ABSTRACTThe deposition of natural dust in an area of 53 ha, situated in the northern Negev desert, is investigated in detail both in the wind tunnel (dust storm simulations over a topographic scale model) and in the field. The wind tunnel results and the field results show a high degree of agreement, indicating that scale‐model simulation may be considered an important technique for future loess and desert research. More dust settles on windward slopes than on leeward slopes, which is in contradistinction with the widespread wind shadow concept. Air‐flow separation zones immediately downwind of steep windward slopes have an important impact on dust deposition too.In the case of dust deposition on topographic scale models, a restricted height distortion of the model will not necessarily lead to serious problems. In addition, wind tunnel blockage percentages up to 13% may be allowed in order to obtain acceptable dust deposition patterns for the scale model.A mean gross dust deposition of about 200–250 g m−2year−1is calculated for the northern Negev desert for 1987. Thus, if the settled dust can be protected against erosion in the cultivated areas in the Negev, the dust content of the top soil will markedly increase with time. However, it has to be borne in mind that cultivation activities themselves may also contribute to a higher soil erosion and, hence, to a higher dust content in the atmosphere. At any rate, a higher dust content in the top soil will unquestionably have an important positive effect on agricultural yields.From the air dust concentration data and the dust deposition data, a deposition velocity of 4.7cms−1can be calculated for