Deposition of Silt by the Irrigation Waters of the Nile at Khartoum

Abstract

Vast quantities of silt are brought down from the volcanic highlands of Abyssinia each year by the flood waters of several rivers: the Blue Nile and its tributaries, the Atbara, the Gash and the Baraka. The last two, which spread out on reaching the Sudan to form inland deltas, are the most heavily laden with silt, and may carry over 10,000 parts of silt per million (1 per cent.) in suspension, making the water look like liquid chocolate. As the waters of these rivers spread out, the silt is deposited to form a visible layer on the land. The Blue Nile and the Atbara form part of the Nile system and carry up to 4000 parts per million of silt in suspension, most of which is borne away to the Mediterranean. Some is deposited in parts of the flood plain, which however is rather small and discontinuous in the Sudan, and further deposition occurs on land which is not flooded but which is low enough to be irrigated by the Nile waters, either by the traditional shaduf or saqia, or more extensively by pump. Such land may be irrigated all the year round, thus receiving a deposit of silt every two or three weeks. It was thought to be of interest to determine just how much silt is added to these lands each year, to what extent continuous irrigation may be able to raise the level of the land, and what the composition of the deposited material is. Experiments.?The work was carried out on the New Farm of the University of Khartoum, 5 miles north of the junction of the Blue Nile and White Nile. The water used for irrigation comes from the east side of the river and for most of the year it consists entirely of Blue Nile water which flows side by side with the White Nile for some miles before mixing. Samples of water were taken in 4-litre tins at three points in the irrigation system: at the head of the main canal, at its tail and at a minor canal where the water is led directly on to the irrigation plots. Of these, only the last is necessary to determine the amount of silt deposited on the land, but it was thought to be of interest to know what happened to the silt during its course from the river to the irrigation plots. The water samples were allowed to stand for a day or more, then the bulk was poured off, and the remainder transferred to a beaker and again allowed to stand before pouring off. The sediment was then dried in an oven and weighed. Samples of irrigation water were taken every month and continued, with a few lapses, from March 1954 to February 1957, so that the results for most months may be regarded as fairly representative of average conditions. Samples in January, February and July were unfortunately only taken once during that time, and for the months of May and June the samples were not taken by the writer but by his labora? tory assistant. Results.?Averages of the weighings (Table on p. 221) show several interesting features. The high amounts of silt are restricted to the months of August and Sep? tember, though the river is in flood for a much longer time. The rise is not sudden as might be expected, but develops slowly from June onwards. The fall is also gradual, being apparent until October, after which the amount of silt in the water remains steady and low at about o-i g. per litre or less until the following June. During the flood months of August and September the bulk of the silt brought up by the pump is deposited in the canals, only about a third of the original amount reaching the plots. This silting up of the canals is a regular occurrence, and the silt has to be periodically dug out. During the remainder of the year, deposition is slight in the main canal, while in the minor channels there is frequently more silt than was contained in the water at the head of the canal. The reason for this is that the soil is