Research project S0/20/075 (Research action S0)
Context and objectives
The Antwerp port is the home to the largest petrochemical complex in Europe and worldwide second only to Houston, Texas. Four oil refineries jointly occupy 175 ha of port area, offering a total refining capacity of close to 36 million tons and volume production of 16 million tones of chemicals per year. The development of the chemical products results in waste plume of different air pollutants to the urbane area closed by, and causes increase in respiratory illnesses to its population.
Thermal Imaging Spectrometry (TIS) shows promise dual-use in a wide range of military, governmental, and private industry applications such as identification and quantisation of emissions from hazardous waste sites. By using remote infrared detection techniques most species, except homonuclear diatomic species, can be detected and quantified due to their unique infrared spectral properties in the wavelength region 3-13 µm. Gases that absorb infrared radiation in this atmospheric window are affecting the sun-earth radiative balance, which in turn is affecting the temperature on earth. Hence, TIR imaging spectroscopy for pollutant detection is found to be more useful than VIS-SWIR spectroscopy since the background, in the thermal, corresponds strongly with temperature and therefore valuable in the endmemeber decomposition, in which a quantity rather than a probability is of interest.
The objective of this project is to detect the presence and the concentration of polluted gas compounds in the atmosphere using the AHS-160 airborne spectrometer over the chemical industry situated in the port of Antwerp.
Methodology
• In June 2005, AHS-160 flight campaign acquired VIS to LWIR data in two operational periods during the same day (morning and afternoon). The airborne data were calibrated and verified using numerous ground truth measurements that were collected using the field thermal imaging reflectometer (SOC 400T), AHS spectrometer and traditional in-situ measurements collected in AQMSs situated in the port.
• Different spectra match-filters were applied for the detection of the thin plume in the VNIR bands over homogeneous background as water and grass.
• Orthogonal background technique applied to the LWIR bands was used to separate the thermal radiance contrast of the plume over the heterorganic background of the industrial area. After the detection of the gas plume in the scene, Linear Unmixing technique was used to calculate how much the plume pixel mixed with pure SO2.
• Gaussian plume model was applied to predict the SO2 concentration, released at a specific source point.
Results
MWIR-LWIR:
• The AHS-160 bandwidths are not sufficient for detection of gas compound based on their spectra abundance;
• Using the CIBR ratio, it is possible to detect pollutant plume in the LWIR wavelengths;
• The LWIR information is complimentary to the VNIR data (detection over heterorganic background).
VNIR:
• It is possible to detect pollutant plume using CIBR in the VNIR wavelengths over homogeny background (no mix pixel).
