Fig 1.
Archaeological sites with Solutrean, Magdalenian, Late or Early Iberomaurusian occupation layers compiled for this study on an ombrotype map of the Worldwide Bioclimatic Classification System (1996–2015) for Iberia (original scale 1:1 Mio.) and Africa (original scale 1:25 Mio.) [17].
For names of the sites refer to S1 Table.
Fig 2.
Number of Solutrean (SOL), Magdalenian (MAG) and Iberomaurusian (IBM) sites in different ombrotypes of the Worldwide Bioclimate Classification System (1996–2015) [17] in the respective study areas (IP divided by 40°N, North-West Africa between 30°N and 0°).
Fig 3.
Schematic explanation of site classification: Single versus multi-layered sites and sites with one or multiple techno-complexes.
Fig 4.
Summed probability curves of all calibrated radiocarbon dates from Solutrean, Magdalenian and Iberomaurusian sites in the Western Mediterranean, together with climatic data for the Late Glacial period.
The upper climate curve (Cold Water surges) shows the southwards shift of the North Atlantic ice as recorded by C37.4 alkenone of core MD-01 2444 [77]. The climate curve in the middle reflects the variation in sea surface temperature of the Mediterranean Sea as reconstructed from relative proportions of di- and tri- alkenones of core ODP-976 [78]. The lower curve shows the δ18O record of the GISP2 ice core [79]. Grey columns display the timing of Heinrich Stadials 2 and 1 and their subdivision in an early, main and late phase as suggested by the pollen record of Alborán Sea core MD95-2043 [80]. The temporal limits for the Younger Dryas (11.7 to 12.9 ka BP) are taken from the recent compilation of delta18O and Ca2+ records of Greenland ice cores [1]. The SCDPD curve of the Western Mediterranean matches climate change from the Greenland intimate curve [1]. Unfavourable climate conditions during cold phases are reflected by drops of the SCDPD curve.
Fig 5.
Summed probability curves of calibrated AMS radiocarbon dates only from Solutrean, Magdalenian and Iberomaurusian sites in the Western Mediterranean, together with climatic data for the Late Glacial period.
Fig 6.
Site numbers for the regions Iberia (north), Iberia (south) and Morocco per techno-complex Solutrean, Magdalenian, Early and Late Iberomaurusian.
Category “Iberomaurusian”: the attribution to the Iberomaurusian is clear, but the period (Early or Late) is unknown due to lack of dating. As many sites were occupied during multiple techno-complexes, the sum of the site numbers in the data table do not add up to the total site numbers.
Fig 7.
Comparison of the regional distribution of single- and multi-layered sites during Solutrean and Magdalenian on the IP.
Category “unknown”: the attribution to Solutrean/Magdalenian is clear, but there is no information on the number of stratigraphic units.
Table 1.
Nearest Neighbour Statistics of the Late Pleistocene Settlement in the Western Mediterranean.
Fig 8.
Ripley’s K analysis of sites by region.
Fig 9.
Kernel density estimate of the Solutrean and of the Magdalenian in Iberia.
Included are Palaeocoastlines of -120m for the Solutrean and of -80m for the Magdalenian [81]. Digital Elevation Model from [82].
Fig 10.
Kernel density estimate of the Early/Unknown Iberomaurusian and of the Late/Unknown Iberomaurusian in Morocco.
Included are Palaeocoastlines of -120m for the Early/Unknown Iberomaurusian and of -80m for the Late/Unknown Iberomaurusian [81]. Digital Elevation Model from [82].
Fig 11.
Mating Networks of Solutrean and Magdalenian settlement in Iberia with population densities of 0.01, 0.025, and 0.05 people per km2.
Included are Palaeocoastlines of -120m for the Solutrean and of -80m for the Magdalenian [81]. Digital Elevation Model from [82].
Fig 12.
Mating Networks of Early/Unknown and Late/Unknown Iberomaurusian settlement in Morocco with population densities of 0.01, 0.025, and 0.05 people per km2.
Included are Palaeocoastlines of -120m for the Early/Unknown Iberomaurusian and of -80m for the Late/Unknown Iberomaurusian [81]. Digital Elevation Model from [82].