The History of Fire in Afromontane Forests and the Future of Fire Management

Forest fire burning in afromontane forest in 2022. Source: Powell et al. (2023).

In the previous two blog posts I discussed the history of ecological management techniques used by people in savannas and grasslands and in rainforests in Africa. In this blog post, I will be addressing the history of landscape management in one last ecosystem: the afromontane forest. Understanding the history of land management in all three environments can offer us a lens through which to imagine the future of policy and management of these ecosystems for a sustainable future. 

Afromontane Forests 

While the exact role of humans in forest retreat in the last few thousand years in the tropical rainforests in Central and West Africa may still be up for debate, there is sizable evidence for human modification of afromontane forests in East Africa and southern-central Africa.

Areas with Afromontane forests in Africa. Area I: West
African and Cameroon Highlands. Area II: Arabian
and Ethiopian highlands. Area III: Western Rift.
Area IV: Eastern Rift.Area V: Southern Right.
Area VI: Eastern Highlands Area VII:Drakensberg.
Source: Wikipedia

One of the unique forest ecosystems of Africa are the Afromontane environments. They are cool humid forests which are very similar to the rainforests seen in the Guineo-Congolian rainforest, with a few key differences in the types of trees. Additionally, afromontane regions tend to be fragments, with sma sections of forests disconnected from other regions. Afromontane forests tend to be bounded by lower elevation more deciduous woodlands or savannas and afroalpine environments at higher elevations consisting of savannas and shrublands. Afromontane forests are very sensitive to fires. The trees in afromontane environments are not tolerant to being burned, unlike many of their nearby counterparts at lower elevation. As such past and present fires on afromontane regions threaten their extent.

In the Lake Malawi basin in southern-central Africa, evidence from 85,000 years ago shows significant ecosystem reorganization associated with a reduction in afromontane environments. This ecosystem reorganization is predated by changes in the charcoal found, which started around 150,000 years ago. Approximately 100,000 years ago, the area experienced aridification, which caused the loss of the afromontane regions. However, when the dry spell ended roughly 85,000 years ago, an arboreal environment recovered, but ino in the same way as it was 400,000 years ago. This new environment had much lower diversity, but included trees that were much more fire tolerant than the prior afromontane environment. This  time period is also associated with the amount of charcoal being produced remaining high. There is a substantial increase in archaeological sites in the area around 85,000 to 60,000 years ago. The expansion of people into the area combined with the altered fire regime and the atypical forest recovery has been interpreted as evidence that people during this time period were practicing burning of the forest for things such as land clearance. This is supported further by the formation of an alluvial fan produced from the higher amounts of sediment being washed away due to topsoil being left unprotected in the aftermath of burnings. From 40,000 to 20,000 years ago, things remained more stable, with charcoal amounts not changing much and sediment runoff remaining high. This likely represented a period of more stable land use by people who had successfully modified the environment to their needs. The correlations between human activity, charcoal, and changes in forest composition strongly support the notion that people were modifying and replacing the afromontane rain forests through the use of controlled landscape fires, starting 85,000 years ago.

Afromontane forest at Table Mountain, Cape Town, South Africa.
Source: Wikipedia

A similar story is found around Lake Edward in East Africa much later. In this area, cycles of wet and dry conditions were producing retreats and recoveries of the afromontane forests around the lake from 6,000 years ago and onward. Here there is dramatic evidence for the role of human controlled fire regimes in altering forest composition. There were decreases in the afromontane forests after 4,700 years ago, but it was still wooded and had few forest fires. This dramatically changes around 2,000 years ago where there is a dramatic decline in the afromontane forests that are now replaced by grasses, and there is a dramatic increase in the amount of fires in the landscape. While the initial drop may have been associated with a drop, there is no recovery of the afromontane forest that follows. This indicates that there was something happening on the landscape preventing this recovery, likely in the form of people. Much like in the rainforests at similar times, this period around is associated with substantial expansion of Iron Ages cultures. The technologies of the iron age, both metallurgy and agriculture and cultivation, required more extensive clearing of the land for fuel and for resource extraction.

There is further evidence for the long history of interaction between people and the forests around them through fire. The type of trees growing in forests and their fire resilience seems to have a strong connection to the density of the people living in and around them. Lowland forests tend to have much higher populations of people associated with them and also have a much higher tolerance to fires. As we have seen in the archaeological record described above, evidence suggests that when human populations move into these forest areas, the populations of the forests also shift to be more similar to the low-land forest and thus more fire tolerant. This correlation seems to indicate that people in these regions have been using fire as a landscape tool in forests, causing the forests to become more tolerant to fires as a response. 

Forest Fires

Afromontane forests and rainforests in general are very intolerant to forest fires and take over 100 years to recover from forest fire2. Paleoecological data indicates that for the most part, forest fires were fairly rare in these forests. However, in recent years, these forest landscapes have been encountering increasing amounts of fire, making forest fires a major concern for the preservation of these habitats.

Forests that have been cleared for agriculture
prior to being burned in Cameroon.
Source: Neumann et al. (2012).

Afromontane forests are threatened by clearing for farming and by clearing for logging3. On top of that, the recent years have been drier than  needed to perfectly maintain the rainforest environment. As such, the forest has been drying out. This makes them much more likely to ignite and burn when they encounter fire. Some of these forest fires starting the Afromontane forests were from natural sources, but a number came from human activity3. Swaths of forests are cut down for agriculture to support expanding populations and then the cut forest gets burned. Especially under dry conditions the fires can spread from the downed trees to the living ones causing forest fires. As discussed in a previous blog post  I discussed how fire is part of the natural seasonality cycling of savannas and how these fires are often lit intentionally by people to try to reduce any potential damages. Unfortunately, due to increased wood harvesting for fuel and timber along the margins of the forests, the natural protections that prevented savanna fire from spreading to the forest have been depleted3. This means that in recent years we have seen more of these savanna fires able to spread into forests. 

Forest fires in Congo Basin Rainforest 2016 captured from
satellite imagery from COPERNICUS. Source: Mongabay

Similar stories of forest fires are happening in the tropical rainforests of West and Central Africa, such as in the Congo Basin5. These forests have seen a substantial increase in the number of forest fires. Like the afromontane forests, these tropical rainforests are not adapted to fire, and fires burning in them are extremely damaging to local wildlife and the people living there.Tropical rainforests also play a significant role in mediating against climate change by storing half of all carbon stored in plants globally6. When there are wildfires, the burning of the trees causes the release of that stored carbon into the atmosphere in the form of CO2.  When the forests begin to regrow, the productivity of plant biomass in the burned area will increase by around four times the amount typically6. However, this respiratory increase does completely out-way the increase from the fire, meaning that in the long run, these forests are net emitters of CO2 when they burn. Additionally, the fires are devastating to these ecosystems as it takes over 100 years for them to recover after being burned3, which is significantly longer than the 5 year time intervals seen in savannas4.

Future of fires and management  in forests compared to grasslands

The way that forest ecosystems respond to fire is drastically different to how savanna systems respond, and it is important that when we talk about the future of fire use on the landscape that we separate the two. 

The number that claims that 70% of the world's wildfires are burning in Africa does not disaggregate this information, which is why that statistic can be misleading. Yes, 70% of the world's wildfires are burning in Afric, but most of them are burning in Savannas, and these fires are of little concern. However, there is a decent percentage of those fires which are burning in tropical rainforests and afromontane forests, as well as low land forests, and these fires should be raising concern. Just like in the Amazon, we are losing rainforest in Africa, and these rainforests are vital habitats for life and extremely productive carbon sinks7.

Fire management being done to protect
afromontane forests at Mount Moco.
Source: The Rutherford Foundation

There has been a long history of extensive fires in the savannas that predate human involvement, although humans have altered the fire regime, moving the seasonality of fires into the early dry season, reducing potential damages to property and trees of these fires8 (for more see previous blog post). All proposed management schemes for savannas must acknowledge that fires are a must have part of the ecosystem. You cannot have savannas without frequent fires.

The same is not true for the humid forests, both the rain forests and afromontane forests. Fire has not been a long and frequent feature in the environments and it is not beneficial to their long term success. They have throughout history seen declines and re-establishments due to climatic and anthropogenic reasons throughout history7. Now they face another period of extreme threat as places get dryer due to climate change and human population expansion continues to add more pressure to the forests. Management plans for these humid forests should include ways to reduce forest fires and using traditional knowledge to use fire in controlled ways as needed, potentially as a way to disconnect fuel loads or reduce them3.

As we talk about the future of fire use in landscape management, it is important to recognize that despite traditional prevailing thought amongst European scholars, fire in the landscape is not inherently bad8. It brings with it rebirth and fertility and when used properly in environments it is designed for, like in the savannas, it is more a bringer of life than a bringer of destruction. Unfortunately, a lot of traditional fire use strategies for the savanna have been outlawed because of the lasting influence of European colonialism on the laws in many countries in Africa7. People tending to the land must either defy these laws and start fires in secret or face the potentially devastating consequences of late season fires, fires which are more likely to spread to nearby forested areas to cause actual serious ecological damage. 

Prescribed burning of savannas in Zambia's Kafue
National Park to prevent damaging late season fires.
Source: The Nature Conservatory

Despite the extensive evidence that savannas are natural landscapes in Africa and fire is a necessary part of them, many environmental groups, especially international groups, are still calling for the foresting of savannas as a move towards greater sustainability8. Since savannas exist in climates that also can support fully wooded areas, almost all savannas in Africa are candidate places to make wooded. The question is should we make wooded places that were previously open savanna?

These environmentalists will argue that by switching to wooded climates, the amount of carbon being stored in plants in the environment will increase and that carbon will be held on to for longer periods of time than it is in the savannas. Also since forests do not require frequent burning like savannas, there would be less CO2 and other greenhouse gasses released from fires in Africa, which at the moment is sizable due to the large area of savanna that burns each year9. 

However, these arguments often fail to acknowledge that foresting these areas that have been savannas for tens of thousands of years is habitat destruction. There are countless plant and animal species which call the savannas of Africa home that would not be able to survive in wooded environments. It also fails to account for the increasing population size, and thus population pressure within Africa. Savannas are the location of most of Africa's agriculture and cultivation, and by foresting these areas, these vital food resources would be stripped from the people living there5. 

Any future environmental policies directed at the rainforests and savannas of Africa must take into account both types of ecosystems' unique relationship with fire and weight the different important features of each environment, both ecologically and culturally to the people that live there.

References 

1.  Thompson, Jessica C., David K. Wright, Sarah J. Ivory, Jeong-Heon Choi, Sheila Nightingale, Alex MacKay, Flora Schilt, Erik Otárola-Castillo, Julio Mercader, Steven L. Forman, Timothy Pietsch, et al. “Early human impacts and ecosystem reorganization in southern-central Africa.” Science Advances 7, no. 19 (2021).  https://doi.org/10.1126/sciadv.abf9776

2. Ivory, Sarah J. and Russell, James. “Lowland forest collapse and early human impacts at the end of the African Humid Period at Lake Edward, equatorial East Africa.” Quaternary Research 89, no. 1 (2018): 7-20. https://doi.org/10.1017/qua.2017.48 

3.  Powell, Luke L., Pedro Vaz Pinto, Michael S. L. Mills, Ninda L. Baptista, Kerllen Costa, Klass-Douwe B. Dijkstra, Amandio Luis Gomes, Patricia Guedes, Timoteo Julio, Ara Monadjem, et al. “The last Afromontane forests in Angola are threatened by fires.” Nature Ecology & Evolution (2023). https://doi-org.libproxy.berkeley.edu/10.1038/s41559-023-02025-9

4. Machete, Reason L. and Dintwe, Kebonyethata. "Cyclic Trends of Wildfires over Sub-Saharan Africa," Fire 6, no. 2 (2023): 71. https://doi.org/10.3390/fire6020071

5. Neumann, Katharina, Koen Bostoen, Alexa Hohn, Stefanie Kahlheber, Alfred Ngomanda, and Bathelemy Tchiengue. "First Farmers in the Central African rainforest: A view from Southern Cameroon." Quaternary International 249, no. 6 (2012): 53-62. https://doi.org/10.1016/j.quaint.2011.03.024

6. Fischer, Rico. “The Long-Term Consequences of Forest Fires on the Carbon Fluxes of a Tropical Rainforest in Africa.” Applied Sciences 11 (2021): 4696. https://doi.org/10.3390/app11104696

7. Malhi, Yadvinder, Stephen Adu-Bredu,Rebecca A. Asare, Simon L. Lewis, Phillippe Mayaux.”African rainforests: past, present and future.” Phil Trans R Soc B 368, (2013): 20120312. http://dx.doi.org/10.1098/rstb.2012.0312African rainforests

8. Bond, William and Zaloumis, Nicholas P. “The deforestation story: testing for anthropogenic origins of Africa's flammable grassy biomes.” Phil. Trans. R. Soc. B 371, no. 1696 (2016): 20150170 http://doi.org/10.1098/rstb.2015.0170

9. Archibald, Sally.  "Managing the human component of fire regimes: lessons from Africa" Phil. Trans. R. Soc. B 371, no. 1696 (2016): 20150346. http://doi.org/10.1098/rstb.2015.0346