The paper "On the Slope-Aspect Correction of Multispectral Scanner Data" by P.M. Teillet, B. Guindon, and D.G. Goodenough examines the impact of topography on the radiometric properties of multispectral scanner (MSS) data, particularly in the context of remote sensing of forests in mountainous regions. The study focuses on two test areas in British Columbia, Canada: the Anderson River near Vancouver Island and Gun Lake near Bralorne. Both areas have diverse forest types and varying elevations, ranging from 275 to 1500 meters at Anderson River and from 670 to 1990 meters at Gun Lake. The authors formulate Lambertian and non-Lambertian illumination corrections, considering atmospheric effects and topographic variations. They determine terrain slope and aspect values from a digital elevation model and atmospheric parameters from a model atmosphere computation. The Lambertian approximation, which neglects sky irradiance and atmospheric path radiance, results in a cosine correction similar to those used for horizontal terrain. However, this extension to sloped terrain is found to be inadequate, especially for larger incidence angles. The study also explores semi-empirical functions based on cosines of incident and reflected illumination angles to remove the topographic effect. Correlations and linear regressions between topographic parameters and MSS radiance values are investigated for different forest types at each site. The analysis includes multitemporal Landsat MSS data at 50-meter resolution and 11-channel airborne MSS data at 10 and 50-meter resolutions. Slope-aspect correction algorithms are implemented in software at the Canada Centre for Remote Sensing, and geometric rectification is performed to relate image geometry to map coordinates. Feature selection based on divergence criteria shows that terrain parameters perform well in distinguishing between forest classes. However, maximum likelihood classification results for MSS data corrected for slope-aspect effects using various functions show little to no significant improvement over uncorrected data. The paper discusses these findings to better understand the physical principles and image processing methodologies involved.The paper "On the Slope-Aspect Correction of Multispectral Scanner Data" by P.M. Teillet, B. Guindon, and D.G. Goodenough examines the impact of topography on the radiometric properties of multispectral scanner (MSS) data, particularly in the context of remote sensing of forests in mountainous regions. The study focuses on two test areas in British Columbia, Canada: the Anderson River near Vancouver Island and Gun Lake near Bralorne. Both areas have diverse forest types and varying elevations, ranging from 275 to 1500 meters at Anderson River and from 670 to 1990 meters at Gun Lake. The authors formulate Lambertian and non-Lambertian illumination corrections, considering atmospheric effects and topographic variations. They determine terrain slope and aspect values from a digital elevation model and atmospheric parameters from a model atmosphere computation. The Lambertian approximation, which neglects sky irradiance and atmospheric path radiance, results in a cosine correction similar to those used for horizontal terrain. However, this extension to sloped terrain is found to be inadequate, especially for larger incidence angles. The study also explores semi-empirical functions based on cosines of incident and reflected illumination angles to remove the topographic effect. Correlations and linear regressions between topographic parameters and MSS radiance values are investigated for different forest types at each site. The analysis includes multitemporal Landsat MSS data at 50-meter resolution and 11-channel airborne MSS data at 10 and 50-meter resolutions. Slope-aspect correction algorithms are implemented in software at the Canada Centre for Remote Sensing, and geometric rectification is performed to relate image geometry to map coordinates. Feature selection based on divergence criteria shows that terrain parameters perform well in distinguishing between forest classes. However, maximum likelihood classification results for MSS data corrected for slope-aspect effects using various functions show little to no significant improvement over uncorrected data. The paper discusses these findings to better understand the physical principles and image processing methodologies involved.