Radiometric modeling of mechanical draft cooling towers to assist in the extraction of their absolute temperature from remote thermal imagery
Determining the internal temperature of a mechanical draft cooling tower (MDCT) from remotely-sensed thermal imagery is important for many applications that provide input to energy-related process models. The problem of determining the temperature of a MDCT is unique due to the geometry of the tower and due to the exhausted water vapor plume. The radiance leaving the tower is dependent on the optical and thermal properties of the tower materials (i.e., emissivity, BRDF, temperature, etc.) and
... rature, etc.) and also the internal geometry of the tower. The tower radiance is then propagated through the exhaust plume and through the atmosphere to arrive at the sensor. The expelled effluent from the tower consists of a warm plume with a higher water vapor concentration than the ambient atmosphere. Given that a thermal image has been atmospherically compensated, the remaining sources of error in extracted tower temperature due to the exhausted plume and the tower geometry must be accounted for. A temperature correction factor due to these error sources will be derived through the use of three-dimensional radiometric modeling. A range of values for each important parameter are modeled to create a target space (i.e., look-up table) that predicts the internal MDCT temperature for every combination of parameter values. This LUT, along with user knowledge of the scene, provides a means to convert the imagederived apparent temperature into the estimated absolute temperature of a MDCT. Preliminary results indicate that temperature error corrections of approximately 1 -9 Kelvin can be achieved with the range of MDCT parameters encompassed by the LUT.