Examining Change and Long-Term Trends in the Marine Environment Using Satellite-Based Time Series
Abstract- The oceans and coastal areas are dynamic environments in which variability occurs at a wide range of temporal scales, from seconds to years to decades and longer. Some very good time series now exist at specific locations, that permit characterization of this variability as well as of long...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Text |
| Language: | English |
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.158.7998 http://www.aslenv.com/reports/RemoteSensing-Oceans2009.pdf |
| Summary: | Abstract- The oceans and coastal areas are dynamic environments in which variability occurs at a wide range of temporal scales, from seconds to years to decades and longer. Some very good time series now exist at specific locations, that permit characterization of this variability as well as of longer-term trends, but for much of the world ocean the in situ data is sparse and such characterization is not possible, or possible only by extrapolation. For these less well-studied areas, satellite imagery and gridded products created from a combination of satellite and in situ data are the only available sources of continuous historical information. Satellite imagery provides regular, spatially synoptic, global information, at spatial resolutions on the order of 1km and temporal resolutions typically measured in days. Some sensors offer higher spatial resolutions but at lower temporal resolution. With accumulated time series of up to 30 years or longer for some satellites and other gridded datasets, we can now begin to use this technology to identify and track long-term change, as well as to characterize shorter-term variability. In this paper we will discuss the analysis of long-term change and trends in marine and coastal environments using examples from recent projects. We will illustrate vegetation changes in and around the Anderson River delta on the Beaufort Sea coast of the Arctic Ocean using 30m spatial resolution Landsat imagery acquired over a 31-year period from 1972 to 2003. Using a temporal classification approach, we map both interannual variability and long-term losses on the mudflats of the delta itself, and gains on the nearby tundra. In a second example, we illustrate how chlorophyll and sea surface temperature (SST) vary locally within the northeast Pacific, and how temporal patterns and trends also vary regionally, using 24 years of weekly 4km resolution sea surface temperature from the |
|---|