that treat ML models as first-class data and queryable objects. The research aims to unify data management, model execution, and performance optimizations to improve latency, scalability, and integration with broader data ecosystems. An important component is engaging students and developing educational materials that span databases and AI systems.

This project develops graph-based representations that fuse heterogeneous data streams to improve inference and reasoning in complex systems.

DISCOVER develops decision intelligence tools for causal exploration and hypothesis verification using interactive analytics.

SWSIE is an NSF Engines initiative that accelerates sustainability-driven innovation in the southwestern United States by connecting research with industry, government, and community partners. The program funds interdisciplinary teams tackling challenges in clean energy, water sustainability, climate adaptation, and economic development. By fostering partnerships, SWSIE supports translational research that moves technologies from concept to regional impact, while building workforce capabilities. The engine emphasizes inclusive innovation strategies to strengthen regional resilience and competitiveness.