Degradation drivers, loss of ecosystem function and management for restoration

Degradation in drylands, especially when the pressure exceeded critical thresholds, implies losses of ecosystem functioning and diversity, and the capacity of the system to recover the original values of these altered properties determines the resilience of the system. In CASCADE we have observed that the degradation drivers considered in the project severely impacted, on occasions beyond recovery thresholds, ecosystem properties and services in the CASCADE field sites, with higher losses along the gradient of aridity represented by the field sites (Valdecantos and Vallejo 2015).

There are ecosystem properties such as the spatial distribution of vegetation that, when changed, may indicate overpassing or proximity to this eventual threshold. The intensity, both in terms of pressure and time, of degradation can affect the resilience of an ecosystem, hampering or even impeding the reversal. Whisenant (1999) proposed the existence of two degradation thresholds beyond which the natural recovery of ecosystem is extremely difficult or impossible. At the lower degree of pressure of the degradation driver, the first one is controlled by biotic interactions and the system still maintains the capacity to capture and retain resources and can be considered as a functional system. In these cases, it is only required an appropriate manipulation of the biotic component (mostly vegetation) to increase ecosystem function. If the pressure or degradation increases, a second threshold controlled by abiotic interactions can be exceeded, primary processes are not functional any longer, and the recovery of ecosystem functions requires the manipulation of the physical environment.

In all these cases restoration actions, acting as accelerated succession (Hilderbrand et al., 2005), should be envisaged to recover the integrity of the site although a complete restoration is not always possible without perpetual management (Lindig-Cisneros et al., 2003). However, some studies suggested that there is no evidence that the lower the functionality of a given ecosystem, the lower the restoration success or the higher the economic input needed (Cortina et al. 2006; Maestre et al. 2006).

In addition to these considerations, ecosystem management for restoration has to include the expected climate change scenarios as successful approaches in the past might not be effective in the future. Global climatic change represents an additional factor of uncertainty not only in the outcomes of forest plantations, but also on the very subsistence of current dryland landscapes. Alkemade et al. (2011) predicted that up 25% of the species currently present in natural landscapes of the Mediterranean Basin will disappear by 2100, being the Mediterranean shrublands one of the ecosystems in Europe most threatened by climate change projections.

Ecosystem functions assessed

Biodiversity represents a structural feature of ecosystems with direct influence in all other services (MA 2005). Monitoring biodiversity in different states of the ecosystem, identifying the local extinction of keystone species and the appearance of exotics, is extremely important as its changes may have irreversible consequences in ecosystem goods and services (Hooper et al., 2005). Restoring biodiversity and maximizing ecosystem services are priorities in the EU Biodiversity Strategy (Lammerant et al., 2013). The ecosystem services we have assessed include: i) water cycle regulation, that is a central ecosystem service for maintaining fresh water resources, controlling floods and, hence, protecting people living downstream (Vörösmarty et al., 2005), ii) nutrient cycling, regulated by a great variety of organisms and its alterations have deep impacts on ecosystem functioning (Lavelle et al., 2005), iii) soil conservation as its loss could be an irreversible process at the human and ecological scale, and its retention contribute to maintain primary productivity and to prevent harmful effects because of soil erosion (de Groot et al., 2002), iv) C sequestration in different compartments of the ecosystem and v) fire risk reduction (for the Ayora study site only).

Restoration of fire-affected areas

Within the framework of CASCADE, the two fire-affected ecosystems of the project considered restoration actions at two different time scales: within the first year after the fire when vegetation reestablishment is still very low (tree trunk removal or logging in Várzea) or several decades after the fire when the forest did not recover but a continuous shrubland was established (selective clearing and planting in Ayora). Salvage logging after fire in pine forests consists in the removal of all burned tree trunks and is one of the most common emergency actions carried out in the Mediterranean in the very early months after forest fires (de las Heras et al. 2012; Moreira et al. 2013). The main objectives of this practice are, especially, to have some return with the market value of the wood, but also to reduce fuel, to avoid erosion once the trees fall down some years after the fire, to reduce aesthetic impact, and, in case of weaken but still alive trees, to avoid pests spread (Vallejo et al. 2012). Potential negative impacts of this practice include the reductions of growth of regenerating seedlings, reductions of deadwood associated fauna, elimination of perches for birds dispersing seeds from neighbor undisturbed habitats, and reduction of microclimatic heterogeneity (Vallejo et al. 2012). There is also a risk to increase erosion associated to wood removal after fire but this impact is highly dependent on the soil properties of the area (Bautista et al. 2004).

Selective clearing of vegetation is one of the preferred management options aimed at sharply reducing fire hazard in Mediterranean fire-prone communities (Baeza et al., 2003). As compared to prescribed or controlled burning, also proposed and accepted as fuel control technique, vegetation clearing offers more positive effects especially related both to the protection of soil surface to erosion (especially when the vegetation remains are chipped and left on the ground). and resource export off-site and to the lag in the build-up of large fuel loads in the community (Baeza and Vallejo, 2008). The combination of this fuel control technique with the plantation of seedlings of late-successional species and with the ability to rapidly resprout after further disturbances (Valdecantos et al., 2009) may increase, at the same time, the resistance and the resilience to forest fires.

In the previous assessment of ecosystem services as a function of fire as degradation driver, we observed marked reductions in most ecosystem properties and services at the short term after fire (Valdecantos and Vallejo 2015). But at the long-term, burned areas recovered functionality to values similar to the Reference pine forest, with a spatial arrangement of vegetation that better conserve the resources.

Restoration of over-grazed areas

Grazing has deep impacts on ecosystem structure, composition and functioning (Milchunas and Lauenroth, 1993). Grazing exclusion is a worldwide extended practice to recover important ecosystem properties affected by overgrazing such as plant cover, vegetation and litter biomass, diversity, infiltration rate, soil fertility and soil biological properties (see Rong et al. 2014). For instance, it has been proposed as an effective management action to promote services such as soil C sequestration in areas severely affected by desertification (Li et al., 2012; Wang et al., 2016). The time elapsed since the avoidance of animals to graze as well as the ecosystem properties assessed determine the magnitude and significance of the effects of grazing exclusion. Under areas that were transformed from forest to grazed lands, fencing results in heavy and rapid forest encroachment by an increase of woody vegetation (Su et al., 2015).

In contrast, areas where the stocking rates are very low are susceptible to woody vegetation encroachment compromising grassland ecosystem types and threaten the biotic component, both plants and animals (Archer and Predick, 2014). However, there are no conclusive evidences that ecosystem services are compromised by woody vegetation encroachment while the recovery of the targeted ecosystem service after shrub management is only ephemeral and may depend on other factors. For instance, Alberti et al. (2011) observed that soil C pool reduces with clearing encroached pasturelands in moist areas but increases under dry environments.

We have observed that the CASCADE field sites affected by grazing showed a generalized decrease in diversity as compared to the reference states of the ecosystems but differences between the three grazed field sites were observed (Valdecantos and Vallejo 2015). Plant pattern in the grazed states was markedly different than in the ungrazed ones modifying the resource sink capacity of the system. LFA derived indices were lower in all Degraded sites than in their respective References suggesting a worsening of soil surface conditions and, hence, soil, water and nutrient conservation. Ecosystem services have shown important losses due to grazing in the order Randi>Messara>Castelsaraceno following a decreasing order of aridity.

Restoration of areas under multiple pressures

Albatera, with an aridity index of 0.16 and affected by multiple stressors, showed the highest relative losses of all individual and combined ecosystem services of all CASCADE field sites. The main ecosystem properties affected by degradation were those related to the sink/source spatial pattern and biodiversity. The assessment and quantification of the spatial distribution and arrangement of vegetation and, in general, of sink and source areas is especially relevant to address the restoration potential of drylands as this features have been described to determine seedling survival and growth of planted seedlings in restored semiarid sites (Urgeghe and Bautista, 2015).

Note: For full references to papers quoted in this article see

» References