We appreciate Swezey's interest in our paper, but we have to disagree with his reinterpretation of our data.
Swezey's statement that our data suggest that “Ogolien” dunes of western Mauritania range in age from 27–8 ka is apparently based upon the total range of all of the optically stimulated luminescence (OSL) dates and their errors. The error bars in fact reflect the precision of individual OSL dates, and it is erroneous to assume that eolian activity spanned this entire period. OSL dates are also but one type of data and, like any other chronometric data, need to be placed within a stratigraphic/geomorphic context. Most significant here is that the dunes occur in three different sizes, which are organized into three distinct trends that reflect three different wind regimes. We interpret these trends as the result of three episodes of eolian activity, and argue that their geomorphic expression is important in understanding how dune fields respond to constructional events separated in time. This interpretation is further supported by the OSL ages, which, when analyzed statistically, fall into distinct clusters on a probability density curve.
Our OSL dates in the 25–15 ka range are from the large, northeast-trending linear dunes. These dates suggest that this dune trend, which is the one locally termed Ogolien, corresponds to the eolian constructional event of the Last Glacial Maximum (e.g., Sarnthein, 1978). OSL dates clustering between 10 and 13 ka occur within the much smaller, north-northeast–trending linear dunes. A significant result of our study is that the Younger Dryas event (12.5–11.5 ka) may be manifested in the region by this second dune trend. There is little data on the environmental conditions between the Last Glacial Maximum and the Younger Dryas in the region, but the two trends of linear dunes clearly reflect separate generations of eolian construction, and there is no geomorphic and stratigraphic evidence that supports the second dune trend as a continuation of the Ogolien episode.
The termination of the second eolian constructional event and the manifestation of the African Humid Period in the region are well documented by interdune lacustrine deposits that overlie eolian strata and onlap the relict dunes. The relict dunes additionally show a paleosol zone indicating stabilization. The oldest lacustrine deposit in the area that we know of is our sample of organic material (TX-7182) that is 14C dated at 9410 ± 150 yr B.P. (ca. 10.5 ka), and which is included in the review by Swezey (2001). This predates the ca. 7.1 ka age from Barusseau et al. (1989) given by Swezey in his discussion of the onset of humid conditions (i.e., termination of the Ogolien event) in the Mauritanian coastal region. The third (and present) episode of eolian activity in the area began ca. 5 ka, in agreement with 14C dates for the cessation of lacustrine deposits (e.g., Kocurek et al., 1991; Swezey, 2001).
References to the dunes and interdune lacustrine deposits of the region are indeed extensive, especially by our French colleagues, but unfortunately restrictions on the length of papers in Geology preclude any extensive discussion of previous work. Kocurek et al. (1991) provide the reader with a detailed discussion of previous work, including those (and others) noted by Swezey. Similarly, we used the review paper by Swezey (2001) as a vehicle to address and refer to the regional data, and could not cite the extensive battery of works that the Swezey paper summarizes.