Satellite detection of complex aerosols, in particular, Oil smoke over water and dust over land, is generally difficult. On 1 March 1991, a smoke plume generated by burning Kuwaiti oil wells and a dust storm over the southwestern Arabian Peninsula, provide the opportunity to study both of these effects. Utilizing NOAA-11 AVHRR data, a two dimensional scatter plot analysis technique was employed to determine and classify the radiative signatures of the smoke and dust. A two dimensional mask routine was then used to assess the reliability of the scatter plot analyses and spatially display the results. A channel 1 to channel 2 radiance ratio and a channel 4 brightness temperature combination provided the best separation of the smoke signature from water. The dust plume was unambiguously represented by a channel 5 brightness temperature minus channel 4 brightness temperature image and a channel 4 brightness temperature combination. Together the 2D scatter plot technique and 2D mask form the groundwork for a possible detection algorithm.