Description of the study site

Location

The Ayora Study Site is located in Eastern Spain within the Autonomous Community of Valencia (39º05 '- 40º5' N, 0º51' - 1º59' W). The Ayora Valley is on the border of the provinces of Albacete and Alicante, running between the Sierra Palomera and Mugrón mountain ranges to the west and Cortes de Pallás and the Caroig Peak. A multitude of forest tracks crossing the valleys and bordering the mountains shape the landscape of Ayora. The region is of inland Mediterranean landscapes and contains a rich variety of fauna and flora.

Topography

The Ayora Region covers an area of approximately 3,300 km². It is characterized by a diversified and rugged topography with moderate to steep slopes, partially covered with agricultural terraces that are currently abandoned and degraded. Altitude ranges from 20 m to nearly 1,200 m. The central part consists of an undulating area with an average elevation of about 600-700 m, interrupted by steep slopes and the valleys of the rivers Rio Grande and Escalona which drain the region in an easterly direction into the Jucar river and to the coastal plains. West of this central area, running straight from south to north, lies the Ayora-Cofrentes valley.

Geology and Soils

The dominant soil groups in the area are Regosols developed over marls and limestone colluviums. Secondary formations of shallow Leptosols and Luvisols, developed over limestone, can also be found. In the study area the dominant soils are Leptosols, with the exception of the north-west part where they give way to Cambisols. Calcaric soil material is dominant across the whole area.

Land Use

Over the last 30 years there have been significant changes in land use in Ayora and the region has been subject to widespread land abandonment. This has led to the encroachment of shrubs and accumulation of flammable biomass which has often served as fuel for wildfires and led to changes in the fire regime. Rangeland is the major land cover in the area, accounting for 67% of the study site. Besides these changes, land use appears to have been relatively stable since the 1990s.

Following a large fire event in 1979, a number of management interventions and mitigation schemes were introduced such as: afforestation efforts (P. halepensis, less frequently P. pinaster and Q. ilex); mechanical fuel load reduction and reduction of competition among plants; the creation of fire breaks; and the installation of water tanks. However, both remote sensing-based analysis and field inspection of these actions have rated them as mostly unsuccessful; wide areas were burned at least one more time after the 1979 fire event.

Climate

The climate in Ayora is typically Mediterranean, with hot, dry summers and mild winters. Specifically, the dominant climate type is dry Mesomediterranean with mean temperatures from 13 to 17 °C. The mean annual rainfall is between 350 and 700 mm and has a bimodal distribution with peaks in late autumn/early winter and in April/May. The distribution of precipitation differentiates two zones: a colder and drier northern zone (with 380-500 mm) and a southern zone with greater marine influence and milder temperatures and higher precipitation (500 -700 mm). Although rainfall occurs in all months the dry season, between July and August/September, limits the vegetative period to 8–11 months. Potential evapotranspiration reaches its yearly maximum during summer, causing a negative regional water balance, extremely low moisture content in the vegetation, and a very high risk of fire. Mean monthly precipitation, measured at the Ayora meteorological station, has a relatively stable long term average with mean annual precipitation of about 500 mm (or about 42 mm/month).

The temperature record shows only a slight upward trend and remains stable at an annual mean of 16.6 °C. The potential evaporation is estimated to be 1,325 mm.

Flora

Rangelands, accounting for 67% of the total land use, are largely dominated by dense shrublands and, to a lesser degree, by pine forests. These shrublands consist mainly of Quercus coccifera oaks with Ulex parviflorus and/or Rosmarinus officinalis. The pine forests are dominated by Pinus halepenensis and more rarely Pinus pinaster. Today, the typical Mediterranean woodland formations of Quercus ilex, P. pinaster and Pinus halepensis are only found in small remote and isolated areas. Agriculture is concentrated in the more suitable areas of the eastern plains and along the Ayora-Cofrentes valley. The mountainous areas, which were once under traditional, labour intensive cultivation, are now abandoned. This has led, particularly at the end of the 1960s and the beginning of the 1970s, to high fuel concentrations on once cultivated areas, enhancing the natural fire-prone environmental conditions of the region.

A synoptic view of vegetation health and the associated function of ecosystems can be derived from analysis of archival and on-going sequences of NDVI which shows a slight upward trend since the 1980s.

Selected vegetation – soil system

Forest and agriculture represent the main land uses in the study sites. During the 20th century many cultivated areas were abandoned and recolonized by early secondary successional species or, frequently, planted with pines. Colonization by later successional species (e.g. Quercus spp) used to be rather slow and always dependent on the presence of nearby preserved natural ecosystems and animals that disperse the seeds. As a consequence, a common landscape in the mid- and long- term in these former agricultural lands is a pine forest with a well-developed understory where seeder species are dominant. These communities are characterized by a large accumulation and density of standing biomass and consequently high fire risk.

Wildfires are the main degradation driver in the Ayora study site. There was an exponential increase from the 1960s to the 1980s both in the number of fires and the burned surface. Therefore, the landscape is a mosaic of pine forests and shrublands that differ in their composition and structure. The tree layer, when present, consists of Pinus halepensis (Aleppo pine) and P. pinaster (maritime pine) either in pure or mixed stands. Seeder species dominate the shrub layer, the main species being Rosmarinus officinalis (rosemary), Ulex parviflorus (gorse) and Cistus albidus (rockrose). Resprouters like Quercus coccifera (kermes oak) are also present within the shrubland but in lower proportion than seeders.

Leptosols, stony and shallow soils formed on limestone and dolomite, dominate in the uplands while Regosols, less stony and deeper soils formed on marl, dominate the lower colluvial areas.

Socioeconomic status

The small amount of economic activity taking place in the area is mostly associated with livestock. Before the 1979 wildfire, animals grazed the rangeland. Afterwards, as forest gave way to dense shrublands, sheep were mostly confined to croplands that were easier to navigate. In limited cases, administrative subsidies allowed grazing in firebreaks in order to assist in their maintenance. In contrast, goats are able to feed in shrublands but the number of animals is much lower than that of sheep (ca 3,000 vs 12,000). Additionally, beekeeping is an important economic activity in the region. The 1979 wildfire also resulted in a reduction of the logging exploitation of the forest from 24,000 to 2,000 m³ of wood per year. At present, more than 19,000 ha are cultivated in the Ayora valley, mainly cereals and olive, fruit and almond groves. This represents 17% of the total surface area of the region. The population employed in agriculture decreased from 20 to 5.6% in 15 years.

Timeline of events

The most relevant events in Ayora are two fire incidents in 1979 and 1991 which burned approximately 30,000 and 50,000 ha respectively. The land tenure status changed significantly, especially after the 1979 fire. Of lesser importance is the introduction of the Common Agricultural Policy (CAP) in Spain with its entrance in the EU in 1986.

Main Causes of Land Degradation

Human induced drivers

Forest fires constitute the main disturbance regime in the Ayora Study Site and have had strong effects on the landscape configuration. In the period 1975-2000 several wildfires took place in the area. Of these, the 1979 wildfire was the biggest and most devastating one. As a consequence, within the study site we may find different fire recurrences in such a small period of time. These forest fires have altered the composition of the forest, which prior to 1979 was dominated by Pinus halepensis, Pinus pinaster and, to a minor extent, Quercus ilex. At present, shrubland and regenerating pine (represented mainly by Pinus halepensis) occupy most of the surface; there are only small and isolated surfaces with the normal characteristic wooded vegetation of the area (Quercus ilex, Pinus pinaster and Pinus halepensis).

The recent increase in fire incidences, combined with the loss of ecosystem resilience due to past agricultural use and grazing, has driven a change in the composition of vegetation communities, from woodland and shrubland dominated by resprouting species to shrubland dominated by seeding species. When these shrublands are burned, further changes in plant species composition may occur, with the transition process influenced by fire recurrence. Thus, in the study site, fire recurrence is considered to drive important shifts in plant communities (from woodland to shrubland, from shrubland to open herb-sedgeland), and associated ecosystem functioning such as reduced post-fire regeneration rates leading to higher soil erosion risk. Therefore, short-interval fire cycles drive positive feedbacks that result in ecosystem degradation. Under these circumstances, the disappearance of key species, such as pines (due to repeated fires before producing viable seeds) and oaks (charcoaling and change of land use for cropping), represents a change in the ecosystem that needs external inputs through restoration to be reversed.

Natural Drivers

Climatic data show a general trend towards decreasing precipitation and increasing aridity in inland areas of the region of Valencia. In addition it has been suggested that there is a significant increase in the seasonal concentration of rainfall. These trends have had consequences for the fire regime; the main degradation driver of the region. A shift in the fire regime around the early 1970s has been reported, and researchers have suggested that while fire was not related to climatic conditions before 1970, since then it has been strongly driven by droughts.

Temporal analysis of drought periods carried out in different weather stations in the area does not show clear conclusions. The Standardized Precipitation Index for the period 1950-2000 shows negative (drought increases over time), positive (drought decreases) and neutral trends. The long-period drought assessment that was carried out for the study area for the period 1960-2007 shows one period of extreme drought (around 1980) and two moderate to severe periods (around 1986 and 2001). Prolonged drought periods have not been seen in recent years and the SPI moves mostly in the area of normal climate. The Aridity Index shows the area to have a semi-arid character over the period 1960-2008, although values vary substantially between years.

Indirect causes

Growth of cities (offsite urbanization) has resulted in the abandonment of fields and therefore an uncontrolled growth of forests and shrubland. As a consequence, fuel accumulated and increased its continuity and, hence, the risk of fire occurrence, intensity and extension. A high agricultural population density led to over-exploitation and land-use transitions before the 50s until the rural depopulation of the 80s. The present low population density has lead to a lack of sufficient management and this in turn to higher fire risk due to uncontrolled growth of forests and shrubland.

Today labour is again available in the area but people (especially from the cities) are less aware of the risk of fire as their livelihood depends less on forest health and products. The urban population tends to visit villages and their surrounding natural areas mostly for recreation. This may be particularly dangerous in summertime when the population of rural areas increases significantly, vegetation is very dry and any negligence or malpractice can cause a terrible fire. Current consumption patterns have changed; the high demand for regional agricultural and forest products has been abandoned. Today the area is undergrazed and  grazing is not profitable anymore. Both these factors have resulted in changes of vegetation composition.

Today land in the region is both publicly and privately owned. The government (local or regional administrations) is in charge of managing public land and this is only possible when there is available budget. Lack of resources can also lead to lack of good management, thus increasing the fire risk. Before the great wildfire of 1979, good infrastructures to fight against the fire were lacking. Afterwards, new plans for fire prevention and fighting were designed and implemented to some extent. Additionally, Forest Law 3/1993 induced the implementation of conservation interventions leading to more and better managed conservation practices such as afforestation efforts, and therefore lower fire risk. Unfortunately, the current economic crisis in Spain leads to a lack of investment in forest management thus exposing Ayora to wildfires once again.

The NDVI trend analysis for the Ayora reveals a break in the increasing trend that is exhibited until 1996, followed by a decrease in the average NDVI, along with a pronounced change in the variability. There is also a change of the seasonal component, with a single peak in early summer and minimum values during the winter preceding the break. On the other hand, after the trend break two peak and two minimum greenness seasons appear. Both these profiles are typical of Mediterranean woody vegetation, but a single peak may signify dependence on water availability and favorable temperatures while shrublands have minimal photosynthetically active vegetation cover and thus change in NDVI within a season is low. The timing of the seasonality change follows the occurrence of several fires (e.g. in 1979, 1984, 1991, 1994) that have resulted to a regime shift from pine forest to seeder shrubland. In addition to this observation, the timing of the seasonality change also coincides with one of the few emerging drought events in the area, highlighting the incremental effect of aridity, loss of resources and fire frequency as illustrated in the cusp catastrophe model (see »Unifying framework). It is therefore possible that the Forest Law 3/1993 was established too late to control a degradation process that was already under way.

Note: For an overview of the historical drivers of change and their analysis in all study sites see »Drivers of change in the study sites.

An area under pressure from fire

Total plant cover in Ayora has been reduced by more than 10% from the reference pine forest to the secondary mature shrubland (89.6 and 78.8%, respectively). The Reference plots are characterized by a mature overstory of maritime pine (Pinus pinaster, 47.9%) and, in a lesser extent, Aleppo pine (P. halepensis, 11.4%) and an understory dominated by Rosmarinus officinalis (26.4%), Ulex parviflorus (13.3%) and grasses (especially Brachypodium spp, 14.6%).The degraded community is a shrubland dominated by R. officinalis (44.3%) and Erica multiflora (15.7%), with an herbaceous layer with Brachypodium spp (22.0%) and Helichtotrichon filifolium (10.8%). Only 19 out of the 41 species found are common in the Reference and the Degraded states.

PCA analysis of specific plant cover data separated the Reference and the Degraded plots based on the values obtained on the first axis. The unburned Reference plots showed positive values on the first axis and the burned Degraded plots showed negative values. Pinus pinaster and Juniperus phoenicea are the species with higher positive weight on axis 1 (eigenvalues 0.914 and 0.804, respectively) and, hence, characteristics of the References while the shrub Quercus coccifera and the grasses Carex humilis, Helichtotrichon filifolium and Brachypodium retusum were negatively extracted on the second axis (-0.811, -0.789, -0.787 and -0.756, respectively) and abundant on the Degraded plots. The References sites showed higher variability of composition than the Degraded ones, which were much closer to each other in the PCA graph.

The highest contrast between biomass fractions of the ecosystem was observed in tree canopy biomass as the Degraded plots showed almost no tree recovery after fire. Biomass of the understory layer was 40% higher in the Degraded sites than in the Reference but differences were not significant (t=1.915, p=0.128). Rosmarinus officinalis was the species of the undestory that showed the highest biomass accumulation in the Reference plots (6.9 Mg ha-¹), with less than 10% of this biomass as dead material. The second species with higher biomass was Ulex parviflorus with 3.0 Mg ha-¹, 50% of it as necromass. The same two species accumulated the highest amount of biomass in the Degraded communities; rosemary has a biomass of 14.0 Mg ha-¹, in this case with less than 5% as dead biomass, and gorse presented 2.7 Mg ha-¹ but 76% of it as dead material conferring a high fire risk to the degraded community. Due to the important component of the overstory on the total biomass of the ecosystem, significant differences were observed in this variable (87.6 and 18.9 Mg ha-¹, in the Reference and Degraded sites, respectively; t=-18.013, p<0.001). Root biomass showed the same trend as the understory biomass, with a not significant increase in the Degraded in relation to the Reference plots. Litter accumulation was very similar in both states of the ecosystem (≈ 25 Mg ha-¹), though composition was different (data not shown).

The three diversity indices (species richness, diversity, evenness) decreased in the Degraded sites in relation to the Reference ones but only the total number of plant species showed significant differences (t=-2.688, p=0.055). Shannon’s diversity index and evenness also showed reductions in the Degraded plots but differences were not significant statistically (t=-1.968, p=0.120 and t=-1.195, p=0.298, respectively).

In Ayora the interpatches included bare soil but also the herbaceous matrix and litter, very abundant in the two levels of pressure as commented above.

The proportion of land considered Interpatch was higher in the Reference state than in the Degraded (50.3 vs 33.1%, respectively) but differences were not significant. A similar trend was observed in the average length of the interpatches, which was slightly higher in the Reference than in the Degraded. Vegetated patches showed similar width (≈ 2.9 m) but different length, higher in the Degraded than in the Reference (1.8 vs 1.1 m, respectively). This is a consequence of the huge continuity of the mature secondary shrubland of the Degraded plots. However, none of these variables showed significant differences between ecosystem states.

The three ecosystem function indices derived from LFA assessment showed very similar values in both states. Stability was around 71%, infiltration 55% and nutrient cycling 53% both in the Reference and Degraded ecosystems. As a consequence, the services with high weight of these indices (Water and Soil conservation and Nutrient cycling) did not show significant losses in the Degraded burned sites. Nevertheless, all these services were slightly lower in the Degraded than in the Reference. The largest loss of ecosystem services were recorded in C sequestration and Biodiversity (t=-2.206, p=0.092; and t=-2.311, p=0.082, respectively). The loss of the tree layer by the fire and the small recovery of pine species are responsible for this lower amount of C stored in the ecosystem. The lower percentage of interpatches and their smaller size in the Degraded than in the Reference sites could have affected the diversity indices and, hence, the biodiversity service. In Ayora field site, the mature secondary shrubland developed more than thirty years after the fire did not show a global loss of the assessed ecosystem services as compared to the Reference pine forest.

The summary of the ecological variables evaluated in this study shows a 60% increase in the length of the patches of shrubs in the Degraded plots as well as in the biomass of the understory layer. The highest loss of ecosystem properties was found in the biomass of the overstory, as pines did not recover after the fire in the Degraded sites, and, as a consequence, in total aboveground biomass.

Note: For an overview of the structural and functional changes and their analysis in all study sites see »Structural and functional changes.

Stakeholders in Ayora perceived changes in vegetation, wildlife and rural practices. Regarding changes in vegetation, stakeholders noted the general loss of natural vegetation, the expansion of shrublands, and the accumulation of biomass causing “fuel load build-up”. Perceived changes in wildlife included the loss of small game species but the increase in big game species and an increase in pests. Land abandonment, undergrazing and soil erosion were also perceived. Stakeholders agreed that the main driver of change was wildfires, as well as the lack of, or poor, environmental management practices, and the use of windmills.

Table: Drivers of change identified by the stakeholders in Ayora, Spain.

Decision makers mentioned that an increase in environmental research and management plans has occurred, however, these were not delivering the expected results, or, as a government representative asserted, “they are not always completely implemented”.

Permanent (as opposed to transient) land users and managers quoted more drivers of change and recognised changes in wildlife and agricultural practices as drivers, as well as wildfires and depopulation.

The forecasted expected changes were shared by all stakeholders, as they anticipated an increase in climate-related events. Their views can be summarized by a transient land user who predicted: “More fires, invasion of and the introduction of big game species (wild sheep, wild boar, Barbary sheep (Ammotragus lervia)), steady disappearance of small game species, more installation of industries within the forest, and a reduction of precipitation”. Two thirds of stakeholders representing land users detailed that they have not adapted to the changes or that they have absorbed the loss as part of their response. They mentioned that they are “resigned” to the detrimental changes as they will happen regardless of their efforts. Two stakeholders mentioned the relocation of their activities and protecting the land as their response to changes: “Resignation, changing game habits and protecting the land as much as possible”. In Ayora, there were 11 proposals to improve environmental conditions, however according to the stakeholders, only 5 measures were being implemented.

Stakeholders’ perceptions of the alternative management options, as well as the policy/measures required highlighted different perspectives within the different stakeholder groups: while land users focused on environmental management and cultural measures, government representatives focused on administrative and economic measures.

Table: Perceptions of alternative land management options and policies required by stakeholder groups in Ayora.

Note: For an overview results of the workshops on identifying adaptation strategies in all study sites and the concluding recommendations see »Adaptation strategies.

Values of soil organic C and nutrients in relation to fire recurrence, and ANOVA results for the soil variables of the Valencia site are shown in Figure 1 and Table 1. Most soil variables showed a decreasing trend from unburned soils to soils burned with the highest fire recurrence. However differences between recurrence levels were only significant for NH₄ and NH₄:N ratio, which had similar and highest values for long unburned areas and decreasing values with increasing fire recurrence. Microsite had a significant effect on soil organic C and total N. Thus, Q. coccifera microsites showed the highest value for both variables, while R. officinalis microsites were lowest in soil organic C and intershrub microsites were lowest in total N (data not shown). Q. coccifera microsites also showed the highest value for potentially mineralizable N but differences with the other microsites were only marginally significant.

Figure 1
Table 1

Two components of the PCA performed with the soil organic matter quantity matrix had eigenvalues higher than one and captured 68% of the variance in the data (Figure 2). Most variables were highly correlated to the first component, while the second component had the highest correlation with dissolved organic carbon. Also two components of the PCA performed with the soil organic matter quality matrix had an eigenvalue higher than one, explaining over 67% of its total variance. DOC:SOC and P_ava:SOC showed the highest correlations with the first component, while NH₄:N and NO₃:N were most strongly (and negatively) related to the second component (Figure 2). The two components of the PCA performed with the quantity matrix, separated unburned and one fire from two and three fires, with higher values for the first group (Figure 2). However, differences between these two groups were not significant due to the high variability between blocks (Figure 3). Microsite also had no significant effect on any of the two components, but the interaction Recurrence X Microsite was highly significant for the first component. The independent ANOVAs performed with this component for the three microsites assessed (patches of Q. coccifera, patches of R. officinalis and intershrub) gave a significant effect of fire recurrence in the case of the intershrub, which decreased gradually from unburned to three fires (Figure 4).  Recurrence X Microsite was also marginally significant for the second component, and the independent ANOVAs performed with this component for the three microsites gave a significant effect of fire recurrence in the case of the Q. coccifera patch, which decreased gradually from long unburned to twice burnt plots (Figure 4). The second component of the PCA performed with the quality matrix, also separated unburned and one fire from two and three fires. However, none of the factors nor their interaction had significant effects on any of the two components, again due to the high variability between the three areas analysed (data not shown).

 
Figure 2
Figure 3
Figure 4

Note: For a full description the context within which this work at Ayroa was done and the methods used, see

» Critical changes preceding a catastrophic shift

Plant cover in Ayora ecosystems ranged between 78.8 and 89.6% with significant differences between the degraded and the other two systems (Figure 1 left). We found fifty-two vascular plant species in the sites, eight of them (subshrubs except the shrub Rhamnus lycioides) were exclusively present in the reference forests. The average number of plant species in the restored and the reference sites were significantly higher than in the degraded plots (23.7, 22.7 and 17.0, respectively; Figure 1 right). Eleven species corresponding to both shrubs and subshrubs were only found in the restored plots, two of them (Pistacia lentiscus and Rhamnus alaternus) were introduced by planting. The reference communities correspond to pine forests of Pinus pinaster and P. halepensis (47.9 and 11.4% of cover, respectively) with two dominant species in the understory (Rosmarinus officinalis - 26.4% - and Ulex parviflorus - 13.3%) and scarce abundance of grasses (Brachypodium phoenicoides and B. retusum with 8.3 and 6.3%, respectively). The degraded shrublands are dominated by the shrubs R. officinalis (44.3%) and Erica multiflora (15.7%) and larger herbaceous layer (B. retusum and Helictotrichon filifolium, 21.3 and 10.8%, respectively). The species dominance inverted in the restored plots, with B. retusum being the most abundant species (40.2%) and R. officinalis dominating the shrub layer (26.2%). Other species with cover percentages above 10% were the shrub U. parviflorus (12.7%) and the grasses B. phoenicoides (12.5%) and Stipa offneri (10.7%). 

The two first axes of the principal component analysis accounted for 43.1% of the total variance and plots were clearly separated, especially, along the first component (Figure 2). The reference plots showed the highest positive scores of PC1, the restored plots released the most negative ones while the degraded showed intermediate values but closer to the restored than to the reference sites. We also observed high variability in species composition and contribution within the three spatially replicates restored plots as showed by wide range of values along the second axis. Degraded areas were plotted in a tight group according with these two axes.

Diversity indexes, especially Shannon-Wiener’s, were also affected by restoration (Figure 3). The H’ value was significantly higher in the restored plots (3.0) than in the degraded (2.2) with the reference areas showing intermediate values. Evenness values were quite similar but in this case differences were not statistically significant. 

Biomass accumulation in the different layers of the ecosystem was also dependent of the ecosystem state (Figure 4). Obviously, the biomass in the overstory was extremely higher in the reference than in the other two states where pine recovery was almost null. Regarding the understory (both shrubs and grasses), the degraded plots showed higher biomass than the restored areas. In fact, understory biomass in untreated degraded shrublands were two-fold that in the cleared plots. As a consequence, and because the contribution of the overstory in these two types of shrublands is negligible, the total aboveground biomass accumulated in the degraded was significantly higher than in the restored plots. 

The litter layer in both the reference and the degraded areas were very similar although the origin of the plant material was quite different (mostly pine needles in the reference and fine and coarse shrub debris in the degraded). The restored plots showed significant reductions, above 40%, of litter accumulation (Figure 5 left). Root biomass in the most superficial 20 cm of soil did not show significant changes due to the state of the ecosystem and ranged between a minimum of 15.8 Mg ha-¹ in the reference to a maximum of 20.6 Mg ha-¹ in the degraded plots (Figure 5 right). 

In relation to the distribution and morphology of the sink and source areas within the landscape, the considered interpatches in Ayora included bare soil but also the matrix of perennial grasses and litter, very abundant in the three ecosystem states as it has already been mentioned. The length and percentage of interpatches in the forest systems were slightly higher than in the two shrubland systems. Interpatches averaged 1.0, 0.9 and 0.7 m in the reference, the degraded and the restored plots, while their cover was 52% in the forest and 33-34% in the two shrubland types (Figure 6). The size of patches was increased by restoration with an average width of 4.6 m while both in the reference and in the degraded plots patches were 2.9 m wide.

The three indexes derived from the LFA assessment showed a significant reduction in the restored system while no differences were observed between the reference and the degraded states (Figure 7). The more pronounced reduction was observed in the nutrient cycling that fell from 53% to 36% in the restored plots. Also the infiltration was sharply reduced, with values around 55% in the reference and degraded sites and 40% in the restored. The reduction in the stability index was slightly lower but however differences between states were also significant. 

Half of the ecosystem services were significantly affected by the state of the ecosystem: C sequestration, biodiversity and fire risk reduction (Figure 8). The other three services (soil and water conservation and nutrient cycling), and the combination of all of them as well, showed the same trend to decrease from the reference to the degraded state while the restored showed intermediate values. The fire and the extremely limited post-fire recovery of pines, combined with the selective clearing of flammable shrubs, resulted in a significant reduction of C sequestration in the restored plots. On the contrary, restoration improved biodiversity of the degraded shrublands even beyond values of the reference state of the ecosystem. The most significant effect of the restoration actions was in reducing fire risk in relation to both the reference and the degraded plots. This was the main objective pursued with the restoration conducted in the degraded areas ten years before this assessment.

The summary of changes of ecosystem properties due to restoration (Figure 9) showed postive effects on variables related to community composition and structure (diversity indexes and patch-interpatch distribution and morphology) while negative chenges were mainly associated to biomass accumulation (litter, understory and nutrient cycling and infiltration). However, large amounts of biomass in the understory in the degraded areas increases the C sequestred in the system but also increases flammability and, hence, fire risk. 

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

» References

One workshop was held in Spain covering both study sites. It involved 14 stakeholders, representing environmental NGOs, researchers, land managers and governmental institutions (Figure 1). Government representatives worked in conservation, forestry and agricultural areas. The Local Forest Association was invited, however they did not attend.

Discussions focused on forest fires and land abandonment principles. Land tenure in the study sites is usually held by small proprietaries that do not usually live within the area. Land management is designed and carried out by governmental institutions. Therefore, the workshop focused on bringing together stakeholders representing those institutions relevant to identify the feasibility and barriers of the principles.

Comments on the principles for the land abandonment context

The CASCADE team explained to the workshop participants that principles were elaborated mostly for land managers rather than owners. Stakeholders mentioned that implementation of the principles requires cooperation and legal frameworks, allowing land managers to apply the principles on private property.

For descriptions of the principles discussed here, see »Guidelines for land managers: the land abandonment context_EN

Most stakeholders in Spain agreed with most of the land abandonment principles. However, they stated that they could not consider them as guidelines to prevent land abandonment as the principles do not consider a holistic approach to socio-environmental development, taking into account aspects such as social integration, forest use regulation or burning of crop residues. Stakeholders mentioned that a multitude of factors are causing land abandonment in the area, which in turn is linked with an increased risk of forest fires. Furthermore, other socio-economic issues in the region (such as land tenure structures and processes), decrease the feasibility of implementing the principles. Therefore, stakeholders considered that managing these areas is highly complex, from both administrative and spatial perspectives. Stakeholders also mentioned various barriers to implement the principles (Table 1) linked to the land abandonment context.

Table 1. Causes and consequences of land abandonment mentioned by stakeholders.

Stakeholders also expressed disagreement with the terminology used in some of the principles, although not with the principle per se. Some of the stakeholders suggested how they could be clarified. For example, an NGO representative suggested using the term “cropland or agriculture abandonment” instead of “land abandonment” in general, as most land abandonment affects croplands.

Comments on the principles for the forest fire context

For descriptions of the principles discussed here, see »Guidelines for land managers: the forest fire context.

In Spain, the forest fire principles were clear and stakeholders agreed with them in their context, however they were also perceived as too general (Table 2). In response to principle 2.1 “Avoid afforestation with single or flammable species” the representative of the wildlife department disagreed with not considering species for restoration, due to them being flammable. Some species like Juniperus spp. are highly flammable but they are key species in certain ecosystems. Equally, clarifications were also suggested regarding terminology that seemed too broad to be meaningful to stakeholders. For example, in “sustain and increase diversity of endemic plants” (principle 2), the term “endemic” was considered potentially confusing by the Department of Wildlife representative, as most endemic species nowadays are characteristic of degraded environments. The suggestion was therefore made to refer to vegetation as “Indigenous”. Furthermore, referring to vegetation as “fuel” was perceived to be a broad and potentially confusing use of the term, as it was recognised that although some vegetation can act as fuel in a fire situation, many plant species also have an ecological role.

Between the recommendations under principle 3 “Sufficient soil cover shortly after a fire reduces risk of soil erosion”, are mulching and maintaining soil cover in fuel breaks. Stakeholders were dubious about the benefits of mulching due to the scarcity of management experiences, although they recognised the potential benefit of this technique to avoid land degradation. The lack of experience in using or being in contact with this technique meant that they were uncertain about its costs and benefits, as there are few experiences about mulching application as a management technique and these are restricted to small areas especially after forest fires. Different comments were made about these experiences and the type of mulch (hay or forest residues) but there was consensus about the beneficial role of this technique as an emergency land restoration action. Stakeholders also asked the CASCADE team about cheaper options than mulching, and cropland residues were mentioned by stakeholders in this regard. Finally, stakeholders recognised that the term “firebreak” (cortafuegos) is an outdated technical term that fails to reflect the ecological configuration of the area, which presents an array of agricultural and forested patches, rather than a continuous of forest with non-forested sections in a heterogeneous mosaic landscape. Therefore, they felt a holistic view of the forest and the measure was needed for fire management.

In Spain the lack of technical support or information were not seen as a barriers to implementing the principles. Stakeholders however recognised that land tenure is an obstacle both for coordination and implementation of management measures. The small holding sizes of the properties means that there is a multitude of stakeholders that need to agree if measures are going to be taken. In some cases, the owners are not identified by land managers, or it is difficult to contact them, therefore measures cannot be carried out easily as it would involve interfering with private land. This issue is amplified in the case of land abandonment, and in some cases there is opposition to governmental intervention by forest owners. Stakeholders also considered that the needs of the population should be a priority to prevent land abandonment, because if there is a lack of schools, jobs or medical services, land abandonment will continue in the region. They proposed that the management principles for land abandonment should consider social participation in a bottom-up approach, as it is currently happening in the region after large forest fires. Equally, plans should include climate change projections in designing management treatments (prioritization of areas, selection of species and so on). Stakeholders also commented that the role of and impacts on the area’s fauna should be included too.

Table 2. Agreements and disagreements with the forest fire principles by Spanish stakeholders. Y = agreement

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

» References