Response of plant functional types to changes in the fire regime in Mediterranean ecosystems: A simulation approach

Abstract. In the Mediterranean basin, the climate is predicted to be warmer and effectively drier, leading to changes in fuel conditions and fire regime. Land abandonment in the Mediterranean basin is also changing the fire regime through the increase in fuel loads. In the present study, two simulation models of vegetation dynamics were tested in order to predict changes in plant functional types due to changes in fire recurrence in eastern Spain. The two modelling approaches are the FATE-model (based on vital attributes) and the gap model BROLLA (based on the gap-phase theory). The models were arranged to simulate four functional types, based mainly on their regenerative strategies after disturbance: Quercus (resprouter), Pinus (non-resprouter with serotinous cones), Erica (resprouter), and Cistus (non-resprouter with germination stimulated by fire). The simulation results suggested a decrease in Quercus abundance, an increase in Cistus and Erica, and a maximum of Pinus at intermediate recurrence scenarios. Despite their different approaches, both models predicted a similar response to increased fire recurrence, and the results were consistent with field observations.     

Spatial pattern of seedlings 1 year after fire in a Mediterranean pine forest

Summary The spatial distribution of seedlings of the dominant perennial plant species (Pinus halepensis, Cistus salviifolius, Rhus coriaria) and may annual species was studied after a wild fire in an eastern Mediterranean pine forest. The spatial distribution of all seedlings is affected by the location of the old burned pine trees. Seedling density of Pinus and Cistus is higher at a distance from the burned pine canopy and lower near the burned pine trunk. It is also higher beneath small burned pine trees than under big ones. Rhus seedling density is higher under big burned pine trees and also near the burned trunks. Seedlings of Pinus, Cistus and Rhus growing under the burned canopy of big pine trees tend to be taller than seedlings under small ones or outside the burned canopy. Most annual species germinate and establish themselves outside the burned canopies, and only a few annual species are found beneath them. It is suggested that variation in the heat of the fire, in the amount of ash between burned pine trees of different sizes, and in the distance from the burned canopy are responsible for the observed pattern of seedling distribution. The possible ecological significance of the spatial pattern of seedlings distribution and their differential growth rate are discussed.     

Spatial variation of seedling distribution in an east Mediterranean pine woodland at the beginning of post-fire succession

Most of the area in pine woodlands is occupied by perennial seeders that regenerate from seeds in the first winter after the fire and by annuals. Control of the germination in the regenerating vegetation after wildfire is therefore a primary ecological component of the post-fire succession in this ecosystem. The aim of the study presented here was to determine the distribution of Pinus, Cistus and other plants seeds around burned Pinus halepensis trees, and to measure the conditions related to seed germination in the upper soil layers in the same locations. The study was carried out in a 50-year old planted Pinus halepensis woodland that was burned down by a wildfire in July 1995. The variation of seedbank density was determined by collecting samples under the canopies of burned trees and in a nearby open area. Pine seedbank density decreased and that of Cistus and annuals increased with increasing distance from the burned trunks. Most pine seeds were present in the ash layer while those of the other plants were in the soil. In situ germination experiments showed that seedling density decreased with distance from the burned trunks while the proportion of pines in the seedling population increased. This was a result of seedbank variation and germination inhibition by the high pH conditions caused by the ash. The establishment of sparse pine seedling under the dead tree canopies insured their rapid development without interference by other plants and played a key role in the regeneration and stability of the pine woodland community. The concomitant mass germination of the perennial seeders in the rest of the area prevented invasion by annuals.     

Time of germination and establishment success after fire of three obligate seeders in a Mediterranean shrubland of central Spain

Aim Determining how differences in time of germination can affect plant establishment in plant communities that, after a disturbance, must reestablish from seeds under climatic conditions subject to extremes, such as the Mediterranean. Although early germination may be beneficial for survival in summer, when drought is severe, this may expose the seedlings to winter extremes, thus to higher mortality. Understanding how sensitive is the establishment of different species to temporal patterns of germination will help to understand the factors that control species distribution and community stability in disturbance‐prone environments, as well as its sensitivity to changes in weather patterns as climate changes. Methods An experimental fire was made in early fall in an old Cistus–Erica shrubland in Toledo (central Spain). After fire, germination, survival and growth of the three dominant seeder species (Cistus ladanifer, Erica umbellata and Rosmarinus officinalis) were monitored during the first 3 years after fire. Seedlings were tagged to identify their time of emergence, and divided into cohorts according to their month of germination. Differences in survival of the various cohorts were evaluated by means of a Wilcoxon (Gehan) statistic. Height of surviving, tagged plants was compared among cohorts by means of a Kolmogorov–Smirnov test. Results The year following fire was one of the driest on record, while the next one was one of the wettest. Germination was more abundant during the first than during the second year. Establishment was mainly from first‐year germination, as the majority of second‐year germinated seedlings died. Temporal patterns of germination within a year and between years varied between species. Seedling mortality was highest immediately following germination, not in summer. Mortality was related to time of germination: during a given period of time, the mortality of younger seedlings was higher than that of older ones. However, survival was not highest for the first cohorts. In general, the earlier the seedlings germinated the more vigourous they became, more clearly so for Cistus than for Rosmarinus, but differences tended to disappear with time. Overall, time of germination varied between species and affected differently seedling survival and vigour of the various species. Rosmarinus and Cistus had sufficient survivors to reestablish the initial population. Erica, despite abundant germination, suffered a strong population reduction. Main conclusions Mediterranean shrub species differ in their temporal patterns of germination and survival after fire. The effect of time of germination is complex: germinating early is advantageous since old seedlings fared better than younger ones when confronted with the same rigours. However, germinating early might expose the seedlings to greater hazards and the first cohort might not survive best. The temporal window for establishment is narrow and mainly restricted to the first year after fire. Second year seedlings, irrespective of most favourable conditions, survived very little. Missing the window of establishment might virtually lead to a population collapse, despite having very high germination, as found for Erica.     

A study of variation in the pollen spectra of honeys sampled from the Baixa Limia‐Serra do Xurés Nature Reserve in north‐west Spain

The pollen content of eleven honey samples from ten different apiaries in the Baixa Limia – Serra do Xurés Nature Reserve and other honey commercialised by the cooperative as “Mel do Xurés” (north‐west Spain) was subjected to quantitative and qualitative melissopalynological analysis. The quantitative analysis found that ten samples belonged to Maurizio's Class III and one to Class IV. According to the qualitative analysis, four samples were classified as unifloral honey with Erica, four samples as multifloral honey with Erica pollen as the principal component and three samples as multifloral honey with Cytisus‐type pollen and Erica as the principal component pollen. The pollen spectra differ between the diverse honeys analysed, with a common denominator being Erica and Cytisus‐type pollen being abundant in all. For the rest of the samples, the pollen spectra were mainly the same, but with different relative percentages among secondary elements. Thus, either as a secondary or an important element, 91% of the honeys contained Quercus, 82% Castanea sativa Miller, 45% Rubus, 36% Cistus and 27% Lithodora prostrate (Loisel) Griseb,. In particular, we record for the first time the presence of Ribes and Ilex aquifolium L. pollen in Spanish honeys as an important minor or minor pollen component.     

An overview of Cistus ectomycorrhizal fungi

The genus Cistus comprises a group of about 20 shrub species found in wide areas throughout the whole Mediterranean region to the Caucasus. Being one of the main constituents of the Mediterranean-type maquis, this plant genus is peculiar in that it has developed a range of specific adaptations to resist summer drought and frequent disturbance events, such as fire and grazing. In addition, it can form both ectomycorrhizas and arbuscular mycorrhizas. In this paper, we review the information available on the ectomycorrhizal fungi of Cistus across its entire geographic range, as gathered and critically sifted from both published literature sources and personal observations. Although the resulting data matrix was based primarily on accounts of sporocarp inventories in the field, existing knowledge on the features of Cistus natural and synthesized ectomycorrhizas was also included and discussed. In total, more than 200 fungal species belonging to 40 genera have been reported so far to be associated with Cistus. An analysis of the pattern of ectomycorrhizal diversity and host specificity revealed that members of the Cortinariaceae and Russulaceae make the most of both Cistus-aspecific and Cistus-specific mycobionts. Further studies are needed to expand our preliminary knowledge of the mycorrhizal ecology and biology of Cistus and its fungal associates, focusing on topics such as mycobiont diversity, host specificity, fungal succession, mycorrhizal influence on stress tolerance, and impact of disturbances, while comparing the findings with those from other ecosystems.     

Aprpaoches to airborne pollen in SE spain. First survey in murcia: one year of pollen monitoring (1993–94)

First data from a pollen survey carried out in the city of Murcia (SE Spain) are given in this paper. Using a Burkard Volumetric Spore Trap, daily slides were prepared and 80 pollen types belonging to 51 families andAlternaria spores were identified and counted. Special attention was paid to 14 relevant taxa: Cupressaceae,Pinus, Genisteae,Olea, Morus, Acer, Platanus, Plantago, Quercus, Urticaceae, Poaceae, Chenopodiaceae,Artemisia andAlternaria. The main sources of airborne particles wereAlternaria (27.7%), Cupressaceae (13.5%),Olea (9.36%), Chenopodiaceae (8.31%) and Urticaceae (5.8%). Annual variations in pollen abundance and length of the flowering seasons are given for individual species and are related to environmental factors. Results indicate a main pollen season from March to June and a second minor season in September to October. The relatively high concentrations of Genisteae and the appearance of anArtemisia winter season were noted.     

Facing drought in a Mediterranean post-fire community: tissue water relations in species with different life traits

Bulk shoot water potential, the osmotic component and the bulk modulus of elasticity were measured throughout one growing season in four species co-occurring in a post-fire Mediterranean community in southern Italy: Pinus halepensis, Phillyrea latifolia, Cistus salvifolius and Rosmarinus officinalis. A severe drought occurred throughout the measurement period. Large seasonal fluctuations have been observed for both predawn and afternoon water potential in all species. Although minimum values down to –4 MPa have been measured, plant water potential always recovered to less negative values after drought. Daily amplitude of water potential decreased with increasing plant water stress in all species. In Cistus and Rosmarinus less ability for short-term control of plant water status has been assessed. Osmotic potential at full turgor did not display clear seasonal patterns, with no consistent ranking of species by their osmotic values. In most cases, no osmotic adjustment (lowering of osmotic potentials) and no change in tissue elastic properties were observed in response to increasing summer drought and intensity of water stress.     

Woody species tolerance to expansion of the perennial tussock grass Ampelodesmos mauritanica after fire

Abstract. In Mediterranean shrublands, post‐fire accumulation of above‐ground biomass of resprouters is faster than that of seeders. This suggests that resprouters may have a competitive advantage. To test this hypothesis, we used a removal experiment to study the effect of the presence of the dominant tussock‐grass Ampelodesmos mauritanica on the resprouting shrubs Erica multiflora and Globularia alypum and on the seeders Rosmarinus officinalis and Pinus halepensis three and four years after a wildfire. Water potential of target plants was also measured to see if Ampelodesmos removal increased water availability. Ampelodesmos marginally reduced growth of all target species but did not influence survival or water potential of any target species. Our results suggest that the effect of climatically influenced water stress was stronger than the effect of Ampelodesmos neighbours. Plant‐plant interactions in this Mediterranean community are weak after fire and the magnitude of the Ampelodesmos effect does not differ between seeders and resprouters.     

A long‐term record of Quercus decline,logging and fires in a southern Swedish Fagus‐Picea forest

Abstract. We reconstructed forest development and disturbance events (fire and logging) during the last 1000 yr with tree‐ring data, pollen and charcoal analysis from a semi‐natural Fagus sylvatica‐Picea abies forest (ca. 1 km²) in the hemiboreal zone. According to pollen analysis, Quercus robur together with Pinus sylvestris was abundant in the forest until the turn of the 18th/19th centuries when these species disappeared completely (Quercus) or nearly completely (Pinus) and were replaced by Fagus and Picea. The disappearance of Quercus was corroborated by the remarkable discovery of a single Quercus stump that had been cut in the 18th century and had become overgrown and preserved by a very old Picea. In total 11 fires were dated from 1555 to 1748 from fire scars in several Pinus stumps cut 100 ‐ 200 yr ago. Since the last fire in 1748, no Quercus or Pinus have regenerated in the core of the reserve apart from single pines in neighbouring managed forest (80 yr ago). During the period of documented fires Fagus was protected from fires in a refuge made up of large boulders. Picea colonized the region at the time when the fires ceased 250 yr ago. We hypothesize that most of the fires were probably of human origin because of their patchiness and high frequency compared to the natural background levels of lightning ignitions in the region. On a 300‐yr time scale, logging and fire suppression seem to strongly overshadow the effect of climate change on forest composition and dynamics.     

Modeling the Effects of Fire on the Long-Term Dynamics and Restoration of Yellow Pine and Oak Forests in the Southern Appalachian Mountains

We used LANDIS, a model of forest disturbance and succession, to simulate successional dynamics of forests in the southern Appalachian Mountains. The simulated environments are based on the Great Smoky Mountains landscapes studied by Whittaker. We focused on the consequences of two contrasting disturbance regimes—fire exclusion versus frequent burning—for the Yellow pine (Pinus L., subgenus Diploxylon Koehne) and oak (Quercus L.) forests that occupy dry mountain slopes and ridgetops. These ecosystems are a conservation priority, and declines in their abundance have stimulated considerable interest in the use of fire for ecosystem restoration. Under fire exclusion, the abundance of Yellow pines is projected to decrease, even on the driest sites (ridgetops, south‐ and west‐facing slopes). Hardwoods and White pine (P. strobus L.) replace the Yellow pines. In contrast, frequent burning promotes high levels of Table Mountain pine (P. pungens Lamb.) and Pitch pine (P. rigida Mill.) on the driest sites and reduces the abundance of less fire‐tolerant species. Our simulations also imply that fire maintains open woodland conditions, rather than closed‐canopy forest. For oaks, fire exclusion is beneficial on the driest sites because it permits oaks to replace the pines. On moister sites (north‐ and east‐facing slopes), however, fire exclusion leads to a diverse mix of oaks and other species, whereas frequent burning favors Chestnut oak (Q. montana Willd.) and White oak (Q. alba L.) dominance. Our results suggest that reintroducing fire may help restore decadent pine and oak stands in the southern Appalachian Mountains.     

Soil seed bank,fire season,and temporal patterns of germination in a seeder-dominated Mediterranean shrubland

Soil seed banks play a major role in the post-fire regeneration of Mediterranean shrublands. They vary throughout the year in species composition, abundance, and readiness to germinate. After fire, germination occurs mainly during the following fall to spring. Time of germination can determine recruitment success. It is unclear what factors control post-fire germination and its timing. We tested the effects of season and fire on the readily germinable soil seed bank of a seeder-dominated shrubland. Plots were burned early and late in the summer season (ES, LS). Soil samples were collected before and after fire, and germinated in a chamber simulating successively autumn, winter, and spring conditions. Samples were kept moistened at all times. Fire intensity was similar between ES and LS. Several species of Cistus and herbs, mostly annuals, were dominant. Most germination occurred during the simulated-autumn period, with little subsequent germination during the following two periods. Germination speed (T ₅₀) during simulated-autumn was similar for shrubs and herbs, and independent of season or fire. Germination was lower for two shrubs (Rosmarinus officinalis, Cistus salvifolius) and higher for herbaceous dicots in LS than in ES soils. Fire reduced monocots and enhanced Cistus. Germination period significantly interacted with fire and season in some groups or species, altering the simulated-autumn germination peak. We demonstrate that the seed bank can germinate swiftly under simulated-autumn conditions. Hence, water availability is the main controlling factor of germination. Fire season differentially affected some species or groups, and could affect the post-fire regeneration.     

The relationship between fire history and an exotic fungal disease in a deciduous forest

Exotic diseases have fundamentally altered the structure and function of forest ecosystems. Controlling exotic diseases across large expanses of forest has proven difficult, but fire may reduce the levels of diseases that are sensitive to environmental conditions. We examined Cornus florida populations in burned and unburned Quercus–Carya stands to determine if burning prior to anthracnose infection has reduced the impacts of an exotic fungal disease, dogwood anthracnose, caused by Discula destructiva. We hypothesized that fire has altered stand structure and created open conditions less conducive to dogwood anthracnose. We compared C. florida density, C. florida health, and species composition and density among four sampling categories: unburned stands, and stands that had burned once, twice, and 3 times over a 20-year period (late 1960s to late 1980s). Double burn stands contained the greatest density of C. florida stems (770 stems ha⁻¹) followed by triple burn stands (233 stems ha⁻¹), single burn stands (225 stems ha⁻¹) and unburned stands (70 stems ha⁻¹; P < 0.01). We observed less crown dieback in small C. florida trees (<5 cm diameter at breast height) in burned stands than in unburned stands (P < 0.05). Indicator species analysis showed that burning favored species historically associated with Quercus–Carya forests and excluded species associated with secondary succession following nearly a century of fire suppression. Our results suggest that fire may mitigate the decline of C. florida populations under attack by an exotic pathogen by altering forest structure and composition. Further, our results suggest that the burns we sampled have had an overall restorative effect on forest communities and were within the fire return interval of the historic fire regime. Consequently, prescribed fire may offer a management tool to reduce the impacts of fungal disease in forest ecosystems that developed under historic burning regimes.     

Influence of rocky landscape features and fire regime on vegetation dynamics in Appalachian Quercus forests

Question : How do interactions between rocky landscape features and fire regime influence vegetation dynamics? Location : Continental Eastern USA. Methods : We measured vegetation, disturbance and site characteristics in 40 pairs of rocky and non‐rocky plots: 20 in recently burned stands, and 20 in stands with no evidence of recent fire (‘unburned’ stands). Two‐way analysis of variance (ANOVA) was used to assess the main and interaction effects of fire and rock cover on plant community composition. Results : In burned stands, rock cover had a strong influence on vegetation. Non‐rocky ‘matrix’ forests were dominated by Quercus, and had abundant ground cover and advance regeneration of early and mid‐successional tree species. Burned rocky patches supported greater density of fire‐sensitive species such as Acer rubrum, Sassafras albidum and Nyssa sylvatica and had little advance regeneration or ground cover. Quercus had fewer fire scars and catfaces (open, basal wounds) on rocky patches, suggesting that rocky features mitigate fire severity. In unburned stands, differences between rocky and non‐rocky patches were less distinct, with both patch types having sparse ground cover, little tree regeneration, and high understorey densities of relatively shade tolerant A. rubrum, N. sylvatica and Betula lenta. Conclusion : Under a sustained fire regime, heterogeneity in rock cover created a mosaic where fire‐adapted species such as Quercus dominate the landscape, but where fire‐sensitive species persisted in isolated pockets of lower fire severity. Without fire, species and landscape richness may decline as early‐mid successional species are replaced by more shade tolerant competitors.     

The effect of fire on flower visitation rate and fruit set in four core-species in east Mediterranean scrubland

The recovery of vegetation following fire has been studied intensively in Mediterranean-type ecosystems. Little attention, however, has been given to floral traits, and almost no data have been collected on the effects of fire on pollinator activity and fruit-set. This paper reports the effects of fire on flower visitation rates and the possibly related fruit-set. We compared visitation rates of the main pollinators on four plant core-species in burned and adjacent unburned areas. Measurements were performed at an unburned phrygana (scrub lands), and at a burned area (5–7 years post-fire). Bumble bees and solitary bees were the main taxa of visitors, while few honeybees were recorded. Solitary bees were almost absent from the burned area. Fruit-set was significantly higher in the unburned area for three out of the four plant species. The lower fruit-set in the burned area was possibly the result of low activity of solitary bees which are the main effective pollinators of the examined species. We hypothesize that the populations of the solitary bees were diminished or extirpated either directly by the fire, or indirectly by the scarcity of nectar in the early post-fire years due to dominance of young pine and Cistus spp. seedlings. The short foraging range of the solitary bees and their slow invasion rate into the burned area may explain our results.     

Woodland restoration and forest structure affect nightjar abundance in the Ozark Highlands

Chuck-will's-widow (Antrostomus carolinensis) and eastern whip-poor-will (Antrostomus vociferus) are nightjars in eastern North America that have declined 69% and 67%, respectively, in abundance since 1966, resulting in conservation concerns for these species. We investigated relationships between nightjar abundance and landscape composition, forest structure, and application of tree thinning and prescribed fire because of regional interest in woodland restoration and nightjar conservation. We conducted nocturnal nightjar surveys at 385 points in southern Missouri, USA, in 2014 and 2015 and related counts to pine (Pinus spp.) and hardwood basal area, canopy closure, percent forest cover, and percent of area thinned or burned within 500 m of survey points. We modeled abundance of chuck-will's-widow and eastern whip-poor-will using time-removal models that included a detection process and an abundance process within a hierarchical Bayesian framework. We detected 534 eastern whip-poor-will and 186 chuck-will's-widow during surveys. Our data supported global models that included all 6 vegetation and management variables for both species. Chuck-will's-widow abundance was negatively related to hardwood basal area and peaked at intermediate values of percent area burned and percent forest cover. Eastern whip-poor-will abundance was negatively related to hardwood basal area and canopy cover, positively related to percent forest cover and percent of area burned, and peaked at low to moderate levels of percent of area thinned. Relationships to forest structure and management activities generally supported the conclusion that woodland restoration benefits nightjars and that chuck-will's-widow select landscapes with less forest cover than eastern whip-poor-will.     

Succession of secondary forests affected by pine wilt disease in western Japan followed on vegetation maps

Abstract. Succession, changes in the distribution pattern of forest vegetation, and Pinus forest survival following pine wilt disease were clarified based on phytosociological analysis and vegetation maps. Survival of Pinus forests was restricted to the early successional stages, which were located on ridges and the upper part of slopes. Subsequent to pine wilt disease, the succession progressed from early to late substages of Pinus forest, mixed deciduous and evergreen Quercus, to evergreen Quercus forest. Succession occurs in abandoned pine forests which apparently are in a bad state and are vulnerable to attacks by pine wilt disease.     

Canopy disturbance patterns and regeneration of Quercus species in two Ohio old-growth forests

This study was conducted to determine the abundance of Quercus species, the spatial pattern of Quercus regeneration, the current canopy disturbance pattern, and their interrelationship in two old-growth deciduous forests in Ohio (Goll Woods and Sears-Carmean Woods). Acer saccharum and Fagus grandifolia had the greatest density and basal area in both forests, yet the largest trees (by basal area) present at each site were Quercus spp. Quercus spp. appeared to be decreasing in abundance in both sites. Though Quercus seedlings were common, few Quercus saplings or subcanopy trees were present. The current disturbance regimes were dominated by small canopy gaps created by death of 1–2 trees; canopy gaps 100 m² in size were rare and only 2.5–2.8% of the forest area was covered by recognizable canopy gaps. No significant differences in the density of Quercus seedlings or saplings were found between gaps and non-gap areas at either site. Though no significant barrier to seedling establishment appeared to exist, the present disturbance regimes are not well suited for the growth of Quercus into the subcanopy size class or the recruitment of Quercus into the canopy. The most frequent gapmakers in Goll Woods were Tilia americana and Acer saccharum, and those in Sears-Carmean Woods were A. saccharum and F. grandifolia. The species most frequent as gap fillers were A. saccharum (in both sites) and F. grandifolia (in Goll Woods). These results suggest that A. saccharum will continue to increase in abundance, and Quercus decrease in abundance, in these two old-growth stands.     

Modern pollen and vegetation relationships in the mountains of southern California,USA

The relationship between modern pollen assemblages and modern vegetation along two elevational transects within the Transverse and Peninsular Ranges of southern California, USA, is demonstrated using cluster analysis of the pollen data. Cluster analysis separates the Sonora Desert vegetation, Valley grassland/agricultural land and chaparral vegetation types on the San Jacinto Mountains transect. Chaparral is not easily separated on the San Bernardino Mountains transect, probably due to the presence of Quercus dumosa (scrub oak) there. The lower montane Quercus – Pinus (oak – pine) community is distinct from other forest types, and can be subdivided palynologically based upon relative importance of Quercus, Pinus and Cupressaceae [primarily Calocedrus decurrens (incense cedar)] pollen. Subdivisions include Quercus – Pinus – Cupressaceae, Quercus – Cupressaceae – Pinus and Quercus – Pinus assemblages. Higher elevation Pinus – Abies (pine – fir) and Pinus-dominated communities are also differentiated from one another, although the subalpine vegetation type only occurs on the San Bernardino Mountains transect. Though the study area presently straddles a transition between winter-wet and summer-wet climatic regimes, differences between the pollen assemblages in the two mountain ranges are minimal. Pollen assemblages from lower elevations document the effects of human activities, primarily agriculture, on the modern pollen rain of the region, with the occurrence of introduced citrus (Citrus sp.) and shade (Eucalyptus sp.) trees and weedy disturbance indicators (e.g., Brassicaceae).     

Old-field secondary succession in SE Spain: can fire divert it?

In the Mediterranean Basin, most cultivated areas were abandoned in the last century and are now in various stages of old-field succession. The aim of this work was to analyse the successional trajectories of these ecosystems, and to assess possible deviations in these pathways due to fire occurrence at high or low recurrence levels. Old-fields abandoned either about 50 or about 100 years ago were selected in SE Spain. Within the 50-year-old abandoned fields, plots were established which had been burned by 1, 2 and 3 fires in the last 25 years. Cover values of vascular species were sampled and then analysed by means of multivariate analysis. Euclidean distances between resulting communities were used as an indicator of the possible deviation from the unburned successional pathway. Our results pointed to the possibility that different successional pathways may exist depending on fire occurrence and recurrence. In the absence of fire, the vegetation is dominated by pioneer species, mainly Pinus. With the passage of time this vegetation will become dominated by later successional tree species (Quercus). However, when early-successional communities are affected by fire, the succession can be diverted. A single fire is enough to change Pinus forests into alternative stable states dominated by Rosmarinus officinalis shrub communities, where the colonisation of species in later successional stages is arrested. This deviation increases in high fire recurrence regimes where the vegetation changes to dwarf shrubs and herbs.