Changes in the deep oceanic circulation, a major component of the global thermohaline circulation, may have an important impact on the Earth's climate. A proxy of the strength of bottom currents, initially proposed by McCave et al. (1995) and largely used to reconstruct past changes in the deep ocean circulation, is the mean size of sortable silt ( 𝐴𝐴 𝑆𝑆𝑆𝑆 ), which covers the 10-63 μm range of the non-cohesive silt fraction. Its use as a proxy for bottom-current strength relies on the fact that hydrodynamic processes of sorting in the viscous sublayer tend to act on particles greater than 10 μm (McCave & Hall, 2006). In sedimentary records from polar or subpolar environments, the sediment is in many cases a mixture of material supplied and sorted by bottom currents and of IRD (L. Wu et al., 2020). Because IRD generally cover the entire grain-size spectrum (J. Andrews, 2000; J. T. Andrews & Principato, 2002), they may influence 𝐴𝐴 𝑆𝑆𝑆𝑆 values. McCave and Andrews (2019) proposed to use the volume (%) of the >250 μm fraction obtained from laser granulometer as an IRD proxy. In the Drake Passage, which is characterized by high bottom-current flow speed, studies suggested an extension of the sortable range up to 125 μm (Lamy et al., 2015;S. Wu et al., 2021). In that case, IRD are thus also represented by medium to coarse sand fractions (e.g., >250 μm). Other studies consider the whole sand content (63 μm-2 mm; Hass, 2002;Hoffmann et al., 2019) as a tracer of IRD. These different size ranges of current-sorted and unsorted particles depend on environments and hydrodynamic conditions. If the limit between current-sorted and vertically advected particles may therefore be shifted toward grain-size >63 μm in environments with high flow speed conditions (S. Wu et al., 2021), this boundary may be shifted toward finer grain-size (<63 μm) in lower hydrodynamical conditions or areas with higher flux of material supplied. In that case, IRD may bias 𝐴𝐴 𝑆𝑆𝑆𝑆 values (Hass, 2002;Jonkers et al., 2015). Despite these variable limits of grain-size range to trace IRD, the influence of IRD on 𝐴𝐴 𝑆𝑆𝑆𝑆 values is poorly constrained. On the one hand, McCave and Andrews (2019) suggest that fine (i.e., silt-clay) IRD do not behave differently from any other particles and are current-sorted. On the other hand, other studies (Hass, 2002;Jonkers et al., 2015) consider that the contribution of unsorted IRD on 𝐴𝐴 𝑆𝑆𝑆𝑆 values should be removed to obtain a true record of past flow speed. For that, various approaches have been proposed.
The degree of correlation between𝐴𝐴 𝑆𝑆𝑆𝑆 and the percentage of sortable silt (i.e., the percent of the 10-63 μm in the <63 μm fraction, SS%) is defined as the index of sorting. High values (r > 0.5) would support 𝐴𝐴 𝑆𝑆𝑆𝑆 as paleo-current strength proxy, while low values could result from a significant contribution of IRD, without allowing the possibility of correcting for this IRD influence