Simulating Fine-Scale Marine Pollution Plumes for Autonomous Robotic Environmental Monitoring.

Marine plumes exhibit characteristics such as intermittency, sinuous structure, shape and flow field coherency, and a time varying concentration profile. Due to the lack of experimental quantification of these characteristics for marine plumes, existing work often assumes marine plumes exhibit behavior similar to aerial plumes and are commonly modeled by filament based Lagrangian models. Our previous field experiments with Rhodamine dye plumes at Makai Research Pier at Oahu, Hawaii revealed that marine plumes show similar characteristics to aerial plumes qualitatively, but quantitatively they are disparate. Based on the field data collected, this paper presents a calibrated Eulerian plume model that exhibits the qualitative and quantitative characteristics exhibited by experimentally generated marine plumes. We propose a modified model with an intermittent source, and implement it in a Robot Operating System (ROS) based simulator. Concentration time series of stationary sampling points and dynamic sampling points across cross-sections and plume fronts are collected and analyzed for statistical parameters of the simulated plume. These parameters are then compared with statistical parameters from experimentally generated plumes. The comparison validates that the simulated plumes exhibit fine-scale qualitative and quantitative characteristics similar to experimental plumes. The ROS plume simulator facilitates future evaluations of environmental monitoring strategies by marine robots, and is made available for community use.