SEARCH

SEARCH BY CITATION

Keywords:

  • forest decline;
  • neighborhood models;
  • Quercus suber;
  • regeneration dynamics;
  • soil-borne pathogens;
  • soil texture;
  • species coexistence

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Soil-borne pathogens are a key component of the belowground community because of the significance of their ecological and socio-economic impacts. However, very little is known about the complexity of their distribution patterns in natural systems. Here, we explored the patterns, causes and ecological consequences of spatial variability in pathogen abundance in Mediterranean forests affected by oak decline.
  • We used spatially explicit neighborhood models to predict the abundance of soil-borne pathogen species (Phytophthora cinnamomi, Pythium spiculum and Pythium spp.) as a function of local abiotic conditions (soil texture) and the characteristics of the tree and shrub neighborhoods (species composition, size and health status). The implications of pathogen abundance for tree seedling performance were explored by conducting a sowing experiment in the same locations in which pathogen abundance was quantified.
  • Pathogen abundance in the forest soil was not randomly distributed, but exhibited spatially predictable patterns influenced by both abiotic and, particularly, biotic factors (tree and shrub species). Pathogen abundance reduced seedling emergence and survival, but not in all sites or tree species.
  • Our findings suggest that heterogeneous spatial patterns of pathogen abundance at fine spatial scale can be important for the dynamics and restoration of declining Mediterranean forests.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Soil-borne pathogens are a key component of the belowground community because of the significance of their ecological and socio-economic impacts. For instance, several species of Phytophthora and Pythium, two well-known genera of soil-borne oomycete pathogens, are common causes of agricultural diseases (Erwin & Ribeiro, 1996; Martin & Loper, 1999) and are involved in the massive decline of Quercus, Castanea, Eucalyptus and other trees in forests worldwide (Brasier et al., 1993; Brasier, 1996; Rizzo et al., 2005; Romero et al., 2007; Cahill et al., 2008). Not surprisingly, an understanding of when and where soil-borne pathogens are more likely to cause destructive epidemics has long been an important topic of agricultural research. In natural systems, however, much less is known about the complexity of their distribution patterns, which remains one of the most challenging aspects of studying belowground organisms (Ettema & Wardle, 2002; Reinhart & Clay, 2009).

The pathogen landscape can be affected by a variety of abiotic and biotic factors (Martin & Loper, 1999; Agrios, 2005). Among these factors, vegetation is a major determinant of the spatial distribution of soil pathogens both across and within plant species (Wardle, 2002). Plant species can affect soil-borne pathogen populations directly by providing living host tissue, or indirectly by generating environmental conditions that affect their reproductive activity (Augspurger, 1990). In forest ecosystems, for example, pathogen populations can benefit from the wetter microclimatic conditions found in the shaded understory relative to open environments (Gómez, 2004; Matías et al., 2011). However, understory environments tend to have more fertile soils than gaps and sustain a larger microbial community, which could negatively affect soil-borne pathogens through competition for resources and colonization space (Weste & Marks, 1987; Aponte et al., 2010). Depending on the relative importance of the different mechanisms, the net effect of a given woody species on soil pathogen abundance might range from highly positive to largely negative. However, species-specific effects could be obscured by intraspecific variation in plant traits, such as size or tolerance to infection (Packer & Clay, 2000, 2003; Reinhart & Clay, 2009). Clearly, further research is needed in order to determine whether and how the mosaic of plant species and gaps in the forest canopy translate into a mosaic of soil pathogen abundance and composition.

Just as adult plants can drive the abundance and activity of soil-borne pathogens in forests, pathogens can, in turn, shape the regeneration dynamics of the plant community (Packer & Clay, 2003; O’Hanlon-Manners & Kotanen, 2006). Seedlings are particularly vulnerable to pathogens because their roots are structurally simple and poorly lignified (Augspurger, 1984, 1990; Romero et al., 2007). Moreover, because pathogens vary in pathogenicity for different tree species (Augspurger & Wilkinson, 2007; Moralejo et al., 2009; Reinhart et al., 2010), they can affect the composition of the seedling bank. For example, it has been proposed that shade-intolerant tree species are more susceptible to soil-borne diseases than are shade-tolerant species, and that such susceptibility might be a key mechanism excluding them from the forest understory (O’Hanlon-Manners & Kotanen, 2004; McCarthy-Neumann & Kobe, 2008). If differential responses to soil pathogens exist, interactions with soil-borne pathogens may contribute to species coexistence across heterogeneous forests.

The objective of this article was twofold. First, we aimed to advance the understanding of the pathogen landscape by developing spatially explicit neighborhood models that explain the importance of abiotic (soil texture) and biotic (tree and shrub community) drivers of soil-borne pathogen abundance in Mediterranean forests affected by cork oak (Quercus suber) decline. We built on established methods for the characterization of neighborhood processes (Canham & Uriarte, 2006), and applied these methods for the first time on soil organisms in close association with plants. A main advantage of the neighborhood approach is that it allows the linking of soil pathogen abundance with the distribution of neighboring individuals of the whole woody community. It therefore captures the complexity of natural plant communities, in which a particular volume of soil is not necessarily occupied by just one host species. The second objective of our study was to explore the consequences of soil-borne pathogen abundance on seedling emergence and survival of dominant tree species with varying shade tolerance (Quercus canariensis > Q. suber > Olea europaea var. sylvestris). For this, we conducted an in situ field experiment in which seeds of the three tree species were sown and monitored in the same locations in which pathogen abundance was quantified. To the best of our knowledge, this is the first study that simultaneously analyzes the spatial relationship among abiotic soil properties, adult plants (trees and shrubs), and the pathogen and seedling community in a multispecies natural context.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Study site and species

The study was conducted in the Alcornocales Natural Park, a hotspot of biodiversity in southern Spain (Médail & Quézel, 1999). The climate is subhumid Mediterranean, with most rainfall (95%) occurring from October to May. Soils are generally sandy, acidic and nutrient poor, derived from a bedrock dominated by Oligo–Miocene sandstones, but interspersed with soils richer in clay derived from layers of marl sediments. The Alcornocales Natural Park contains the largest and best conserved Q. suber forests of Europe (Anonymous, 2005). In the drier lowlands of the park, Q. suber forms mixed open woodlands with the evergreen and shade-intolerant O. europaea var. sylvestris, whereas, in wetter areas, Q. suber coexists with the deciduous shade-tolerant Q. canariensis, forming closed forests. The shrubby understory is diverse and rich in endemic taxa (Ojeda et al., 2000).

A severe decline affecting Quercus species (especially evergreen oaks Q. ilex and Q. suber) has been reported since the early 1990s in the park and throughout the Mediterranean Basin (Brasier, 1992, 1996). Several abiotic (e.g. drought) and biotic (e.g. insects and pathogens) factors are potentially involved in this decline (Tuset & Sánchez, 2004). However, in the study area, two main oomycete soil-borne pathogens (Phytophthora cinnamomi and Pythium spiculum) have been isolated from symptomatic Q. suber trees and are suggested to be the main drivers of the decline of the species (Brasier, 1996; Sánchez et al., 2002, 2006; Romero et al., 2007).

Field sampling of soils and plants

We selected six study sites, three in open woodlands of Q. suber and O. europaea var. sylvestris (hereafter woodland sites) and three in closed forests of Q. suber and Q. canariensis (hereafter closed forest sites), distributed across the whole Natural Park (see site descriptions in Supporting Information Table S1). At each site, we established a 60 × 50 m2 permanent plot in a topographically uniform area. The topography was kept constant in order to avoid confounding effects for the analysis of the impact of soil texture and plants on pathogens. Each plot was subdivided into 30 (10 × 10 m2) subplots. During the spring of 2010 (April–May), we took two soil samples (0–20 cm) at the center of each subplot, one for texture and one for pathogen analysis. Soil samples were taken within 1 m of where the seeds were planted for the sowing experiment (see ‘Seed sowing experiment’ section below), rapidly placed in a cooler and transported to the laboratory for the assessment of texture and pathogen abundance (see ‘Laboratory methods’ section below).

To characterize local neighborhoods, we identified and mapped all live and standing dead trees (including stumps) with a diameter at breast height (dbh) > 2 cm and all shrubs in the 60 × 50 m2 permanent plots, as well as in a buffer zone (15 m wide for trees and 5 m wide for shrubs) around each plot. Tree neighborhoods of similar size have been shown to capture the most important aspects of neighborhood interactions in temperate forests (Gómez-Aparicio et al., 2008a; Coates et al., 2009). Although we did not have any reference from which to choose the maximum shrub neighborhood, we considered a size of 5 m to be sufficiently large based on the small size of most shrubs in these forests (height usually < 3 m). We measured the dbh of each of the trees mapped (n = 1341 trees). As a result of the multistem growth form, shrub size was characterized by measuring the two diameters of the elliptical projection of its crown (n = 3005 shrubs). In addition, we evaluated the health status of Q. suber individuals by a visual estimation of crown defoliation on a standardized semi-quantitative scale widely used in the region to monitor oak decline (e.g. García et al., 2011): healthy reference trees; slightly defoliated trees (< 50% crown defoliation); highly defoliated trees (> 50% defoliation); and dead trees (including stumps). No other tree or shrub species in the study area showed symptoms of decline.

Laboratory methods

Soil texture  Soil samples were air dried and sieved through a 2-mm mesh sieve to remove root material and stones. Particle size analysis was undertaken using the Bouyoucos hydrometer method (Gee & Bauder, 1986). Total sand (i.e. fine + coarse sand, 0.05–2 mm) was used as a representative measurement of the soil texture (see a similar approach in Gómez-Aparicio et al., 2008b).

Pathogen abundance  Aliquots of 10 g from each soil sample were processed as described in Romero et al. (2007), preparing soil suspensions in 100 ml of water–agar 0.2%. Aliquots of 1 ml taken from the soil suspensions were plated onto NARPH Petri dishes (20 dishes per sample; Romero et al., 2007). Colonies growing on the plates were morphologically identified and counted. As soil samples had been dried previously, it was assumed that each colony obtained resulted from the germination of at least one resistant spore (oospore or chlamydospore). Results were expressed as colony-forming units per gram of dry soil (cfu g−1).

The identification of the isolated colonies was conducted by microscope observations after incubation on carrot–agar medium (Dhingra & Sinclair, 1995) at 24°C in the dark for 4–6 d and staining with acid fuchsine in lactophenol. Colonies were classified into three groups: P. cinnamomi, characterized by clustered hyphal swellings and smooth cell-walled chlamydospores (Erwin & Ribeiro, 1996; Romero et al., 2007); Py. spiculum, which showed characteristic ornamented oospores (Paul et al., 2006; Romero et al., 2007); and Pythium spp., characterized by the absence of septa in narrow branched hyphae (< 4 μm thickness). Phytophthora cinnamomi and Py. spiculum are the main soil-borne pathogens involved in the decline of Quercus species in southern Spain (Sánchez et al., 2006; Romero et al., 2007; Jiménez et al., 2008). The Pythium spp. group represents a mix of Pythium species of unknown pathogenicity, and therefore may include both virulent and avirulent (i.e. saprophytic) species (see a similar approach in Reinhart & Clay, 2009 and Reinhart et al., 2010). Although this group does not necessarily cause any pathogenic effect on trees, we refer to the three different oomycete categories considered as ‘pathogen species’ for simplicity.

Seed sowing experiment

During winter 2009–2010 (December–January), we conducted a sowing experiment in the six study sites. Surface-sterilized seeds of the two dominant tree species were sown at the center of each of the 30 subplots. Seeds were sown at a depth of 2 cm in two adjacent 30 × 30 cm2 quadrats per subplot. Each quadrat contained three lines of seeds separated by 7.5 cm from each other and from the border of the quadrat. Each line was randomly assigned for sowing either three Quercus or six Olea seeds. The larger number of Olea seeds was chosen based on their lower probability of germination (Goyiatzis & Porlingis, 1987; Rey et al., 2004). Sown quadrats were protected with 1-cm mesh hardware to exclude seed predators. As a whole, we sowed 1620 seeds of Q. suber, 1620 seeds of O. europaea and 810 seeds of Q. canariensis. Seedling emergence was monitored in early June 2010 to ensure that most seedlings had emerged (Pérez-Ramos & Marañón, 2012). Seedlings were revisited in early October 2010 to record survival after the first summer in the field, the main period of seedling mortality in Mediterranean systems (Gómez-Aparicio, 2008; Pérez-Ramos et al., 2012). Unfortunately, the emergence of O. europaea was virtually nil in all sites (data not shown), which precluded us from testing the effect of pathogen abundance on the emergence and survival of this tree species.

Statistical analysis

Neighborhood models of soil pathogen abundance  We used likelihood methods and model selection for the analysis of our data (Johnson & Omland, 2004; Canham & Uriarte, 2006). Following the principles of likelihood estimation, we estimated model parameters that maximized the likelihood of observing the pathogen abundance measured in the field given a suite of alternative neighborhood models.

We fitted separate models for each combination of forest type (woodland and closed forest) and pathogen species (P. cinnamomi, Py. spiculum and Pythium spp.). Our analyses of soil-borne pathogen abundance estimated four terms: (1) average potential pathogen abundance (PPA, in cfu g−1) at each of the three sites, and three multipliers that quantified the effects on the average PPA of: (2) local abiotic conditions (expressed in terms of soil texture); (3) the characteristics of the tree neighborhood (expressed in terms of the size, spatial distribution, species and health status of the trees); and (4) the characteristics of the shrub neighborhood (expressed in terms of shrub size). Our full model had the following form:

  • image(Eqn 1)

We also tried a linear model framework in which the different effects were summed (for a similar approach, see Baribault & Kobe, 2011), but, in general, it showed a poorer performance than the multiplicative model framework (data not shown). PPASite is an estimated parameter that represents the expected pathogen abundance at each site in the absence of sand in the soil texture and of trees and shrubs in the neighborhood (i.e. the abiotic, tree and shrub effects = 1). The three effects in Eqn 1 were modeled using Weibull functions:

  • image(Eqn 2)
  • image(Eqn 3)
  • image(Eqn 4)

where b, c and d are parameters estimated by the analyses determining the sign and magnitude of the abiotic, tree and shrub effects, respectively.

The abiotic effect was modeled as a function of soil texture quantified as the proportion of sand content. Texture was chosen to represent the abiotic driver of pathogen abundance because it is a relatively stable soil property that influences key environmental variables for pathogens (e.g. water availability) and is not easily modified by plants, therefore being independent of the biotic effects in the equation.

The tree effect was modeled as a function of a tree neighborhood index (NITree). This index quantifies the net effect of j = 1,...,n neighboring trees of i = 1,...,s species on pathogen abundance, and was assumed to vary as a direct function of the size (dbh) and as an inverse function of the distance to neighbors following the form:

  • image(Eqn 5)

where α, β and γ are estimated parameters that determine the shape of the effect of the dbh (α) and distance to neighbors (β and γ) on pathogen abundance. Instead of setting α, β and γ arbitrarily, we tested two different versions of Eqn 5, fixing α to values of zero or unity and letting β and γ vary. A value of α = 1 implies that the effect of a neighbor is proportional to its dbh, whereas a value of α = 0 means that the tree influence on soil pathogen varies as a function of tree density, regardless of size.

We were particularly interested in exploring whether tree effects varied between individuals of different species or health status. For this purpose, we multiplied the net effect of an individual tree by a per capita coefficient (λ) that ranged from − 1 to 1, and allowed for differences between neighbors in their effects (negative or positive) on a target pathogen. We tested four different groupings of neighbor species in Eqn 5 with increasing complexity: a model in which all trees were considered to be equivalent (i.e. fixing λ = 1); a species-specific model that calculated two separate λ values, one for Q. suber and one for the coexisting tree species (either O. europaea or Q. canariensis); a model that also took into account the health status of Q. suber trees, and therefore calculated four separate λ values (healthy Q. suber, slightly defoliated Q. suber, highly defoliated Q. suber and the coexisting tree species); and a model that not only considered alive trees of different species and health status, but also the legacy effect of dead Q. suber trees, calculating five separate λ values.

The shrub effect was modeled as a function of the shrub neighborhood index (NIShrub). This index is a simplified version of the NITree, and quantifies the net effect of j = 1,...,n neighboring shrubs of i = 1,...,s species on pathogen abundance following the form:

  • image(Eqn 6)

NIShrub was assumed to vary as a direct function of the size (crown area) of neighbor shrubs in a neighborhood of 5 m radius. We decided not to include distance in the calculation of the index, given the already restricted area over which shrubs were mapped and to keep the number of parameters in the models manageable.

Finally, in order to test whether any of the three effects studied (i.e. texture, trees and shrubs) varied among sites of a given forest type, we tried variations of the full model in which the slopes of each effect (i.e. parameter b, c or d) were allowed to vary among sites.

Effect of soil-borne pathogens on seedling emergence and survival  We fitted models that estimated seedling emergence or survival at each subplot as a direct function of the pathogen abundance in the soil. We tried both a multiplicative and a linear model framework, the latter offering a better fit to the data. Thus, for each combination of forest type, tree species and pathogen species, seedling emergence and survival were predicted as:

  • image(Eqn 7)
  • image(Eqn 8)

where PSESite and PSSSite are the potential seedling emergence and survival, respectively, at each site in the absence of pathogens, and b and c are the slopes of the regressions determining the pathogen effect. We explored the existence of site-dependent pathogen effects by fitting models that allowed the parameters b and c to vary among sites of a given forest type.

Parameter estimation and model selection  Following the principle of parsimony, we followed the strategy of systematically reducing the number of parameters in the full model to the simplest model that was not a significantly worse fit than any more complicated model. We used the Akaike Information Criterion corrected for small sample sizes (AICc) to select the best model, with lower AICc values indicating stronger empirical support for a model (Burnham & Anderson, 2002). Pathogen abundance values were modeled using a Poisson error distribution, and seedling emergence and survival using a binomial error distribution. We used simulated annealing, a global optimization procedure, to determine the most likely parameters (i.e. the parameters that maximized the log-likelihood) given our observed data (Goffe et al., 1994). The slope of the regression (with a zero intercept) of observed on predicted pathogen abundance was used to measure bias (with an unbiased model having a slope of unity), and the R2 of the regression was used as a measure of the goodness-of-fit. We used asymptotic two-unit support intervals to assess the strength of evidence for individual parameter estimates (Edwards, 1992). All analyses were performed using software written specifically for this study employing Java (Java SE Runtime Environment v6, Sun Microsystems Inc., Santa Clara, CA, USA).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Neighborhood models of soil pathogen abundance

All of the models produced unbiased estimates of soil-borne pathogen abundance (i.e. slopes of predicted vs observed abundance were all very close to 1.0) and explained a percentage of the variation in the data that ranged from 0.07 to 0.43 (Table 1). The full model (i.e. including the effect of texture, trees and shrubs) was the best fit in five of the six forest type–pathogen species combinations. The only exception was Py. spiculum in closed forests, for which a simpler alternative model that ignored the effect of texture and shrubs (‘No texture + shrub model’ in Table 1) had a much lower AICc score (i.e. was much better supported statistically) than the full model. Site-dependent models were never a better fit to the data than more simple site-independent models (results not shown for simplicity), which implies that soil and plant effects on pathogen abundance can be considered to be consistent across sites of the same forest type.

Table 1.   Comparison of the alternative models for the three pathogen species (Phytophthora cinnamomi, Pythium spiculum and Pythium spp.) in the two forest types using the Akaike Information Criterion corrected for small sample sizes (AICc)
Forest typePathogen speciesAICcType of tree effectNPnSlopeR2
FullNo textureNo treeNo shrubNo texture + shrubNull
  1. The full model includes the effect of texture, tree neighbors and shrub neighbors on each pathogen species. The ‘No texture’, ‘No tree’, ‘No shrub’ and ‘No texture + shrub’ models ignore the effect of these factors, respectively. The ‘Type of tree effect’ column indicates whether the best model considered all tree species as equivalent (‘Equiv.’), differentiated among species (‘Sp’) and health status (‘H’), or considered the legacy effect of dead trees (‘D’). The most parsimonious model (indicated in bold) is that with the lowest AICc. NP is the total number of parameters in the best model, and n is the sample size. The slope and R2 for the relationship between predicted and observed pathogen abundance are also given.

WoodlandsP. cinnamomi16 59717 57618 91017 73517 75818 914Sp + H + D12901.000.35
Py. spiculum614639650619642654Equiv.7601.000.07
Pythium spp.203524722510219424862565Sp + H + D11600.960.43
Closed forestsP. cinnamomi424144766020581359116038Sp + H + D11901.080.36
Py. spiculum128126131121118129Equiv.5601.070.18
Pythium spp.178922372270207022532298Sp + H + D11601.010.28

The proportion of sand in the soil always had a negative effect on pathogen abundance (i.e. negative b parameter; Table S2). The magnitude of the texture effect (indicated by the magnitude of the b parameter) was larger in woodlands than in closed forests for all three pathogen species (Fig. 1). Within forest types, the texture effect also varied among pathogen groups (i.e. support intervals for the b parameter did not overlap), being larger for Pythium spp. > P. cinnamomi > Py. spiculum (Fig. 1).

image

Figure 1. Predicted effect of soil texture (proportion of sand) on the potential abundance of Phytophthora cinnamomi (solid line), Pythium spiculum (dashed line) and Pythium spp. (dashed–dotted line) in woodlands (a) and closed forests (b). The texture effect on potential abundance is calculated using Eqn 2, and values of the b parameter are reported in Supporting Information Table S2. y values of < 1 indicate a negative effect of sand proportion on potential pathogen abundance.

Download figure to PowerPoint

Much of the variation in soil-borne pathogen abundance was explained by the tree neighborhood from which the soil was sampled. Thus, exclusion of the tree effect from the full model (‘No tree’ model in Table 1) always caused a much larger increase in AICc than exclusion of either the texture or shrub effect (Table 1). In all models, α = 1 offered a better fit to the data than α = 0, indicating that the tree influence on pathogen abundance was proportional to its size. The effect of distance to neighbors on pathogen abundance (controlled by parameters β and γ in Eqn 5; Table S2) was, however, not consistent among pathogen species. The decline in distance varied from very steep in P. cinnamomi to virtually null in Pythium spp., for which the abundance was only proportional to the host density (Fig. 2).

image

Figure 2. Predicted change in the tree neighborhood index (NITree) as a function of distance to a neighbor for the three studied pathogen species in woodlands and closed forests. NITree is calculated using Eqn 5 and values of the γ and β parameters given in Supporting Information Table S2 (λ = 1 and α = 0 for simplicity of presentation of results). To facilitate comparison among pathogen species, NITree values are shown standardized (i.e. divided by the maximum value for the species).

Download figure to PowerPoint

For P. cinnamomi and Pythium spp. in both forest types, models that discriminated among living trees of different species and health status, and included the legacy effect of dead trees (i.e. calculated five different λ values), provided a much better fit to the data (i.e. had lower AICc) than simpler models that ignored species or health differences (Table 1). In woodlands, λ values varied from very positive in highly defoliated Q. suber trees to largely negative in O. europaea (Table S2). This is because neighborhoods dominated by healthy Q. suber trees had a lower abundance of P. cinnamomi and Pythium spp. than those dominated by symptomatic Q. suber trees, but a higher abundance than neighborhoods dominated by dead Q. suber and O. europaea trees (Fig. 3). Similarly, in closed forests, neighborhoods dominated by healthy Q. suber trees had lower P. cinnamomi and Pythium spp. abundance than those dominated by symptomatic Q. suber trees. In this forest type, however, healthy Q. suber neighborhoods also had lower pathogen abundance than neighborhoods of the coexisting species Q. canariensis (Table S2, Fig. 3). Finally, for Py. spiculum, models that grouped all tree species as equivalent always had the largest empirical support (i.e. lower AICc; Table 1). It is likely that the substantially lower abundance of Py. spiculum relative to the other two pathogen groups limited the capacity of the models to detect complex spatial patterns for this species. In both forest types, the abundance of Py. spiculum varied positively with tree abundance in its neighborhood (Table S2, Fig. 3).

image

Figure 3. Predicted effects of variation in neighbor identity and quantity on abundance (measured in colony-forming units per gram of dry soil) of Phytophthora cinnamomi in woodlands (a), P. cinnamomi in closed forests (b), Pythium spiculum in woodlands (c), Py. spiculum in closed forests (d), Pythium spp. in woodlands (e) and Pythium spp. in closed forests (f). Neighbor types are given by the best model for each pathogen species (Table 1). For Py. spiculum, the best model considered all trees as a single group. Pathogen abundance is calculated using Eqns 1–6 and optimum texture values, standard 30-cm tree neighbors (the average tree size across study sites) at 2-m distance from target soils and no shrubs. For each combination of pathogen species and forest type, only the site with the largest potential pathogen abundance (PPASite in Supporting Information Table S2) is represented.

Download figure to PowerPoint

The effect of shrubs on pathogen abundance varied strongly among forest types, being negative in woodlands, but positive (P. cinnamomi and Pythium spp.) or neutral (Py. spiculum) in closed forests (Fig. 4). The magnitude of the effect did not vary among pathogen species in most cases, as indicated by the overlapping values of the d parameter (Table S2).

image

Figure 4. Predicted effect of variation in the shrub neighborhood index (NIShrub) on the potential abundance of Phytophthora cinnamomi (solid line), Pythium spiculum (dashed line) and Pythium spp. (dashed–dotted line) in woodlands (a) and closed forests (b). The shrub effect on potential abundance is calculated using Eqn 4 and values of the d parameter reported in Supporting Information Table S2. y values of < 1 indicate a negative effect of shrubs on potential pathogen abundance, whereas values > 1 indicate a positive effect.

Download figure to PowerPoint

Effect of soil-borne pathogens on seedling emergence and survival

Among the nine combinations of forest type–seedling species–pathogen species tested, we only found support for an effect of P. cinnamomi on the emergence of Q. suber seedlings in woodlands (Table 2). This effect varied among sites, as indicated by the fact that a site-specific model was a better fit to the data than a simpler linear model (Table 2). Thus, P. cinnamomi had a large negative effect on Q. suber emergence in two sites (Cinchao and Picacho) and a neutral effect (i.e. support interval for the b parameter overlaps zero; Table S3) in one site (Ahumada, Fig. 5).

Table 2.   Comparison of alternative models analyzing the effect of pathogens (Phytophthora cinnamomi, Pythium spiculum and Pythium spp.) on Quercus seedling emergence and survival
Forest typeSeedling speciesPathogen speciesAICcNPnSlopeR2
Site-specificLinearNull
  1. The most parsimonious model (indicated in bold) is that with the lowest Akaike Information Criterion corrected for small sample sizes (AICc). The ‘Site-specific’ model considers differential pathogen effects among sites, the ‘Linear’ model a homogeneous pathogen effect among sites, and the ‘Null’ model the absence of pathogen effects. NP is the total number of parameters in the best model, and n is the sample size. The slope and R2 for the relationship between predicted and observed emergence/survival are given for best models other than the null.

Emergence
 WoodlandsQ. suberP. cinnamomi363.01368.56369.136901.000.12
Py. spiculum244.34242.65240.65360  
Pythium spp.259.53256.24255.72360  
 Closed forests Q. suberP. cinnamomi392.02388.01386.19388  
Py. spiculum258.01255.47251.97258  
Pythium spp.296.66292.25290.25258  
Q. canariensisP. cinnamomi423.27416.38414.48388  
Py. spiculum293.34291.79289.88258  
Pythium spp.277.07275.80273.84258  
Survival
 WoodlandsQ. suberP. cinnamomi190.68194.81195.336861.040.16
Py. spiculum145.89142.23140.60257  
Pythium spp.121.55119.29117.31256  
 Closed forestsQ. suberP. cinnamomi230.42226.33224.60381  
Py. spiculum143.98141.00139.06251  
Pythium spp.159.03156.00154.02256  
Q. canariensisP. cinnamomi282.49278.44279.273820.980.14
Py. spiculum181.08178.67176.69252  
Pythium spp.192.79190.59191.773560.970.04
image

Figure 5. Probability of emergence (a) and survival (b) of Quercus suber seedlings in the three woodland sites as a function of the abundance (colony-forming units per gram of dry soil) of Phytophthora cinnamomi in the soil. Sites: Ahumada (dashed–dotted line), closed circles; Cinchao (solid line), open circles; Picacho (dashed line), crosses.

Download figure to PowerPoint

We found support for an effect of P. cinnamomi on the survival of Q. suber seedlings in woodlands, but not in closed forests (Table 2). The model that incorporated site effects had a lower AICc than a simpler model omitting these effects. Thus, P. cinnamomi had a negative effect on Q. suber survival in only one of the three woodland sites (Ahumada), which happened to be the only site in which P. cinnamomi effects on seedling emergence were not found (Table S3, Fig. 5). Although models incorporating pathogen effects were the most parsimonious fit in two other situations – effects of P. cinnamomi and Pythium spp. on Q. canariensis survival – the differences in AICc from the null model were < 2 units (Table 2), and therefore do not provide strong support for a pathogen effect on survival of this species.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Our results indicate that pathogen abundance in the forest soil is not randomly distributed, but exhibits spatially predictable patterns influenced by both abiotic (soil texture) and, particularly, biotic (tree and shrub species) factors. The relative importance of each factor on soil-borne pathogen abundance varied among forest types and/or pathogen species, revealing the complexity of the pathogen landscape. We also found that the spatial variability in the pathogen community had significant ecological consequences by affecting the performance of tree seedlings under natural field conditions, but only for particular combinations of species and sites. Our findings suggest that heterogeneous spatial patterns of pathogen abundance at fine spatial scale can have important implications for the dynamics and restoration of declining Mediterranean oak forests.

Drivers of soil-borne pathogen abundance: the role of soil texture

Our models showed a consistent negative effect of soil sand content on pathogen abundance, presumably as a result of the direct influence of texture on water availability. Sandy soils have lower water-holding capacity and higher percolation rates than do clayish soils (Brady & Weil, 2008). Therefore, they are less prone to suffer temporal waterlogging conditions that strongly benefit pathogen abundance and disease development (Hendrix & Campbell, 1973; Weste & Marks, 1987). The texture effect was much larger in woodlands than in closed forests for the three pathogen species, probably because the closed forest soils were all very sandy (Table S1). Indeed, the sandier soils of closed forests could also explain why they showed lower loads of all pathogen species than did woodlands (Fig. 3). Our results therefore suggest that texture is an important abiotic driver of soil-borne pathogen variation at both the local and landscape scale.

Drivers of soil-borne pathogen abundance: the role of the tree community

One main finding of this study is the strong empirical support found for a spatial concordance among the distribution and health status of trees of different species and the abundance of soil-borne pathogens in the soil. Phytophthora cinnamomi and Pythium spp. were much more abundant under declining Q. suber trees, particularly those already showing a high defoliation level (> 50%), than under healthy Q. suber trees. Although our observational approach does not allow the separation of the cause and effect in the tree–pathogen interaction, the finding of a concomitant increase in the abundance of pathogens in the soil and defoliation in the canopy is consistent with the predictions of the hypothesis of decline development in oak forests, which proposes a tree–pathogen feedback process (Brasier, 1996). According to this hypothesis, the loss of fine roots by soil-borne pathogens may translate into a loss of leaf area aboveground. The opening of the canopy triggers a series of environmental changes (e.g. higher soil temperature, reduced organic matter content and microbial activity) that might, in turn, favor pathogen development, giving rise to a feedback loop in which pathogens under the trees produce changes at the canopy level that favor the build-up of larger pathogen loads, eventually killing the tree. However, the fact that our models supported a negative effect of dead trees on soil pathogen abundance (negative λ, Table S2) suggests that, once a tree dies, its legacy is a gap with lower pathogen abundance than the surrounding forest matrix. These gaps could play a role of refuge for the establishment of susceptible species, as reported for canopy gaps in tropical and cool temperate forests (Augspurger, 1984; O’Hanlon-Manners & Kotanen, 2004, 2006; Reinhart et al., 2010).

Our results indicate that tree species can play very different roles in the pathogen landscape. Thus, among the species co-existing with the susceptible Q. suber in our study sites, O. europaea neighborhoods seem to suppress pathogens (pathogen abundance was lower under O. europaea than in neighborhoods without trees), whereas Q. canariensis seems to act as a reservoir without showing any apparent disease symptoms (Fig. 3). These results help us to understand the nature of plant–plant interactions mediated by pathogens in these forests. Specifically, they suggest that, whereas O. europaea trees could indirectly benefit Q. suber by acting as refuges for its recruitment, the presence of Q. canariensis could result in apparent competition (Cobb et al., 2010) by promoting pathogens that harm Q. suber more strongly than itself. These complex indirect effects, although largely ignored in the literature, show the need to use a community approach when trying to explain patterns of spatial variation in disease dynamics for particular tree species (e.g. Janzen–Connell effects; Mordecai, 2011).

Our neighborhood approach allowed valuable insights to be gained into the role of tree size and distance as determinants of pathogen abundance. This type of information is extremely rare in the literature on plant–pathogen interactions, as most studies do not quantify pathogen abundance, but measure disease expression directly, which can be affected by additional factors, such as host susceptibility or environmental conditions (e.g. Augspurger & Kelly, 1984; Gilbert et al., 1994; Packer & Clay, 2000, 2003; but see Reinhart & Clay, 2009). First, our models indicate that the tree effect on pathogen abundance is not independent of its size, with larger trees hosting larger pathogen communities. This result provides empirical support for the hypothesis of the importance of dbh as a source of intraspecific variation in plant–pathogen interactions (Reinhart & Clay, 2009), and calls for the inclusion of this plant trait as a covariable in experimental and observational studies of pathogen abundance and disease. Second, our models show that the decline in the net effect of a neighbor tree within the 15-m neighborhood varies strongly among pathogen species, from rather sharp for P. cinnamomi (tending to zero within 5–6 m) to virtually null for Pythium spp. (Fig. 2). This result suggests that pathogen species can vary strongly in their scale of spatial variation, with some showing heterogeneous patterns at smaller scales than others.

Drivers of soil-borne pathogen abundance: the role of the shrub community

We found that not only the tree community, but also the shrub community, had a strong effect on soil-borne pathogen abundance, supporting previous studies that have emphasized the relevance of the understory as a driver of the soil microbial community (Nilsson & Wardle, 2005; Wu et al., 2011). However, the sign of the shrub effect was not consistent among forest types, being negative in woodlands but positive in closed forests. A probable explanation for this difference is that the net effect of the understory is driven by the identity of the dominant shrub species, which differs among forest types. Thus, despite their similar species composition, species relative abundance changed from a dominance of Pistacia lentiscus in woodlands to a dominance of Erica spp. in closed forests. These two species vary strongly in their litter quality and effects on soil fertility: Pistacia lentiscus forms islands of fertility rich in organic matter (Armas & Pugnaire, 2009), whereas Erica spp. produces low-quality litter and is indicative of acidic nutrient-poor soils (Van Vuuren & Berendse, 1993; Zas & Alonso, 2002). Because an acidic pH, low nutrient content and low organic matter favor soil-borne pathogen growth and disease expression (Weste & Marks, 1987; Jönsson et al., 2003; Serrano et al., 2012), soils under Erica spp. could be expected to provide more favorable conditions for pathogen build-up than soils under Pistacia lentiscus. Although this hypothesis remains to be tested, our results highlight the strong variability in the understory effects on pathogen populations that can be expected even in forests with similar shrub species composition.

Pathogen effects on seedling emergence and survival: implications for regeneration dynamics

Our models of seedling emergence and survival indicate that, under natural field conditions, the spatial variability of soil-borne pathogen abundance translates into spatial variation in seedling performance, but not for all species or forest types. Indeed, we only found support for a negative effect of P. cinnamomi on emergence and survival of Q. suber seedlings in woodlands. The fact that we detected negative effects of P. cinnamomi, but not of Py. spiculum or Pythium spp., on seedling performance is probably influenced by its much larger abundance in the studied forests, but could also be indicative of the larger aggressiveness of this species (Romero et al., 2007). Differences in P. cinnamomi abundance, much greater in woodlands than in closed forests, could also explain its stronger effects in the former forests. These findings therefore suggest that certain thresholds in pathogen abundance need to be overcome before they translate into measurable effects on seedling performance in the field. Interestingly, these thresholds seem to be higher than those expected based on pathogenicity trials, where much lower P. cinnamomi abundances (< 500 cfu g−1) are lethal for Mediterranean Quercus seedlings (Sánchez et al., 2002; Serrano et al., 2012).

Our results do not support the hypothesis of lower susceptibility to pathogens in shade-tolerant than shade-intolerant species (Augspurger, 1990; McCarthy-Neumann & Kobe, 2008). Unfortunately, the lack of emergence of O. europaea precluded the exploration of interspecific differences in pathogen susceptibility in woodlands. However, in closed forests, neither Q. suber nor Q. canariensis seedlings were negatively affected by pathogens, despite their differences in shade tolerance. A question that remains to be answered is whether differences among the two Quercus species will emerge at larger pathogen abundances, such as those found in woodlands, where Q. suber emergence and survival are severely impaired by P. cinnamomi (Fig. 5). To date, our findings do not support a significant role of pathogens as promoters of species coexistence through species-specific effects at the seedling level. On the contrary, because adults of Q. suber are much more susceptible to pathogen attack than are adults of Q. canariensis, soil-borne pathogens seem more likely to play a role in species exclusion than in coexistence in the studied forests.

Although our results indicate important net impacts of pathogens on tree seedlings under natural conditions, the rather low explanatory power of the emergence and survival models should be taken as evidence that pathogens are just one of many other relevant abiotic and biotic drivers of natural patterns of recruitment. These drivers could interact with each other, a given abundance of soil-borne pathogens having implications for seedling performance only under specific environmental situations, such as low light availability or low mycorrhizal abundance (Hood et al., 2004; Morris et al., 2007). Further studies that explore the simultaneous effect of multiple abiotic and biotic drivers of seedling performance are clearly needed to advance our understanding of the factors affecting the expression of disease in forest ecosystems.

Concluding remarks

This study provides new insights into the highly complex spatial distribution of soil-borne pathogens and reveals the degree to which soil characteristics and the woody plant community can explain pathogen abundance in forest soils. Because we have shown that the spatial variation in pathogen abundance can affect the recruitment of susceptible tree species, such as Q. suber, these findings might be useful in the restoration of forests affected by pathogen-driven decline, which frequently involves the planting of seeds or seedlings of susceptible species to replace dead trees in the future (Tuset & Sánchez, 2004). Specifically, our results could help to choose those planting microsites in which seedling emergence and survival would have a lower probability of being impaired by soil-borne pathogens, hence maximizing the economic and ecological benefits of restoration efforts.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank the director and technicians of the Los Alcornocales Natural Park for facilities and support to carry out the field work. We are also indebted to Pierre Callier, Eduardo Gutiérrez and Ana Pozuelos for invaluable laboratory and field assistance. This study was supported by the Ministerio de Ciencia e Innovación (MICIIN) project INTERBOS (CGL2008-4503-C03-01), the Organismo Autónomo de Parques Nacionales/Ministerio de Medio Ambiente, Rural y Marino (OAPN/MIMARM) project DECALDO (091/2009) and European Fondo Europeo de Desarrollo Regional (FEDER) funds. B.I. was supported by a Formación de Personal Investigador (FPI)-MICIIN grant, J.M.A. by a Junta para la Ampliación de Estudios (JAE)pre-CSIC grant and I.M.P.R. by a JAEdoc-Consejo Superior de Investigaciones Científicas (CSIC) contract.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Agrios JN. 2005. Plant pathology, 5th edn. London, UK: Elsevier Academic Press.
  • Anonymous. 2005. PORN/PRUG/PDS Parque Natural Los Alcornocales. Sevilla, Spain: Junta de Andalucía, Consejería de Medio Ambiente.
  • Aponte C, Marañón T, García LV. 2010. Microbial C, N and P in soils of Mediterranean oak forests: influence of season canopy cover and soil depth. Biogeochemistry 101: 7792.
  • Armas C, Pugnaire FI. 2009. Ontogenetic shifts in interactions of two dominant shrub species in a semi-arid coastal sand dune system. Journal of Vegetation Science 20: 535546.
  • Augspurger CK. 1984. Seedling survival of tropical tree species: interactions of dispersal distance, light gaps and pathogens. Ecology 65: 17051712.
  • Augspurger CK. 1990. Spatial patterns of damping-off disease during seedling recruitment in tropical forests. In: Burton J, Leather S, eds. Pests, pathogens and plant communities. Oxford, UK: Blackwell Scientific, 131143.
  • Augspurger CK, Kelly CK. 1984. Pathogen mortality of tropical tree seedlings: experimental studies of the effects of dispersal distance, seedling density, and light conditions. Oecologia 61: 211217.
  • Augspurger CK, Wilkinson HT. 2007. Host specificity of pathogenic Pythium species: implications for tree species diversity. Biotropica 39: 702708.
  • Baribault TW, Kobe RK. 2011. Neighbour interactions strengthen with increased soil resources in a northern hardwood forest. Journal of Ecology 99: 13581372.
  • Brady NC, Weil RR. 2008. The nature and properties of soils, 14th edn. Upper Saddle River, NJ, USA: Pearson-Prentice Hall.
  • Brasier CM. 1992. Oak tree mortality in Iberia. Nature 360: 539.
  • Brasier CM. 1996. Phytophthora cinnamomi and oak decline in southern Europe: environmental constraints including climate change. Annales des Sciences Forestières 53: 347358.
  • Brasier CM, Robredo F, Ferraz JFP. 1993. Evidence for Phytophthora cinnamomi involvement in Iberian oak decline. Plant Pathology 42: 140145.
  • Burnham KP, Anderson DR. 2002. Model selection and multimodel inference: a practical information-theoretic approach. New York, NY, USA: Springer.
  • Cahill DM, Rookes JE, Wilson BA, Gibson L, McDougall KL. 2008. Phytophthora cinnamomi and Australia’s biodiversity: impacts, predictions and progress towards control. Australian Journal of Botany 56: 279310.
  • Canham CD, Uriarte M. 2006. Analysis of neighborhood dynamics of forest ecosystems using likelihood methods and modeling. Ecological Applications 16: 6273.
  • Coates KD, Canham CD, LePage PT. 2009. Above- versus below-ground competitive effects and responses of a guild of temperate tree species. Journal of Ecology 97: 118130.
  • Cobb RC, Meentemeyer RK, Rizzo DM. 2010. Apparent competition in canopy trees determined by pathogen transmission rather than susceptibility. Ecology 91: 327333.
  • Dhingra OD, Sinclair JB. 1995. Basic plant pathology methods. Boca Raton, FL, USA: CRC Press.
  • Edwards AWF. 1992. Likelihood, 2nd edn. Baltimore, MD, USA: Johns Hopkins University Press.
  • Erwin DC, Ribeiro OK. 1996. Phytophthora diseases worldwide. St Paul, MN, USA: APS Press.
  • Ettema CH, Wardle DA. 2002. Spatial soil ecology. Trends in Ecology and Evolution 17: 177183.
  • García LV, Ramo C, Aponte C, Moreno A, Domínguez MT, Gómez-Aparicio L, Redondo R, Marañón T. 2011. Protected wading bird species threaten relict centennial cork oaks in a Mediterranean Biosphere Reserve: a conservation management conflict. Biological Conservation 144: 764771.
  • Gee GW, Bauder JW. 1986. Particle-size analysis. In: Klute A, ed. Methods in soil analysis, Part 1. Physical and mineralogical methods. Madison, WI, USA: American Society of Agronomy and Soil Science Society of America, 383409.
  • Gilbert GS, Hubbell SP, Foster RB. 1994. Density and distance to adult effects on canker disease of trees in a moist tropical forest. Oecologia 98: 100108.
  • Goffe WL, Ferrier GD, Rogers J. 1994. Global optimization of statistical functions with simulated annealing. Journal of Econometrics 60: 6599.
  • Gómez JM. 2004. Importance of burial and microhabitat on Quercus ilex early recruitment: non-additive effects on multiple demographic processes. Plant Ecology 172: 287297.
  • Gómez-Aparicio L. 2008. Spatial patterns of recruitment in a Mediterranean tree (Acer opalus subsp. granatense): linking the fate of seeds, seedlings, and saplings in heterogeneous landscapes at different scales. Journal of Ecology 96: 11281140.
  • Gómez-Aparicio L, Canham CD, Martin PH. 2008a. Neighborhood models of the effects of the invasive Acer platanoides on tree seedling dynamics: linking impacts on communities and ecosystems. Journal of Ecology 96: 7890.
  • Gómez-Aparicio L, Pérez-Ramos IM, Mendoza I, Quero JL, Matías L, Castro J, Zamora R, Marañón T. 2008b. Oak seedling survival and growth along resource gradients in Mediterranean forests: implications for regeneration under current and future environmental scenarios. Oikos 117: 16831699.
  • Goyiatzis DG, Porlingis IC. 1987. Temperature requirements for the germination of olive seeds (Olea europaea L.). Journal of Horticultural Sciences 62: 405411.
  • Hendrix FF, Campbell WA. 1973. Pythiums as plant pathogens. Annual Review of Plant Pathology 11: 7798.
  • Hood LA, Swaine MD, Mason PA. 2004. The influence of spatial patterns of damping-off disease and arbuscular mycorrhizal colonization on tree seedling establishment in Ghanaian tropical forest soil. Journal of Ecology 92: 816823.
  • Jiménez JJ, Sánchez JE, Romero MA, Belbahri L, Trapero A, Lefort F, Sánchez ME. 2008. Pathogenicity of Pythium spiculum and Pythium sterilum on feeder roots of Quercus rotundifolia. Plant Pathology 57: 369.
  • Johnson JB, Omland KS. 2004. Model selection in ecology and evolution. Trends in Ecology and Evolution 19: 101108.
  • Jönsson U, Jung T, Rosengren U, Nihlgard B, Sonesson K. 2003. Pathogenicity of Swedish isolates of Phytophthora quercina to Quercus robur in two different soils. New Phytologist 158: 355364.
  • Martin FN, Loper JE. 1999. Soilborne plant diseases caused by Pythium spp: ecology, epidemiology, and prospects for biological control. Critical Reviews in Plant Sciences 18: 111181.
  • Matías L, Gómez-Aparicio L, Zamora R, Castro J. 2011. Disentangling the complex relationships among resources and early plant community recruitment: an experimental approach. Perspectives in Plant Ecology, Evolution and Systematics 13: 277285.
  • McCarthy-Neumann S, Kobe RK. 2008. Tolerance of soil pathogens co-varies with shade tolerance across species of tropical tree seedlings. Ecology 89: 18831892.
  • Médail F, Quézel P. 1999. Biodiversity hotspots in the Mediterranean Basin: setting global conservation priorities. Conservation Biology 13: 15101513.
  • Moralejo E, García-Muñoz JA, Descals E. 2009. Susceptibility of Iberian trees to Phytophthora ramorum and P. cinnamomi. Plant Pathology 58: 271283.
  • Mordecai EA. 2011. Pathogen impacts on plant communities: unifying theory, concepts, and empirical work. Ecological Monographs 81: 429441.
  • Morris WF, Hufbauer RA, Agrawal AA, Bever JD, Borowicz VA, Gilbert GS, Maron JL, Mitchell CE, Parker IM, Power AG et al. 2007. Direct and interactive effects of enemies and mutualists on plant performance: a meta-analysis. Ecology 88: 10211029.
  • Nilsson MC, Wardle DA. 2005. Understory vegetation as a forest ecosystem driver: evidence from the northern Swedish boreal forest. Frontiers in Ecology and the Environment 8: 421428.
  • O’Hanlon-Manners DL, Kotanen PM. 2004. Evidence that fungal pathogens inhibit recruitment of a shade-intolerant tree, white birch (Betula papyrifera), in understory habitats. Oecologia 140: 650653.
  • O’Hanlon-Manners DL, Kotanen PM. 2006. Losses of seeds of temperate trees to soil fungi: effects of habitat and host ecology. Plant Ecology 187: 4958.
  • Ojeda F, Marañón T, Arroyo J. 2000. Plant diversity patterns in the Aljibe Mountains (S. Spain): a comprehensive account. Biodiversity and Conservation 9: 13231343.
  • Packer A, Clay K. 2000. Soil pathogens and spatial patterns of seedling mortality in a temperate tree. Nature 404: 278281.
  • Packer A, Clay K. 2003. Soil pathogens and Prunus serotina seedling and sapling growth near conspecific trees. Ecology 84: 108119.
  • Paul B, Bala K, Belbahri L, Calmin G, Sánchez-Hernández ME, Lefort F. 2006. A new species of Pythium with ornamented oogonia: morphology, taxonomy, ITS region of its rDNA, and its comparison with related species. FEMS Microbiology Letters 254: 317323.
  • Pérez-Ramos IM, Marañón T. 2012. Community-level seedling dynamics in Mediterranean forests: uncoupling between the canopy and the seedling layers. Journal of Vegetation Science. doi: 10.1111/j.1654-1103.2011.01365.x.
  • Pérez-Ramos IM, Urbieta IR, Zavala MA, Marañón T. 2012. Ontogenetic conflicts and rank reversals in two Mediterranean oak species: implications for coexistence. Journal of Ecology 100: 467477.
  • Reinhart KO, Clay K. 2009. Spatial variation in soil-borne disease dynamics of a temperate tree, Prunus serotina. Ecology 90: 29842993.
  • Reinhart KO, Royo AA, Kageyama SA, Clay K. 2010. Canopy gaps decrease microbial densities and disease risk for a shade-intolerant tree species. Acta Oecologica 36: 530536.
  • Rey P, Alcántara JM, Valera F, Sánchez-Lafuente AM, Garrido JL, Ramírez JM, Manzaneda AJ. 2004. Seedling establishment in Olea europaea: seed size and microhabitat affect growth and survival. Ecoscience 11: 310320.
  • Rizzo DM, Garbelotto M, Hansen EA. 2005. Phytophthora ramorum: integrative research and management of an emerging pathogen in California and Oregon forests. Annual Review of Phytopathology 43: 309335.
  • Romero MA, Sánchez JE, Jiménez JJ, Belbahri L, Trapero A, Lefort F, Sánchez ME. 2007. New Pythium taxa causing root rot on Mediterranean Quercus species in South-West Spain and Portugal. Journal of Phytopathology 155: 289295.
  • Sánchez ME, Caetano P, Ferraz J, Trapero A. 2002. Phytophthora disease of Quercus ilex in south-western Spain. Forest Pathology 32: 518.
  • Sánchez ME, Caetano P, Romero MA, Navarro RM, Trapero A. 2006. Phytophthora root rot as the main factor of oak decline in southern Spain. In: Brasier C, Jung T, Oßwald W, eds. Progress in research on Phytophthora diseases of forest trees. Farnham, UK: Forest Research, 149154.
  • Serrano MS, De Vita P, Fernández-Rebollo P, Sánchez ME. 2012. Calcium fertilizers induce soil suppressiveness to Phytophthora cinnamomi root rot in Quercus ilex. European Journal of Plant Pathology 132: 271279.
  • Tuset JJ, Sánchez G. 2004. La Seca: el decaimiento de encinas, alcornoques y otros Quercus en España. Madrid, Spain: Ministerio de Medio Ambiente.
  • Van Vuuren MMI, Berendse F. 1993. Changes in soil organic matter and net nitrogen mineralization in heathland soils after removal, addition or replacement of litter from Erica tetralix or Molinia caerulea. Biology and Fertility of Soils 15: 268274.
  • Wardle DA. 2002. Communities and ecosystems: linking the aboveground and belowground components. Princeton, NJ, USA: Princeton University Press.
  • Weste G, Marks GC. 1987. The biology of Phytophthora cinnamomi in Australasian forests. Annual Review of Phytopathology 25: 207229.
  • Wu JP, Liu ZF, Wang XL, Sun YX, Zhou LX, Lin YB, Fu SL. 2011. Effects of understory removal and tree girdling on soil microbial community composition and litter decomposition in two Eucalyptus plantations in South China. Functional Ecology 25: 921931.
  • Zas R, Alonso M. 2002. Understory vegetation as indicators of soil characteristics in northwest Spain. Forest Ecology and Management 171: 101111.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Table S1 Description of the main characteristics of the six study sites located in the South (S), Center (C) and North (N) of the Los Alcornocales Natural Park

Table S2 Parameter estimates and two-unit support intervals (in parentheses) for the best model selected for each combination of soil pathogen species and forest type

Table S3 Parameter estimates and two-unit support intervals for the best models of Phytophthora cinnamomi effects on the emergence and survival of Quercus suber and Quercus canariensis seedlings

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

FilenameFormatSizeDescription
NPH_4108_sm_TableS1-S3.docx55KSupporting info item