Methodology |
Using remote sensing, water is easy to detect compared to other
features, such as soil and vegetation. Based on the absorption and
reflection properties, the blue wavelength can penetrate more deeply
in clear water than any other wavelength (Figure 1) and more likely
reflected back. That's the reason a water body looks blue as the
blue wavelength is strongly reflected back from the water surface.
There is an increase in absorption with the increase in wavelength
and almost total absorption in red and near-infrared wavelengths.
There are several satellite sensors specifically designed for monitoring of coastal and oceanic waters. The estimation of a certain water quality parameter using remote sensing, depends on the sensor’s characteristic sensitivity to certain wavelengths or spectral bands. In complex coastal regions, because the concentration of water constituents is spatially variable, conventional methods for studying water quality by point sampling are expensive, time consuming and spatially incomplete. Remote sensing is therefore important in providing a synoptic view for detailed retrieval over large regions.
Hong Kong has a complex marine environment due to terrestrial discharges from the Pearl River Delta in the west, urban pollutants in the centre and the clearer waters of the South China Sea in the east. Monitoring this complex near-shore environment requires retrieval of water quality parameters such as Chl-a and SS concentrations at high spatial and temporal resolutions.
Earth’s atmosphere, composed of different gasses e.g. Nitrogen (78%), Oxygen (21%), Argon (1%) and variable concentration of other minor components, has always been a great concern in remote sensing. The EMR signal received by the remote sensing sensor passes through the atmosphere twice i.e. from sun to target and from target to the sensor (Figure 3). During this process the signal strength is affected by absorption and scattering, and this atmospheric interference is high when target is a non-bright surface such as water.
Absorption occurs when the EMR intermingles with water vapors, carbon dioxide and some other atmospheric gases, this reduces the strength of the signal, hence it is called a multiplicative factor of the signal. Scattering is most likely to occur due to the interaction of gas molecules/aerosols with EMR and redirects the incident signal from its actual path, therefore it is called an additive factor in the signal strength. The atmospheric scattering, affects much the visible bands while has very small effect on NIR bands.
The spectral radiance measured by satellite sensors is affected by absorption and scattering by atmospheric particles. Therefore, to obtain constant and accurate ground signatures it is necessary to remove atmospheric artifacts using an efficient and reliable atmospheric correction method. Atmospheric correction is important when using ratio transformations e.g. the Normalized Difference Vegetation Index (NDVI) as well as applications dependent on subtle differences in the Surface Reflectance (SR) such as crop phenology, or the retrieval of WQPs. Numerous methods for atmospheric correction of satellite images have been presented. These fall into two types, namely physical methods including LOWTRAN (Low resolution atmospheric TRANsmission), MODTRAN (Moderate resolution atmospheric Transmission) and 6S (Second Simulation of the Satellite Signal in the Solar Spectrum), and image based methods including DOS (Dark Object Subtraction) and the ELM (Empirical Line Method). Physical methods use a Radiative Transfer Model (RTM) to estimate SR while image based methods obtain relevant parameters from the image.
Monitoring of Hong Kong's coastal waters is carried out by the Hong Kong EPD from a scientific vessel equipped with a differential global positioning system. The water samples are collected from 76 fixed monitoring stations from three depths, namely near the surface (1 m below surface), the middle layer and near the sea bed (1 m above sea floor). Samples are collected in a 500 ml Nalgene bottle and refrigerated for transport, then analyzed for extraction of Chl-a and SS concentrations. The Chl-a concentration is determined using an 'In-house' GL-OR-34 method based on the American Public Health Association (APHA) 10200H 2 spectro-photometric method and the SS concentration is determined using an 'In-house' GL-PH-23 method based on the APHA 2540D weighing method. The in-situ "Surface" data of Chl-a and SS concentrations for the dates coincident with Landsat TM, ETM+ and HJ-1 A/B CCD images were retrieved from the EPD water quality parameters database.
In order to capture the short term changes in coastal waters of Hong Kong and to develop a more rigorous model, a time window of ± 2 hours of the image acquisition time was used to identify collocated satellite and in-situ water samples. Water sampling locations affected by clouds, scan line errors on ETM+ images, and those with ship wake effects were not included. In considering the dynamic nature of water, the mean surface reflectance of each sampling station was extracted from a window of 3×3 pixels rather than a single pixel. Of these, 200 observations (from year 2000–2010) were used in regression model development and 40 observations (from year 2011–2012) were used for validation.
Nazeer M., Nichol J. E. (submitted), Characterization and delineation of optically different zones for the coastal waters of Hong Kong using collocated satellite and in-situ datasets, Remote Sens..