|HOME SCHEDULE AUTHOR INDEX SUBJECT INDEX|
High spatial resolution sea surface climatology from Landsat thermal infrared data.
Fisher, Jeremy*,1, Mustard, John1, 1 Brown University, Providence, Rhode Island
ABSTRACT- The annual cycle of sea surface temperature (SST climatology) documents processes important for estuarine ecology. However, detailed thermal spatial patterns in coastal areas are generally more complex than can be modeled or inferred from in situ measurements, and global climatologies derived from satellite data are too coarse to resolve estuarine processes. We derived high spatial resolution sea surface climatologies for southeastern New England, including Narragansett Bay, using an extensive database of thermal measurements from Landsat Thematic Mapper sensors. Average daily temperatures of water bodies across a 28,800 km2 region were mapped at 60 meter spatial resolution using a curve fitting analysis. This method reveals that isolated water bodies warmed faster and to a higher temperature than deeper, well-mixed waters. High amplitude seasonal temperature changes tend to be well correlated with faster response times (earlier maximum and minimum temperatures), with the notable exception of Mount Hope Bay in NE Narragansett Bay, which is subjected to anthropogenic thermal input. Narragansett Bay is shown to have a mean annual temperature of 11.86 ± 0.41 °C, while the Mount Hope Bay system is consistently warmer at 12.30 ± 0.21 °C. Generally, ocean temperatures have extremes from 6 to 17 °C (winter to summer, respectively), while lakes on average range from -2 to 25 °C. This method provides a mechanism to study temperature change and effects over broad areas while still capturing high resolution details important at the estuarine scale. This climatology is a unique use of Landsat thermal infrared information, and provides a new view of coastal and estuarine scale dynamics from a multi-temporal viewpoint.
Key words: remote sensing, sea surface temperature, climatology, high resolution