SPATIAL STATISTICAL ANALYSIS ON THE TERRITORY OF ROYAL NURI NECROPOLIS AND NAPATAN REGION, SUDAN

This work aims at creating a model of unsupervised learning based on the spatial statistical analysis of common characteristics of the necropolis of Nuri and the Napatan Region. Nuri (N 18 33 50, E 31 55 00), together with the sites of Gebel Barkal, El-Kurru, Sanam, and Zuma has been inscribed in the List of the World Heritage Sites in July 2003 during the 27th session of the World Heritage Committee. These five archaeological sites, stretching over more than 60 km in the Nile valley, are testimony of the Napatan (900 to 270 B.C.) and Meroitic (270 B.C. to 350 A.D.) cultures, of the second kingdom of Kush. On site there are tombs, with and without pyramids, temples, living complexes and palaces. In particular, the necropolis of Nuri is located on the left bank of the Nile, facing the town of Karima, about two kilometres from the river. The complex occupies an area of over 700m2 for a total of 82 tombs, all excavated by George A. Reisner from 1917 to 1922. Most of the tombs have pyramidal superstructures. Heights are between 20 and 30 m. Here it is preserved the largest pyramid built in Sudan, that of Taharqa (690-644 B.C.), which heigh is about 30 m. The first burial in Nuri is from the year 664 B.C. and the last from around 310 B.C. The tombs contain one, two or more burial chambers, some decorated other ones plain. Other structures at Nuri include funerary chapels, a church and houses. In this work, the archaeological applications involving the use of high-resolution georeferenced, panchromatic and multispectral images shot from QuickBird, are first aimed at analyzing the patches and their spatial distribution and at testing the potential of remote sensing data as a means for identification, analysis, documentation and monitoring of the archaeological sites. Within the spatial data analysis techniques, the position of some elements observed in the space is used as a relevant factor to study the geographical distribution, the variability of a phenomenon in the space and measure, the spatial correlation between some specific properties that can be geometric and/or radiometric. The statistical geospatial indexes provide us synthetic measures on the spatial arrangement of observed phenomenon that can be random, uniform or cluster groups. The goal of this work is to estimate if in other areas of Sudan territory is possible to correlate a space cluster model inferred from observations found in the necropolis of Nuri.