Reply to comment by E. Williams on “Atmospheric controls on the annual cycle of North African dust”

Abstract

[01] The aim of Engelstaedter and Washington [2007a] (hereafter EW) was to analyze the long-term mean annual cycle of dust over North Africa and to assess the role of the large-scale wind in controlling the annual cycle. Total Ozone Mapping Spectrometer (TOMS) aerosol index (AI) [Herman et al., 1997] was used as it is the only spatially complete data set of dust loadings which is long enough to compute a climatology. EW report on the location of maximum AI loadings in the course of the annual cycle and show that these are found in close spatial association with the near surface wind divergence minima and are all in the central Sahara. These maxima colocate with the Saharan summer heat low and where dry convection extends beyond 5 km in summer in one of the deepest boundary layers on the planet. [2] Williams [2008] notes that in the EW ‘‘review’’ we neglect a potent source of dust production in North Africa, namely haboobs which result from moist convection in the northern fringe of the Intertropical Convergence Zone (ITCZ) and which also produces upper level cirrus. He argues that the cirrus artificially reduces the TOMS AI values in the lower troposphere. [3] There are important points of departure between EW and Williams [2008]. First, EW is not a review paper. It is the first paper to focus on the climatology of the annual cycle of dust in North Africa using a spatially complete data set. It was not the intention of EW to partition the role of specific dynamic systems in dust production. In order to investigate the role of specific dust generating systems in North Africa, including systems such as Sharav Cyclons [Moulin et al., 1998], African easterly waves [Jones et al., 2003], density currents [Knippertz et al., 2007], and also haboobs, an entirely different sampling framework would be necessary and simple averaging would not be appropriate. Some of these dust production mechanisms may be more important than others. Our view of haboobs and gust fronts associated with moist convection is that they may be an important system of dust generation in the boreal summer over the Sahara but on the basis of currently published data and observations we cannot tell. EW do not assess their role because it was not the aim of the paper to do so. But Williams [2008] provides no evidence either, other than to refer to research undertaken at a time (summer 2006) when EW was already in press and which identifies dust generation from cold outflows based on short-term monitoring distant from the central Sahara (a long-standing problem in the way that Saharan dust has been sampled [Engelstaedter et al., 2006]). It therefore remains the case that the relative role of dust generating systems in the central Sahara is unknown. It will be an important step to unravel this problem although there are dangers of doing so from sampling in the Sahel only. This will undoubtedly bias the perspective toward the role of systems in the Sahel which Engelstaedter and Washington [2007b] have already warned to have prejudiced the long-term view of North African dust sources away from the data scarce central Sahara toward the relatively well sampled Sahel.