Bush Encroachment

by

Timothy Gordon O'Connor, Nicola Stevens

• LAST REVIEWED: 24 May 2017
• LAST MODIFIED: 24 May 2017
• DOI: 10.1093/obo/9780199363445-0069

Introduction

Savannas are a significant biome within the tropical and subtropical regions of Africa, Australia, North America, South America, and Southeast Asia including India. On a global scale, savannas encompass a rainfall regime from 300 to 2000 millimeters per annum, although the majority of area considered savanna is on the drier end of this spectrum. They are typically defined as an admixture of a near-continuous grass layer and a discontinuous tree layer. Grasses are mostly perennial and possess a C₄ photosynthetic pathway. Savanna vegetation is used mostly for support of livestock production but also underpins much of the protected area network of Africa and Asia that supports iconic large mammal wildlife consisting of grazers, browsers, and mixed feeders. The “balance” or coexistence between trees and grasses is a pivotal concept for understanding and use of savanna. Interest in the relation between these two components was founded in early observation that livestock production declined when woody plants increased at the expense of grasses. “Bush encroachment” is a colloquial term used to describe this process in Africa that elsewhere may be described as woody thickening, woody encroachment, brush encroachment, or an increase in woody cover. The term also now encompasses invasion of open tropical or subtropical C₄ true grassland by indigenous woody species. Similar invasion of vegetation such as tundra by woody species, as well as an increase of alien (exotic) woody plants in savanna or grassland, has not been considered in this article, although many of the concepts and issues may prove to be similar. Bush encroachment is a vegetation dynamic of global importance that may result in a biome change with ramifications for the organization, functioning, and use of affected ecosystems. A loose framework of “Changes, Causes, Consequences, and Control” has been adopted for summary of its state of knowledge. Key questions addressed are the extent to which woody cover has increased across the savannas and tropical grasslands of the globe, whether similar agents and processes have been responsible for change across space and over time, what the nature and scale of effects on ecosystem organization and functioning have been, whether there should be a response to this dynamic, which responses are effective, and what consequences they in turn may have on system functioning?

Global Extent and Purported Drivers of Encroachment

A starting point for addressing a dynamic of global interest is to assess the extent, rate, and pattern of change over a large spatial scale and to identify likely drivers of change, which requires a remote-sensing approach. Three approaches used since bush encroachment was recognized early in the 20th century are lateral photographs, aerial photographs, and satellite imagery. Lateral photographs provide a record of landscape-level change since around the 1870s. Shantz and Turner 1958 captured the imagination of the ecological world with their repeat photography showing that encroachment had occurred within many African savannas or grasslands. Using this approach, Rohde and Hoffman 2012 achieved quantitative rigor for assessing the influence of global (climate, atmospheric [CO₂]) and local (e.g., fire, grazing) drivers of change. Use of this approach elsewhere has been limited. Aerial photography has been widely employed since the 1930s for quantitatively describing the extent of change using first desk- then computer-based cartographic methods usually of a study landscape. Roques, et al. 2001 and Asner, et al. 2003 compare adjacent land uses in order to identify local drivers responsible for change. Directional change across land uses was used by Fensham, et al. 2005 to indicate an effect of a widespread driver such as rainfall change; similarly Stevens, et al. 2016 infers an effect of increasing atmospheric [CO₂]. The emerging norm of satellite imagery with ever-improving resolution offers analysis of a vast spatial scale at a high frequency, offset only by a limited length of record. Use of imagery was initially of individual locations, but Skowno, et al. 2016 conducts a national-level analysis. Most satellite-based studies assess woody cover using NDVI, but Tian, et al. 2017 shows that new methods can estimate wood biomass. Meta-analysis conducted at a regional level by O’Connor, et al. 2014 and at a cross-continental level by Stevens, et al. 2017 has broadened perspective of purported drivers of encroachment, although the remote-sensing study Tian, et al. 2017 presents a global perspective for tropical drylands. Although the study illustrates that encroachment is occurring across C₄ grassy ecosystems usually driven by multiple agents, it also highlights emerging regional patterns of encroachment that need to be interpreted in the context of evolutionary history, climate, historical and current land management, and the abiotic environment.

• Asner, G. P., S. Archer, R. F. Hughes, R. J. Ansley, and C. A. Wessman. 2003. Net changes in regional woody vegetation cover and carbon storage in Texas drylands, 1937–1999. Global Change Biology 9:316–335.

  DOI: 10.1046/j.1365-2486.2003.00594.x

  Using a combination of historical aerial photography, field surveys, and remote sensing, the trajectory of woody cover change is measured across 400 km² of semi-arid rangeland in northern Texas. When integrated across the regional landscape, woody cover increased by 30 percent over a sixty-three-year period.

• Fensham, R. J., R. J. Fairfax, and S. R. Archer. 2005. Rainfall, land use and woody vegetation cover change in semi-arid Australian savanna. Journal of Ecology 93:596–606.

  DOI: 10.1111/j.1365-2745.2005.00998.x

  Another example of insight derived from analysis of aerial photographs. Increases in woody cover of over storey and understorey of an extensive area of Australian savanna over four decades was positively related to rainfall and negatively related to starting amount of woody cover with no effect of fire or grazing management.

• O’Connor, T. G., J. R. Puttick, and M. T. Hoffman. 2014. Bush encroachment in southern Africa: Changes and causes. African Journal of Range and Forage Science 31:67–88.

  DOI: 10.2989/10220119.2014.939996

  A summary of the extent of change recorded for southern Africa based on twenty-three landscape-level studies, with an accompanying review of purported ultimate and proximate causes most relevant to southern Africa. Use of results gained from long-term experiments featured prominently, illustrating consideration of environmental context (climate, soils) is imperative for addressing this dynamic.

• Rohde, R. F., and M. T. Hoffman. 2012. The historical ecology of Namibian rangelands: Vegetation change since 1876 in response to local and global drivers. Science of the Total Environment 416:276–288.

  DOI: 10.1016/j.scitotenv.2011.10.067

  A regional contemporary example of vegetation change including bush encroachment is presented for a semi-arid African country based on lateral photographs. The paper shows that rigorous quantitative analysis of land use as a purported driver of change is feasible if there is sufficient number of photographs over a time span that other tools cannot match.

• Roques, K. G., T. G. O’Connor, and A. R. Watkinson. 2001. Dynamics of shrub encroachment in an African savanna: Relative influences of fire, herbivory, rainfall and density dependence. Journal of Applied Ecology 38:268–280.

  DOI: 10.1046/j.1365-2664.2001.00567.x

  Based on changes in woody cover on aerial photographs over forty-three years, this study offers a quantitative analysis of the relative contribution of purported agents driving bush encroachment, illustrating their interdependent and contingent effects. The study capitalized on rainfall variation over time and the juxtaposition of land uses whose fire and herbivory regimes differed.

• Shantz, H. L., and B. L. Turner. 1958. Photographic documentation of vegetational changes in Africa over a third of a century. Tempe: Univ. of Arizona Press.

  The cliché “a picture speaks a thousand words” is exemplified by this classic collection of “then and now” lateral photographs first taken c. 1920 and re-photographed in the 1950s. This publication clearly showed that woody vegetation was increasing in certain locations across Africa.

• Skowno, A. L., M. W. Thompson, J. Hiestermann, B. Ripley, A. G. West, and W. J. Bond. 2016. Woodland expansion in South African grassy biomes based on satellite observations (1990–2013); general patterns and potential drivers. Global Change Biology.

  DOI: 10.1111/gcb.13529

  Provides the first complete statement using remote sensing products of woody cover change in South Africa from 1990–2013. Over a twenty-three-year period, an increase of 27,000 km² in extent of woodland occurred. Expansion was greatest in grassy regions receiving more than 500 mm (0.31 percent y^-1). Conservation areas with elephants present were the only examples of land use in which a decline in woodland cover took place.

• Stevens, N., B. F. N. Erasmus, S. Archibald, and W. J. Bond. 2016. Woody encroachment over 70 years in South African savannahs: Overgrazing, global change, or extinction aftershock? Philosophical Transactions of the Royal Society B 371:20150437.

  DOI: 10.1098/rstb.2015.0437

  Measurement of woody plant encroachment across four land uses in South Africa between 1940 and 2010 revealed a doubling of woody cover across a stark rainfall gradient for all land uses except in low rainfall savannas with elephants present. The authors conclude that a global factor, most likely elevated atmospheric [CO₂], may be driving encroachment.

• Stevens, N., C. E. R. Lehmann, B. P. Murphy, and G. Durigan. 2017. Savanna woody encroachment is ubiquitous across three continents. Global Change Biology 23:235–244.

  DOI: 10.1111/gcb.13409

  This cross-continental meta-analysis of increases in woody cover revealed rapid increases of sub-humid savannas in Brazil likely owing to fragmentation and fire suppression, a limited but consistent rate of encroachment of Australian savanna, which was a third of the accelerating rate of encroachment of African savanna. Pattern of encroachment across Australia and Africa suggest rainfall, changing land uses and rising atmospheric [CO₂] as likely causes.

• Tian, F., M. Brandt, Y. Y. Liu, and R. Fensholt. 2017. Mapping gains and losses in woody vegetation across global tropical drylands. Global Change Biology 23:1748–1760.

  DOI: 10.1111/gcb.13464

  The paper reports a global effort. Above-ground wood biomass of tropical drylands was estimated using vegetation optical depth derived from dry-season satellite imagery between 2000 and 2012. Wood biomass increased over large areas (22.7 percent) in the Sahel, Namibia, South Africa, Texas and northern Mexico, and eastern Australia but decreased (13.3 percent) within the Gran Chaco, Western Australia, and eastern sub-equatorial Africa.