extraction and carbon storage (Nasi et al. 2012). Yet, hunting is a non-negligible
side-effect, as it nearly always accompanies logging in tropical forests of SSA,
having dramatic consequences on animal populations and diversity (Poulsen et al.
2011). Defaunation is indeed occurring at an alarming rate in central SSA (Nasi et al.
2011; Abernethy et al. 2016), potentially having substantial impacts on ecosystem
functioning, and specifically tree regeneration and recruitment as demonstrated
in other parts of the world (for example, see Terborgh et al. (2008) for
southeastern Peru).
Tropical savannas cover 20% of the global land surface, account for 30% of
terrestrial net primary productivity, and sustain 20% of the human population, while
hosting most of the remaining megafauna (Scholes and Archer 1997; Veldman et al.
2015a; Bond 2016). Tropical savannas cover the overwhelming majority of SSA,
while provide important services to local populations (IPBES 2018). Furthermore,
savannas and other grassy ecosystems harbor important biodiversity, including the
largest herbivores on Earth (Parr et al. 2014; Bond 2016).
Savannas are commonly used in SSA for agriculture, pastoralism, and fuelwood
extraction, catering for the livelihoods of many local communities (Scholes and
Archer 1997; IPBES 2018) (see also Chaps. 3 and 5, Vol. 2). Recent land use change
has been threatening savanna ecosystems across SSA, with many savannas identified
as potentially interesting areas for large-scale conversion to industrial/cash
crops (Searchinger et al. 2015) (Chap. 3, Vol. 1), biofuel feedstock production
(Alexandratos and Bruinsma 2012) (Chap. 2, Vol. 1; Chap. 5, Vol. 2), and forest
plantations for forest restoration and CO 2 sequestration (Laestadius et al. 2012).
Contrary to tropical forests, savannas rely on chronic disturbances for maintaining an
open and diverse ecosystem (as detailed above). In fact, savannas require antagonist
management strategies such as recurrent or prescribed fires in high rainfall areas
(Lehmann et al. 2014; Osborne et al. 2018), which are not taken into account by the
forest scientific community (Veldman 2016).
Mesic savannas are located in areas where forest can also occur, and are especially targeted for the types of land conversion outlined above (Searchinger et al.
2015; Aleman et al. 2016). These types of savannas are indeed often mistaken as
“degraded” landscapes, or even as secondary successional stages following deforestation, which are of little value for biodiversity conservation, see arguments in
Bond and Parr (2010), Parr et al. (2014), and Veldman et al. (2015b). This is mainly
due to the fact that there is no recognition regarding their status as an alternative
stable state (Bond and Zaloumis 2016). This is clearly illustrated in the ongoing
debate on how to define forests.
2 A recent study that reassessed the extent of forests
in dryland biomes worldwide (Bastin et al. 2017), mistakenly considered most
2 The FAO defines forests as areas of more than 0.5 ha that have tree cover >10% (Pan et al. 2013).
Other authors have proposed higher thresholds such as tree cover >65% (e.g., Hirota et al. 2011;
Staver et al. 2011a).
9 Long-Term Vegetation Change in Central Africa: The Need for an Integrated. . .
283
side-effect, as it nearly always accompanies logging in tropical forests of SSA,
having dramatic consequences on animal populations and diversity (Poulsen et al.
2011). Defaunation is indeed occurring at an alarming rate in central SSA (Nasi et al.
2011; Abernethy et al. 2016), potentially having substantial impacts on ecosystem
functioning, and specifically tree regeneration and recruitment as demonstrated
in other parts of the world (for example, see Terborgh et al. (2008) for
southeastern Peru).
Tropical savannas cover 20% of the global land surface, account for 30% of
terrestrial net primary productivity, and sustain 20% of the human population, while
hosting most of the remaining megafauna (Scholes and Archer 1997; Veldman et al.
2015a; Bond 2016). Tropical savannas cover the overwhelming majority of SSA,
while provide important services to local populations (IPBES 2018). Furthermore,
savannas and other grassy ecosystems harbor important biodiversity, including the
largest herbivores on Earth (Parr et al. 2014; Bond 2016).
Savannas are commonly used in SSA for agriculture, pastoralism, and fuelwood
extraction, catering for the livelihoods of many local communities (Scholes and
Archer 1997; IPBES 2018) (see also Chaps. 3 and 5, Vol. 2). Recent land use change
has been threatening savanna ecosystems across SSA, with many savannas identified
as potentially interesting areas for large-scale conversion to industrial/cash
crops (Searchinger et al. 2015) (Chap. 3, Vol. 1), biofuel feedstock production
(Alexandratos and Bruinsma 2012) (Chap. 2, Vol. 1; Chap. 5, Vol. 2), and forest
plantations for forest restoration and CO 2 sequestration (Laestadius et al. 2012).
Contrary to tropical forests, savannas rely on chronic disturbances for maintaining an
open and diverse ecosystem (as detailed above). In fact, savannas require antagonist
management strategies such as recurrent or prescribed fires in high rainfall areas
(Lehmann et al. 2014; Osborne et al. 2018), which are not taken into account by the
forest scientific community (Veldman 2016).
Mesic savannas are located in areas where forest can also occur, and are especially targeted for the types of land conversion outlined above (Searchinger et al.
2015; Aleman et al. 2016). These types of savannas are indeed often mistaken as
“degraded” landscapes, or even as secondary successional stages following deforestation, which are of little value for biodiversity conservation, see arguments in
Bond and Parr (2010), Parr et al. (2014), and Veldman et al. (2015b). This is mainly
due to the fact that there is no recognition regarding their status as an alternative
stable state (Bond and Zaloumis 2016). This is clearly illustrated in the ongoing
debate on how to define forests.
2 A recent study that reassessed the extent of forests
in dryland biomes worldwide (Bastin et al. 2017), mistakenly considered most
2 The FAO defines forests as areas of more than 0.5 ha that have tree cover >10% (Pan et al. 2013).
Other authors have proposed higher thresholds such as tree cover >65% (e.g., Hirota et al. 2011;
Staver et al. 2011a).
9 Long-Term Vegetation Change in Central Africa: The Need for an Integrated. . .
283