Succession used to be a fundamental concept in ecology, but more recently has been overtaken by the concept of land-use change. Each concept allows us to understand different things and ask different questions.

“Succession” refers to one thing coming after another, replacing it (as in “With Edward the Confessor.. perished the last English King, as he was succeeded by Waves of Norman Kings, Tudors, Stuarts, Hanoverians…” [1]). The question in ecology is what are the things that come one after the other, why, and in what order. Although, in fact, numerous answers were given to these questions based on observations from different fieldwork methods and study sites, successional theory has generally been divided into two schools, Clemensian and Gleasonian. Clements is associated with the idea of linear and predictable transitions of plant communities towards a fixed climax state. He observed that plants modify their environments (for example by creating shade) in ways that do not favour their own reproduction, but do favour the reproduction of another community of plants, which then replaces them. Gleason is now associated with the idea that individual plant species move in and out of communities of species and habitat areas independently, showing up in places that fit their environmental needs. Most species distribution modelling, which is used to understand how species will adapt to climate change, or be affected by land-use changes, is based on Gleasonian assumptions (e.g. Kattwinkel et al. 2009). Much fieldwork looking at restoration ecology, by contrast, has found Clementsian plant-plant interactions leading to forest succession or allowing forest restoration (e.g. Root-Bernstein et al. 2017). At the same time, the Clementsian ‘climax’ concept has been severely critisized—if there is an ‘end’ to the successional sequence it isn’t more stable or longer-lasting than any of the other steps (e.g. Willis et al. 2004).

Both Gleasonian and Clementsian concepts of succession have been left behind and largely replaced by ecosystem-specific models of grass-acacia competition, arid-land degradation, or shrub encroachment, drawing on mathematical and simulation modelling and concepts such as historicity (the immediate past affects the immediate future), tipping points and thresholds of accumulated change at which a place shifts irreversibly from one ecosystem type to another, and, simply, outcomes of multi-factorial interactions involving climate, soil, competition, disturbance, and herbivory, for example. This new repertoire of multifactorial interactions and hard-to-predict system-level dynamics is not usually refered to as “succession.” This seems to be in part because the aspiration to have a complete characterization of environmental change over time has been lost. The new concepts question whether we can even ever answer why one ecosystem or habitat follows another and in what order, since randomness, feedbacks, and emergence (a new phenomenon that emerges at a large scale from the interactions of other things at a smaller scale) are the kinds of mechanisms at play. These mechanisms do not allow for predictive what order or intepretive why answers. Ecosystems emerge dynamically and transiently from small-scale interaction rules. Finally, replacing the question of what things come one after the other we have the new concept of land-use and land-use change.

Land-use change was introduced to the universe of ecological concepts via the Kyoto Protocol, which was adopted in 1997 and came into force in 2005 (https://en.wikipedia.org/wiki/Kyoto_Protocol). The Kyoto Protocol is a set of activities that states were supposed to engage in to limit or reduce climate change. These activities included monitoring Land Use, Land Use Change and Forestry (LULUCF), and detecting and quantifying afforestation, reforestation, and degradation (ARD). Monitoring and quantifying LULUCF is carried out using different kinds of satellite sensors, which are then interpreted using image analysis tools and GIS (geographical information system) mapping analysis. Even as the resolution in space and frequency in time of different kinds of satellite data has increased since the late 1990s, this kind of data limits the kinds of questions that can be asked. Satellite data gives a view from above of a slice in time: we can ask questions about land cover class rather than species presence or composition (since most species cannot be detected from above), and about states rather than processes. Although research using these technologies, analytical methods and framework goes far beyond reporting to the Kyoto Protocol, the key questions that are asked remain the one important to Kyoto: “which land cover class is getting bigger and which is getting smaller?” This can be explained with reference to other compatible data, such as large-scale climate data or mappable features such as distance to roads or elevation. There is, however, no theory about why some land cover classes get bigger or smaller, or about what constitutes a land-cover class to begin with. A land-cover class is something that your image analysis programme can recognise reliably. They are things like “forest” “grassland” “arable land” “urbanized surface” “water”. The answer to why “grassland” becomes “arable land” is always “anthropogenic activity” and the reverse is always due to the cessation of anthropogenic activity. This analytical framework has also proved a good fit for modelling things like how ecosystem services change with different hypothetical landscape organizations of land-uses (e.g. Bateman et al. 2013).

Some authors (Prach & Walker 2011) argue that succession is still important for understanding contemporary issues in ecology, including biodiversity loss, climate change, invasive species and restoration. Considering these issues through the lens of succession can help to formulate questions that elucidate species-species interaction mechanisms, recognise varieties of community formation, and interpret and predict possible successional trajectories, which together, they argue, can allow for properly targeted and contextualized conservation interventions.

Another approach that has been moving along slowly for many years is the IUCN Red List of Ecosystems (https://www.iucn.org/theme/ecosystem-management/our-work/red-list-ecosystems). The IUCN maintains a Red List of species, their state of conservation and risk of extinction. The Red List of Ecosystems uses a similar framework and procedure to assess the risk of collapse of ecosystems (https://iucnrle.org/static/media/uploads/references/key-documents/Red%20List%20of%20Ecosystems%20Criteria%20Summary%20Sheet/summary_rle_categorie&criteria_2.2_en.pdf). There are very few published reports to date, but they appear to use a range of data sources, including botanical records, environmental histories, and expert opinions. In their paper explaining and justifying the Red List of Ecosystems (https://iucnrle.org/static/media/uploads/references/key-documents/scientific-foundations/keith-etal-2013-scientific-foundations-red-list-ecosystems-en.pdf), the authors suggest that ecosystems as units are unambiguously relevant to ecology and conservation, unlike ecosystem services or proxy data such as functional trait databases. The loss of an ecosystem is clearly bad, they argue. They define ecosystems following Tansley, one of the early ecologists interested in succession, as “i) a biotic complex or assemblage of species; ii) an associated abiotic environment or complex; iii) the interactions within and between those complexes; and iv) a physical space in which these operate.” There is no master list of ecosystems, which are defined according to their characteristic species, abiotic environment, interactions, and distribution. They note that unlike with species extinction, “ecosystems may not disappear, but rather transform into novel ecosystems with different characteristic biota and mechanisms of self-organisation”. I find this formulation rather odd—the ‘may’ is science-speak for ‘according to some people but not established as a general fact’. What exactly could take the place (literally) of an ecosystem that wasn't another ecosystem is not entirely clear (unless we rule out anthropogenic habitats as ecosystems, but how anthropogenic habitats work in this scheme is not fully clear either). In any case, the goal of the Red List is to assess how close the ecosystem is to collapse, in order to conserve or restore it before it disappears—not to understand what it would be replaced with or why. On the one hand, this approach is interesting because it is more sophisticated than LULUCF, in the sense of being more sensitive to ecology and incorporating more forms of data and knowledge. On the other hand, it seems to be unduly concerned with ecological stasis, and with the maintenance of the ecosystems we find now rather than those that may have already disappeared, or those that might be adaptations to future climte change, for example (‘novel ecosystems’). It is unclear how dynamic processes and cycles of ecosystem change can be incorporated into the ecosystem assessments—though the process also seems sufficiently flexible to potentially be able to consider such issues.

Cities are involved in ecosystem change, whether we think of it as succession, land-use change, or ecosystem collapse. We will look at that in the next blog post.

--Meredith Root-Bernstein 30/9/2019

[1] Sellar & Yeatman. 1930. 1066 and all that. Methuen. N.B. This book is highly recommended, but has nothing to do with urban ecology.

References:

Bateman, I. J., Harwood, A. R., Mace, G. M., Watson, R. T., Abson, D. J., Andrews, B., ... & Fezzi, C. (2013). Bringing ecosystem services into economic decision-making: land use in the United Kingdom. science, 341(6141), 45-50.

Kattwinkel, M., Strauss, B., Biedermann, R., & Kleyer, M. (2009). Modelling multi-species response to landscape dynamics: mosaic cycles support urban biodiversity. Landscape Ecology, 24(7), 929-941.

Prach, K., & Walker, L. R. (2011). Four opportunities for studies of ecological succession. Trends in Ecology & Evolution, 26(3), 119-123.

Root‐Bernstein, M., Valenzuela, R., Huerta, M., Armesto, J., & Jaksic, F. (2017). Acacia caven nurses endemic sclerophyllous trees along a successional pathway from silvopastoral savanna to forest. Ecosphere, 8(2), e01667.

Willis, K. J., Gillson, L., & Brncic, T. M. (2004). How" virgin" is virgin rainforest?. science, 304(5669), 402-403.