Cookies on Forest Science Database

Like most websites we use cookies. This is to ensure that we give you the best experience possible.


Continuing to use  means you agree to our use of cookies. If you would like to, you can learn more about the cookies we use.

Forest Science Database

Supporting your research in forest and wood science

>>> Sign up to receive our Environmental Sciences e-newsletter, book alerts, and offers <<<

News Article

Woody Plant Encroachment in Sub-Saharan Africa

Climatic and local scale human interventions are used to explain WPE over the span of 30 years

Deforestation is a widely studied contributor to global greenhouse gas emissions and it is known to be increasing thanks to regular and high resolution mapping. This research has meant the drivers, often human induced, are also well understood. A countering force to this removal of forested land is woody plant encroachment (WPE) whereby woody shrubs and small trees begin to re-colonise areas of grassland and pasture. This effect has strong repercussions for ecosystem functioning, carbon sequestration and local livelihoods; increasing carbon capture and providing greater fuelwood availability whilst also reducing the carrying capacity of the land for pasture. However this phenomenon is not very well studied; previous mapping efforts have been low resolution and unable to synthesise local and global drivers.

This study has sought to address this by collecting data relating to the drivers of WPE stretching back the last 30 years. Woody plant cover was mapped at 30m resolution for Sub-Saharan Africa excluding closed forest, defined as >40% cover by trees >5m. A boosted regression tree (BRT) which included over 60 explanatory variables was designed to unearth which had the most influence on WPE and how the variables interacted. These drivers were able to explain 78% of the spatial variation in WPE, excluding CO2 levels, with warmer and wetter conditions being the key climatic drivers. CO2 increases also contributed to WPE but this is a global phenomenon and so did not contribute to spatial variation. Human interventions also played a part in WPE, with a decrease in the use of burning and reduction in the number of browsing herbivores also allowing the plants to spread.

The results showed that over the last 30 years there has been an 8% increase in woody vegetation in sub-Saharan African non-forest biomes. 7.5 million km2 (55%) experienced significant net gains in woody plant cover, including Cameroon, central African republic, south Sudan and Uganda. However, 2.2 million km2 (16%) showed significant decrease in cover, these areas included Sahel, East Africa and Madagascar.  The greatest levels of encroachment occurred in regions with 30-60% woody cover in 1986. Areas with >75% woody cover showed the greatest losses in woody plant coverage, likely a result of human influence. Also noted was the small difference in WPE between protected and non-protected areas.

These result show that local drivers often curtailed encroachment brought on by wider climatic changes. For instance herbivory was shown to aid WPE even when at high or low levels, this variation in impact was thought to be the result of different livestock managements. A decrease in the number of browsing herbivores was also thought to reduce grazing pressure on woody plants. This means human intervention e.g. fire regimes and grazing, may be able to manage WPE, overruling the weaker impacts of climatic drivers. These efforts could also be targeted in areas at risk of WPE due to predicted climatic changes i.e. more rain and higher temperatures. Finally, this study has ramifications for discussions on both desertification in Africa and how carbon balances are calculated as it presents the usage of high resolution data in the mapping of an understudied source of carbon sequestration.

Read the full paper here: Venter, Z.S., Cramer, M.D. and Hawkins, H.J., 2018. Drivers of woody plant encroachment over Africa. Nature communications, 9(1), p.2272.

To find over 550 similar records use the following search string in the Forest Science Database: ("woody plants" OR "woody vegetation”) AND (enchroach* OR spread*) AND (pastures OR grasslands)