DDDAS-TMRP: Collaborative Research: Adaptive Data-Driven Sensor Configuration, Modeling and Deployment for Oil, Chemical and Biological Contamination Near Costal Facilities

An interdisciplinary team of 9 highly qualified researchers from four institutions will conduct collaborative research in measurement methods, mathematical algorithms for DDDAS, and system software integration focused on developing DDDAS to greatly improve monitoring and modeling of contamination transport in large bodies of water. The outcome of the proposed project will be the development of a programmable, networked portable low cost mil-spec sensor based system using DDDAS in extreme aqueous environments. Intellectual Merit of the Proposed Activity. We are developing a variable light wave sensor array that we will integrate into an ocean observational system. This system will be superior to most near coastal ocean models, which are typically wind driven but not contamination transport driven, in that our new model will be both. We will accomplish this through the dynamic injection of observed ocean data into multiscale mathematical models and computer simulations. Our project will create research topics in multiscale mathematics, statistics, and software application integration with a flexible, Grid-based database and problem solving environment. We will develop hardware and software with which we will perform both lab and ocean tests of the project. Both academic and industrial partners will be involved. The research in this project will be extendable to other application environments. Broader Impacts Resulting from the Proposed Activity. The Exxon Valdez oil spill, at its time the largest oil spill ever, no longer ranks among even the top 50 largest oil spills globally. Oil spills are still among the biggest potential threats to coastal water and water supplies. Our data-driven simulation will also be useful in proposing countermeasures to general biochemical and chemical contaminants, including those that could be used in a terrorist attack through coastal waters. Our data-driven real time methodology will reduce threats to daily life, especially in areas where water supply critically depends on seawater desalination, and it will help automate remediation plans and early warning systems. We will also positively impact both the fishing and coastal tourism industries (and a cascading list of businesses that rely on both) by forecasting calamities from algae blooms (red tide) to spilled chemicals. Methodology. The project will follow an integrated approach that addresses technical issues at each step of the process: I) the dynamic simulation instructs the sensors what to look for and reprograms it for those analytes, 2) the sensors report to the simulation the new observed data, and 3) the simulation then incorporates the new data, updates its predictions, and reprograms the sensors as necessary in a closed loop. We will reduce the amount of human intervention needed to monitor spills and other contamination events, making DDDAS viable for sensors going to locations that are difficult to communicate with the sensors in real-time (e.g., an unreliable satellite link or another planetary body in the future). The work will build on the successful results of research previously funded by the NSF, including the SURA Coastal Ocean Observation and Prediction and two ITR projects to develop algorithms, error controls, and middleware to optimally manage provably scalable computing resources for Grid computing. Expected Mutual Benefit for Participants. We have expertise in DDDAS, Grid computing, multiscale methods, sensors, middleware, user interfaces, and problem solving environments, all of which are essential to the success of the project. The project will build on and expand existing collaborations, which are expected to continue beyond the funding period of the project. Exchanges of graduate students and postdoctoral scientists among institutions during the course of the project will further strengthen collaborations and will give students a broader research experience. Participating institutions have a history of working with minority and disadvantaged persons and will build on existing programs to recruit a diverse group of researchers to the project. TPI6363666