Examples of the savannas of Ghana: 1) Guinean savanna (Left, Mole National Park)  2) Guinean savanna in the High Forest Zone-Interior Savanna Zone transition (Middle, Kogyae Strict Nature Reserve), 3)Coastal savanna (Right, Shai Hills Resource Reserve).

“When is a forest a savanna and why does it matter”? That is the title of a paper that was published in the journal “Global Ecology and Biogeography” in 2011, by one of my favourite savanna ecologists (Jayashree Ratnam)1. This article got my attention and full interest. I wondered if a savanna can ever be a forest and vice-versa. Although, the debate about whether a “savanna” is a “forest” is quite old, it still persist! But this is not a worthless debate about semantics. Classifying a forest as a savanna and vice-versa have real world management implications and consequences.

As a plant ecologist interested in the vegetations of Ghana (and West Africa), I believe it is still important to continue this discussion to dispel the old and archaic notion that savannas in Ghana are degraded forests. I also believe it is critically important to substantially improve our understanding of the functioning and dynamics of “forest” and “savanna” biomes in Ghana so we can capture their high economic and conservation value more accurately. Furthermore, from a conservation perspective, misreading savannas as degraded forests, undermines conservation efforts and prevents effective management.

So, off we go! settle in and let us learn about the savannas of Ghana, their origins, ecology, conservation values, potential roles in climate change mitigation and biodiversity conservation. 

Figure 1: Vegetation Types and Zones in Ghana (Adapted from White 1983)

1) Ghana hosts three distinctive savanna types.

Savanna is the most dominant vegetation type in Ghana, covering about two-thirds of the country. Three distinct savanna types are found in Ghana (see Fig.1), two of which (SS = Sudan Savanna, GS = Guinean Savanna) are widespread in West Africa. However, the most extensive portion of the rare and unique Coastal Savanna (CS) of West Africa – which includes the palm savannas of Ivory Coast and Liberia – occur in Ghana (“Accra Plains + Southern Volta”). Savanna landscapes in Ghana encompass some of the most important conservation areas (e.g., Mole National Park) and tourist attraction sites (e.g., Paga Crocodile Pond, Larabanga Mosque) and clearly have critical roles to play in the socio-economic development of the country.

2) Savannas in Ghana are not forest!

Yes, I might have misled you a bit in the beginning (It is “April fool” afterall ☹). But “No”, savannas are not forests! The irony, though, is that the current forest definition in Ghana is so broad that most savannas in Ghana are technically forest! Ghana defines forest as “any land with a minimum area of 1 hectare, with a minimum tree height of 5 meters, and minimum canopy cover of 15%3. Although Ghana also distinguishes the HFZ (High Forest Zone) from the ISZ (Interior Savanna Zone) and CSZ (Coastal Savanna Zone), technically speaking, most healthy savanna landscape in Ghana will likely meet the minimum tree cover and tree height thresholds of forests4. This is no trivial issue because definitions matter!

3) Savannas cannot be managed as forest?

That is the crux of the problem! A savanna cannot be managed as a forest and vice-versa. But if your definition of forest is so broad that it circumscribes savannas, then you are likely to manage these savannas as forests. However, managing savannas as forest is akin to considering palm or cocoa plantations as forest! This is because savannas generally have different species composition and structure to forests (see Fig.2). A savanna system is often composed of 1) sparse tree overstorey, 2) continuous grassy understorey, 3) regular fire occurrences, and 4) a diverse assemblage of large herbivores. These components interact and shape each other and ensure the proper functioning of a savanna ecosystem. For instance, abundant grass cover is important for maintaining both diverse large herbivore assemblage (source of food) and frequent fires (fuel for fire). Large herbivore and frequent fire, in turn, promote the dominance of sun-loving grasses by restricting tree establishment and hence maintaining the vegetation in open state (see this excellent video by Prof. William Bond for more details). In contrast, forests have closed canopies that exclude sun-loving grasses and prevent frequent fires. Exclusion of grasses and fire is important in forests because seedlings of forest species are poorly adapted to both grass competition and fire.

Figure 2: Differences between forest (left, closed canopy, grasses  and fire excluded) and savanna (right, open canopy, grassy understorey, large herbivores and fire): Source: Osborne et al, 2018, New Phytologist.

4) Most of Ghana’s savannas are not derived from degraded forest.

There is this pervasive, but erroneous, notion that the current savannas of Ghana (or West Africa) are “derived” from degraded forest. The argument has been that forest was the main vegetation over much of the current savanna landscape and that human disturbance (particularly use of fire) led to forest degradation and their subsequent replacement by savannas.  

However, recent scientific knowledge does not support this notion, rather evidence suggest that these savannas are ancient and achieved dominance over much of Africa prior to the emergence of modern humans6. For example, sun-loving savanna grasses (also called “C4 grasses”) appeared 30 million years ago whereas savanna expansion occurred approximately 8 million years ago. Thus, evolution of savanna grasses and savanna expansion occurred long before the evolution of modern humans, which is believed to have happened some 200,000-300,000 years ago.

This is not to say that savanna vegetation cannot replace degraded forest and that humans have no role at all. Indeed, it is well known that periodic forest or savanna expansion and retraction characterize the forest-savanna boundary in much of West Africa 7,8. There is also enough evidence that, in modern times, savanna has pushed into the HFZ in Ghana. This savanna invasion seems to have been first facilitated by the severe drought and bush fires of the early 1980s and subsequently by human activities. Therefore, the forest-savanna transition zone (here, I am referring to landscapes in the HFZ-ISZ and HFZ-CSZ transitions of Fig. 2) represents the crucial landscape where forest restoration may be important, particularly in the context of climate adaptation and mitigation in Ghana.

 5) Ghana’s savanna landscapes are underappreciated in spite of their high conservation values.

Savannas in Ghana contain high conservation values that must be protected. As such, it would be important to increase the attention and discussion around savannas among conservation professionals and with the public. Most savanna landscapes in Ghana contain multiple habitat types with each presenting its own unique species composition and hence contributing to high regional biodiversity. For instance, the CSZ host not only savanna vegetation but also dry forests, thickets, grasslands, mangroves, and wetland systems. These diverse habitats within the savanna landscapes are home for several regionally and nationally vulnerable, threatened, and endangered flora (e.g., Talbotiella gentii) and fauna (e.g., the African savanna elephant, Loxodonta Africana; African White-backed Vulture, Gyps africanus). Therefore, several important conservation areas, such as; 1) National Wildlife Parks (e.g., Mole National Park), 2) Ramsar Sites (e.g., Songor Ramsar site), and 3) Important Bird Areas (e.g., Shai Hills Resource Reserve) are located within Ghana’s savanna landscapes.  

6) Ghana savannas can play an important role in climate change mitigation but should be concurrently managed for biodiversity conservation.

There is a clear interest in using nature-based solutions (NBS) to mitigate global climate change.  Tree planting is preferred NBS choice and is strongly promoted by international development organizations. Indeed, several global initiatives (e.g., the Bonn Challenge) seeks to convert substantial areas of savannas to tree plantations. Although massive tree planting in the HFZ (including the forest-savanna transitions) are desirable, extending such initiatives to the ISZ and CSZ is likely to lead to irreversible negative ecological consequences (Yes, not all tree planting is good). As pointed out above, healthy savanna landscapes depend on continuous grass cover and fire. Excessive tree planting can have negative consequence on grass cover and hence alter fire regime with cascading impacts on savanna biodiversity. Thus, there is the need to balance climate change mitigation objectives (which highlight tree planting) with biodiversity objectives (which requires careful balance between tree, grasses, fire, and large herbivores). 

I am not suggesting that all forms of tree planting in savanna landscapes are bad. As indicated above, most savanna landscapes also host closed systems (dry forest and thickets). Tree planting in Ghanaian savannas can focus on restoring or maintaining these components of the savanna landscape. Also, we can certainly increase tree cover (not “afforestation) in our savannas using indigenous species of ecological and economic importance (e.g., the shea tree, Vitellaria paradoxa) when there is clear evidence that recent human activities have led to substantial loss of tree cover . However, care must be taken not to disrupt the balance between trees, grasses, fire, and large herbivores.

Thanks to Esther Rojas-Garcia for reviewing an initial draft of this post and helping make it a bit more readable (I hope)! If you found this post interesting, kindly subscribe to my blog to get email updates of new posts!


  1. Ratnam, J. et al. When is a ‘forest’ a savanna, and why does it matter? Glob. Ecol. Biogeogr.20, 653–660 (2011).
  2. White, F. Vegetation of Africa—A Descriptive Memoir to Accompany the UNESCO/AETFAT/UNSO Vegetation Map of Africa. (UNESCO, 1983).
  3. Forestry Commission. Ghana REDD + Strategy. (2016).
  4. Torello-Raventos, M. et al. On the delineation of tropical vegetation types with an emphasis on forest/savanna transitions. Plant Ecol. Divers. 6, 101–137 (2013).
  5. Osborne, C. P. et al. Human impacts in African savannas are mediated by plant functional traits. New Phytol. 220, 10–24 (2018).
  6. Bond, W. & Zaloumis, N. P. The deforestation story : testing for anthropogenic origins of Africa ’ s flammable grassy biomes. Philos. Trans. R. Soc. B Biol. Sci. (2016).
  7. anssen, T. A. J. et al. Extending the baseline of tropical dry forest loss in Ghana (1984–2015) reveals drivers of major deforestation inside a protected area. Biol. Conserv. (2018). doi:10.1016/j.biocon.2017.12.004
  8. Adejuwon, J. O. & Adesina, F. A. The nature and dynamics of the forest-savanna boundary in south-western Nigeria. in Nature and Dynamics of Forest-Savanna boundaries (eds. Furley, P. A., Procter, J. & Ratter, J. A.) 331–351 (Chapman and Hall, 1992).

%d bloggers like this: