Harmattan dust deposition and particle size in Ghana

Abstract

In Ghana, a dust-laden Harmattan wind blows from the Sahara from November to March. Some of the dust is trapped in the vegetation, in the lakes and other inland waters, while the rest is blown further away into the Ivory Coast or into the Atlantic Ocean. Several methods have been used to trap the Harmattan dust, mainly bowls with or without water, but also plates or sheets of various materials have been used. This paper compares three different methods to trap the Harmattan dust and describes the differences in amount and particle size distribution of dust trapped in various agroecological zones. The investigation shows that bowls with water trap more dust than plastic mats with 1.5 cm straw which again trap more dust than wooden plates. The bowl values can be used as a measure of the total Harmattan dust deposition, the mat values can be used as a measure of how much Harmattan dust is retained by vegetation, while the plates provide a measure of how much dust is retained on vegetation free areas. The amount of Harmattan dust captured is largest in the north, the amount of dust retained differs significantly from year to year, and a severe Harmattan in the north is not necessary coincident with severe Harmattan in the south. The dust particle size becomes finer towards the south and with increasing amounts of organic matter, and the particle size distribution shows that the major part of the dust trapped is long-term suspended material with a significant input of local dust. The dust particle size is coarser on the mats than in the bowls due to re-suspension from the mats. The presence of diatoms in the samples from Bawku indicates that a substantial part of the dust originates from former lakes in the Sahara. The amount of dust retained in the north gives a deposition rate of about 15 mm per 1000 years. This may explain why loess layers have not developed in Ghana.