HÂþ»­

Ìý

B.Sc. (Honours) Thesis

(PDF - 90.7 Mb)

Satellite imagery can provide information on the spatial distribution of surficial
suspended sediment over broad scales in coastal environments. An outstanding challenge is to
determine the extent to which surficial sediment distributions can be linked to sediment
processes occurring near the seabed. Recent research indicates that dense suspended sediment
suspensions at the bottom of tidal channels off the southwest coast of South Korea limit upward
turbulent mixing of sediment to the sea surface.

The goal of this project is to determine whether this subsurface sediment process is
detectable with reflectance at the sea surface measured by the Landsat 8 satellite. The project’s
hypothesis is that the magnitude and variance of surficial reflectance will be lower over channels
than over ridges, due to the presence of dense suspensions in channels that limit vertical mixing.
On the ridges this process would not occur, because dense suspensions would flow into adjacent
channels under the influence of gravity. As a result, reflectance would be higher and more
variable at shallower depths. To assess this hypothesis, Acolite processing software was used to
perform atmospheric corrections on Landsat 8 images, and sea surface reflectance at 655 nm and
865 nm were used as proxies for suspended sediment concentration in a total of 15 cloud-free
images collected over the years 2013-2018.

Reflectance in both bands was extracted over a tidal channel and over an adjacent tidal
ridge using SeaDAS. The reflectance from these two points were assessed for statistical
correlation with depth and other environmental variables, including sea level, wind speed, and
recent precipitation levels. Results indicate that depth exerts primary control on mean and
standard deviation reflectance, consistent with the project’s hypothesis. Secondary controls on
reflectance are wind speed and sea level. I propose that higher windspeeds are associated with
larger reflectance due to resuspension over fringing tidal flats and that the correlation of higher
water levels with reduced reflectance is caused by sediment supply limitation.
In order to see if these methods can identify other areas with similar suspended sediment
processes, two secondary sites – located in Gyeonngi Bay, South Korea and the Gulf of
Khambhat in India have been analyzed. Preliminary results show similar primary and secondary
reflectance patterns. While further field work is needed to study the effects of tidally-dependant
suspended sediment stratification on surface sediment concentration and associated reflectance,
results indicate that remote sensing may be applicable to detection of subsurface suspended
sediment processes.

Keywords: South Korea, Heuksan Mud Belt, remote sensing, suspended sediment stratification, tidal seas.
Pages: 110
Supervisor: Paul Hill