Regional-scale photochemical modeling system is widely used for simulating the tropospheric ozone problem. A regional scale modeling exercise is currently underway for the near non-attainment areas in Texas to simulate the high ozone episode of September 13-21, 1999. A nested grid system was developed for the large model domain with finer grids encompassing the major urban areas in Texas. This modeling system uses emissions of ozone precursors (such as NOx and VOCs) from various types of sources (point, area, mobile and biogenic) as inputs. Except for point source emissions, all other emissions should be allocated into proper grids overlaying the modeling region. Area and non-road source emission are important inputs to photochemical air quality models. Since most area and non-road source emissions are calculated at the county-level, these emissions should be geographically allocated to the computational grid cells of the model prior to running the model. In allocating the area and non-road source emissions, usually the emissions are distributed over the county equally rather than apportioning the emissions using the appropriate surrogates and allocating it on a sub-county grid system. This paper focuses on the spatial allocation of emissions from the non-road and area sources in Corpus Christi urban airshed. This airshed comprised of two counties, namely Nueces and San Patricio. Spatial allocations were performed using Geographical Information System (GIS) software -ArcView and ArcInfo. A 4 x 4 km grid system was developed in Lambert-Conformal conic projection scheme and overlaid on the two counties for this purpose. Proper emission surrogates were identified and developed for allocation of these sources into proper grids. Based on the spatial distribution ratios of these different surrogates within the grid system covering the two counties, emissions from corresponding sources were allocated in the grids. Spatial surrogate data were acquired in the form of GIS shapefiles from various federal, state, and local agencies. Though United States Environment Protection Agency (US EPA) suggests default spatial surrogates such as population for specific emission sources, we have used additional surrogates such as housing and water bodies that better represented the corresponding emission sources. Since the use of GIS facilitates spatial visualization of emissions, allocation of emissions using GIS software provided for the adequate quality assurance of spatial texture of the data. Results from this GIS-based study is presented in detail in this paper.