2.2. GIS DEVELOPMENT METHODOLOGY The goal of the GIS development activity was to create GIS layers that were useful for risk assessments of oil field activities. The Russian and U.S. participants agreed on data layers (e.g., infrastructure, geology, vegetation) to meet this requirement. To support the production of the GIS layers, the data acquisition phase included obtaining and developing charts, maps, data tables (such as chemical and contaminant toxicity testing), civilian images, and NSS imagery of the study site (described in detail in section 4). In addition to archived data, recent imagery was obtained to document the present conditions and the high temporal rate of change in the study site. Following the data acquisition phase, each of the GIS layers was extracted from NSS imagery or from other unclassified data sources. Note that after the GIS information was extracted from NSS data, the GIS layer itself was unclassified even though the source of the GIS layer was still classified. The GIS extraction methodology followed the usual system for GIS data input, including some additional safeguards for the NSS data. From the outset it was intended that NSS data augment other data sources, including civilian satellite data, unclassified maps, research reports, and relevant data compilations. Because of differences in the U.S. and Russian data, it was anticipated that there would be differences in the GIS layers derived from NSS data. These differences were due to temporal differences of U.S. and Russian NSS data, both in year and season; resolution differences between NSS imagery of each country; and interpretation and digitization differences while extracting the GIS layers. The reconciliation of differing GIS layers was done at a joint U.S.-Russian meeting at which paper maps of each side’s GIS layers were compared side by side and overlain on a light table. The U.S. and Russian layers were compared for content and detail. In most cases one of the two layers was obviously superior. That layer was chosen as the base (primary) layer. Then features in the other layer (that were not in the base layer) were added to create a combined U.S./Russian layer. Areas of disagreement, such as the existence and location of some pipelines, were decided by consensus since inspection of the source imagery was not possible. In some cases the existence of features was validated by using the records of a 1996 joint Russian-U.S. on- site inspection. With regard to spatial registration between the Russian and U.S. GIS layers, it was found that even though the UTM grid was agreed upon, the projections used by both sides were slightly different and the control points used by both sides were not identical. The spatial error was found to be a 50-100 meter translation, which was corrected by resampling the GIS layers. After the resampling, each of the layers and sublayers were precisely spatially registered. One of the advantages of this system is that newly acquired information can be easily incorporated into the existing GIS to support monitoring or further risk assessment activities.
