human-impacted ecosystems in central Africa, using as a background the land-cover
map of Mayaux et al. (2004) for Africa.
After the Last Glacial Maximum (21,000 cal yr BP), tropical forests were reduced
to only few refugia in central Africa. This is suggested both through the direct paleodata (Dechamps et al. 1988; Maley 1989) and indirect information from endemism
centers (Maley 1989). After the Younger Dryas, which was a short but intensely dry
period (from ~12,900 to ~11,700 cal yr BP, see Table 9.1, Sect. 9.2.1), rainfall levels
started increasing at the early stages of the Holocene (Fig. 9.1e). During that period,
rainfall levels were even higher than current levels, with this period called the
African Humid Period (de Menocal et al. 2000; Shanahan et al. 2015). As a result,
tropical forests were more extensive across central Africa, even reaching the
Adamawa Plateau in Cameroon (sites 1–2, Fig. 9.1a (Vincens et al. 2010; Lebamba
et al. 2016)) and the Niari Valley in the Republic of Congo (site 11, Fig. 9.1a
(Vincens et al. 1998)).
Figure 9.1b–e presents information about the different proxies and determinants
outlined in Sect. 9.2.2.1. Figure 9.1b, c highlights two examples of changes in pollen
assemblages, (a) a site that had undergone major changes in vegetation composition
during the LGM and the 3000 cal yr BP crisis (Fig. 9.1b): Lake Barombi Mbo from
Giresse et al. (1994) (site 4, Fig. 9.1a) and (b) a site that had experienced a transition
from forest to savanna during the 3000 cal yr BP crisis (Fig. 9.1c): Lake Mbalang
from Vincens et al. (2010) (site 1, Fig. 9.1a).
Figure 9.1d presents the frequency of dated archaeological evidence in sites
binned every 50 years. This is used to illustrate the changes in human impact across
the Congo basin using available data since 5000 cal yr BP (Oslisly et al. 2013;
Morin-Rivat et al. 2014). The period indicated in red corresponds to a period of
population increase across the Congo basin.
The long-term dynamics of sea surface temperature (SST) illustrates changes in
climate, and specifically rainfall (Fig. 9.1e) (Weldeab et al. 2007). It is worth noting
the period spanning the 3000 cal yr BP crisis, which is a period characterized by
abrupt changes in SST as indicated in the gray area (Fig. 9.1e).
9.3.1.2 Trends During the Late Holocene
The African Humid Period ended abruptly ~4000 cal yr BP (Fig. 9.1e). The
subsequent period was characterized by low rainfall and major droughts that lasted
until 1200 cal yr BP (Vincens et al. 1999). This period is called the “third millennium
rainforest crisis” and is divided into two major phases (Maley 2002).
The first phase started shortly after 4000 cal yr BP, and is associated with an
abrupt decrease in rainfall (see increase in SST in Fig. 9.1e). This trend affected the
peripheral areas of the Congo basin (see sites 1–2 and the Niari Valley in Fig. 9.1a,
c), and was responsible for the opening of the coastal savannas in central Africa (see
site 7, Fig. 9.1a) (Elenga et al. 1992; Ngomanda et al. 2009) and the Dahomey Gap in
west Africa (Salzmann and Hoelzmann 2005, not seen in Fig. 9.1a). During the same
period, savanna vegetation was also heavily modified, as suggested by the pollen
9 Long-Term Vegetation Change in Central Africa: The Need for an Integrated. . .
293
map of Mayaux et al. (2004) for Africa.
After the Last Glacial Maximum (21,000 cal yr BP), tropical forests were reduced
to only few refugia in central Africa. This is suggested both through the direct paleodata (Dechamps et al. 1988; Maley 1989) and indirect information from endemism
centers (Maley 1989). After the Younger Dryas, which was a short but intensely dry
period (from ~12,900 to ~11,700 cal yr BP, see Table 9.1, Sect. 9.2.1), rainfall levels
started increasing at the early stages of the Holocene (Fig. 9.1e). During that period,
rainfall levels were even higher than current levels, with this period called the
African Humid Period (de Menocal et al. 2000; Shanahan et al. 2015). As a result,
tropical forests were more extensive across central Africa, even reaching the
Adamawa Plateau in Cameroon (sites 1–2, Fig. 9.1a (Vincens et al. 2010; Lebamba
et al. 2016)) and the Niari Valley in the Republic of Congo (site 11, Fig. 9.1a
(Vincens et al. 1998)).
Figure 9.1b–e presents information about the different proxies and determinants
outlined in Sect. 9.2.2.1. Figure 9.1b, c highlights two examples of changes in pollen
assemblages, (a) a site that had undergone major changes in vegetation composition
during the LGM and the 3000 cal yr BP crisis (Fig. 9.1b): Lake Barombi Mbo from
Giresse et al. (1994) (site 4, Fig. 9.1a) and (b) a site that had experienced a transition
from forest to savanna during the 3000 cal yr BP crisis (Fig. 9.1c): Lake Mbalang
from Vincens et al. (2010) (site 1, Fig. 9.1a).
Figure 9.1d presents the frequency of dated archaeological evidence in sites
binned every 50 years. This is used to illustrate the changes in human impact across
the Congo basin using available data since 5000 cal yr BP (Oslisly et al. 2013;
Morin-Rivat et al. 2014). The period indicated in red corresponds to a period of
population increase across the Congo basin.
The long-term dynamics of sea surface temperature (SST) illustrates changes in
climate, and specifically rainfall (Fig. 9.1e) (Weldeab et al. 2007). It is worth noting
the period spanning the 3000 cal yr BP crisis, which is a period characterized by
abrupt changes in SST as indicated in the gray area (Fig. 9.1e).
9.3.1.2 Trends During the Late Holocene
The African Humid Period ended abruptly ~4000 cal yr BP (Fig. 9.1e). The
subsequent period was characterized by low rainfall and major droughts that lasted
until 1200 cal yr BP (Vincens et al. 1999). This period is called the “third millennium
rainforest crisis” and is divided into two major phases (Maley 2002).
The first phase started shortly after 4000 cal yr BP, and is associated with an
abrupt decrease in rainfall (see increase in SST in Fig. 9.1e). This trend affected the
peripheral areas of the Congo basin (see sites 1–2 and the Niari Valley in Fig. 9.1a,
c), and was responsible for the opening of the coastal savannas in central Africa (see
site 7, Fig. 9.1a) (Elenga et al. 1992; Ngomanda et al. 2009) and the Dahomey Gap in
west Africa (Salzmann and Hoelzmann 2005, not seen in Fig. 9.1a). During the same
period, savanna vegetation was also heavily modified, as suggested by the pollen
9 Long-Term Vegetation Change in Central Africa: The Need for an Integrated. . .
293