Signal Processing Enhances Environmental Sensing

S ensors and other data sources, combined with sophisticated signal processing techniques, promise to help scientists better observe and analyze various types of environmental data. Biologist Nathan Merchant, for example , has created a method for tracking ships and monitoring underwater noise levels in protected marine mammal habitats. Merchant, senior scientist for underwater noise at the U.K. Centre for Environment , Fisheries, and Aquaculture Science (CEFAS), developed the system with coresearchers " Underwater noise levels have risen significantly over time in step with human activity, " Merchant says. These changes in the acoustic environment affect marine mammals, which rely on sound as their primary sensory mode. " The disturbance caused by man-made noise can disrupt crucial activities, such as hunting for food, affecting the animals' health. " To help understand the impact noise might exert on dolphins and their population levels, the researchers conducted a study on Moray Firth, Scotland's largest inlet (Figure 1). Moray Firth is home to a population of bottlenose dolphins as well as numerous types of seals, porpoises, and whales. The protected habitat also hosts construction yards that supply Scotland's rapidly expanding offshore wind farm industry. Projected increases in wind farm construction are expected to drive more shipping through the habitat—something many scientists believe could eventually negatively impact resident marine mammals. " Various types of ships emit noise at different levels and frequencies, therefore it is vital to know which kinds of vessels are crossing the habitats and migration routes of marine mammals, " Merchant says. Merchant and his fellow researchers recently monitored underwater noise levels using hydrophone sensors (underwater microphones), ship-tracking data, and shore-based time-lapse photography. The techniques created a ship-noise assessment toolkit. " In this project, we used signal processing techniques to integrate several different data sources into one package: time-lapse video, underwater sound recordings , and ship-tracking data, " Merchant says. Using signal processing techniques in combination with video editing software, the researchers combined video, audio, and spatial data to produce a synchronized audiovisual representation of the soundscape, including shipping activity and weather conditions across the marine mammal habitat. The main challenge the researchers faced was processing each data source in a way that would supply a common time resolution for the audiovisual output. " We used geolocation tools to map the ship tracking data through time, which involved temporal and spatial interpolation of the raw data, which has a fairly coarse, ~10 …