Infrared Observations of Temperature Modulations on the Hudson River

The thermal boundary layer at the surface of a river is constantly disrupted and renewed by physical processes associated with convection, turbulence, wind stress, heat flux, and other environmental factors. These disruptions cause temperature modulations in the surface layer which can be measured with an infrared (IR) sensor. Over the course of two ten-day periods in August and November of 2010, we imaged the Hudson River from atop a nearby cliff using a large-format, mid-wave IR sensor. Time series imagery was collected for 5 to 10 minute periods, every 30 minutes for the entirety of each experiment. In the field of view, several in situ instruments were mounted to a steel piling driven into the river bed. Above and below the water surface, an array of instruments were installed to measure heat flux, wind speed, air and water temperature, current velocity, humidity, radiance, and conductivity. In this analysis, we investigate the relationship between the temperature modulations present in the IR imagery, which are associated with coherent features advecting with the mean flow, and the environmental parameters measured from our in situ instruments. The IR imagery from these experiments show a diverse range of temperature modulation patterns, on scales of 20cm to several tens of meters, often masked by the presence of surface waves. At low grazing angles, the IR images of the water surface are comprised of a combination of emitted radiance from temperature modulations on the surface and reflected radiance from the sky above. To separate out the emitted signal from the reflected signal, we employ a Fourier space filtering technique to exclude the variance in the imagery due to the surface waves. We find the remaining emitted signal to be correlated with wind speed and the air-water temperature difference, and weakly or uncorrelated with stratification and mean current speed. We report on both the signal processing technique used to extract the emitted signal from low grazing angle IR imagery and the observed relationships between these signals and environmental parameters. Approved for Public Release, Distribution Unlimited