Environmental Engineering Reference
In-Depth Information
VNIR imagery mentioned above were all used within an operational context during
the DWH incident.
Due to the large area of the spill, viewangle effects, and the limited size of the areas
imaged bymany of the satellites used, daily repeat satellite coverage of the entire spill
was sometimes incomplete. To partly address the limitation of individual satellite
passes, each daily estimate of the area time series used in this study corresponds
to a three-day average. The variability in this time series may be due to several
factors, including wind forcing, ocean dynamics, changes in the flow rate of oil from
the well, recovery efforts (e.g. skimming efforts), oil washing ashore, and lack of
measurements due to cloud coverage. A limitation of the MPSRs was the inability
to differentiate a thin sheen from very thick oil. Also variations in the wind speed
affect the interpretations of the satellite data. For example, high winds affect both
the true and apparent extent of oil coverage. High surface wind speeds over the slick
tended to decrease satellite-derived surface oil extent because wind and wave forcing
caused oil droplets to become increasingly dispersed into the surface mixed layer
and, consequently, become undetectable by satellite. Rough seas and deep convection
caused oil to be difficult to view in satellite imagery. In addition, rough seas and high
winds also enhanced natural dispersion, which resulted in apparent reduction of the
surface oil extent. On the other hand, during calm wind periods, relatively thin oil
sheens on the sea surface could be detected using satellite observations. Persistent
(three days or longer) limitations in satellite coverage or optimal viewing conditions
also affected the assessment of the extent of the detected surface oil, tending to create
an underreporting of the amount of oil coverage. In addition, the satellite techniques
presented here did not show onshore oil or, generally, oil in wetlands. Therefore, any
oil that washed ashore or moved into the wetlands did not appear in the MPSRs and,
thus, decreased the actual surface oil extent. At any rate, the detection of oil onshore
and in wetlands falls outside the scope of this chapter.
The final MPSR analysis product was an outline of an 'anomaly' presumed to
be oil. In fact, the surface oil varied substantially in type, thickness, concentration,
and percent coverage. Identification and prediction of the location of thicker oil is
of crucial importance to the response, as this oil may be recoverable via skimmers
or targeted for in situ burning, and also poses the greatest threat to shorelines. Thin-
ner and patchy relatively small oil slicks and lower concentrations of oil are also
important in evaluating possible effects on the ecosystem. False-positives were also
an issue, with transparent, presumably biological, sheens and patches of sargassum
frequently erroneously identified as oil. Therefore, satellite analysis was critical to
directing overflights, and overflight observations provided essential feedback for the
validation of the satellite analysis. Ultimately, satellite and overflight data were used
daily to initialize the surface oil distribution for modelling purposes by overlaying
analyses from individual satellite passes and observations from multiple overflight
tracks to create a time-dependent surface oil distribution.
The response to the DWH incident also demonstrated the benefits of combining
SAR and VNIR, especially the latter under sunglint conditions, when oil slicks
increase the specular reflectance of the sea surface, and the oil covered areas have
more brightness than the regular oil-free ocean surface, making it easier to extract
Search WWH ::
Custom Search