Woody plant encroachment

Woody plant encroachment (also called woody encroachment, bush encroachment, shrub encroachment, shrubification, woody plant proliferation, or bush thickening) is a natural phenomenon characterised by the increase in density of woody plants, bushes and shrubs, at the expense of the herbaceous layer, grasses and forbs. It predominantly occurs in grasslands, savannas and woodlands and can cause regime shifts from open grasslands and savannas to closed woodlands. The term bush encroachment refers to the expansion of native plants and not the spread of alien invasive species. It is thus defined by plant density, not species. Woody encroachment is often considered interpreted as a symptom of land degradation. The phenomenon is observed across different ecosystems and with different characteristics and intensities globally.

Causes include land-use intensification, such as overgrazing, as well as the suppression of wildfires and the reduction in numbers of wild herbivores. Elevated atmospheric CO2 and global warming are found to be accelerating factors. To the contrary, land abandonment can equally lead to woody encroachment.

The impact of woody plant encroachment is highly context specific. It can have severe negative impact on key ecosystem services, especially biodiversity, animal habitat, land productivity and groundwater recharge. Across rangelands, woody encroachment has led to significant declines in productivity, threatening the livelihoods of affected land users. Various countries actively counter woody encroachment, through adapted grassland management practices, controlled fire and mechanical bush thinning.

In some cases, areas affected by woody encroachment are classified as carbon sinks and form part of national greenhouse gas inventories. The carbon sequestration effects of woody plant encroachment are however highly context specific and still insufficiently researched. Depending on rainfall, temperature and soil type, among other factors, woody plant encroachment may either increase or decrease the carbon sequestration potential of a given ecosystem. In its Sixth Assessment Report of 2022, the Intergovernmental Panel on Climate Change (IPCC) states that woody encroachment may lead to slight increases in carbon, but at the same time mask underlying land degradation processes, especially in drylands. The UNCCD has identified woody encroachment as a key contributor to rangeland loss globally.

Ecological definition and etymology
Woody plant encroachment is the increase in abundance of indigenous woody plants, such as shrubs and bushes, at the expense of herbaceous plants, grasses and forbs, in grasslands and shrublands. The term encroachment is thus used to describe how woody plants outcompete grasses during a given time, typically years or decades. This is in line with the meaning of the term encroachment, which is "the act of slowly covering more and more of an area". Among earliest published notions of woody plant encroachment are publications of R. Staples in 1945, O. West in 1947 and Heinrich Walter in 1954.

Although the terms are used interchangeably in some literature, woody plant encroachment is different from the spread of invasive species. As opposed to invasive species, which are deliberately or accidentally introduced species, encroacher species are indigenous to the respective ecosystem and their classification as encroachers depends on whether they outcompete other indigenous species in the same ecosystem over time. As opposed to alien plant invasion, woody plant encroachment is thus not defined by the mere presence of specific plant species, but by their ecological dynamics and changing dominance.

In some instances, woody plant encroachment is a type of secondary succession. This applies to cases of land abandonment, for example when previous agricultural land is abandoned and woody plants re-establish. However, this is distinctly different from woody plant encroachment that occurs due to global drivers, e.g. increased carbon dioxide in Earth's atmosphere, and unsustainable forms of land use intensification, such as overgrazing and fire suppression. Such drivers disrupt the ecological succession in a given grassland, specifically the balance between woody and herbaceous plants, and provide a competitive advantage to woody plants. The resulting process that leads to an abundance of woody plants is sometimes considered an ecological regime shift (also ecological state transition) that can shift drylands from grassy dominated regimes towards woody dominated savannas. An increase in spatial variance is an early indicator of such regime shift. Depending on the ecological and climatic conditions this shift can be a type of land degradation and desertification. Progressing shrub encroachment is expected to feature a tipping point, beyond which the affected ecosystem will undergo substantial, self-perpetuating and often irreversible impact.

Research into the type of woody plants that tend to become encroaching species is limited. Comparisons of encroaching and non-encroaching vachellia species found that encroaching species have a higher acquisition and competition for resources. Their canopy architecture is different and only encroaching tree species reduce the productivity of perennial vegetation.

By definition, woody plant encroachment occurs in grasslands. It is thus distinctly different from reforestation and afforestation. However, there is a strong overlap between vegetation greening, as detected through satellite-derived vegetation indices, and woody plant encroachment. Grasslands and forests, as well as grasslands and shrublands, can be alternative stable states of ecosystems, but empirical evidence of such bistability is still limited.

Global extent
The UNCCD identifies woody encroachment as a key contributor to rangeland loss globally. Woody encroachment occurs on all continents, affecting and estimated total area of 500 million hectares (5 million squarekilometres). Its causes, extent and response measures differ and are highly context specific. Ecosystems affected by woody encroachment include closed shrublands, open shrublands, woody savannas, savannas, and grasslands. It can occur not only in tropical and subtropical climates, but also in temperate areas. Woody encroachment occurs at 1 percent per decade in the Eurasian steppes, 10–20 percent in North America, 8 percent in South America, 2.4 percent in Africa and 1 percent in Australia.

In Sub-Saharan Africa, woody vegetation cover has increased by 8% during the past three decades, mainly through woody plant encroachment. Overall, 750 million hectares of non-forest biomes experienced significant net gains in woody plant cover, which is more than three times the area that experienced net losses of woody vegetation. In around 249 million hectares of African rangelands, long-term climate change was found to be the key driver of vegetation change. Across Africa, 29 percent of all trees are found outside classified forests. In some countries, such as Namibia and Botswana, this percentage is above 80 percent and likely linked to woody encroachment. In Southern Africa, woody encroachment has been identified as the main factor of greening, i.e. of the increase in vegetation cover detected through remote sensing.

In Southern Europe an estimated 8 percent of land area has transitioned from grazing land to woody vegetation between 1950 and 2010.

In the Eurasian Steppe, the largest grassland globally, climate change linked woody plant encroachment has been found to occur at around 1% per decade.

In the Arctic Tundra, shrub plant cover has increased by 20 percent during the past 50 years. During the same time period, shrub and tree cover increased by 30 percent in the savannas of Latin America, Africa and Australia.

Causes
Woody encroachment is assumed to have its origins at the beginning of Holocene and the start of warming, with tropical species expanding their ranges away from the equator into more temperate regions. But it has occurred at unparalleled rates since the mid-19th century. As such, it is classified as a type of grassland degradation, which occurs through direct and indirect human impact during the Anthropocene.

Susceptibility of ecosystems
There is evidence that some characteristics of ecosystem render them more susceptible to woody encroachment than others. For example, coarse-textured soils promote woody plant growth, while fine-textured soils limit it. Moreover, the likelihood of woody encroachment is influenced by soil moisture and soil nutrient availability, which is why it often occurs on downslope locations and coolers slopes. The causes of woody encroachment differ significantly under different climatic conditions, e.g. between wet and dry savanna.

Various factors have been found to contribute to the process of woody plant encroachment. Both local drivers (i.e. related to land use practices) as well as global drivers can cause woody plant encroachment. Due to its strong link to human induced causes, woody plant encroachment has been termed a social-ecological regime shift. Research shows that both legacy effects of specific events, as well as plant traits can contribute to encroachment. There is still insufficient research on the interplay between the various positive and negative feedback loops in encroaching ecosystems.

Land use
Where land is abandoned, the rapid spread of native bush plants is often observed. This is for example the case in former forest areas in the Alps that had been converted to agricultural land and later abandoned. In Southern Europe encroachment is thus linked to rural exodus. In such instances, land use intensification, e.g. increased grazing pressure, is found to be effective against woody encroachment. More recently, it is observed that land use cessation is not the only driver of woody encroachment in aforement regions, since the phenomenon occurs also where land continued to be used for agricultural purposes.

In other regions, land use intensification, and the related fragmentation of landscapes, is the main cause of woody plant encroachment, especially in the following forms:
 * Overgrazing: In the context of land intensification, a frequently cited cause of woody plant encroachment is overgrazing, commonly a result of overstocking and fencing of farms, as well as the lack of animal rotation and land resting periods. Overgrazing plays an especially strong role in mesic grasslands, where bushes can expand easily when gaining a competitive advantage over grasses, while woody encroachment is less predictable in xeric shrublands. Seed dispersal through animals is found to be a contributing factor to woody encroachment.  While overgrazing has in the past frequently been found to be a main driver of woody encroachment, it is observed that woody encroachment continues in the respective areas even after grazing reduced or even ceases.
 * Absence of large mammals: linked to the introduction of rangeland agriculture as well as unsustainable hunting practices, the reduction of large mammals such as elephant and rhino (in Africa) or elk (in Northern America) is a contributing factor to woody encroachment.
 * Fire suppression: A connected cause for woody plant encroachment is the reduction in the frequency of wildfires that would occur naturally, but are suppressed in frequency and intensity by land owners due to the associated risks and the fragmentation of landscapes. When the lack of fire reduces tree mortality and consequently the grass fuel load for fire decreases, a negative feedback loop occurs. It has been estimated that from a threshold of 40% canopy cover, surface grass fires are rare.  At intermediate rainfall, fire can be the main determinant between the development of savannas and forests.  In experiments in the United States it was determined that annual fires lead to the maintenance of grasslands, 4-year burn intervals lead to the establishment of shrubby habitats and 20-year burn intervals lead to severe woody plant encroachment. Moreover, the reduction of browsing by herbivores, e.g. when natural habitats are transformed into agricultural land, fosters woody plant encroachment, as bushes grow undisturbed and with increasing size also become less susceptible to fire. Already one decade of land management change, such as the exclusion of fires and overgrazing, can lead to severe woody plant encroachment. The global increase in atmospheric CO2 contributes to the reduction of wildfires, as it decreases flammability of grass.
 * Competition for water: a positive feedback loop occurs when encroaching woody species reduce the plant available water, providing a disadvantage for grasses, promoting further woody encroachment. According to the two-layer theory, grasses use topsoil moisture, while woody plants predominantly use subsoil moisture. If grasses are reduced by overgrazing, this reduces their water intake and allows more water to penetrate into the subsoil for the use by woody plants. Moreover, research suggests that bush roots are less vulnerable to water stress than grass roots during droughts.
 * Population pressure: population pressure can be the cause for woody plant encroachment, when large trees are cut as building material or fuel. This stimulates coppice growth and results in shrubbiness of the vegetation.

Climate change
While changes in land management are often seen as the main driver of woody encroachment, some studies suggest that global drivers increase woody vegetation regardless of land management practices. For example, a representative sampling of South African grasslands, woody plant encroachment was found to be the same under different land uses and different rainfall amounts, suggesting that climate change may be the primary driver of the encroachment. Once established, shrubs suppress grass growth, perpetuating woody plant encroachment.

Predominant global drivers include the following:
 * Atmospheric CO2: climate change has been found to be a cause or accelerating factor for woody plant encroachment. This is because increased atmospheric CO2 concentrations fosters the growth of woody plants. Woody plants with C3 photosynthetic pathway thrive under high CO2 concentrations, as opposed to grasses with C4 photosynthetic pathway.    Also tolerance to herbivory is found to be enhanced during the plants' recruitment stage under increased CO2 concentrations.
 * Rainfall patterns: a frequently cited theory is the state-and-transition model. This model outlines how rainfall and its variability is the key driver of vegetation growth and its composition, bringing about woody plant encroachment under certain rainfall patterns. For example, if rainfall intensity increases, deep soil water typically increases, which in turn benefits bushes more than grasses. Both the amount of rainfall and its timing are important and distinct factors. Changes in precipitation can foster woody encroachment. Increased precipitation can foster the establishment, growth and density of woody plants. Also decreased precipitation can promote woody plant encroachment, as it fosters the shift from mesophytic grasses to xerophytic shrubs.
 * Global warming: woody encroachment correlates to warming in the tundra, while it is linked to increased rainfall in the savanna. Species such as Vachellia sieberiana thrive under warming irrespective of the competition with grasses. The Intergovernmental Panel on Climate Change (IPCC) in its report "Global warming of 1.5°C" states that high-latitude tundra and boreal forests are at particular risk of climate change-induced degradation, with a high likelihood of shrub encroachment under continued warming. In other ecosystems, such as sub-Sahara grasslands, rising aridity may cause woody plants to be more prone to hydraulic failure.
 * Droughts: droughts contribute to woody plant encroachment, if they reduce the perennial grass cover and the latter recovers slowly, providing shrubs with an competitive advantage with regard to the acquisition of deep-soil water. Drought, in combination with high levels of grazing pressure, can function as the tipping point for an ecosystem, causing woody encroachment.

Impact on grassland ecosystems
Woody encroachment constitutes a major global shift in plant composition, structure and function, wit far-reaching impact on the affected ecosystems. The accelerating rate of woody encroachment across grasslands globally may lead to an abrupt decline of this biome type, owing to human impact. For example, the Great Plains biome is found to be at the brink of collapse due to woody encroachment, with 62% of Northern American grassland lost to date.

Encroachment is commonly identified as a form of land degradation, with severe negative consequences for various ecosystem services, such as biodiversity, groundwater recharge, carbon storage capacity and herbivore carrying capacity. Nevertheless, negative impact is not universal. Impacts are dependent on species, scale and environmental context factors. Woody encroachment can have significant positive impacts on ecosystem services as well. Affected ecosystem services fall into the category of provisioning (e.g. forage value), regulating (e.g. hydrological regulation, soil stability) and supporting (nutrient cycling, carbon sequestration, biodiversity, primary production). There is a need for ecosystem-specific assessments and responses to woody encroachment. Generally, the following context factors determine the ecological impact of woody encroachment:
 * Prevailing land use : While positive ecological effects can occur in unmanaged landscapes or certain land-uses, negative ecological effects are observed especially in landscapes used for livestock grazing.
 * Density of woody plants : Plant diversity and ecosystem multifunctionality typically peaks at intermediate levels of woody cover and high woody covers generally have negative impacts.
 * Environmental conditions : Arid ecosystems show more negative responses to woody encroachment than non-arid ecosystems. In arid ecosystems woody encroachment is sometimes regarded as a form of land degradation and an expression of desertification Due to its ambiguous role in these dry ecosystems, it has been termed "green desertification". To the contrary, in ecosystems of the Mediterranean region and in Alpine grasslands, encroachment can enhance ecosystem functionality and reverse desertification trends. An key difference is that during woody encroachment the herbaceous cover in the inter-canopy zones can remain intact, while during desertification these zones degrade and turn into bare soil devoid of organic matter.

Biodiversity
Woody encroachment causes widespread declines in the diversity of herbaceous vegetation through competition for water, light, and nutrients Bush expands at the direct expense of other plant species, potentially reducing plant diversity and animal habitats. These effects are context specific, a meta-analysis of 43 publications of the time period 1978 to 2016 found that woody plant encroachment has distinct negative effects on species richness and total abundance in Africa, especially on mammals and herpetofauna, but positive effects in North America. However, in context specific analyses also in Northern America negative effects are observed. For example, piñon-juniper encroachment threatens up to 350 sagebrush-associated plant and animal species in the US. A study of 30 years of woody encroachment in Brazil found a significant decline of species richness by 27%. Shrub encroachment may result in increase vertebrate species abundance and richness. However, these encroached habitats and their species assemblages may become more sensitive to droughts. As encroachment is not a stable state, but characterised by changing bush densities, it is important to identify how different density threshold affect plant and animal species.

Evidence of biodiversity losses includes the following:
 * Grasses: encroachment results in substantial loss of herbaceous diversity, with a loss of richness that is not replaced. Studies in South Africa have found that grass richness reduces by more than 50% under intense woody plant encroachment. In North America, a meta-analysis of 29 studies from 13 different grassland communities found that species richness declined by an average of 45% under woody plant encroachment. Rare species and those with lower stature, are at risk of going extinct. Among the severely affected flora is the small white lady's slipper. Generally, large bushes are found to coexist with the herbaceous layer, while smaller shrubs compete with it.
 * Mammals: woody plant encroachment has a significant impact on herbivore assemblage structure and can lead to the displacement of herbivores and other mammal types that prefer open areas. Among other factors, predation success of various mammals is negatively impacted by bush encroachment. Among the species found to lose habitat in areas affected by woody plant encroachment are cats such as cheetah, white-footed fox , as well as antelopes such as the Common tsessebe, Hirola and plains zebra. In Latin America the habitat of the almost extinct Guanaco is threatened by woody encroachment. In some rangelands, woody plant encroachment is associated with a decline in wildlife grazing capacity of up to 80%. Among rodent species, those specialists on grasslands typically decline in abundance under woody encroachment, while those specialised on forests might increase in abundance. Also burrowing mammals can lose habitat when woody encroachment occurs.
 * Birds: the impact of woody encroachment on bird species must be differentiated between shrub-associated species and grassland specialists. Studies find that shrub-associated species benefit from woody encroachment up to a certain threshold of woody cover (e.g. 22 percent in a study conducted in North America), while grassland specialist populations decline.  Experiments in Namibia have shown that foraging birds, such as the endangered Cape vulture, avoid encroachment levels above 2,600 woody plants per hectare. In Southern Africa, woody encroachment drives population decline of 20% of the common open ecosystem bird species, on average at a rate of 50% population decline over fifty years. In North American grasslands, bird population decline as a result of woody encroachment has been identified as a critical conservation concern.  Among the birds negatively affected by woody plant encroachment are the Secretarybird, Grey go-away-bird, Marico sunbird, lesser prairie chicken,  Greater sage-grouse, Archer's lark,  Northern bobwhite and the Kori bustard.
 * Insects: woody plant encroachment is linked to species loss or reduction in species richness of insects with preference for open habitats. Affected species include butterfly, ant and beetle.

Vegetation structure
Encroachment often creates connected bare plant interspaces where water and wind erosion can occur.

Groundwater recharge and soil moisture
Woody plant encroachment is frequently linked to reduced groundwater recharge, based on evidence that bushes consume significantly more rainwater than grasses and encroachment alters water streamflow. Woody encroachment generally leads to root elongation in the soil and the downward movement of water is hindered by increased root density and depth. The impact on groundwater recharge differs between sandstone bedrocks and karst regions as well as between deep and shallow soils.

Besides groundwater recharge, woody encroachment increases tree transpiration and evaporation of soil moisture, due to increased canopy cover. Woody encroachment leads to the drying up of stream flows. Further, woody plant control can effectively improve the connectivity of water resources. Although this is strongly context dependent, bush control can be an effective method for the improvement of groundwater recharge.

While water loss is common in closed canopy woodlands (i.e. sub-humid conditions with increased evapotranspiration) in semiarid and arid ecosystems recharge can also improve under encroachment, provided there is good ecohydrological connectivity of the respective landscape.

There is limited understanding how hydrological cycles through woody encroachment affect carbon influx and efflux, with both carbon gains and losses possible. Moreover, there is evidence that woody encroachment enhances bedrock weathering, with unclear consequences for soil erosion and subsurface water flows.

However, concrete experience with changes in groundwater recharge is largely based on anecdotal evidence or regionally and temporally limited research projects. Applied research, assessing the water availability after brush removal, was conducted in Texas, US, showing an increase in water availability in all cases. Studies in the United States moreover find that dense encroachment with Juniperus virginiana is capable of transpiring nearly all rainfall, thus altering groundwater recharge significantly. An exception is shrub encroachment on slopes, where groundwater recharge can increase under encroachment. Further studies in the US indicate that also stream flow is significantly hampered by woody plant encroachment, with the associated risk of higher pollutant concentrations.

Studies in South Africa have shown that approximately 44% of rainfall is captured by woody canopies and evaporated back in to the atmosphere under woody encroachment. This effect is strongest with fine-leaved species and in events of lower rainfall sizes and intensities. It was found that up to 10% less rain enters the soil overall under woody encroachment. A meta-analysis of studies in South Africa further finds that woody encroachment has low water loss effect in areas with limited rainfall. Streamflow can increase after targeted removal of invasive and encroaching species, as showcased in South Africa.

Carbon sequestration
The impact of bush control on the carbon sequestration and storage capacity of the respective ecosystems is an important management consideration. Against the background of global efforts to mitigate climate change, the carbon sequestration and storage capacity of natural ecosystems receives increasing attention. Grasslands constitute 40% of Earth's natural vegetation and hold a considerable amount of the global Soil Organic Carbon. Shifts in plant species composition and ecosystem structure, especially through woody encroachment, lead to significant uncertainty in predicting carbon cycling in grasslands. Research on the changes to carbon sequestration under woody plant encroachment and its control is still insufficient. The Intergovernmental Panel on Climate Change (IPCC) states that woody plant encroachment generally leads to increased aboveground woody carbon, while below-ground carbon changes depend on annual rainfall and soil type. The IPCC points out that carbon stock changes under bush encroachment have been studied in Australia, Southern Africa and North America, but no global assessment has been done to date.

Total ecosystem carbon: considering above-ground biomass alone, encroachment can be seen as a carbon sink. However, considering the losses in the herbaceous layer as well as changes in soil organic carbon, the quantification of terrestrial carbon pools and fluxes becomes more complex and context specific. Changes to carbon sequestration and storage need to be determined for each respective ecosystem and holistically, i.e. considering both above-ground and below-ground carbon storage. Generally, elevated CO2 leads to increased woody growth, which implies that the woody plants increase their uptake of nutrients from the soil, reducing the soil's capacity to store carbon. In contrast, grasses increase little biomass above-ground, but contribute significantly to below-ground carbon sequestration. It is found that above-ground carbon gains can be completely offset by below-ground carbon losses during encroachment. It is generally observed that carbon increases overall in wetter ecosystems under encroachment and can reduce in arid ecosystems under encroachment. Some studies find that carbon sequestration can increase for a number of years under woody encroachment, while the magnitude of this increase is highly dependent on annual rainfall. It is found that woody encroachment has little impact on sequestration potential in dry areas with less than 400mm in precipitation. This implies that the positive carbon effect of woody plant encroachment may decrease with progressing climate change, particularly in ecosystems that are forecasted to experience decreased precipitation and increased temperature. Woody encroachment is further linked to fluvial erosion that in turn leads to the loss of previously stabilised organic carbon from legacy grasslands. Moreover, encroached ecosystems are more likely than open grasslands to lose carbon during droughts. Among the ecosystems expected to lose carbon storage under woody encroachment is the tundra.

Factors relevant for comparisons of carbon sequestration potentials between encroached and non-encroached grasslands include the following: above-ground net primary production (ANPP), below-ground net primary production (BNPP), photosynthesis rates, plant respiration rates, plant litter decomposition rates, soil microbacterial activity. Also plant biodiversity is an important indicator, as plant diversity contributes more to soil organic carbon than the quantity of organic matter.
 * Above-ground carbon: woody plant encroachment implies an increase in woody plants, in most cases at the expense of grasses. Considering that woody plants have a longer lifespan and generally also more mass, woody plant encroachment typically implies an increase in above-ground carbon storage through biosequestration. Studies however find that this is dependent on climatic conditions, with aboveground carbon pools decreasing under woody encroachment where mean annual precipitation is less than 330mm and increasing where precipitation is higher. A contributing factor is that woody encroachment decreases above-ground plant primary production in mesic ecosystems.
 * Below-ground carbon: globally, the soil organic carbon pool is twice as large as the plant carbon pool, making its quantification essential. Soil organic carbon makes out two-thirds of total soil carbon. Comparisons of grasslands, shrublands and forests show that forest and shrubland hold more above-ground carbon, while grasslands boast more soil carbon. Generally, herbaceous plants allocate more biomass below-ground than woody plants. The impact of woody encroachment on soil organic carbon is found to be dependent on rainfall, with soil organic carbon increasing in dry ecosystems and decreasing in mesic ecosystems under encroachment. Degradation of grasslands has in some areas led to the loss of up to 40% of the ecosystem's soil organic carbon. An important factor is that under woody plant encroachment the increased photosynthetic potential is largely offset by increased plant respiration and respective carbon losses. In tropical savanna soils, most soil organic carbon is derived from grass, not woody plants.  For example, research in South Africa found that soil organic carbon from tree input matched grass-derived soil organic carbon only after 70 years of fire exclusion, challenging the view that increased tree density leads to SOC improvements.


 * Soil organic carbon changes need to be viewed at landscape level, as there are differences between under canopy and inter canopy processes. When a landscape becomes increasingly encroached and the remaining open grassland patches are overgrazed as a result, soil organic carbon may decrease. In South Africa, woody plant encroachment was found to slow decomposition rates of litter, which took twice the time to decay under woody plant encroachment compared to open savannas. This suggests a significant impact of woody encroachment on the soil organic carbon balance. In pastoral lands of Ethiopia, woody plant encroachment was found to have little to no positive effect on soil organic carbon and woody encroachment restriction was the most effective way to maintain soil organic carbon. In the United States, substantial soil organic carbon sequestration was observed in deeper portions of the soil, following woody encroachment.
 * An important factor is that rooting depth increases with woody encroachment, on average by 38 cm and up to 65 cm. Deeper rooting may promote the accumulation of organic carbon in the deep soil layers, but at the same time also lead to a positive priming effect, i.e. the stimulation of microbial activity and decomposition of organic matter. The trajectory of deep soil carbon under woody encroachment will depend on the balance of increased SOC accumulation and priming losses.
 * A meta-analysis of 142 studies found that shrub encroachment alters soil organic carbon (0–50 cm), with changes ranging between -50 and 300 percent. Soil organic carbon increased under the following conditions: semi-arid and humid regions, encroachment by leguminous shrubs as opposed to non-legumes, sandy soils as opposed to clay soils. The study further concludes that shrub encroachment has a mainly positive effect on top-soil organic carbon content, with significant variations among climate, soil and shrub types. There is a lack of standardised methodologies to assess the effect of woody encroachment on soil organic carbon.
 * A meta-analysis of 142 studies found that shrub encroachment alters soil organic carbon (0–50 cm), with changes ranging between -50 and 300 percent. Soil organic carbon increased under the following conditions: semi-arid and humid regions, encroachment by leguminous shrubs as opposed to non-legumes, sandy soils as opposed to clay soils. The study further concludes that shrub encroachment has a mainly positive effect on top-soil organic carbon content, with significant variations among climate, soil and shrub types. There is a lack of standardised methodologies to assess the effect of woody encroachment on soil organic carbon.
 * A meta-analysis of 142 studies found that shrub encroachment alters soil organic carbon (0–50 cm), with changes ranging between -50 and 300 percent. Soil organic carbon increased under the following conditions: semi-arid and humid regions, encroachment by leguminous shrubs as opposed to non-legumes, sandy soils as opposed to clay soils. The study further concludes that shrub encroachment has a mainly positive effect on top-soil organic carbon content, with significant variations among climate, soil and shrub types. There is a lack of standardised methodologies to assess the effect of woody encroachment on soil organic carbon.

Land productivity
Woody plant encroachment directly impacts land productivity, as widely documented in the context of animal carrying capacity. In the western United States, 25% of rangelands experience sustained tree cover expansion, with estimated losses for agricultural producers of $5 billion since 1990. The forage lost annually is estimated to be equal to the consumption of 1.5 million bison or 1.9 million cattle. In Northern America, each 1 percent of increase in woody cover implies a reduction of 0.6 to 1.6 cattle per 100 hectares. In the Southern African country Namibia it is assumed that agricultural carrying capacity of rangelands has declined by two-thirds due to woody plant encroachment. In East Africa there is evidence that an increase of bush cover of 10 percent reduced grazing by 7 percent, with land becoming unusable as rangeland when the bush cover reaches 90 percent.

Tourism potential
Touristic potential of land is found to decline in areas with heavy woody plant encroachment, with visitors shifting to less encroached areas and better visibility of wildlife.

Rural livelihoods
Woody encroachment is often considered to have a negative impact on rural livelihoods. In Africa, 21% of the population depend on rangeland resources. Woody encroachment typically leads to an increase in less palatable woody species at the expense of palatable grasses. This reduces the resources available to pastoral communities and rangeland based agriculture at large. Woody encroachment has negative consequences on livelihoods especially arid areas, which support a third of the world population's livelihoods. Woody plant encroachment is expected to lead to large scale biome changes in Africa and experts argue that climate change adaptation strategies need to be flexible to adjust to this process.

Others
In the United States, woody encroachment has been linked to the spread of tick-borne pathogens and respective disease risk for humans and animals. In the Arctic tundra, shrub encroachment can reduce cloudiness and contribute to a raise in temperature. In Northern America, significant increases in temperature and rainfall were linked to woody encroachment, amounting to values up to 214mm and 0.68 °C respectively. This is caused by a decrease in surface albedo.

Targeted bush control in combination with the protection of larger trees is found to improve scavenging that regulates disease processes, alters species distributions, and influences nutrient cycling.

Studies of woody plant encroachment in the Brazilian savanna suggest that encroachment renders affected ecosystems more vulnerable to climate change.

Quantification and monitoring
There is no static definition of what is considered woody encroachment, especially when encroachment of indigenous plants occurs. While it is simple to determine vegetation trends (e.g. an increase in woody plants over time), it is more complex to determine thresholds beyond which an area is to be considered as encroached. Various definitions as well as quantification and mapping methods have been developed.

Data collection can typically involve mapping and morphological characterisation of trees and shrubs, phytosociological survey of permanent plots, grid-point intercept survey of permanent plots, line-intercept surveys along transects as well as allometric shrub measurements along transects. In Southern Africa, the BECVOL method (Biomass Estimates from Canopy Volume) finds frequent application. It determines Evapotranspiration Tree Equivalents (ETTE) per selected area. This data is used for comparison against climatic factors, especially annual rainfall, to determine whether the respective area has a higher number of woody plants than is considered sustainable.

Remote sensing imagery is frequently used to determine the extent of woody encroachment. Shortcomings of this methodology include difficulties to distinguish species and the inability to detect small shrubs. Moreover, UAV (drone) based multispectral data and Lidar data are frequently used to quantify woody encroachment. The combination of colour-infrared aerial imagery and support-vector machines classification, can lead to high accuracy in identifying shrubs. The probability of woody plant encroachment for the African continent has been mapped using GIS data and the variables precipitation, soil moisture and cattle density. An exclusive reliance on remote sensing data bears the risk of wrongly interpreting woody plant encroachment, e.g. as beneficial vegetation greening. Hyperspectral vegetation indices (HVIs) can be developed to accurately separate shrub cover from green vegetation. Google Earth images have been successfully used to analyse woody encroachment in South Africa. In Namibia, the so-called Bush Information System is based on synthetic-aperture radar satellite data.

Increasingly, machine-learning techniques and applications based on artificial intelligence are used to investigate woody plant encroachment.

Rephotography is found to be an effective tool for the monitoring of vegetation change, including woody encroachment and forms the basis of various encroachment assessments.

Methods to overcome the limited availability of photographic evidence or written records include the assessment of pollen records. In a recent application, vegetation cover of the past 130 years in a woody plant encroachment area in Namibia was established.

Vegetation mapping tools developed for the use by individual land users and support organisations include the American Rangeland Analysis Platform, and the Namibian Biomass Quantification Tool.

Restoration
Brush control is the active management of the density of woody species in grasslands. Although woody encroachment in many instances is a direct consequence of unsustainable management practices, it is unlikely that the introduction of more sustainable practices alone (e.g. the management of fire and grazing regimes) will achieve to restore already degraded areas. Encroached grasslands can constitute a stable state, meaning that without intervention the vegetation will not return to its previous composition.

For decisions on appropriate control measures, it is essential that both local and global drivers of woody encroachment, as well as their interaction, are understood. Restoration must be approached as a set of interventions that iteratively move a degraded ecosystem to a new system state. Responsive measures, such as mechanical removal, are needed to restore a different balance between woody and herbaceous plants. Once a high woody plant density is established, woody plants contribute to the soil seed bank more than grasses and the lack of grasses presents less fuel for fires, reducing their intensity. This perpetuates woody encroachment and necessitates intervention, if the encroached state is undesirable for the functions and use of the respective ecosystems. Most interventions constitute a selective thinning of bush densities, although in some contexts also repeat clear-cutting has shown to effectively restore diversity of typical savanna species. In decision making on which woody species to thin out and which to retain, structural and functional traits of the species play a key role. Bush control measures must go hand in hand with grazing management, as both are crucial factors influencing the future state of the respective ecosystems. State and Transition Models have been developed to provide management support to land users, capturing ecosystem complexities beyond succession, but their applicability is still limited.

The restoration of degraded grasslands can bring about a wide range of ecosystem service improvements. It can therewith also strengthen the drought resilience of affected ecosystems. Bush control can lead to biodiversity improvements regardless of the predominant land use.

Types of interventions
The term bush control, or brush management, refers to actions that are targeted at controlling the density and composition of bushes and shrubs in a given area. Such measures either serve to reduce risks associated with woody plant encroachment, such as wildfires, or to rehabilitate the affected ecosystems. It is widely accepted that encroaching indigenous woody plants are to be reduced in numbers, but not eradicated. This is critical as these plants provide important functions in the respective ecosystems, e.g. they serve as habitat for animals. Efforts to counter woody plant encroachment fall into the scientific field of restoration ecology and are primarily guided by ecological parameters, followed by economic indicators.

Three different categories of control measures can be distinguished: There is an increasing focus on the carbon sequestration impact, which differs among control measures. The application of chemicals, for example, can lead to higher carbon losses than mechanical shrub thinning.
 * Preventive measures: application of proven good management practices to prevent the excessive growth of woody species, e.g. through appropriate stocking rates and rotational grazing in the case of rangeland agriculture. It is generally assumed that preventative measures are a more cost-effective method to combat woody encroachment than treating ecosystems once degradation has occurred. Certain land uses and animal species can aid in preventing woody plant encroachment, for example elephants. Research on degradation tipping points, suggests soil organic carbon and carbon isotopes as early-warning indicators in potentially encroached areas.
 * Responsive measures: the reduction of bush densities through targeted bush harvesting or other forms of removal (bush thinning).
 * Maintenance measures: repeated or continuous measures of maintaining the bush density and composition that has been established through bush thinning.

Natural bush control
The administration of controlled fires is a commonly applied method of bush control. The relation between prescribed fire and tree mortality, is subject of ongoing research. The success rate of prescribed fires differs depending on the season during which it is applied. In some cases, fire treatment slows down woody encroachment, but is unsuccessful in reversing it. Optimal fire management may vary depending on vegetation community, land use as well as frequency and timing of fires. Controlled fires are not only a tool to manage biodiversity, but can also be used to reduce GHG emissions by shifting fire seasonality and reducing fire intensity.

Fire was found to be especially effective in reducing bush densities, when coupled with the natural event of droughts. Also the combination of fire and browsers, called pyric herbivory, is shown to have positive restoration effects. Cattle can in part substitute for large herbivores. Moreover, fires have the advantage that they consume the seeds of woody plants in the grass layer before germination, therefore reducing the grasslands sensitivity to encroachment. Prerequisite for successful bush control through fire is sufficient fuel load, thus fires have a higher effectiveness in areas where sufficient grass is available. Furthermore, fires must be administered regularly to address re-growth. Bush control through fire is found to be more effective when applying a range of fire intensities over time. Fuel load and therewith the efficacy of fires for bush control can reduce due to the presence of herbivores.

Long-term research in the South African savanna found that high-intensity fire did reduce encroachment in the short-term, but not in the mid-term. In a cross-continental collaboration between South Africa and the US, a synthesis on the experience with fire as a bush control method was published.

Rewilding ecosystems with historic herbivores can further contribute to bush control.

Variable livestock grazing can be used to reduce woody encroachment as well as re-growth after bush thinning. A well documented approach is the introduction of larger herds of goats that feed on the wood plants and thereby limiting their growth. There is evidence that some rural farming communities have used small ruminants, like goats, to prevent woody plant encroachment for decades. Further, intensive rotational grazing, with resting periods for pasture recovery, can be a tool to limit woody encroachment. Overall, the role of targeted grazing systems as biodiversity conservation tool is subject of ongoing research.

Chemical bush control
Wood densities are frequently controlled through the application of herbicides, in particular arboricides. Commonly, applied herbicides are based on the active ingredients tebuthiuron, ethidimuron, bromacil and picloram. In East Africa, first comprehensive experiments on the effectiveness of such bush control date back to 1958–1960. There is however evidence that applied chemicals can have negative long-term effects and effectively prevent the recruitment of desired grasses and other plants. The application of non-species-specific herbicides is found to result in lower species richness than the application of species-specific herbicides. Further, aboricide application can negatively affect insect populations and arthropods, which in turn is a threat for bird populations. Scientific trials in South Africa showed that the application of herbicides has the highest success rate when coupled with mechanical bush thinning.

Mechanical bush control
Cutting or harvesting of bushes and shrubs with manual or mechanised equipment. Mechanical cutting of woody plants is followed by stem-burning, fire or browsing to suppress re-growth. Some studies find that mechanical bush control is more sustainable than controlled fires, because burning leads to deeper soil degradation and faster recovering of shrubs. Bush that is mechanically harvested is often burnt on piles, but can also serve as feedstock for value addition, including firewood, charcoal, animal feed, energy and construction material. Mechanical cutting is found to be effective, but requires repeat application. When woody branches are left to cover the degraded soil, this method is called brush packing. Some forms of mechanical woody plant removal involve uprooting, which tends to lead to better results in terms of the restoration of the grass layer, but can have disadvantages for chemical and microbiological soil properties.

Economics
As woody encroachment is often widespread and most rehabilitation efforts costly, funding is a key constraint. In the case of mechanical woody plant thinning, i.e. the selective harvesting, the income from downstream value chains can fund the restoration activities.

An example of highly commercialised encroacher biomass use is charcoal production in Namibia. There are also efforts to use encroaching woody species as source of alternative animal fodder. This involves either making use of the leaf material of encroaching species,    or milling the entire plant.

In the same vein, the World Wildlife Fund has identified invasive and encroaching plant species as a possible feed stock for Sustainable Aviation Fuel in South Africa.

Also Payment for Ecosystem Services and specifically Carbon Credits are increasingly explored as a funding mechanism for the control of woody encroachment. Savanna fire management is found to have potential to generate carbon revenue, with which rangeland restoration in Africa can be funded.

Challenges
Grassland restoration has generally received less attention than forest restoration during recent decades.

Literature emphasises that a restoration of woody plant encroachment areas to a desired previous non-encroached state is difficult to achieve and the recovery of key-ecosystem may be short-lived or not occur. Intervention methods and technologies must be context-specific to achieve their intended outcome. Current efforts of selective plant removal are found to have slowed or halted woody encroachment in respective areas, but are sometimes found to be outpaced by continuing encroachment. A meta-analysis of 524 studies on ecosystem responses to both encroachment and the removal of woody plants, finds that most efforts to restore the respective ecosystems fail, while the success rate predominantly depends on encroachment stage and plant traits. It was further found that different control methods have different effects on specific ecosystem services. For example, mechanical removal of woody plants can enhance forage value, while reducing hydrological regulation. In contrast, chemical removal can enhance hydrological regulations at the expense of plant diversity. This implies that there are trade-offs to be considered for each set of control measures.

When bush thinning is implemented in isolation, without follow-up measures, grassland may not be rehabilitated. This is because such once-off treatments typically target small areas at a time and they leave plant seeds behind enabling rapid re-establishment of bushes. A combination of preventative measures, addressing the causes of woody plant encroachment, and responsive measures, rehabilitating affected ecosystems, can overcome woody plant encroachment in the long-run.

In grassland conservation efforts, the implementation of measures across networks of private lands, instead of individual farms, remains a key challenge. Due to the high cost of chemical or mechanical removal of woody species, such interventions are often implemented on a small scale, i.e. a few hectares at a time. This differs from natural control processes before human land use, e.g. widespread fires and vegetation pressure by free roaming wildlife. As a result, the interventions often have limited impact on the continued dispersal and spread of woody plants. For this reason, a key strategy developed in Northern America is termed "defending the core". It involveds the systematic expansion of healthy areas of grasslands to the outside, i.e. thinning of bush stands at the perimeter.

Countering woody encroachment can be costly and largely depends on the financial capacity of land users. Linking bush control to the concept of Payment for ecosystem services (PES) has been explored in some countries.

Managing the woody cover alone does not guarantee productive ecosystems, as also the cover and diversity of desired grass species must form part of the management considerations.

National carbon accounting and related tradeoffs
Grassland conservation can make a significant contribution to global carbon sequestration targets, but compared to sequestration potential in forestry and agriculture, this is still insufficiently explored and implemented. Detailed accounting for the effect of woody encroachment on global carbon pools and fluxes is unclear. Given scientific uncertainties, it varies widely how countries factor woody encroachment and the control thereof into their national Greenhouse Gas Inventories.

In early carbon sink quantifications, woody encroachment was found to account for as much as 22% to 40% of the regional carbon sink in the USA. In the US, woody encroachment is however seen as a key uncertainty in the US carbon balance. The sink capacity is found to decrease when encroachment has reached its maximum extent. Also in Australia, woody encroachment constitutes a high proportion of the national carbon account. Australia's carbon plan is however criticised for ignoring the carbon potential of the soil, which in drylands is found to be seven to one hundred times larger than that of vegetation. In South Africa, woody encroachment was estimated to have added around 21.000 Gg CO2 to the national carbon sink, while it has been highlighted that especially the loss of grass roots leads to losses of below-ground carbon, which is not fully compensated by gains of above-ground carbon.

It is suggested that the classification of encroached grasslands and savannas as carbon sinks may often be incorrect, underestimating soil organic carbon losses. Beyond difficulties to conclusively quantify the changes in carbon storage, promoting carbon storage through woody encroachment can constitute a trade-off, as it may reduce biodiversity of savanna endemics and core ecosystem services, like land productivity and water availability.

Several tradeoffs must be considered in land management decisions, such as a possible carbon-biodiversity tradeoff. It can have severe negative consequences, if woody encroachment or the invasion of alien woody species, is accepted and seen as a way to increase ecosystem CO2 sink capacities. In its 2022 Sixth Assessment Report, the Intergovernmental Panel on Climate Change (IPCC) identifies woody encroachment as a contribution to land degradation, through the loss of open ecosystems and their services. The report further stipulates that while there may be slight increases in carbon, woody encroachment at the same time masks negative impacts on biodiversity and water cycles and therewith livelihoods.

Carbon focused restorations approaches remain vital and can be balanced with the need to enhance other ecosystem services through spatially mixed management strategies, leaving encroached patches and in thinned areas.

Conflicting climate change mitigation measures
Woody encroachment can be exacerbated when affected ecosystems become the target of misguided afforestation. It is found that grasslands are frequently misidentified as degraded forests and targeted by afforestation efforts. According to an analysis of areas identified to have forest restoration potential by the World Resources Institute, this includes up to 900 million hectares grasslands. In Africa alone, 100 million hectares of grasslands are found to be at risk by misdirected afforestation efforts. Among the areas mapped as degraded forests are the Serengeti and Kruger National Parks, which have not been forested for several million years. Over half of all tree-planting projects in Africa are implemented in savannah grasslands.

Research in Southern Africa suggests, that tree planting in such ecosystems does not lead to increased soil organic carbon, as the latter is predominantly grass-derived. Also the Intergovernmental Panel on Climate Change (IPCC) states that mitigation action, such as reforestation or afforestation, can encroach on land needed for agricultural adaptation and therewith threaten food security, livelihoods and ecosystem functions.

Encroachment control as adaptation measure
Some countries, for example South Africa, acknowledge inconclusive evidence on the emissions effect of bush thinning, but strongly promote it as a means of climate change adaptation. Geographic selection of intervention areas, targeting areas that are at an early stage of encroachment, can minimise above-ground carbon losses and therewith minimise the possible trade-off between mitigation and adaptation. The Intergovernmental Panel on Climate Change (IPCC) reflects on this trade-off: "This variable relationship between the level of encroachment, carbon stocks, biodiversity, provision of water and pastoral value can present a conundrum to policymakers, especially when considering the goals of three Rio Conventions: UNFCCC, UNCCD and UNCBD. Clearing intense woody plant encroachment may improve species diversity, rangeland productivity, the provision of water and decrease desertification, thereby contributing to the goals of the UNCBD and UNCCD as well as the adaptation aims of the UNFCCC. However, it would lead to the release of biomass carbon stocks into the atmosphere and potentially conflict with the mitigation aims of the UNFCCC." The IPPC further lists bush control as relevant measure under ecosystem-based adaptation and community-based adaptation.

Websites

 * Stockholm Resilience Centre – Regime Shifts DataBase: Bush Encroachment
 * Panorama.Solutions – Rangeland Restoration through Bush Control
 * The Rangelands Partnership – Global Rangelands Portal
 * Wrangle – World Rangeland Learning Experience

Articles

 * Rural 21 Magazine – Namibia's bush business
 * Daily Maverick: Biological vandalism— the world's wild savannas may be doomed, but few pay attention