The Gap between Theory and Practice of “Ecological Thresholds” and “Alternative Stable States” in Restoration

Restoration ecology is a growing science within the dynamic of the ecosystem itself. Learning concepts about “ecological thresholds” and “alternative stable states” will hugely contribute to ecological restoration outcomes. However, some scientists believe that the restorationists can not apply these concepts uniformly in various ecosystems with different characteristics. Therefore, this essay notably critically reviews the gap between the theory of “ecological thresholds” and “alternative stable states” and their practices in restoration.

Restoration ecology is experiencing dramatic growth as an academic science. The society for Ecological Restoration (SER) (2004) defined ecological restoration as “an intentional activity that initiates or accelerates the recovery of an ecosystem with respect to its health, integrity and sustainability” (p.1) and “the process of assisting the recovery of an ecosystem that has been degraded, damaged, or destroyed” (p.3). The goal of restoration is to return the ecosystem to its historic trajectory and re-establish the pre-existing biotic, including community structure and species composition. These goals differentiate restoration from rehabilitation which only notices ecosystem processes, productivity and services reparation.
Whereas restoration ecology is a science, the practise is called ecological restoration or restoration practise. Ideally, restoration ecology should provide good concepts, models, tools, and methodologies for restorationists to support them in practices. Sometimes, the restoration ecologist is also the practitioner, and vice versa.
Nowadays, identifying and understanding the concepts of “ecological thresholds” and “alternative stable states” are important to restoration practices. A natural ecosystem, which tends to be complex, nonlinear, uncertain and often unpredictable, causes different responses of disturbances and cannot be predicted easily. These concepts could be used to describe the changes in natural environment, and as a result, could give wider knowledge for restoration stakeholders for achieving their goals. Recently, the use of the concepts are increasing and being involved to policy decisions. For example, the concepts have been applied to discuss issues about natural resource and public land management in the United States Senate since 2003 (Bestelmeyer, 2006).

The idea of “thresholds” has been acknowledged among the scientists since the 18th century (Hugget, 2005). To illustrate, if the temperature is lower than 0° C, water can be turned into ice. The temperature is a threshold which causes water (a liquid) to change to ice (a solid), illustrating a transition between different states (Bennet and Radford, 2003).
Since thirty years ago, the ecologists have applied “thresholds” in ecosystems, such as forests, agricultures, freshwaters, and grasslands. They want to use “thresholds” as a major concept for the development of ecological conservation and sustainable natural resources management (Hugget, 2005).
There are several definitions that have been used to identify ecological thresholds. A world-wide collective research scientists from social to ecological subjects, The Resilience Alliance, has attempted to list examples of ecological thresholds. One of their definition is a threshold is a point that divides alternate states which when passed causes a system change to a significantly different state (Meyers and Walker, 2003, in Hugget, 2005).
In many other ecological thresholds definition, the notice of change in state which happens around specific points or zones has been common. For example, Friedel (1991) in Hugget (2005) described thresholds as “boundaries in space and time between two different states” (p.2). Muradian (2001) in Hugget (2005) viewed threshold as a variable which causes a change from one stable state to another independently. The threshold is closely connected to ecological discontinuities which occur to respond to simultaneous change in an independent variable.
According to Hugget (2005), the most practical definition of an ecological threshold was developed by Bennet and Radford (2003), who described an ecological threshold in natural systems as “a point at which relatively rapid change occurs from one ecological condition to another” (p.1). Small changes in influential factors can cause large responses in nature that has more complex systems than what they may seem (Bennet and Radford, 2003).
Another theory that relates to restoration ecology is the alternative stable states. This theory is based on the idea that ecosystems in the natural world are dynamic uncertain and unpredictable. As a result, frequent and continual disturbances force ecosystems to go through alternative unpredictable responses rather than remain in its original condition (Hobbs & Norton, 2004). These disturbances constantly change the assemblages and patterns of the previous environment (Fiedler et al. 1997, Pickett et al. 1992, Pickett and White 1985, Sousa 1984 in Hobbs and Norton 2004). This view is known as nonequilibrium paradigm, in which alternative stable states may exist and the endpoints after disturbances are unpredictable (Hobbs and Norton, 2004). Unpredictable course of succession happens because of the contingency of the disturbances and the naturalness of the biophysical condition.
Suding et al. (2004) describe some state combinations, which happened continually at a specific spatial extent and temporal scale, are alternative stable states. These can be defined as “multiple (alternative) basins of attraction within a system. In such a system, a given habitat or environment would be able to support two or more different assemblages of species, and these assemblages would be stable (self-replacing)” (Suding and Hobbs, 2009, p.1). A relationship between thresholds and alternative stable states in ecosystems is described in Figure 1 (Hobbs and Norton, 2004).

Examples of ecological thresholds and alternative stable states concepts in ecosystems
To describe ecological thresholds and alternative stable states theories, Bestelmeyer (2006) took an example in a transition from savanna state to a shrub-encroached woodland state. This change occurred because of continuous grazing and drought, which caused a gradual loss of the grass. As a result, there were the lack of fuel connectivity and fire disturbance. The shrubs began to dominate the area replacing the grasses as a primary species. The shift to response fire disturbance, grazing and competition between shrub and grass is known as a biotic threshold. Recovering to an original state across the threshold would be done by reducing overgrazing and removing shrub. However, shrub dominance could increase the rate of erosion and soil degradation. The soil degraded gradually over time to a point that even restoration efforts, such as grass seeding and shrub removing, could not recover to a primary state. This condition was marked by a point as an abiotic threshold (Whisenant, 1999, in Bestelmeyer, 2006).
Laycock (1991) used an example of sagebrush-grass vegetation in North America to illustrate both concepts. The original environment probably consisted of a productive understory of grasses and forbs (Laycock, 1978 in Laycock, 1991). The amount of original sagebrush in the local area would have been reduced by natural fire triggers, which happened periodically. Young et al. (1979) in Laycock (1991) noted that when domestic livestock was introduced in large numbers in the late 19th century, overgrazing reduced the palatable herbaceous plants. The grazing pressure during the short growing season decreased the understory. On the other hand, the amount of sagebrush increased because this particular species, since the Pleistocene, had not been subjected to herbivore grazing pressure. Thus, the ecosystem was turned into a new state which was dominated by sagebrush, across the natural fire as the threshold trigger.
Similar example of both theories with fire as a threshold, can be illustrated in Jameson’s example of pinyon juniper woodlands (Jameson, 1987, in Laycock ,1991). In the subalpine zone in Colorado, there was a forest dominated by Englemann spruce (Picea englemani Parry). After a fire, bristlecone pine (Pinus aristata Engelm.) was found growing on the livestock grazing area. However, Englemann spruce started to establish again on different area, ungrazed land inside a city watershed (Laycock, 1991). This means the forest ecosystem changed into new state with different dominated species across the fire as the threshold.

Application of ecological thresholds and alternative stable states concepts in restoration

Jeppesen et al. (1990) evaluated fish manipulation as a restoration tool in lake ecosystems. They did empirical studies to find the link among the concentration of phosphorus (P) in lake water, the phytoplankton composition, the cover of submerged macrophytes and fish stock. These studies were combined with the fish manipulation experiments in shallow, eutrophic, and temperate Danish lakes. The results found a threshold level for long term effects in shallow and temperate lake > 10 ha is 100 µg P 1-1 and supported the hypothesis of alternative stable state from clear or turbid water stages. However, the threshold level maybe higher in smaller lakes, < 3 ha, due to more suitable conditions for fish and submerged macrophytes. In lakes with the higher level of P concentration and nutrient levels than the threshold level, removing fish will cause temporary changes in trophic structure. The changes will become the most noticeable in shallow and green algal lakes. On the other hand, the shifts will be less noticeable in lakes with heavy blooms of large cyanobacteria in summer. If fish manipulations are simultaneously repeated, the changes of trophic structure may only happen in the long term. To summarize, Jeppesen et al. (1990) said that as a restoration tool, fish manipulation is likely to be more efficient in shallow lakes rather than deeper ones.



The concept of ecological threshold and alternative stable states are very important for restoration ecology because of three reasons. To begin with, these concepts would be considered as important equipments for many ecological restorationists to identify, understand, and anticipate the dynamic factors. As a result, the restorationists could manage challenges and make valuable options for their successful restoration goals (Bestelmeyer, 2006). Restoration thresholds, for many ecosystems, exist because of human activities which “prevent the system from returning to a less degraded state without the input of management, restoration, and aftercare” (Hobbs and Norton, 2004, p.74). Thresholds usually appear because of two main factors: “one caused by biotic interactions and alterations and the other by abiotic alterations, transformations, or inherent limitations” (Whisenant, 1999, in Hobbs and Norton, 2004, p.75). Different factors mean different efforts. For example, if the causes are abiotic factors, the restorationists should focus on repairing the physical and/or chemical environment as well as removing the degradation factors. On the other hand, if the change happens because of biotic factors, the restorationists should concentrate on removing the degradation factors (biotic manipulations) (Hobbs and Norton, 2004). Reversing ecosystems which have been accrossed biotic thresholds are much more difficult than abiotic ones. The restorationists should correct and maintain the system function before considering biotic composition and structure. This will provide a beneficial framework for their initial assessment (Ludwig et al., 1997; Tongway and Ludwig 1996 in Hobbs and Norton, 2004).

Second, ecological threshold and alternative stable states models could help restorationists to prioritize restoration efforts in their management areas (Suding et al., 2004). The low priority will be given to areas towards abiotic thresholds resulting difficult restoration efforts. Similar priority will be considered to areas that do not indicate degradation although have crossed biotic thresholds (Bestelmeyer, 2006). Finally, there is an increasing awareness about the possibility of threshold behaviour among restoration stakeholders. They realize that good understanding between biotic and abiotic factors can be a pathway to successful restoration management. Feedback between these factors constrains restoration management efforts. Understanding alternative stable states that incorporate ecological thresholds in degraded systems will considerably help to disrupt feedback and respond constraints, especially in dynamic ecosystems. Models of alternative stable states can be used to restore systems which have collapsed to a degradation state through providing an analytical framework for developing innovative management (Suding et al., 2004). In the degraded systems, the restorationists might consider manipulate more than a factor or process.

Problems and Solutions

Even though the concept of ecological threshold and alternative stable states are becoming common in restoration application, there are some problems. In the past, threshold models tended only to be applied in semi arid rangelands and lake ecosystems, but now, they are more common in varied ecosystems (Suding and Hobbs, 2009). Recently, these models were applied uniformly without considering interacting elements, such as human impacts which can introduce new threshold triggers to ecological resilience by altering disturbances in particular systems, such as fires, and changing climatic extremes. There is also the question of whether threshold models can or cannot be applied in environmental management problem solving, although it is a useful theory to provide a beneficial framework about ecosystem function (Groffman et al., 2006).

The gap between theory and application could be because of incomplete knowledge, inadequate tools to evaluate evidence, and unpredictable threshold behaviour in a given management situation. To bridge these gaps, Van de Ven and Johnson (2006) suggested solutions such as improving knowledge transfer and knowledge product with the “engaged scholarship” method. Both researchers and practitioners can cooperate to produce advanced knowledge which combines theory and application.

For implementing thresholds in practice, Hugget (2005) suggested four ways. To begin with, ecological thresholds knowledge may be applied to define species sensitivity into threatened processes, for instance, genetic diversity loss, habitat fragmentation, and flora and fauna pests’ invasion. If this sensitivity is known, then cost-effective plans which specifically target these kinds of thresholds able to developed and utilised. Second, the ability to address ecological thresholds in both natural and human modified ecosystems can assist the restorationists to understand and cope well with the biodiversity conservation and production balance. As a result, the restorationists can accurately develop the strategies of management.  Next, the threshold knowledge will help the managers to set the habitat retention and restoration targets, and possibly landscape recovery. For example, Hugget (2004) in Hugget (2005) addressed potential thresholds, in the northern wheat belt of Western Australia, for remnant area and condition and also habitat patch size and isolation for sedentary declining birds. These identifications were being used by the Department of Conservation and Land Management in Western Australia into a catchment based management plan. Finally, the knowledge can be applied to develop designs for biodiversity conservation landscape, especially in fragmented landscapes. The thresholds can also be identified to help prioritise restoration actions and optimise the use of involving resources in developing and implementing management designs.

Suding and Hobbs (2009) believed that, to be applicable for restoration efforts, threshold models should include multiple particular interacting factors, rates of environmental change, transient dynamics, and realistic disturbance frequencies. However, threshold models could be inapplicable for all of these systems, regarding their dynamics, thus, further calculations about the incorporation of costs and benefits in management plans are necessary.

Furthermore, there is a similar view about a gap between pattern and process of the thresholds term (Bestelmeyer, 2006). For instance, this gap can be illustrated in the relation of vegetation cover and erosion. A large change in the erosion process can result from a small change in vegetation cover pattern. However, soil erosion potential and climate as the relationship between threshold pattern and process, are unlikely to be consistent in time and space.

To overcome this problem, Bestelmeyer (2006) suggested “a classification of thresholds and their relationships” (p.327) which has been illustrated in Figure 2. The figure can help to visualize the link between research and application about threshold. There are consequences to an environmental condition, which is described by the pattern threshold to its process. Even though feedback between the pattern and the process is nonlinear, the processes can have linear links with environmental condition. A degradation threshold happens when there is a specific point of change in the environmental condition. In addition, Bestelmeyer (2006) also said to use the link between the pattern, process, and degradation thresholds to define classification thresholds and address states, but it is difficult due to the lack of sufficient data on their relationship, so guessing will be the most common way for identification. Classification thresholds are divided into preventive management and restoration. Restoration, in contrast to preventive management, should identify and define the simultaneous pattern, process, and degradation thresholds. Unsuccessful restoration may occur when degradation thresholds can be addressed but the pattern and process cannot be identified.


The ecological thresholds and alternative stable states models would help the restorationists as key concepts to design plans through prioritizing ecological management changes and restoration practices. Linking the concepts with the dynamic of ecosystems will bring a broader knowledge for restoration perspective. However, these models could not be applied uniformly. If the restoration actions cost too much and if it is unclear whether a recovery threshold could be removed or not, it would be better to modify the goals or reallocate resources.

Literature cited:

Beisner, B. E., Haydon, D.T. and Cuddington, K. (2003) Alternative stable states in ecology. Frontiers in Ecology and the Environment 1(7):376-382

Bennet, A. and Radford, J., (2003) Know your ecological thresholds. Thinking Bush 2, 1–3

Bestelmeyer, B.T. (2006) Threshold Concepts and Their Use in Rangeland Management and Restoration : The Good, the Bad, and the Insidious. Restoration Ecology 14, 325-329

Groffman, P.M. et al. (2006) Ecological Thresholds : The Key to Successful Environmental Management or an Important Concept with No Practical Application ? Ecosystems 9, 1-13

Hobbs, R. J. and Norton D. A. (2004) Ecological filters, thresholds and gradients in resistance to ecosystem reassembly. Pages 72-95 in V. M. Temperton, R. J. Hobbs, T. Nuttle and S. Halle, editors. Assembly rules and restoration ecology. Island Press, Washington D.C.

Hobbs, R. J. and Norton D. A. (1996) Towards a conceptual framework for restoration ecology. Restoration Ecology 4, 93-110

Hugget, A. (2005) The concept and utility of “ecological thresholds” in biodiversity conservation. Biological Conservation 124, 301–310

Jeppesen, E. (1990) Fish Manipulation as a Lake Restoration Tool in Shallow, Eutrophic, Temperate Lakes 2 : Threshold Levels, Long-term Stability and Conclusions. Hydrobiologia 200/201, 219-227

Laycock, W.A. (1991) Stable States and Thresholds of Range Condition on North American Rangeland – A Viewpoint. Journal Range Management 44, 427-433

Society for Ecological Restoration International Science and Policy Working Group (2004) The SER International Primer on Ecological Restoration. & Tucson : Society for Ecological Restoration International

Suding, K. N., Gross K. L. and Houseman, G. R. (2004) Alternative states and positive feedbacks in restoration ecology. Trends in Ecology and Evolution 19, 46-53

Suding, K. N. and Hobbs R. J. (2009) Threshold models in restoration and conservation: a developing framework. Trends in Ecology and Evolution 24, 271-279

Van de Ven, A.H. and Johnson, P.E. (2006) Knowledge for theory and practice.  Academy of Management Review 31, 802-821