Interregional variation in the floristic recovery of post-agricultural forests


Correspondence author. E-mail:


1. Worldwide, the floristic composition of temperate forests bears the imprint of past land use for decades to centuries as forests regrow on agricultural land. Many species, however, display significant interregional variation in their ability to (re)colonize post-agricultural forests. This variation in colonization across regions and the underlying factors remain largely unexplored.

2. We compiled data on 90 species and 812 species × study combinations from 18 studies across Europe that determined species’ distribution patterns in ancient (i.e. continuously forested since the first available land use maps) and post-agricultural forests. The recovery rate (RR) of species in each landscape was quantified as the log-response ratio of the percentage occurrence in post-agricultural over ancient forest and related to the species-specific life-history traits and local (soil characteristics and light availability) and regional factors (landscape properties as habitat availability, time available for colonization, and climate).

3. For the herb species, we demonstrate a strong (interactive) effect of species’ life-history traits and forest habitat availability on the RR of post-agricultural forest. In graminoids, however, none of the investigated variables were significantly related to the RR.

4. The better colonizing species that mainly belonged to the short-lived herbs group showed the largest interregional variability. Their recovery significantly increased with the amount of forest habitat within the landscape, whereas, surprisingly, the time available for colonization, climate, soil characteristics and light availability had no effect.

5.Synthesis. By analysing 18 independent studies across Europe, we clearly showed for the first time on a continental scale that the recovery of short-lived forest herbs increased with the forest habitat availability in the landscape. Small perennial forest herbs, however, were generally unsuccessful in colonizing post-agricultural forest – even in relatively densely forested landscapes. Hence, our results stress the need to avoid ancient forest clearance to preserve the typical woodland flora.


Large parts of the present-day forest area in Europe and eastern North America have been cleared for agriculture and subsequently reforested (spontaneous or planted) during the past centuries (Flinn & Vellend 2005). In several regions, only a small fraction of the actual forest cover can be referred to as ‘ancient’. These ancient forests (AF) have no historical record (mainly cartographical) of agricultural land use and have generally been continuously wooded for at least c. 150–400 years (recent reviews: Hermy et al. 1999; Flinn & Vellend 2005; Hermy & Verheyen 2007). Forest understorey plant species need to (re)colonize post-agricultural forest sites from relict populations in AF or hedgerows, which are often scarce and isolated in contemporary landscapes. Previous work has demonstrated that a large number of forest plants are mostly unsuccessful in (re)colonizing isolated post-agricultural forests so that the community composition bears the imprint of past land use for decades to centuries (Peterken & Game 1984; Whitney & Foster 1988; Dupouey et al. 2002). Today, there is a general consensus that the low dispersal capacity of many forest plants may principally account for their low colonization rates (Verheyen et al. 2003c; Takahashi & Kamitani 2004; Matlack 2005). Also the low recruitment rates into the adult life stage and adult survival pose a bottleneck in the long-term colonization of understorey species (Donohue, Foster & Motzkin 2000; Jacquemyn & Brys 2008; Baeten, Hermy & Verheyen 2009).

Many studies have compared the actual floristic composition of ancient and post-agricultural forest in a specific region or local landscape and classified species into fast- vs. slow-colonizing subclasses (Peterken & Game 1984; Wulf 1997; Honnay, Degroote & Hermy 1998; Graae 2000; Verheyen et al. 2003c, 2006) or calculated species-specific colonization rates (Brunet & Von Oheimb 1998; Bossuyt, Hermy & Deckers 1999; Dzwonko 2001b; Orczewska 2009b). Other studies compared the life-history traits (e.g. seed production, dispersal type, capacities for vegetative spread, etc.) of slow- vs. faster colonizing forest plants (Honnay, Degroote & Hermy 1998; Hermy et al. 1999; Graae & Sunde 2000; Verheyen et al. 2003c). However, a comparison of the colonization rates within species and across regions suggests that forest plants may differ considerably in their capacity to colonize post-agricultural sites (Hermy & Stieperaere 1981; Hermy et al. 1999; Orczewska 2009b). Peterken (1974) has stressed the need to develop region-specific lists of fast- vs. slow-colonizing species.

Yet, a broader biogeographical approach in which the (interactive) effects of local and regional factors such as climate, soils and habitat availability are assessed is still lacking. Hence, little is known about the factors controlling the interregional differences in recovery rates (RR) of understorey forest plant populations into post-agricultural forests on a continental scale. It can be expected that several factors are of influence (cf. Wright & Fridley 2010). First, slow-colonizing forest plants tend to produce heavier seeds, which show higher germinability and better performance of seedlings in warmer regions (De Frenne et al. 2009, 2010b; Graae et al. 2009). Temperature also has a positive effect on the potential growth rates of understorey plants (Farnsworth et al. 1995; De Frenne et al. 2010a). Together, this suggests that variables related to climate may explain differences in colonization behaviour of forest plants among regions (Peterken 1974; De Frenne et al. 2010b; Wright & Fridley 2010). Second, it can be expected that local site conditions affect colonization rates since variation in soil characteristics (e.g. pH, nutrients) and light availability across sites have been shown to affect the abundance and performance of understorey plants (Dupré & Ehrlen 2002; Verheyen et al. 2003a, 2006; Graae et al. 2004; Kolb & Diekmann 2004; Baeten et al. 2010). Moreover, Wright & Fridley (2010) recently showed that the succession rate of woody species (expressed as the years to 10% and 50% woody cover) in old fields across eastern North America is significantly related to latitudinal gradients in temperature (expressed as growing degree-days) and soil fertility (indirectly measured through cation exchange capacity and pre-agroindustrial maize yields). Finally, regional differences in landscape properties such as the habitat availability (proportion of forest cover within a landscape) and the time since agricultural abandonment (age of the post-agricultural forests) can also affect the colonization chances of forest herbs (Grashof-Bokdam & Geertsema 1998; Jacquemyn, Butaye & Hermy 2001; Graae et al. 2004; Brunet 2007). By combining a meta-analysis of ten studies in Europe and eastern North America with a mainland–island metapopulation model, Vellend (2003), for instance, showed that the ratio of the species richness of post-agricultural over AF decreased significantly when the proportion of AF in the landscape dropped below 10%.

To get insight into the broad-scale variation in RR of forest plant populations, we carried out a meta-analysis of 18 studies that determined distribution patterns of in total 90 individual understorey species within ancient and post-agricultural forest patches in forested landscapes across Europe. We also included the functional traits related to colonization capacity (Verheyen et al. 2003c) in the analyses to compare the effects of each predictor variable among species with contrasting life-histories. More specifically, we aimed (i) to quantify the interregional variation in RR of understorey plants in post-agricultural forest in Europe; (ii) to examine the effects of species’ life-history traits on this variation and (iii) to investigate the importance of both local (soil characteristics and light availability) and regional factors (habitat availability, time available for colonization, and climate) potentially altering this RR.

Materials and methods

Data collection

We searched the literature (peer-reviewed papers but also reports and unpublished theses) for field studies investigating the floristic composition in both ancient and post-agricultural deciduous forests in landscapes across Europe. Studies were only included if they met the following criteria (adapted from Verheyen et al. 2003c): (i) the inventoried forests were situated in lowland or lower mountainous (< 600 m a.s.l.) Europe; (ii) frequency data of the forest plants were available for all (or a representative subsample of) ancient and post-agricultural forest patches in the landscape; (iii) the land-use history and sufficient information on the other predictor variables (see below) of both forest types or the region was given and (iv) canopy closure had already occurred in post-agricultural forests (thus, the analyses were restricted to closed-forest ecosystems). Eighteen geographically non-overlapping studies from nine countries (from central Italy in the south to central Sweden in the north and from the UK in the west to Poland in the east; see Fig. 1) complied with these four criteria and yielded 2231 species × study combinations (693 species) from 1098 ancient and 1370 post-agricultural data collection units (patches or plots; Table 1 and S1 in Supporting Information). However, to avoid a biased estimation of the parameters in the statistical models for species with low frequency, we only focused on those species that were cited in more than five studies. This reduced the species × study combinations to 1332 (155 species).

Figure 1.

 Map of the study sites included in this meta-analysis. The numbers refer to the studies in Tables 1 and S1 in Supporting Information.

Table 1.   Summary of the 18 studies included in the meta-analysis (from north to south). The location of each study area (ID) is depicted in Fig. 1. A more extensive overview of the characteristics of each study area is given in Table S1 in Supporting Information
IDPublicationCountryLat. (°N)Long. (°E)Data collection unit*No. of ancient unitsNo. of PA† unitsNo. of species
  1. *If the data collection units were randomly distributed plots, plot size is given. Otherwise whole forest patches were surveyed.

  2. †Post-agricultural.

  3. S, Sweden; DK, Denmark; GE, Germany; UK, United Kingdom; NL, The Netherlands; PO, Poland; B, Belgium; F, France; and I, Italy.

 1Cousins & Eriksson (2008)S59.417.25 plots (4 m2)patch−1543551
 2Graae (2000)DK56.89.6Plots (625 m2)381045
 3Petersen (1994)DK55.710.9Plots (100–400 m2)9916
 4Brunet (2007)S55.513.2Patch13712630
 5Kolb & Diekmann (2004)GE53.49.4Patch714073
 6Peterken & Game (1984)UK53.3−0.3Patch8927364
 7Wulf (2003)GE53.212.0Plots (200–400 m2)30417442
 8Grashof-Bokdam & Geertsema (1998)NL52.37.0Patch5411010
 9Zacharias (1994)GE52.110.6Patch111168
10Jakubowska-Gabara & Mitka (2007)PO52.020.2Plots (100 m2)161831
11Honnay, Degroote & Hermy (1998)B & F51.03.5Patch584666
12Orczewska (2009a)PO51.017.8Patch232265
13Jacquemyn, Butaye & Hermy (2001)B50.94.9Patch3520460
14Verheyen et al. (2003b)B50.84.6Patch1518211
15Dzwonko & Loster (1989)PO49.919.7Patch63632
16Jamoneau (2010)F49.83.6Patch141570
17Sciama (1999); Sciama et al. (2009)F46.75.6Patch626555
18De Sanctis et al. (2010)I41.612.2Patch452423

To explore the effects of life-history traits related to plant colonization on the RR of post-agricultural forest, we used the emergent groups (EG) of plant species (sensuLavorel et al. 1997) identified by Verheyen et al. (2003c) according to 13 reproductive, vegetative and phenological traits: seed mass, seed size, seed shape, seed production, dispersal type, seed bank, germination requirements, age of first reproduction, growth form, life cycle, vegetative spread, maximum height and flowering phenology. Based on these traits, each species was classified into one of four (herbs) or three (graminoids) EG as delineated by Verheyen et al. (2003c) by combining Gower’s Similarity Coefficients, non-metric multidimensional scaling and clustering with Ward’s method (see Verheyen et al. 2003c for more information). For the herbs the four groups were (i) short-lived herbs, (ii) tall perennials with heavy seeds, (iii) tall perennials with light seeds and (iv) small perennials with heavy seeds; for the graminoids the three groups were (i) large, summer-flowering graminoids, (ii) small, summer-flowering vegetatively spreading graminoids, and (iii) early flowering graminoids (the characteristic life-history traits per EG are shown in Table S2 in Supporting Information and the EG per species in Table S3 in Supporting Information). Only those species for which > 50% of the life-history traits were available were used in the analyses. The 18 studies then yielded a final set of 90 species (71 herbs and 19 graminoids) and 812 species × study combinations (Tables 1 and S3 in Supporting Information). The EG were calculated across all species and individual species can thus display a deviating value for some life-history traits. The species nomenclature follows Wisskirchen & Haeupler (1998).

To conduct the interregional analyses, four groups of predictor variables were gathered for each study. These were related to habitat availability, time available for colonization, climate and soil and light characteristics. Similar to Vellend (2003), we focus on habitat loss and not habitat fragmentation sensu stricto (cf. Fahrig 1997, 2003) as information to calculate the absolute patch sizes or patch isolation distances was unavailable for most studies. Therefore, to account for habitat availability and time available for colonization, data on the total proportion of forest in the landscape (ancient plus post-agricultural; TF) and of ancient and post-agricultural forest separately (AF and PAF, respectively) as well as the mean and maximum post-agricultural forest age were collected instead. It should be kept in mind, however, that the amount of forest cover is not necessarily correlated with spatial isolation. Forest cover data were provided by the authors, compiled from the original publications, taken from Vellend (2003) or, in the case of Zacharias (1994) and Jakubowska-Gabara & Mitka (2007), obtained by digitizing maps provided in the original publication using Image J (Rasband, W.S., US National Institutes of Health, Bethesda, MD, USA, The total forest (TF) and AF cover in the landscape encompasses a broad range: 5.4–50.7% and 0.9–27%, respectively. The TF cover was significantly correlated with the AF cover (r = 0.489, P = 0.039) and the post-agricultural forest cover (r = 0.859, P < 0.001) across our studies (n = 18). The mean and maximum age of the post-agricultural forests (further referred to as the mean and maximum colonization time, respectively) was mostly available in the original publications or provided by the authors. We calculated the mean colonization time as the average of the minimum and maximum colonization time within the landscape when the mean was not explicitly mentioned. It should be noted that forest age sensu stricto is not the main focus here. Due to data availability in the different regions, an AF in one region can be younger than a post-agricultural forest in another region (Table S1 in Supporting Information). Although an old post-agricultural forest is probably more likely to be already colonized, we consider forest continuity here (i.e. whether a forest has always been forest or cleared for agriculture at some point). Hence, we take both the mean and maximum colonization time of the post-agricultural forests in a region into account. The mean and maximum colonization time of the post-agricultural forests varied between 24–135 and 54–240 years (excl. Peterken & Game 1984). The British study by Peterken & Game (1984) forms a notable exception because much older detailed maps are available in Britain (Goldberg et al. 2007; mean and maximum colonization time of the post-agricultural forests up to 201 and 370 years in Peterken & Game 1984). Exclusion of this study, however, yielded similar statistical results (data not shown).

Next, climate data were obtained from the NewLocClim 1.10 software (FAO 2005) using nearest-neighbour interpolation with ten weather stations. We gathered latitude, longitude and altitude data of the centre of each study region and used these values to deduce mean annual temperature (MAT; 1961–1990), mean annual precipitation (MAP) and potential evapotranspiration (PET). Also the effects of latitude and longitude themselves were tested for. Subsequently, an aridity index was calculated according to FAO (2005) as the ratio of MAP and PET. The MAT and aridity index varied between 6.6–15.0 °C and 0.74–2.58, respectively.

Finally, as detailed soil data or light measurements were not available for most studies so that we could rigorously test for their effects, we used the mean frequency-weighted Ellenberg indicator values (Ellenberg et al. 1992) for nutrients (mNj), reaction (mRj), moisture (mFj) and light (mLj) of the AF as rough proxies for the local soil nutrient availability, soil acidity, soil moisture and light availability, respectively. Ellenberg values are known to be very good correlates of in situ measured environmental characteristics in AF but not in post-agricultural forests (Dzwonko 2001a). Therefore we only used the Ellenberg values of the AF as proxy for these environmental variables to allow for a general interregional ranking in soil and light characteristics. For mNj, for example, this was calculated as:

image(eqn 1)

with Freqspi,j the frequency of species i in the AF in study j and Ni the Ellenberg N value for species i. The calculated mN values ranged between 2.5 (nutrient-poor soils) and 5.8 (nutrient-rich), the mR between 2.7 (acid soils) and 4.9 (more neutral), the mF between 2.6 (dry soils) and 6.0 (moister) and the mL between 3.1 (forests with low light availability in the understorey) and 4.9 (higher light availability) (Table S1 in Supporting Information). Hereafter, the mN, mR, mF and mL values are referred to as soil nutrient availability, soil acidity, soil moisture and light availability, respectively.

Data analysis

The RR calculated for each species × study combination as the risk ratio with binary data (2 × 2 tables) in standard meta-analytical procedures (Borenstein et al. 2009):

image(eqn 2)

with RRij being the RR for species i in study j and PAFij and AFij the percentages of data collection units (patches or plots; Table 1) occupied by species i in post-agricultural and AF in study j, respectively. Eqn 2 includes a correction (+ 0.01) in both the numerator and denominator to account for zero-percentages in both forest types. Zero values of RR thus correspond to equal percentages of the species in ancient and post-agricultural forest, whereas positive and negative values correspond to a lower and higher affinity to ancient than to post-agricultural forest, respectively. A species with an RR = −1, for instance, showed a percentage in AF that was approximately 2.7 times the percentage in post-agricultural forest.

Subsequently, the effect of the life-history traits of each species (EG) on RR was tested with mixed models in R 2.11.0, using the lmer function of the lme4 library (R Development Core Team 2010). According to Zuur et al. (2009), we first selected the optimal random-effects structure based on a likelihood ratio test between models with a similar fixed component (no predictor variables included), but a different random component. The optimal model included both study and species as non-nested random effects. Modelling the hierarchical nature of the data using two non-nested random effect terms in a mixed model then leads to partial pooling across the different levels (Qian et al. 2010), and hence, this takes the possible autocorrelated characteristics of (i) species from the same study region and (ii) similar species in different regions into account. Next, we compared the null model (only including the two non-nested random effects) with a model that included the EG (species level) (χ2-test statistic with likelihood ratio test; Zuur et al. 2009); these analyses were conducted for all species together and afterwards also separately for herbs and graminoids.

Next, to quantify the interregional variation in the RR for each species, we calculated the coefficient of interregional variation (CIVRR) of species i as inline image, with SDi the standard deviation and inline imagei the mean RR for species i (hence, one CIVRR value per species). A correction factor of 1 was added to the denominator to prevent CIVRR values from skyrocketing for species with an RR close to zero. We then performed a one-way anova using a general linear model (GLM) with Bonferroni post hoc test (using spss 15.0) to investigate whether there were differences in the CIVRR between the different EG; again, herbs and graminoids were analysed separately.

To explore the effects of all the environmental predictor variables on the RR, namely (i) the proportion of total, ancient and post-agricultural forest in the landscape, (ii) the mean and maximum colonization time, (iii) latitude, (iv) longitude, (v) temperature, (vi) aridity index, (vii) soil nutrient availability, (viii) soil acidity, (ix) soil moisture and (x) light availability (all study-level), we again applied mixed modelling in R following a similar approach as above. The model again included study and species as non-nested random effects and the null model (only including random effects) was compared with a model that included one of the predictor variables (on a one-by-one basis mainly to avoid multicollinearity problems) to test the significance of that particular variable (χ2-test statistic with likelihood ratio test; Zuur et al. 2009). We also estimated the percentage of variation explained by adding the predictor variables to the null model through calculations of the ratio of the difference in residuals between the null model and the final model over the residuals of the null model (Hox 2002). Finally, we tested for additive models and interactions among all significant predictor variables with the life-histories of the species (EG) by comparing (i) a model that included the EG plus that particular predictor variable with a model that included only the EG as main effect (i.e. test of the additive effects) and (ii) a full factorial model with a model that included the EG plus that particular predictor variable (i.e. test of the interaction term). Again, the χ2-test statistic with likelihood ratio test was used for this purpose.


Across Europe, many species displayed large interregional variability in their RR in post-agricultural forest. The RR across species ranged between 0.46 (Galium aparine) and −2.29 (Luzula pilosa) and amounted to −0.60 (± 0.04 SE) on average indicating that the 90 investigated understorey species showed an overall c. 1.8 (= e0.60) times higher affinity to ancient than to post-agricultural forest. Looking at individual species, the within-species variation in RR was high: for example RR varied strongly in the herbs Anemone nemorosa [0.03 to −2.72, n = 17], Lamium galeobdolon [0.15 to −2.60; n = 14] and Paris quadrifolia [0.61 to −2.40; n = 12], and the graminoids Festuca gigantea [3.05 to −1.67; n = 11], Holcus mollis [0.10 to −2.89; n = 8] and Melica uniflora [−0.07 to −3.53; n = 13] (see Table S3 in Supporting Information for the mean RR per species). In general, however, large differences were apparent among the different EG (Fig. 2a; Table 2). Species that were rather indifferent to forest continuity (RR close to zero) were members of the ‘short-lived herbs’-EG (e.g. Moehringia trinervia and Geranium robertianum) and the ‘large summer-flowering graminoids’-EG (e.g. Brachypodium sylvaticum and Festuca gigantea). The poorer colonizers (i.e. species with a negative RR), on the other hand, were small perennials with heavy seeds (e.g. Anemone nemorosa and Convallaria majalis) and early flowering graminoids (e.g. Milium effusum and Luzula pilosa). The EG (species-level predictor) explained a large proportion of the variability in the RR for herbs (percentage explained from the species and study level variance: 57.0% and 2.8%, respectively), while the RR of the graminoids was not significantly different between the EG (percentage explained from the species and study level variance: 33.8% and 1.1%, respectively; Table 2). Finally, we found significant differences in the CIVRR between the herbs, but not between the graminoids (Fig. 2b; Table S3 in Supporting Information).

Figure 2.

 The recovery rate (RR) of forest herbs (left) and graminoids (right) (a) and the interregional variation (CIVRR) (b) for the different emergent groups (EG) across Europe. Significance values of the RR were obtained from a mixed model with study and species as non-nested random effect terms; significance values of the CIVRR were obtained from a one-way anova using a GLM with Bonferroni post hoc test. Error bars depict standard errors; numbers of replicates (n) indicate species × study combinations in (a) and species in (b).

Table 2.   Effects of the predictor variables on the recovery rate (RR) of forest plants in post-agricultural forests in Europe for all species and for the herbs and graminoids separately. Results from mixed models with study and species as non-nested random effect terms. χ2-test statistic from likelihood ratio test. Significant effects in italics
Predictor(s)All species (n = 90)Herbs (n = 71)Graminoids (n = 19)
  1. †Post agricultural.

  2. ‡Additive model with the factor variable EG was tested among the significant environmental variables from the first step and compared to a model that included the EG as main effect.

  3. §Interaction with the factor variable EG was tested among the significant variables from the first step (full factorial) and compared to an additive model that included the EG and that particular environmental variable.

  4. (*)P < 0.1, *P < 0.05, ***P < 0.001.

Emergent group (EG)32.56<0.001***26.46<0.001***4.280.117
Total forest cover (TF)4.560.033*4.540.033*1.920.165
Ancient forest cover (AF)0.040.8410.140.7090.030.858
PA† forest cover (PAF)5.890.015*5.290.022*3.640.057(*)
Mean colonization time0.130.7230.330.5660.000.973
Maximum colonization time0.020.8830.050.8180.020.902
Aridity index0.510.4750.820.3640.020.892
Soil nutrient availability0.190.6670.170.6770.020.878
Soil acidity0.040.8450.080.7720.10.758
Soil moisture1.230.2671.290.2560.320.572
Light availability0.370.5420.550.4580.040.841
Additive EG + TF‡4.620.032*4.950.026*  
Interaction EG × TF§12.620.049*10.390.016*  
Additive EG + PAF‡6.430.011*6.140.013*3.550.059(*)
Interaction EG × PAF§7.460.2817.110.069(*)0.050.975

The habitat availability in a given region appeared to be an important landscape property to explain differences in the RR between regions; the species’ RR increased with the proportion of forest (ancient plus post-agricultural; study level predictor) in the landscape (percentage explained from the species and study level variance: 0.9% and 24.9% across species, respectively; Table 2 and Fig. 3a). There was also a significant interaction between the TF cover and the EG across all species and within the group of the herb species; all EG displayed a positive slope between the TF cover and the RR (Table 2; Fig. 3b,c). When analysed for each EG separately, the effect of forest cover was not significant (χ2 < 2.21, P > 0.137) except within the short-lived herbs (χ² = 5.20, P = 0.022). However, the positive trend between the average RR of each EG and the slope of the relationship between the RR and TF cover (slopes from Fig. 3b,c) suggests that EG with a higher average RR tend to profit more from increased habitat availability. The post-agricultural forest cover within a landscape also had a significant positive impact on the RR for all species and for herbs and graminoids separately (positive slope for all EG; percentage explained from the species and study level variance: 0.6% and 31.0% across all species, respectively; Table 2). In contrast, neither the AF cover nor the climatic variables, latitude and longitude, soil nutrient and light availability, soil acidity, soil moisture, and mean and maximum colonization time had a significant effect on the RR (Table 2).

Figure 3.

 The effect of total forest (TF) habitat availability (forest cover within the landscape) on the recovery rate (RR) of understorey species in post-agricultural forest (a) across all species and within the different emergent groups (EG) of the (b) herbs and (c) graminoids. The EG for the herbs were: 1. short-lived herbs, 2. tall perennials with heavy seeds, 3. tall perennials with light seeds and 4. small perennials with heavy seeds. For the graminoids the three groups were 1. large, summer-flowering graminoids, 2. small, summer-flowering vegetatively spreading graminoids, and 3. early flowering graminoids (ranked according to the mean RR). Slope and significance values from mixed models with study and species as non-nested random effect terms which takes the possible autocorrelation of species and studies into account. *P < 0.05.


By applying a quantitative meta-analytical approach across Europe, we provide evidence for large among-region variability in the recovery rate (RR) of understorey species into post-agricultural forest. The slope of the effect of the forest habitat availability within the landscape on the RR varied for the (life-history trait-based) emergent groups (EG). The short-lived herbs showed the highest interregional variability in their RR (CIVRR), and this variability was significantly related to the total amount of forest cover within the landscape. The small perennial herbs with heavy seeds and early flowering graminoids were generally confined to ancient forest (AF) patches throughout Europe. The effect of increased habitat availability appeared to be favourable across all species, but when this effect was analysed per EG, it seems only beneficial for the short-lived herbs. This finding suggests that even a high proportion of forest within the landscape (in the present study up to 51%; Grashof-Bokdam & Geertsema 1998) has no positive effect on the colonization rates into post-agricultural forests by the species with the lowest RR, confining these mostly to AF patches. This also implies that these species strongly depend on temporal habitat continuity, occurring mainly as relict populations in AF or hedgerows, whereas the short-lived herbs can show a metapopulation behaviour between the different forest patches within a largely agricultural and urban matrix (Verheyen et al. 2004). The latter species may colonize new forest patches soon after canopy closure.

Various studies have shown the beneficial effect of forest cover on species migration into post-agricultural forest on a local scale (Honnay et al. 2002; Graae et al. 2004; Verheyen et al. 2006) or for species richness (Vellend 2003). Here, we quantified this effect for the first time on a continental scale by combining 18 independent studies across Europe and also show contrasting effects depending on the life-history traits of the understorey species. Contrary to Vellend (2003), who highlighted the importance of AF cover in the landscape for species richness in post-agricultural forest in Europe and eastern North America, not AF cover per se but the availability of total and post-agricultural forest habitat was found to have a significant positive effect on the RR. Since the total and post-agricultural forest habitat availabilities were strongly correlated in the present study, densely forested landscapes generally have a high proportion of post-agricultural forests. This could be caused by historical afforestation practices in lowland Europe such as the tendency to afforest (mostly little-productive) regions that had lost most of their (ancient) forests until the 1800s. In (often more productive) regions where few AF remained, on the other hand, hardly any afforestation effort was made. Consequently, the low total forest (TF) cover there is associated with low post-agricultural forest cover (e.g. noble estates, state-owned land; Rackham 2003). Most species displayed a high affinity to AF in forest-poor regions whereas only the short-lived herb species appeared to be able to benefit from increased possibilities for migration in densely forested landscapes. Finally, in spite of a different methodology, the present study clearly confirms the findings of Verheyen et al. (2003c) that plant traits associated with a low dispersal capacity result in a high affinity to AF. Our results corroborate that plant height (e.g. tall herbs have greater chances for epizoochorous dispersal), possibilities for long-distance dispersal (e.g. light seeds) and abilities to reproduce easily (e.g. vegetative spread) together account for the interspecific differences (Graae 2002; Verheyen et al. 2003c).

Surprisingly, climate, soil and light characteristics, and colonization time were of no influence to the recovery. Although an effect of, for instance, climate, soil nutrient availability and colonization time has been shown for the performance and abundance of individual species (Kolb & Diekmann 2004; Baeten et al. 2010; De Frenne et al. 2010a; b), this effect was not apparent in the present study. This may be due to multiple factors including long-term adaptations to the local environment (genetic control), a changing species pool across soil types and a complex mosaic of different soil types and age classes in post-agricultural forests within a landscape. First, long-term adaptations to the local environment (both to climate and to soil characteristics) result in a bias in observational studies along geographic transects. For instance, some ecotypes of the same species may be adapted to growing in warmer climates whereas others perform equally well in colder conditions (Joshi et al. 2001) leading to a ‘homogenization’ of colonization rates across climates. Second, the species pool on inherently richer soils (in the present study, high Ellenberg indicator value for nutrients) may differ drastically from the species pool on poorer soils (Zobel & Pärtel 2008), again making an effect of soil nutrient availability on the floristic recovery difficult to extract. The heterogeneity in soil conditions within large study areas (e.g. in Honnay, Degroote & Hermy 1998) reduces the chance of finding an interregional effect. Also, mean Ellenberg indicator values for AF alone may not represent the whole range of regional environmental variation in post-agricultural forests. Besides, also the position of AF within a landscape may a priori be determined by the local soil conditions due to historical deforestation and farming practices (Flinn, Vellend & Marks 2005; Cousins 2009). Finally, there is also large variation in mean and maximum colonization time available since agricultural abandonment between the post-agricultural patches within a landscape, thereby reducing the effect of colonization time (see also Methods or Vellend 2003).

Still, our results clearly suggest that, even within the maximum time horizon studied here (up to 370 years), the species with life-history traits that limit dispersal are most likely to be missing from the floristic community of post-agricultural forests across Europe. In summary, we clearly showed that the degree to which a species manages to colonize post-agricultural forests depends mainly on (i) life-history traits related to dispersal and colonization and (ii) the forest habitat availability in the landscape. Short-lived herbs showed the highest interregional variability. Of this variability, a significant part could be explained by the total and post-agricultural forest habitat availability in the landscape. Hence, short-lived herbs are most likely to be the first colonizers into a newly established forest patch and also show the greatest variability in RR across Europe. Forest-rich landscapes thereby provide the best setting for high permeability from remnant AF or hedgerows into post-agricultural patches for these species. Furthermore, we stress the need to avoid AF clearance to preserve the typical woodland flora with an intrinsically low RR.


We are grateful to the Research Foundation – Flanders (FWO) for funding the Scientific Research Network ‘FLEUR’ ( that made this study possible. We also greatly acknowledge Thilo Heinken for help with the German data sets, Olivier Chabrerie with the French data, Dietmar Zacharias for additional information, and Hans Van Calster, the Editors and two anonymous referees for commenting on earlier drafts of this manuscript. This paper was written while P.D.F. and A.D.S. held a PhD and post-doctoral fellowship from the FWO and L.B. held a PhD fellowship from the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen), respectively.