Seed ecology of Mediterranean hind dune wildflowers

Ecological Engineering 91 (2016) 282–293

Contents lists available at ScienceDirect

Ecological Engineering journal homepage: www.elsevier.com/locate/ecoleng

Seed ecology of Mediterranean hind dune wildflowers Stefano Benvenuti Department of Agriculture, Food and Environment, Pisa University, Via del Borghetto 80, Pisa, Italy

a r t i c l e

i n f o

Article history: Received 20 October 2015 Received in revised form 22 January 2016 Accepted 26 January 2016 Keywords: Dune ecosystem Sand burial Seed bank Seedling emergence Germination

a b s t r a c t There is an increasing need for the landscape conservation of the threatened biodiversity of coastal ecosystems. This work studies the seed ecology (germination, dormancy, self-burial, seedlings emergence and seed bank disposition) of various Mediterranean hind dune wildflowers in a perspective of landscape protection and restoration. The deep dormancy of most species (overall Fabaceae, Brassicaceae and Calystegia soldanella) was eliminated or reduced after some seed treatments selected to simulate some natural events (washing, cold stratification, scarification, etc). “In situ” seed bank analysis was carried out by collecting sand samples in different dune ecosystems of Tuscany and Sardinia. Indeed the seed vertical distribution plays a crucial role in the germination dynamics on buried seeds. The vertical distribution of the seed bank of 20 selected species was confined to the upper 0–12 cm layer of sand dune. Their burial depth was found to be inversely related to the 1000 seed weight of the different species ranging by 0.01 g of Centaurium maritimum to 47.58 of C. soldanella. Roughly 80% of the seed bank is accumulated in the shallowest sand layer (0–3 cm) and the only very little seeds of C. maritimum and Silene colorata were capable to reach the deeper layer of 9–12 cm. Seedling emergence from increasing burial depth has been studied in pots “ex situ”. In spite of the respective seed dormancy-breaking treatments, their germination was progressively inhibited by burial depth increasing in the sand matrix. Calculation of the depth able to halve emergence established an inverse relationship between seed weight and depthmediated inhibition. In addition, the ex situ capacity for self-burial (mediated by winter rains), ranging between 1 and 3.5 cm, showed a similar, but inverse, relationship between seed weight and self-burial performance. Seed bank of herbaceous perennial species appears to be a good indicator of an ecosystem’s health and the perspective of environmental restoration, by using native wildflowers, should be carried out by sowing treated seeds, with a light burial, behind the natural or artificial foredunes. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Coastal dune systems are characterized by strong environmental gradients which determine the coexistence of different plant communities in a relatively small area (Wilson and Gitay, 1995). Unfortunately, many species in this highly biodiverse environment (Grootjans et al., 2004), are threatened by both the reduction of these particular habitats and their anthropogenic alteration (Brown and McLachlan, 2002; Defeo et al., 2009). Bare sand and semi-fixed dunes are ideal conditions for successional young slack habitats that support rare species of coastal dune flora (Lucas et al., 2002). Open patches of these microhabitats generally have the greatest richness, diversity, and productivity of species (Yu et al., 2008). In such coastal environments setting

E-mail address: [email protected] http://dx.doi.org/10.1016/j.ecoleng.2016.01.087 0925-8574/© 2016 Elsevier B.V. All rights reserved.

up successful biodiversity conservation programs entails conserving “biologically rich hotspots” and also including species-poor habitats where there are endangered species or which are unique in some way (Acosta et al., 2009). In addition to this habitat conservation policy, several restoration programs of dune ecosystems have been planned (Grootjans et al., 2002; Lithgow et al., 2013), sometimes carried out as natural engineering projects (Hanley et al., 2014). Non- or minimal intervention as well as totally management systems such as intensive single-species management and habitat re-creation is required when attempting to reach certain objectives in terms of conservation (Rhind and Jones, 2009). In both cases (conservation or restoration) the study of the biology and ecology of dune species plays a crucial role. Primary importance has been given to studies on the species typically facilitate the consolidation of dunes as a consequence of their indisputable importance as stabilizers (Hanley et al., 2014). In contrast much less importance has been given to the hind dune

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

283

Table 1 Botanical characteristics and germplasm collection site of the dune wildflowers selected for the experiments. Species

Botanic family

Life form

Locality of seed collection

Geographical coordinates

Anchusa crispa Viv. Anthemis maritima L. Anthemis mixta L. Armeria pungens (Link) Hoff. et Lk Cakile maritima Scop. Calystegia soldanella (L.) Roem. Et Schult Centaurea aplolepa Moretti Centaurium maritimum L. Dorycnium hirsutum L. Echium arenarium Guss. Glaucium flavum Crantz Matthiola tricuspidata L. Medicago marina L. Ononis variegata L. Pancratium maritimum L. Scabiosa maritima L. Scabiosa rutifolia Vahl Silene colorata Poiret Silene conica L. Solidago litoralis Savi

Boraginaceae Asteraceae Asteraceae Plumbaginaceae Brassicaceae Convolvulaceae Asteraceae Gentianaceae Fabaceae Boraginaceae Papaveraceae Brassicaceae Fabaceae Fabaceae Amaryllidaceae Dipsacaceae Dipsacaceae Caryophyllaceae Caryophyllaceae Asteraceae

Hemicryptophyte Hemicryptophyte Therophyte Chamaephyte Therophyte Geophyte Hemicryptophyte Therophyte Chamaephyte Hemicryptophyte Hemicryptophyte Therophyte Chamaephyte Therophyte Geophyte Hemicryptophyte Therophyte Therophyte Therophyte Hemicryptophyte

Vignola (OT) Torre Mozza (LI) Tirrenia (PI) Vignola (OT) Castiglione della Pescaia (LI) Viareggio (LU) S. Vincenzo (LI) Vada (LI) Marina di Vecchiano (PI) Capo Mannu (OR) Vignola (OT) Marina di Cecina (LI) Marina di Grosseto (GR) Marina di Vecchiano (PI) Follonica (GR) Marina di Grosseto (GR) Capo Mannu (OR) Vada (LI) Castiglione della Pescaia (GR) Marina di Vecchiano (PI)

41◦ 13 N, 9◦ 11 E 42◦ 09 N, 10◦ 67 E 43◦ 56 N, 10◦ 24 E 41◦ 13 N, 9◦ 11 E 42◦ 76 N, 10◦ 87 E 43◦ 86 N, 10◦ 25 E 43◦ 11 N, 10◦ 54 E 43◦ 35 N, 10◦ 45 E 43◦ 80 N, 10◦ 26 E 40◦ 03 N, 8◦ 39 E 41◦ 13 N, 9◦ 11 E 43◦ 27 N, 10◦ 50 E 42◦ 72 N, 10◦ 96 E 43◦ 80 N, 10◦ 26 E 42◦ 09 N, 10◦ 76 E 42◦ 72 N, 10◦ 96 E 40◦ 03 N, 8◦ 39 E 43◦ 35 N, 10◦ 45 E 42◦ 76 N, 10◦ 87 E 43◦ 80 N, 10◦ 26 E

environment (immediately back to the fore dunes) in spite of its high, and often rare biodiversity. This biodiversity strongly depends on the vegetation of the foredune, since they serve as ecosystem engineers in this coastal environment, with an influence on multiple levels of biological organization (Cushman et al., 2010). The wind is not able to carry large amounts of sand particles since the sand is protected by the shrubs, which enables the particular flora evolved to tolerate abiotic stresses (wind, drought, salinity) without the need for robust rooting which instead is required in the foredune in front of the sea (García-Mora et al., 1999). This partially protected environment is the ecological niche of herbaceous species including many fixed dune wildflowers (O’Rourke et al., 2014). Many of these species have mutualism with insect pollinators (Junker and Blüthgen, 2010) thanks to the availability of pollen and nectar (Goulson, 1999) and are sometimes indispensable for the survival of rare butterfly and/or solitary bees (Pyle et al., 1981; Petanidou et al., 1998). Consequently, flowerpollinator mutualism increases the biodiversity level of these areas even more with the flora-micro-fauna threat of disappearance. In addition the landscaping impact (Provoost et al., 2011) of these beautiful hind dune wildflowers is important due to their typical chromatically eye-catching flowers. Since these herbaceous species entrust the survival dynamics to their seeds, it follows that knowledge of their seed ecology plays a crucial role for both conservation and/or restoration. Indeed the survival dynamics of these species, in such erratic environments, strongly depends on their ability to retain, in the sand matrix, viable seeds able to re-colonize them in space and time. In this background plays a crucial role both seed dormancy and their burial capacity. This last, allow even to not dormant seeds, to acquire a “forced dormancy” (Baskin and Baskin, 2004) mediated by hypoxia (Benvenuti and Macchia, 1995). This burial-dependent germination inhibition often implies seed longevity (Thompson et al., 2003) and consequently makes less vulnerable the survival dynamics of the dune species. In this background the self-burial capacity and velocity, of each species, support a seed bank accumulation in the shallowest sand layers making the dune ecosystem capable to a dynamic re-colonization after natural and/or anthropic ecological disturbance. Despite the complexity of experimental studies dedicated to the dune environment there are few studies focused to wildflowers. Indeed the important role that many wildflowers dune take in terms of biodiversity and landscape suggested to explore even their seed ecology in both perspectives: conservation and restoration. The aim of this work was to investigate: (i) the amount of the wildflower’s seed bank and its vertical arrangement in situ in

various hind dunes of Mediterranean ecosystems; and (ii) the seed ecology in terms of dormancy, germination, self-burial capacity, and emergence from increasing sand depths. 2. Materials and methods 2.1. Plant material and dune environments Personal observations since 2000 on coastal flora have identified several Mediterranean dune ecosystems (in Tuscany and Sardinia) characterized by a high floristic biodiversity. In these environments plant community surveys were conducted in order to identify species that can be defined as wildflowers. The selection criterion was to choose herbaceous wild species, common in the above cited dune environments roughly defined as “Crucianellion maritimae” (Del Vecchio et al., 2012), is characterized by chromatically eye-catching flowers that have evolved towards entomophilous pollination (Benelli et al., 2014). Table 1 shows the 20 selected species (nomenclature follows Pignatti, 1982) in the various areas of hind dunes. Between 2012 and 2013 (in the summer and autumn) seeds from each species were collected in the respective dune environments as shown in Fig. 1. The seeds were extracted from the senescent fruits in the

Fig. 1. Landscape of the dune environment (left hindune and right foredune) of a Sardinian site (near Vignola, OT) where sand samples for seed bank analysis were collected.

284

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

laboratory, cleaned, dried in dry room, and kept in glass containers at 20 ◦ C. 2.2. Germination test and seed treatments Laboratory-stored seeds (6 months after collection), were disinfected with sodium hypochlorite (0.5%, v/v) for 10 min before use, air dried, and imbibed on a single sheet of moistened filter paper, and placed in 12 cm Petri dishes (100 seeds each, 3–5 replicates). Seeds were incubated in climatized cabinets at alternate temperatures of 15/20 ◦ C (dark–light respectively, photoperiod 12 h/12 h). White light was generated by neon fluorescent lamps regulated at the intensity of roughly 50 ␮mol m−2 s−1 . Subsequently, as a consequence of the frequent seed dormancy of these wild species, four different seed treatments were carried out: cold stratification, washing, coat elimination (of indehiscent siliques) and scarification. For the cold stratification treatment, seeds were placed in Petri dishes (three replicates) filled with moist sand and maintained at low temperatures (4 ◦ C) in darkness (wrapped in a double of aluminum sheet) for 3 months. Seeds were washed in running water for 3 days and then dried in ventilated air. In the case of indehiscent silique (as in the case of the Brassicaceae studied) the external fruit coat was carefully removed using a scalpel in order to obtain naked seeds. Scarification was performed by rubbing the seeds mechanically with sand paper for roughly 1–2 min. Seeds were then observed with an optical microscope to ensure that the seed coats had been cracked by the scarification. Each species was subjected to all seed treatments although only those that showed the best performance in terms dormancy-breaking are reported in this paper. After each treatment, incubation was carried out in the same climatic conditions as the above-cited untreated seeds. 2.3. Seed weight Seed weight (or fruit as in the case of the achenes of Asteraceae and Scabiosa columbaria) was determined by weighing 1000 seeds chosen randomly according to ISTA rules for seed testing (ISTA, 1999). 2.4. Seedbank evaluation Seedbank investigations were carried out in the dunes where there was the greatest abundance of each of the 20 wildflowers, corresponding to the respective site of seed collection shown in Table 1. In these plots of about 10 m2 (three plots as replications for species) complete randomized sand samples were collected. The dune samplings were performed in autumn–winter 2013 using a custom-designed metal probe already used for seed bank evaluation (Benvenuti, 2007), which is able to obtain small intact cylindrical cores (4 cm in diameter and 15 cm long). For each of the 20 species, 10 soil cores per plot (a total of 30 cores per species) were randomly collected. After they had been extracted and arranged horizontally the samples were then further subdivided into the following layers: 0–3, 3–6, 6–9, 9–12 and 12–15 cm. This division was carried out using a metal blade, taking care not to collapse the cylinder because of the sandy matrix. To facilitate the collection of “sand carrots”, the samples were collected after natural or artificial (common gardening sprinkler) rain whose relative moisture helped the extractability of intact samples. They were then air-dried quickly to prevent undesired germinations. Seed crush test (Sawma and Mohler, 2002) showed that

non-viable seeds were a rare and negligible seedbank amount (less than 1%). Seeds were extracted from sub-samples using a previously adopted washing system (Benvenuti and Macchia, 2006). The extracted material (seeds, coarse sand, plant residues, etc.) was separated manually after air drying using a back-lighted magnifying glass. Seeds were then identified and counted with the aid of an optical microscope (45×). The data were expressed as absolute density (number of seeds per square meter) and relative density (percentage of seeds in a given layer compared to the total). 2.5. Emergence test During winter–spring 2014 (February–May), seeds of each of the 20 species were sown in increasing depths in large pots (50 cm × 100 cm wide × 60 cm in height) filled with sterilized sand quartz (grains 0.08–0.5 mm). Seeds of each species were subjected to the relative treatment (cited previously in the germination test) in order to highlight the best dormancy-breaking performance. These trials were carried out in Pisa, at the experimental site of the Department of Agriculture, Food and Agro-environment (43◦ 70 N 10◦ 43 E) of Pisa University. Seeding depths were as follows: 0 (sand surface), −2, −4, −6, −8, −10 and −12 cm. In each large pot, five groups of 50 seeds were sown for each depth. Irrigation was unnecessary due to the natural rainfall. The thermal trend that occurred during this period ranged from an average of about 5/10 ◦ C in February to 15/25 ◦ C in May. During the experimental period, the emergence dynamics (at cotyledon appearance) were detected weekly. 2.6. Self-burial performance Seeds of all of the 20 species were carefully placed on the surface of the previously described large pots filled with quartz sand. The winter period was chosen as the most suitable time for the experiments for two reasons: (i) it is typically a rainy month (which necessary to test the burial performance); and (ii) low temperatures during this period prevent seed germination, which would have hindered observations of their natural burial. At the end of February, seeds were retrieved from the sand layers. The samplings were performed using the probe described above in order to obtain small intact cylindrical cores. After the two-part opening of the probe, the sand cores were cut transversely using a metallic blade. Thus the desired small 3 cm-long sand cylinders were obtained. Seeds were extracted with the same method as previously described in the seedbank study. The extent of seed burial was obtained by calculating the weighted means according to the following formula: D=

 (n × d ) i i nt

where D is the weighted mean; ni the number of seeds at a given depth; di the mean depth of the soil layer considered, and nt the total number of seeds. 2.7. Calculation of depth of 50% emergence inhibition Polynomial regressions were calculated that showed the best fit of the biological response of weed seed emergence inhibition at increasing burial depths. These equations of the depth-mediated inhibition were used to identify the depths capable of reducing emergence to 50% (with respect to surface germination). They gave emergence-halving depths using a modified “x-intercept” method (Wiese and Binning, 1987). This “x-intercept” method overlaps the

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

various polynomial equations with the line of 50% depth inhibition (y = 50), and the intersection indicates, for each species, the burial depth capable of halving their seedling emergence. 2.8. Statistical analysis The experimental design of the ex situ experiments (germination and emergence test) was a randomized complete block with 3–5 replications. Similarly, in situ seedbank experiments, to compare the number of seeds among the different burial depths, had a completely randomized sampling-scheme. After a homogeneity test of variance, arc-sin transformation was necessary for all percentage data. Data were subjected to ANOVA using the Student–Newman–Keuls test (p < 0.05 and p < 0.01) for means separation (LSD). Both (i) specific burial capacity to halve germination and (ii) self-burial performance of each species were plotted with the corresponding 1000 seed weight and fitted with a Boltzmann polynomial regression. For each statistical analysis, commercial software (CoStat) was used.

285

all the deep-dormant species belonging to the botanical family of Fabaceae (D. hirsutum, M. marina and O. variegata) showed a drastic germination increase (significant, p < 0.01) after seed scarification. A similar mechanical intervention, in this case due to the removal of the external portion of the indehiscent siliques, induced germination in species belonging to the botanical family of Brassicaceae such as C. maritima and M. tricuspidata (85.2% and 78.4%). Eight of the tested species showed a significant (p < 0.05) dormancy reduction after seed washing, due to the removal of the seed inhibitors (Baskin and Baskin, 2004). This occurred in all tested species belonging to the botanical families of Asteraceae, Dipsacaceae and also in A. pungens and P. maritimum. Finally, the remaining species, C. maritimum and G. flavum, as well as those belonging to the botanical families of Boraginaceae (A. crispa and E. arenarium) and Caryophyllaceae (S. colorata and S. conica), showed a germination increase (statistically significant, p < 0.05 or 0.01), after cold stratification.

3. Results

3.2. Seed bank evaluation

3.1. Seed dormancy and germination

Fig. 2A and B shows the seed bank burial architecture of the 20 selected species. As can be seen, the shallowest profile (0–3 cm) contains the most seeds. However this accumulation strongly depends on the species considered since this seed concentration, in the shallowest sand layer, ranges from more than 90% of P. maritimum to about 60% of the C. maritimum. In the layer (3–6 cm) immediately below, this seed quantity tends to decrease very markedly in all species. Their density ranged between approximately 10% and 20% as a function of the species. In some cases (C. maritima, A. pungens, C. soldanella and P. maritimum) the seed burial does not exceed a depth of 6 cm, while in most of the tested species, a small quantity of seeds (about 2–10%), reaches the next sand layer (6–9 cm). Only in cases of species characterized by very small seeds (see Table 3), was a poor seed (roughly 1–3% in S. colorata and C. maritimum respectively) quantity able to reach the underlying sand profile (9–12 cm). In terms of absolute density, values ranged from the lowest quantity of A. pungens, P. maritimum and A. crispa (115, 125 and 145 seed m−2 , respectively) to the highest level of C. maritimum, G. flavum and S. colorata (365, 355 and 325 seed m−2 , respectively).

In the absence of seed treatments most species showed a marked dormancy (Table 2). This poor germination was more evident in species belonging to the botanical family Fabaceae (Dorycnium hirsutum 15.1%, Medicago marina 21.8%, Ononis variegata 18.4%), in Brassicaceae (Cakile maritima 15.7% and Matthiola tricuspidata 16.4%) and also in Centaurium maritimum (22.7%) and Calystegia soldanella (5.4%). In addition, although to a lesser extent, other species showed a rather poor germination as in the case of the two species Boraginaceae (Anchusa crispa, 32.4% and Echium arenarium 33.4%), in Armeria pungens (44.2%) and Glaucium flavum (38.4%). A lesser degree of dormancy was instead shown by other species able to germinate to above 50% as in the case of Anthemis maritima (52.8%), Anthemis mixta (58.2%), Centaurea aplolepa (54.4%), Solidago litoralis (55.4%), Pancratium maritimum (56.4%), Scabiosa maritima (65.4%), Scabiosa rutifolia (57.4%), Silene colorata (68.4%) and Silene conica (77.2%). However, some seed treatments were able to significantly (p < 0.05 or p < 0.01) remove most of this dormancy. For example

Table 2 Germination performance before and after seed treatments (cold stratification, washing, coats elimination and scarification). Asterisks indicate the significance levels of L.S.D. test (**p < 0.01; *p < 0.05). Species

Germination before treatment

Anchusa crispa Anthemis maritima Anthemis mixta Armeria pungens Cakile maritima Calystegia soldanella Centaurea aplolepa Centaurium maritimum Dorycnium hirsutum Echium arenarium Glaucium flavum Matthiola tricuspidata Medicago marina Ononis variegata Pancratium maritimum Scabiosa maritima Scabiosa rutifolia Silene colorata Silene conica Solidago litoralis

32.4 52.8 58.2 44.2 15.7 5.4 53.4 22.7 15.1 33.4 38.4 16.4 21.8 18.4 56.4 65.4 57.4 68.4 77.2 55.4

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

2.7 5.1 5.4 4.2 1.2 0.5 1.2 2.2 1.2 2.9 3.2 1.2 2.2 1.5 5.2 6.2 5.2 6.3 6.2 5.2

Germination after treatment 65.4 68.5 71.1 58.2 85.2 75.7 68.4 42.6 86.1 46.2 49.7 78.4 76.6 72.4 69.5 78.6 72.2 77.8 89.5 68.4

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

4.2** 5.1* 6.2* 5.2* 7.2** 6.5** 4.2* 3.2** 4.2** 3.9* 3.8* 6.2** 6.2** 6.5** 5.8* 6.8* 5.8* 6.8* 5.0* 4.3*

Seed treatment Cold stratification Washing Washing Washing Silique removal Scarification Washing Cold stratification Scarification Cold stratification Cold stratification Silique removal Scarification Scarification Washing Washing Washing Cold stratification Cold stratification Washing

286

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

3.3. Seedling emergence

in the surface layers and then tended to decrease as a consequence of increasing burial depths. However, this trend varied depending on the species. In fact, while in M. tricuspidata and C. maritimum, a 6 cm sowing depth was able to completely inhibit germination

Fig. 3A and B shows the emergence performances of the various species in increasing sand depths. Seedling emergence was highest

a

0-3

Burial depth (cm)

Burial depth (cm)

A

b

3-6

c

6-9

145 seeds m

9-12

-2

Anchusa crispa

12-15

0

20

40

60

80

a

0-3

b

3-6

c

6-9

275 seeds m -2

9-12

Anthemis maritima 12-15

100

0

20

b c

6-9

225 seeds m -2

9-12

Anthemis mixta 12-15 0

20

40

60

80

Burial depth (cm)

Burial depth (cm)

a

0-3 3-6

b 115 seeds m-2

6-9 9-12

Armeria pungens 12-15 0

185 seeds m -2

9-12

Cakile martima 12-15 40

60

80

Burial depth (cm)

Burial depth (cm)

b

20

20

-2

160 seeds m

9-12

Centaurea aplolepa 12-15 20

40

60

135 seeds m-2

6-9 9-12

Calystegia soldanella 12-15 20

80

-2

195 seeds m

9-12

Dorycnium hirsutum 0

20

40

60

Seedbank (%)

100

b c

6-9

d

9-12

365 seeds m-2 Centaurium maritimum

12-15 0

80

Burial depth (cm)

Burial depth (cm)

b

12-15

80

a

3-6

100

a c

60

20

40

60

80

100

Seedbank (%)

0-3

6-9

40

0-3

Seedbank (%)

3-6

100

a

0

Burial depth (cm)

Burial depth (cm)

b

0

80

b

3-6

100

a c

60

Seedbank (%)

0-3

6-9

40

0-3

Seedbank (%)

3-6

100

Seedbank (%)

a

0

80

a

3-6

100

0-3

6-9

60

0-3

Seedbank (%)

3-6

40

Seedbank (%)

Seedbank (%)

100

a

0-3

b

3-6

c

6-9

175 seeds m -2

9-12

Echium arenarium 12-15 0

20

40

60

80

100

Seedbank (%)

Fig. 2. (A) Seed bank disposition (absolute seed density shown as a number) in the dune sand layers of the tested wildflower species. Analyses were carried out in the respective dune environments shown in Table 1. Vertical bars indicate standard errors of the means. (B) Seed bank disposition (absolute seed density shown as a number) in the dune sand layers of the tested wildflower species. Analyses were carried out in the respective dune environments shown in Table 1. Vertical bars indicate standard errors of the means.

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

287

a

0-3

Burial depth (cm)

Burial depth (cm)

B

b

3-6

c

6-9

355 seeds m

9-12

-2

Glaucium flavum

12-15 0

20

40

60

80

a

0-3

b

3-6

165 seeds m -2

c

6-9 9-12

Matthiola tricuspidata 12-15 0

100

20

Burial depth (cm)

Burial depth (cm)

a

0-3

b

3-6

c

6-9

270 seeds m

-2

9-12

Medicago marina 12-15 0

20

40

60

80

225 seeds m-2

c

6-9 9-12

Ononis variegata 12-15 20

Burial depth (cm)

Burial depth (cm)

125 seeds m-2

9-12

Pancratium maritimum 12-15 40

60

80

b

3-6

255 seeds m-2

c

6-9 9-12

Scabiosa maritima 12-15 0

Burial depth (cm)

Burial depth (cm)

b

260 seeds m -2

9-12

Scabiosa rutifolia 12-15 0

20

40

60

20

80

315 seeds m

9-12

Silene conica 12-15 20

40

60

100

b 325 seeds m-2

c

6-9

d

9-12

Silene colorata

12-15 0

80

Burial depth (cm)

Burial depth (cm)

b

0

80

a

3-6

100

a c

60

20

40

60

80

100

Seedbank (%)

0-3

6-9

40

0-3

Seedbank (%)

3-6

100

a

100

a c

80

Seedbank (%)

0-3

6-9

60

0-3

Seedbank (%)

3-6

40

Seedbank (%)

b

20

100

b

0

a

0

80

a

3-6

100

0-3

6-9

60

0-3

Seedbank (%)

3-6

40

Seedbank (%)

Seedbank (%)

-2

a

0-3

b

3-6

235 seeds m -2

c

6-9 9-12

Solidago litoralis 12-15

100

0

Seedbank (%)

20

40

60

80

100

Seedbank (%) Fig. 2. (Continued ).

and emergence, P. maritimum and C. soldanella still showed germination, albeit poor (5% and 3%, respectively), even from a depth of 10 cm. The germination-inhibition (deep-mediated) data fitted with polynomial regressions had a high statistical significance (p < 0.01). Fig. 3A and B highlights that the germination-halving ranged from 3 cm of C. maritimum to 7.8 cm of C. soldanella. All other dune

wildflowers were within this range. Differences of a few cm of burial are able to strongly influence the germination response of each species. These burial values (the depths able to halve the emergence) were plotted with the corresponding 1000-seed weights and fitted with a Boltzmann polynomial regression (Fig. 4). This significant regression (p < 0.05) shows that the smaller sized seeds (1000 seeds

288

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

Table 3 Seed weight of the dune wildflowers tested in the experiments. Means are followed by ±standard errors. Species

1000 seed weight (g)

Anchusa crispa Anthemis maritima Anthemis mixta Armeria pungens Cakile maritima Calystegia soldanella Centaurea aplolepa Centaurium maritimum Dorycnium hirsutum Echium arenarium Glaucium flavum Matthiola tricuspidata Medicago marina Ononis variegata Pancratium maritimum Scabiosa maritima Scabiosa rutifolia Silene colorata Silene conica Solidago litoralis

2.45 1.22 1.34 1.45 15.56 47.58 1.55 0.01 4.15 2.46 1.12 0.23 4.22 4.37 1.68 1.95 2.35 0.27 0.34 0.42

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.25 0.12 0.16 0.25 0.16 5.1 0.15 0.01 0.42 0.27 0.12 0.02 0.51 3.38 1.35 0.21 0.19 0.03 0.04 0.05

less than 0.5 g) are more depth-inhibited (3–5 cm of burial depth were enough to halve seedling emergence) and are thus more suitable to form a permanent seedbank. Consequently the seed fate, in terms of germination or quiescence (also defined as deep-mediated dormancy), makes the different self-burying potential of the various species of crucial importance. 3.4. Natural burial performance A further question that needs to be investigated, is also whether, and to what extent, seed size may be involved in the self-burial dynamics subjected to the kinetic energy of the rains. Fig. 5 shows the rains that occurred during the experimental period (January–February 2014) and the degree of burial of the seeds of each of the 20 species tested. The selected period was very suitable for this investigation since there was consistent rain (429 mm). This rain led to a very varying degree of seed burial in the different species. The values ranged by about 1 cm, in the case of P. maritimum and C. maritimum, to more than 3 cm in the case of C. maritimum, G. flavum, A. mixta and S. colorata. All the other species were within this range. Fig. 6 shows a close (statistically significant, p < 0.05) relationship between seed weight and their self-burial capacity. This Boltzmann polynomial regression highlights that smaller seeds penetrate the sandy matrix more. 4. Discussion 4.1. Seed dormancy and germination Each of the 20 species showed different degrees of seed dormancy (Table 2) both in terms of quantity (low or high) and quality (physical or physiological). This dormancy appears to play a crucial role in the recolonization dynamics of this fragile ecological environment, since it provides an important survival strategy in unpredictable conditions (Allen and Meyer, 1998). Non-synchronized germination represents enables survival by xerophytes in arid dunes by reducing seedling mortality after shortterm rainfall events (Zeng et al., 2010). However, such dormancy appears to be removable by ecological conditions that typically

occur in sand dunes. For example, the physical seed dormancy that was shown by Fabaceae (D. hirsutum, O. variegata and M. marina) and C. soldanella was removed by seed scarification due to seed coat water impermeability also called “hard seeds” (Rolston, 1978). Dormancy can thus be removed in natural conditions by the windy events that typically occur in dune environments (Baas and Sherman, 2006). Indeed the inevitable friction between the wind-moved sand and seed coats tends to gradually eliminate their impermeability with the consequent dormancy removal as already observed in other dune plants of the Fabaceae (Liu et al., 2011) and the Convolvulaceae botanic family (Jayasuriya et al., 2008). Seed dormancy due to siliques indehiscence, shown by C. maritima and M. tricuspidata), as well as other Brassicaceae (Eslami et al., 2010), was easily (mechanically) removed in laboratory from the seed extraction from fruits. In natural environment this dormancy-breaking appears to be elicited by burial in moist sand. Indeed, in such micro-environments, siliques undergo a slow microbial degradation similar to what happens in fruit tissues of other sand species (Yang et al., 2012). This burial in a wet environment appears useful to remove dormancy also in many other species (via stratification) as occurs in the tested Boraginaceae (A. crispa and E. arenarium), Caryophyllaceae (S. colorata and S. conica), and also in C. maritimum and G. flavum. Such a stratification treatment, which is frequent in sand dunes (during cool and rain periods), triggers germination in many species following the physiological mechanisms often related to the synthesis of important plant hormones such as gibberellins (Baskin and Baskin, 2004) and/or due to the inhibitor degradation (Hilhorst and Karssen, 1992). Finally, the seed-washing also induced dormancy-breaking in the remaining eight species (Asteraceae, Dipsacaceae and A. pungens). This treatment also appears to mimic the absolutely typical ecological conditions that occur in the sand matrix crossed by the rainwater in their vertical movement. On the other hand, seed washing eliminates the seed-external inhibitor mechanism already observed in plants of arid environments (Nadjafi et al., 2006). In fact, many species of water stressed areas tend to synchronize germination during rainy periods characterized by cooler temperatures (Pemadasa and Lovell, 1975) since this is when there is the greatest chance of survival of emerging seedlings. 4.2. Seed bank evaluation and natural burial Due to dormancy (albeit removable) and the scarce availability of water to promote germination, a substantial amount of seed bank gradually accumulated in each respective hind dune environment (Fig. 2A and B). In each case a surface seed disposition was found. However the seedbank analysis showed a marked speciesdependent diversification in the seed’s ability to reach the deeper sand layers. This may be due to the selective capacity of natural burial performances (Maun, 2004), which is connected to seed characteristics in terms of shape, coat micro-sculpture and weight (Benvenuti, 2007). Weight was also found to be linked to the seedbank’s persistence in arid Mediterranean dune environments (Yu et al., 2007; Zhao et al., 2011). These long-lived species could have a crucial role in re-colonization dynamics after the ecological dune disturbances (Bekker et al., 1999). At a community level, burial acts as a filter and selectively eliminates susceptible taxa, reduces the relative abundance of less tolerant plants, and increases the abundance of tolerant sanddependent species (Maun, 1998). However, species richness was significantly higher at the top of the soil profile (roughly 0–5 cm), thus confirming the seed bank analysis carried out in other dune ecosystems (Bossuyt and Hermy, 2004; Mason et al., 2007). This rather superficial, seed vertical arrangement, has been found in other dune species such as A. crispa (Quilichini and Debussch,

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

0

0 0

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

20

20

0 0

80

80

60

60

Anthemis mixta

20

40 20

0

0 -2

-4

-6

-8

-10

-12

Seedling emergence (%)

- 4.9

Depth inhibition

Seedling emergence (%)

100

(as % of surface germination)

y = -2.19 x3+2.97x2+3.52 x+0.21

0

80

60

60

Cakile maritima

20 0

40 20 0

0

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

80

Depth inhibition

Seedling emergence (%)

- 6.7

(as % of surface germination)

100

40

80

80

60

60

Centaurea aplolepa

20 0

40 20 0

0

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

-5.8

Depth inhibition

Seedling emergence (%)

100

(as % of surface germination)

y = -0.61 x3+6.59x2+3.81 x+0.21

40

80

60

60 40

Dorycnium hirsutum

20

20

0

0 0

-2

-4

-6

-8

-10

Seeding depth (cm)

-12

Seedling emergence (%)

Seedling emergence (%)

80

Depth inhibition

- 4.7

(as % of surface germination)

100

40

- 4.3

80

80

60

60

40

Armeria pungens

20

40 20

0

0 0

-2

-4

-6

-8

-10

-12

y = 2.14 x 3+9.44x2+3.35 x+0.15 100

R2=0.99

100

- 6.8

80

80

60

60

40 20

40

Calystegia soldanella

20

0

0 0

-2

-4

-6

-8

-10

-12

y = -8.19 x 3-20.42x2-0-43 x+0.04 100

100

- 3.0

R2=0.95

80

80

60

60

40

40

Centaurium maritimum

20

20

0

0 0

-2

-4

-6

-8

-10

-12

Seeding depth (cm)

y = -2.42 x 3+3.09x2+3.73 x+0.23 R2=0.97

-12

100

R2=0.97

Seeding depth (cm)

100

-10

Seeding depth (cm)

100

R =0.99

-8

y = -0.7 x3-0.86x 2+3.02 x+0.19

Seeding depth (cm)

2

-6

Seeding depth (cm)

y = 1.52x 3+3.79x2+0.75 x+0.04 R2=0.98

-4

100

Seeding depth (cm)

100

-2

Seeding depth (cm)

100

40

20

0

Seeding depth (cm)

R2=0.98

40

Anthemis maritima

Depth inhibition

20

60

40

(as % of surface germination)

Anchusa crispa

60

Depth inhibition

40

(as % of surface germination)

60

40

80

Depth inhibition

60

80

(as % of surface germination)

80

- 4.8

Depth inhibition

80

100

R2=0.98

(as % of surface germination)

- 5.3

Depth inhibition

R2=0.99

y = -0.7 x3+6.79x2+4.29 x+0.25 100

Depth inhibition

100

(as % of surface germination)

y = -1.43 x 3+4.13 x 2+3.70 x+0.22 100

(as % of surface germination)

Seedling emergence (%)

A

289

y = -2.66 x 3+2.74 x 2+3.75 x+0.23 100

R2=0.98

100

- 4.8

80

80

60

60

Echium arenarium

40 20 0

40 20 0

0

-2

-4

-6

-8

-10

-12

Seeding depth (cm)

Fig. 3. (A) Emergence performances of the dune wildflowers by increasing burial depths. Filled circles indicate the emergence percentages, while empty circles indicate burial inhibition. The inhibition equations (significant for p < 0.01) and the corresponding R2 values are reported. Arrows indicate the values of 50% depth inhibition. (B) Emergence performances of the dune wildflowers by increasing burial depth. Filled circles indicate the emergence percentages, while empty circles indicate the burial inhibition. The inhibition equations (significant for p < 0.01) and the corresponding R2 values are reported. Arrows indicate the values of 50% depth inhibition.

290

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

0 0

0 0

80 60

Medicago marina

0

40 20 0

0

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

R2=0.98

60

Depth inhibition

Seedling emergence (%)

100

- 5.1

(as % of surface germination)

y = -1.88 x3+2.12 x2+2.95 x+0.17

20

80

60

60

20

40

Pancratium maritimum

20

0

0 0

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

- 7.8

80

Depth inhibition

Seedling emergence (%)

100

(as % of surface germination)

y = -6.64 x 3+4.78 x 2+2.22 x+0.09

40

R2=0.98

60

80 60

Scabiosa rutifolia

0

40 20 0

0

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

- 4.1

Depth inhibition

Seedling emergence (%)

100

(as % of surface germination)

y = -4.02 x 3-3.01 x 2+2.49 x+0.17

20

60

60

40

40

Silene conica

20

0

0 0

-2

-4

-6

-8

-10

-12

Seedling emergence (%)

Seedling emergence (%)

R =0.96

80

Depth inhibition

2

(as % of surface germination)

100

- 4.0

20

100

80 60

80 60

40

Ononis variegata

20

40 20

0

0 0

-2

-4

-6

-8

-10

-12

y = -2.59 x 3+2.04 x 2+3.44 x+0.21 100

100

- 4.8 R2=0.98

80 60

80 60

40

Scabiosa maritima

20 0

40 20 0

0

-2

-4

-6

-8

-10

-12

y = -6.19 x 3-9.25 x 2+1.52 x+0.13 100

100

- 3.8 80

R2=0.95

60

80 60

40

Silene colorata

20

40 20

0

0 0

-2

-4

-6

-8

-10

-12

Seeding depth (cm)

y = -5.35 x 3+7.48 x 2+1.67 x+0.13

80

-12

R2=0.96

Seeding depth (cm)

100

-10

Seeding depth (cm)

100

40

-8

- 4.1

Seeding depth (cm)

80

-6

Seeding depth (cm)

100

R =0.98

-4

y = -5..21 x 3+4.54 x2+2.53 x+0.18 100

Seeding depth (cm)

2

-2

Seeding depth (cm)

100

40

20

0

Seeding depth (cm)

80

40

Matthiola tricuspidata

20

Depth inhibition

0

40

(as % of surface germination)

20

Depth inhibition

40

(as % of surface germination)

Glaucium flavum

20

60

Depth inhibition

40

60

80

(as % of surface germination)

60

R2=0.97

Depth inhibition

60

100

- 4.4

80

(as % of surface germination)

80

y = -3.61 x 3+0.88 x 2+3.59 x+0.23 100

Depth inhibition

R2=0.97

Depth inhibition

80

100

- 4.1

(as % of surface germination)

y = -4.61 x 3-5.12 x 2+2.15 x+0.15 100

(as % of surface germination)

Seedling emergence (%)

B

y = -7.88 x 3+13.70 x 2+0.82 x+0.10 100

100

- 3.9 R2=0.92

80 60

60

40

Solidago litoralis

20

40 20

0

Seeding depth (cm)

80

0 0

-2

-4

-6

-8

-10

-12

Seeding depth (cm) Fig. 3. (Continued ).

2000), Cyperus capitatus (Redondo-Gómez et al., 2011), Calligonium spp. (Ren et al., 2002), Elymus canadensis, Ammophila breviligulata, Cakile edentula and Corrispermum hyssopifolium (Maun and Lapierre, 1986). However, in the case of pioneering foredune species (strongly wind-moved), seeds are able to reach a higher burial depth (Wen-Ming et al., 2004; Liu et al., 2007) however in

these cases due to the wind-mediated seed passive sand “submersion”. The annual “in dune” entry of annually produced seeds, leads to the accumulation of a germplasm stock thus preventing their dissemination in an unsuitable environment and/or their predation. However, this “latent life”, at the seed level, does not appear

R2= 0.69

-2 -4

10.28

Y = - 3.23 -

1 + exp

-6

-X ) ( 4.27 3.37

-8 -10 0

10

20

30

40

50

Fig. 4. Boltzmann polynomial regression (statistically significant, p < 0.05) between 1000 seed weight and their respective emergence capacity (expressed as the ability to halve germination percentage of unburied seeds).

Rain (mm)

Total rainfall= 429 mm

A 40

20

Seed self-burial after rain (cm)

0

Jannuary

February

0 -1

4

-2 -3 1 -4

2

3

56

15 14 10 12 16 9 17 13

7

8

11

B

-1

R2= 0.78

-2

104

Y = -103 -

-3

1 + exp

-X ) ( 32.68 8.27

-4 0

10

20

30

40

50

1,000 seed weight (g)

1,000 seed weight (g)

60

291

0

(cm)

0

Self-burial performances

Burial capable to halve germination (cm)

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

20 19 18

Dune wildflower species Fig. 5. (A) Rain dynamics during the self-burial test and (B) burial performances of dune wildflower seeds after rain tested in various experimental conditions. Numbers indicate the several species: 1. Anchusa crispa, 2. Anthemis maritima, 3. Anthemis mixta, 4. Armeria pungens, 5. Cakile maritime, 6. Calystegia soldanella, 7. Centaurea aplolepa, 8. Centaurium maritimum, 9. Dorycnium hirsutum, 10. Echium arenarium, 11. Glaucium flavum, 12. Matthiola tricuspidata, 13. Medicago marina, 14. Ononis variegata, 15. Pancratium maritimum, 16. Scabiosa maritime, 17. Scabiosa rutifolia, 18. Silene colorata, 19. Silene conica, 20. Solidago litoralis. Vertical bars indicate the standard errors of the means.

to have a strong persistence (Bossuyt and Hermy, 2004) and their botanic complexity does not always reflect the floristic composition of the adult community (Bakker et al., 2005; Messina and Rajaniemi, 2011). In spite of this non long-living seed bank, it is clear that its quantity and diversity prevent both the invasiveness of alien species (French et al., 2011) and the recovery of dune ecosystems after ecological disturbances (Marchante et al., 2011). 4.3. Seedling emergence and deep-mediated inhibition The question remains: does burial inhibit the seeds of the various species in the same way?

Fig. 6. Boltzmann polynomial regression (statistically significant, p < 0.05) between 1000 seed weight and their respective self-burial capacities after rain.

The observation that species with larger seeds (P. maritimum and C. soldanella) are less inhibited by increasing depths, and vice versa, means that seed size and depth-inhibition could be inversely related, as found in other species (Bond et al., 1999), although with some exceptions (Chen and Maun, 1999). A possible explanation is that these species are well adapted to sand burial which is one of the main characteristics of fore dunes. Indeed both these two species, at least in the areas examined (data not shown), coexist in the front and hind dunes highlighting adaptation to diversified microenvironments. The inverse relation between seed weight and depth-mediated inhibition confirms this hypothesis. Consequently, species with small seeds are very depth-inhibited and, overall if also characterized by dormancy, appears particularly suitable for the accumulation of a persistent seed bank according to the results of similar experiments conducted in different environments (Thompson et al., 1993). This is the case of C. maritimum and M. tricuspidata which are, at the same time, characterized by both small and dormant seeds. From an ecological point of view, the quiescent/dormant seed status enables the seeds to wait for the most suitable periods (after wind-mediated dune movements) to have a shallow burial for germination and to complete seedling emergence. Indeed, in cases of a too shallow burial, the consequent drought can prevent germination due to water stress (Tobe et al., 2005). An important adaptation strategy to the dune environment is to be able to persist, as both a seed (Maun and Lapierre, 1986) and an adult plant (Sykes and Wilson, 1990), during sand burial. Since a persistent seed bank is not always formed by species with dormant seeds (Thompson et al., 2003), it is clear that environmentalmediated quiescence (due to sand burial) also plays a crucial role in the accumulation of a seed bank. Sometimes, during the burial period of dune species, the induction of secondary dormancy (Liu et al., 2013; Zhu et al., 2014) has also been found, probably mediated by dark and high temperatures in the case of the thermo-dormancy found in G. flavum (Thanos et al., 1989). A further environmental stress capable of inducing secondary dormancy is due to the erratic salinity. Indeed the proximity to the sea typically involves periodic, wind-dependent, increases in sodium chloride able to inhibit seed germination (Mariko et al., 1992). The question of why smaller seeds show higher depth inhibition naturally arises. Since this inhibition is hypoxia-mediated (Benvenuti and Macchia, 1995) via gas diffusion with the overlying atmosphere (Benvenuti, 2003), the smaller seeds of the most inhibited species (such as C. maritimum, M. tricuspidata, S. colorata, S. maritima, G. flavum and O. variegata) appear to be hypoxia intolerant. Conversely, species such as P. maritimum, C. soldanella and

292

S. Benvenuti / Ecological Engineering 91 (2016) 282–293

Cakile maritima were less inhibited by burial depth probably due to the higher energy reserves needed during the heterotrophic phase of seedling growth during their pre-emergence elongation (Forcella et al., 2000). On the other hand, there are advantages and disadvantages linked to the larger or smaller seed size (Moles and Westoby, 2004) in terms of emergence performances and predation risk, respectively. It was thus considered important to verify if, and how, in the studied dune species, there was a relationship between seed weight and emergence capacity by increasing burial depths. The polynomial regression, shown in Fig. 4, confirmed the direct relationship between seed weight and germination capacity by increasing depths. The poor burial found in the in situ experiments (Fig. 5) occurs rain-mediated, as shown by the tests carried out ex situ in a controlled environment. Indeed, during the self-burial experimental period (two rainy winter months), the seeds reached a burial depth between 1 and 3.5 cm depending on the species. This seed slippage, in spite of the short experimental time, is not very far from the already cited data of the vertical arrangement seed bank. This suggest that rain decreases their kinetic energy in the underlying sand layers and tends to deplete their action above a certain degree of depth, as observed in other species (Chambers and MacMahon, 1994). The fact that the smaller seeds reached deeper sand layers (Fig. 6) appears to be due to the lower frictional forces that the small seeds exert between the sand particles confirming similar performances observed in sandy desert environments (Reichman, 1984). On the other hand, the fact that a suicidal germination (seed germination not followed by seedling emergence) is, at least in agro-ecosystems, a rare event (Benvenuti et al., 2001), suggests that self-burial is a survival strategy used to maintain the state of “latent life” as long as the environmental conditions (i.e. strong winds) can bring the seeds back near to the dune surface. The occurrence that most of the tested species with a higher capacity for self-burial (due to their small size) are therophytes, suggests that this category of annual species evolved in order to re-colonize the environment after destructive ecological events, as occurs in similar plant succession dynamics (Kutiel et al., 2000; Debussche et al., 1996).

5. Conclusions The performance of the wildflowers dunes studied, in terms of dormancy, self-burial capacity and seedling emergence by increasing sand depth, was quite different and it is this inhomogeneity that appears to ensure the stability over time of the dune ecosystem. Indeed, the resilience to ecological disturbances, diverse in terms of quantity, frequency and type, implies different survival strategies in such highly unpredictable environments (Forey et al., 2008). Although the role of the seed bank in the dune recolonization after destructive events remains to be fully elucidated, it is however clear that its biodiversity (Mason et al., 2007) and its diverse ecological performance, are able to ensure stability and resilience (De Luca et al., 2011). In this background, the high seed dormancy of most of the studied species plays a crucial role. Indeed the gradual dormancy-breaking appears to guarantee a high potential for re-colonization of the various ecological niches. On the other hand are just natural events such as rain, wind, cold periods that are able to reduce or eliminate their seed dormancy. For example seed scarification is an event that naturally occurs on the seed coat affected by the sand particles during windy periods. Since the seed bank is concentrated in the more superficial sand layers it follows that this thin dune portion require a special attention to biodiversity conservation. The natural (too windy)

or artificial (trampling) sand moving of these layers implies an alteration of the floristic resilience. The most studied species, is characterized by small seeds and this characteristic enables their burial thus promoting a persistent seed bank. Moreover, such persistence, appears increased by their greater depth-mediated inhibition since an inverse relationship between seed weight and emergence performance was found. From a practical point of view seeding of native species (Avis, 1989), in the eventual perspective of environmental restoration, should thus be carried out by using treated seed (cold stratification, scarification, washing, etc.) as a function of the respective needs. In addition seeding must be carried out with a light burial (a few cm) behind the natural or artificial foredunes. This dune environment is quite stable and characterized by a high biodiversity of species without a particular adaptation to the burial, contrary to what happens in the case of the wind-stressed foredune (García-Mora et al., 1999). Finally, the biodiversity conservation of these hind dune wildflowers strongly depends on the maintenance of this particular environment characterized by moderate wind disturbance, due to the fore dune protection, according to the intermediatedisturbance hypothesis that implies a higher species numbers (Palmer, 1994). The balanced degree of intensity and frequency of such ecological disturbance, that strongly imply sand movement and seed burial, will allow both the survival and/or restoration of this important Mediterranean landscape. References Acosta, A., Carranza, M.L., Izzi, C.F., 2009. Are there habitats that contribute best to plant species diversity in coastal dunes? Biodivers. Conserv. 18, 1087–1098. Allen, P.S., Meyer, S.E., 1998. Ecological aspects of seed dormancy loss. Seed Sci. Res. 8, 183–192. Avis, A.M., 1989. A review of coastal dune stabilization in the Cape Province of South Africa. Landsc. Urban Plan. 18, 55–68. Baas, A.C., Sherman, D.J., 2006. Spatiotemporal variability of aeolian sand transport in a coastal dune environment. J. Coast. Res. 22, 1198–1205. Bakker, C., Graaf, H.F., Ernst, W.H., Bodegom, P.M., 2005. Does the seed bank contribute to the restoration of species-rich vegetation in wet dune slacks? Appl. Veg. Sci. 8, 39–48. Baskin, J.M., Baskin, C.C., 2004. A classification system for seed dormancy. Seed Sci. Res. 14, 1–16. Bekker, R.M., Lammerts, E., Schutter, A., Grootjans, A.P., 1999. Vegetation development in dune slacks: the role of persistent seed banks. J. Veg. Sci. 10, 745–754. Benelli, G., Benvenuti, S., Desneux, N., Canale, A., 2014. Cephalaria transsylvanicabased flower strips as potential food source for bees during dry periods in European Mediterranean basin countries. PLOS ONE 9, e93153. Benvenuti, S., 2003. Soil texture involvement in germination and emergence of buried weed seeds. Agron. J. 95, 191–198. Benvenuti, S., 2007. Natural weed seed burial: effect of soil texture, rain and seed characteristics. Seed Sci. Res. 17, 211–220. Benvenuti, S., Macchia, M., 1995. Hypoxia effect on buried weed seed germination. Weed Res. 35, 343–351. Benvenuti, S., Macchia, M., Miele, S., 2001. Quantitative analysis of emergence of seedlings from buried weed seeds with increasing soil depth. Weed Sci. 49, 528–535. Benvenuti, S., Macchia, M., 2006. Seedbank reduction after different stale seedbed techniques in organic agricultural systems. Ital. J. Agron. 1, 11–21. Bond, W.J., Honig, M., Maze, K.E., 1999. Seed size and seedling emergence: an allometric relationship and some ecological implications. Oecologia 120, 132–136. Bossuyt, B., Hermy, M., 2004. Seed bank assembly follows vegetation succession in dune slacks. Veg. Sci. 15, 449–456. Brown, A.C., McLachlan, A., 2002. Sandy shore ecosystems and the threats facing them: some predictions for the year 2025. Environ. Conserv. 29, 62–77. Chambers, J.C., MacMahon, J.A., 1994. A day in the life of a seed: movements and fates of seeds and their implications for natural and managed systems. Annu. Rev. Ecol. Syst. 25, 263–292. Chen, H., Maun, M.A., 1999. Effects of sand burial depth on seed germination and seedling emergence of Cirsium pitcheri. Plant Ecol. 140, 53–60. Cushman, J.H., Waller, J.C., Hoak, D.R., 2010. Shrubs as ecosystem engineers in a coastal dune: influences on plant populations, communities and ecosystems. J. Veg. Sci. 21, 821–831. Debussche, M., Escarré, J., Lepart, J., Houssard, C., Lavorel, S., 1996. Changes in Mediterranean plant succession: old-fields revisited. J. Veg. Sci. 7, 519–526. Defeo, O., McLachlan, A., Schoeman, D.S., Schlacher, T.A., Dugan, J., Jones, A., Lastrag, M., Scapini, F., 2009. Threats to sandy beach ecosystems: a review. Estuar. Coast. Shelf Sci. 81, 1–12.

S. Benvenuti / Ecological Engineering 91 (2016) 282–293 De Luca, E., Novelli, C., Barbato, F., Menegoni, P., Iannetta, M., Nascetti, G., 2011. Coastal dune systems and disturbance factors: monitoring and analysis in central Italy. Environ. Monit. Assess. 183, 437–450. Del Vecchio, S., Mattana, E., Acosta, A.T., Bacchetta, G., 2012. Seed germination responses to varying environmental conditions and provenances in Crucianella maritima L., a threatened coastal species. C. R. Biol. 335, 26–31. Eslami, S.V., Gill, G.S., McDonald, G., 2010. Effect of water stress during seed development on morphometric characteristics and dormancy of wild radish (Raphanus raphanistrum L.) seeds. Int. J. Plant Prod. 4, 159–168. Forcella, F., Arnold, R.L.B., Sanchez, R., Ghersa, C.M., 2000. Modeling seedling emergence. Field Crops Res. 67, 123–139. Forey, E., Chapelet, B., Vitasse, Y., Tilquin, M., Touzard, B., Michalet, R., 2008. The relative importance of disturbance and environmental stress at local and regional scales in French coastal sand dunes. J. Veg. Sci. 19, 493–502. French, K., Mason, T.J., Sullivan, N., 2011. Recruitment limitation of native species in invaded coastal dune communities. Plant Ecol. 212, 601–609. García-Mora, M.R., Gallego-Fernández, J.B., García-Novo, F., 1999. Plant functional types in coastal foredunes in relation to environmental stress and disturbance. J. Veg. Sci. 10, 27–34. Goulson, D., 1999. Foraging strategies of insects for gathering nectar and pollen, and implications for plant ecology and evolution. Perspect. Plant Ecol. Evol. Syst. 2, 185–209. Grootjans, A.P., Geelen, H.W.T., Jansen, A.J.M., Lammerts, E.J., 2002. Restoration of coastal dune slacks in the Netherlands. In: Ecological Restoration of Aquatic and Semi-Aquatic Ecosystems in the Netherlands (NW Europe). Springer, Netherlands. Grootjans, A.P., Adema, E.B., Bekker, R.M., Lammerts, E.J., 2004. Why coastal dune slacks sustain a high biodiversity. In: Coastal Dunes. Springer, Berlin Heidelberg. Hanley, M.E., Hoggart, S.P.G., Simmonds, D.J., Bichot, A., Colangelo, M.A., Bozzeda, F., Heurtefeux, H., Ondiviela, B., Ostrowski, R., Recio, M., Trude, R., ZawadzkaKahlau, E., Thompson, R.C., 2014. Shifting sands? Coastal protection by sand banks, beaches and dunes. Coast. Eng. 87, 136–146. Hilhorst, H.W.M., Karssen, C.M., 1992. Seed dormancy and germination: the role of abscisic acid and gibberellins and the importance of hormone mutants. Plant Growth Regul. 11, 225–238. ISTA (International Seed Testing Association), 1999. International rules for seed testing. Seed Sci. Technol. 27 (Suppl.), 50–52. Jayasuriya, K.M.G., Baskin, J.M., Baskin, C.C., 2008. Dormancy, germination requirements and storage behaviour of seeds of Convolvulaceae (Solanales) and evolutionary considerations. Seed Sci. Res. 18, 223–237. Junker, R.R., Blüthgen, N., 2010. Floral scents repel facultative flower visitors, but attract obligate ones. Ann. Bot. 105, 777–782. Kutiel, P., Peled, Y., Geffen, E., 2000. The effect of removing shrub cover on annual plants and small mammals in a coastal sand dune ecosystem. Biol. Conserv. 94, 235–242. Lithgow, D., Martínez, M.L., Gallego-Fernández, J.B., Hesp, P.A., Flores, P., Gachuz, S., Rodríguez-Revelo, N., Mendoza-Gonzáleza, G., Álvarez-Molina, L.L., 2013. Linking restoration ecology with coastal dune restoration. Geomorphology 199, 214–224. Liu, H.L., Shi, X., Wang, J.C., Yin, L.K., Huang, Z.Y., Zhang, D.Y., 2011. Effects of sand burial, soil water content and distribution pattern of seeds in sand on seed germination and seedling survival of Eremosparton songoricum (Fabaceae), a rare species inhabiting the moving sand dunes of the Gurbantunggut Desert of China. Plant Soil 345, 69–87. Liu, H.L., Zhang, L.W., Yin, L.K., Zhang, D.Y., 2013. Effects of temperature, dry storage, and burial on dormancy and germination of seeds of 13 desert plant species from sand dunes in the Gurbantunggut Desert, Northwest China. Arid Land Res. Manag. 27, 65–78. Liu, Z., Yan, Q., Liu, B., Ma, J., Luo, Y., 2007. Persistent soil seed bank in Agriophyllum squarrosum (Chenopodiaceae) in a deep sand profile: variation along a transect of an active sand dune. J. Arid Environ. 71, 236–242. Lucas, N.S., Shanmugam, S., Barnsley, M., 2002. Sub-pixel habitat mapping of a costal dune ecosystem. Appl. Geogr. 22, 253–270. Marchante, H., Freitas, H., Hoffmann, J.H., 2011. The potential role of seed banks in the recovery of dune ecosystems after removal of invasive plant species. Appl. Veg. Sci. 14, 107–119. Mariko, S., Kachi, N., Ishikawa, S.I., Furukawa, A., 1992. Germination ecology of coastal plants in relation to salt environment. Ecol. Res. 7, 225–233. Mason, T.J., French, K., Russell, K.G., 2007. Moderate impacts of plant invasion and management regimes in coastal hind dune seed banks. Biol. Conserv. 134, 428–439. Maun, M.A., Lapierre, J., 1986. Effects of burial by sand on seed germination and seedling emergence of four dune species. Am. J. Bot. 73, 450–455. Maun, M.A., 1998. Adaptations of plants to burial in coastal sand dunes. Can. J. Bot. 76, 713–738. Maun, M.A., 2004. Burial of plants as a selective force in sand dunes. In: Coastal Dunes. Springer Berlin Heidelberg, pp. 119–135.

293

Messina, D.S., Rajaniemi, T.K., 2011. Does the seed sank reflect plant distributions in a coastal dune? Northeast. Nat. 18, 107–114. Moles, A.T., Westoby, M., 2004. Seedling survival and seed size: a synthesis of the literature. J. Ecol. 92, 372–383. Nadjafi, F., Bannayan, M., Tabrizi, L., Rastgoo, M., 2006. Seed germination and dormancy breaking techniques for Ferula gummosa and Teucrium polium. J. Arid Environ. 64, 542–547. O’Rourke, A.T., Fitzpatrick, Ú., Stout, J.C., 2014. Spring foraging resources and the behaviour of pollinating insects in fixed dune ecosystems. J. Pollinat. Ecol. 13, 161–173. Palmer, M.W., 1994. Variation in species richness: towards a unification of hypotheses. Folia Geobot. Phytotaxon. 29, 511–530. Pemadasa, M.A., Lovell, P.H., 1975. Factors controlling germination of some dune annuals. J. Ecol. 63, 41–59. Petanidou, T., Ellis-Adam, A.C., Njjs, J.C.M., Oostermeijer, J.G.B., 1998. Pollination ecology of Gentianella uliginosa, a rare annual of the Dutch coastal dunes. Nord. J. Bot. 18, 537–548. Pignatti, S., 1982. Flora d’Italia, vols. 1–3. Edagricole, Bologna. Provoost, S., Jones, M.L.M., Edmondson, S.E., 2011. Changes in landscape and vegetation of coastal dunes in northwest Europe: a review. J. Coast. Conserv. 15, 207–226. Pyle, R., Bentzien, M., Opler, P., 1981. Insect conservation. Annu. Rev. Entomol. 26, 233–258. Quilichini, A., Debussch, M., 2000. Seed dispersal and germination patterns in a rare Mediterranean island endemic (Anchusa crispa Viv., Boraginaceae). Acta Oecol. 21, 303–313. Redondo-Gómez, S., Andrades-Moreno, L., Parra, R., Mateos-Naranjo, E., SánchezLafuente, A.M., 2011. Factors influencing seed germination of Cyperus capitatus, inhabiting the moving sand dunes in southern Europe. J. Arid Environ. 75, 309–312. Reichman, O.J., 1984. Spatial and temporal variation of seed distributions in Sonoran Desert soils. J. Biogeogr. 11, 1–11. Ren, J., Tao, L., Liu, X.M., 2002. Effect of sand burial depth on seed germination and seedling emergence of Calligonum L. species. J. Arid Environ. 51, 603–611. Rhind, P., Jones, R., 2009. A framework for the management of sand dune systems in Wales. J. Coast. Conserv. 13, 15–23. Rolston, M.P., 1978. Water impermeable seed dormancy. Bot. Rev. 44, 365–396. Sawma, J.T., Mohler, C.L., 2002. Evaluating seed viability by an unimbibed seed crush test in comparison with the tetrazolium test. Weed Technol. 16, 781–786. Sykes, M.T., Wilson, J.B., 1990. Dark tolerance in plants of dunes. Funct. Ecol. 4, 799–805. Thanos, C.A., Georghiou, K., Skarou, F., 1989. Glaucium flavum seed germination – an ecophysiological approach. Ann. Bot. 63, 121–130. Thompson, K.B., Band, S.R., Hodgson, J.G., 1993. Seed size and shape predict persistence in soil. Funct. Ecol. 7, 236–241. Thompson, K., Ceriani, R.M., Bakker, J.P., Bekker, R.M., 2003. Are seed dormancy and persistence in soil related? Seed Sci. Res. 13, 97–100. Tobe, K., Zhang, L., Omasa, K., 2005. Seed germination and seedling emergence of three annuals growing on desert sand dunes in China. J. Arid Environ. 59, 151–157. Wen-Ming, B., Xue-Mei, B., Ling-Hao, L., 2004. Effects of Agriophyllum squarrosum seed banks on its colonization in a moving sand dune in Hunshandake Sand Land of China. J. Arid Environ. 59, 151–157. Wiese, A.F., Binning, L.K., 1987. Calculating the threshold temperature of development of various weeds. Weed Sci. 35, 177–179. Wilson, J.B., Gitay, H., 1995. Community structure and assembly rules in a dune slack: variance in richness, guild proportionality, biomass constancy and dominance/diversity relations. Vegetatio 116, 93–106. Yang, X., Baskin, C.C., Baskin, J.M., Zhang, W., Huang, Z., 2012. Degradation of seed mucilage by soil microflora promotes early seedling growth of a desert sand dune plant. Plant Cell. Environ. 35, 872–883. Yu, S., Bell, D., Sternberg, M., Kutiel, P., 2008. The effect of microhabitats on vegetation and its relationships with seedlings and soil seed bank in a Mediterranean coastal sand dune community. J. Arid Environ. 72, 2040–2053. Yu, S., Sternberg, M., Kutiel, P., Chen, H., 2007. Seed mass, shape, and persistence in the soil seed bank of Israeli coastal sand dune flora. Evol. Ecol. Res. 9, 325–340. Zeng, Y.J., Wang, Y.R., Zhang, J.M., 2010. Is reduced seed germination due to water limitation a special survival strategy used by xerophytes in arid dunes? J. Arid Environ. 74, 508–511. Zhao, L.P., Wu, G.L., Cheng, J.M., 2011. Seed mass and shape are related to persistence in a sandy soil in northern China. Seed Sci. Res. 21, 47–53. Zhu, Y., Yang, X., Baskin, C.C., Baskin, J.M., Dong, M., Huang, Z., 2014. Effects of amount and frequency of precipitation and sand burial on seed germination, seedling emergence and survival of the dune grass Leymus secalinus in semiarid China. Plant Soil 374, 399–409.