The Oldest Records of the Nile Floods

Abstract

Because rainfall over all but the northern Delta has long been rare and irregular, Egyptian farmers have depended for at least some 5000 years upon the annual flood of the Nile River to water their fields and prepare the soil for culti? vation. The amount of any particular inundation?at least before the building of the modern system of dams and barrages?determined whether that year would bring plenty or famine or something intermediate. These annual floods are the direct consequence of the summer monsoon rainfall over the catchment basin of the Blue Nile and the Atbara in the highlands of Ethiopia. The maximum level of the flood waters in Egypt thus provides a measure of the amount of this rainfall. From early historic times the Ancient Egyptians regularly measured the maximum height of the yearly flood and recorded the level in their royal annals. Of these records only fragments have come to light, and it has proved impossible to extract information on the relative magnitude of the floods in different eras because the various measures quite obviously do not use the same zero point, and perhaps not even the same scale (Kees, 1961, p. 50). However, there is one quite long series of measurements that, although surviving only in fragments, seems capable of yielding more information than has yet been extracted from it. This record was carved on a large stone stele during the Fifth Dynasty in the twenty-fifth century bc, and includes the level of the inundation for every year back to the reign of King Zer (Djer) early in the First Dynasty?about 3050 bc according to the chronology used by the revised Cambridge Ancient History (Edwards, 1964; Smith, 1962) and in this note. The most valuable surviving fragment, known as the Palermo Stone (Gardiner, 1961, plate III, p. 62Q from its location in the Palermo Museum, was originally published and translated by Schafer (1902). Additional fragments have since been discovered (Gauthier, 1915; Petrie, 1916; Cenival, 1965). All the flood heights preserved in these annals have been listed and discussed by Helck (1966), who points out an apparent decline in flood level after Dynasty I. His paper is concerned mainly with possible relations between certain festivals and the recorded flood levels, however, and he even suggests that the largest value?from year 30(F) of King Den (Wedimu)?may be a fiction created for religious purposes. As Helck does not discuss the data fully from the paleoclimatological point of view, the possibility remains of extracting more information from a different emphasis in the study of these ancient records. To faciliatate comparison with modern flood levels?in respect to their dispersion about the mean?I have converted all the measurements to metres, according to the relation: 0-524 metre = 1 cubit=7 hands/palms = 28 fingers = 2 spans. The value of the 'span* is only a guess, the true value being unknown as the unit was not used except in very early times. The resulting flood levels in metres are plotted against time in Figure 1, on the left. On the right are plotted some modern values measured at the Roda gauge near Cairo (Toussoun, 1925), with an arbitrary relation between the zero points of the two scales, the true relation being unknown. Table I gives