Abstract Aims Pond environmental conditions may differ among years with regards to the season in which ponds begin to fill. We experimentally evaluated how seedling emergence, plant growth and phenology differed among years in which filling occurred in winter, autumn or spring. Methods We collected sediments from a natural temporary pond and located them in aquariums. They were placed in a climatic chamber that simulated annual variation in field environmental temperatures and light conditions. Aquariums were assigned to one of three treatments, which differed in the date on which they were filled with water (autumn, winter and spring). We counted the number of seedlings of different species emerged and recorded data about the presence of flowers, seeds or spores every week. The experiment was finished in June, when we harvested the plants and estimated their biomass. Important Findings In most species, seedling emergences were primarily related to time after filling, and thus synchronized their life cycles with the unpredictably timed wet phase of the ponds. Autumn filling resulted in the highest numbers of seeds/spores. However, winter filling promoted plant growth the most. In the spring filling treatment, more terrestrial plant seedlings emerged and fewer seeds/spores were produced. When ponds are flooded earlier, plants may produce a higher number of propagules. However, in years when inundation is delayed to spring and hydroperiods are short, seedling emergence deplete the seed bank and there is little to no seed production, while terrestrial monocots are able to colonize pond basin. Aquatic plants, temporary ponds, seed bank, seedlings, germination INTRODUCTION Plants living in temporary ponds must accommodate their life cycles to pond inundation duration (i.e. hydroperiod) (Cherry and Gough 2006; Fernández-Aláez et al. 1999). Such ponds display two well-differentiated phases: a wet phase, which usually begins following a period of heavy rain, and a dry phase, which usually occurs in summer in our Mediterranean region. During the wet phase, plant assemblages contain aquatic annual species as well as amphibious or terrestrial species that are able to withstand the flood. The singular plant composition of Mediterranean temporary ponds has been described in several studies (Bagella and Caria 2012; De Bélair 2005; Deil 2005; Fraga 2008; Ferchichi-Ben Jamaa et al. 2010; Fernández-Zamudio et al. 2016; Médail 2004; Pinto-Cruz et al. 2009; Rhazi et al. 2009, 2012; Rouissi et al. 2014). Because these ponds harbor rare and threatened plant species, they have been categorized as a priority habitat by the European Union (NATURA 2000 habitat code 3170). In temporary wetlands, the dry phase is essentially a disturbance event from which aquatic plants must recover (Brock et al. 2003). Temporary pond communities are considered to be resilient because wet-phase community structure can be re-established following the dry phase, largely thanks to long-lived seed banks (Brock et al. 2003). Indeed, aquatic plant communities depend primarily on seed germination to recover, as vegetative propagules (i.e. turions or bulbs) are not usually resistant to the long summer dry phase (Casanova and Brock 1990; Grillas et al. 1993). In the Mediterranean, pond filling dynamics depend on the amount of rain that accumulates after the summer. However, the characteristic unpredictability of rainfall patterns in Mediterranean ecosystems means that there is large interannual variability with regards to when pond filling begins. Although the dry phase typically occurs during the summer, pond filling can begin in the autumn, winter or early spring, with the wet phase lasting most frequently until summer (Díaz-Paniagua et al. 2010). Thus, a given pond may display highly variable hydroperiod duration. The latter affects temperature and light conditions during different parts of the wet phase in important ways. For instance, when a pond begins to fill in the autumn, its hydroperiod is longer than when it begins to fill in the winter. Furthermore, autumn filling means water temperature is warmer at first but then becomes colder. Winter filling means water temperature is colder at first but then becomes warmer. Finally, when a pond fills in the spring, water temperature starts warm and remains warm, but the hydroperiod is shorter. The fact that the same pond can experience such interannual variation means that conditions will be different during seed germination, plant growth and plant reproduction. In this article, we experimentally assessed how filling season affected plants commonly found in Mediterranean temporary ponds in terms of germination, growth and phenology. We exposed pond sediment to simulated pond filling during autumn, winter and spring. We then evaluated differences in plant assemblage composition, seed germination and biomass production. MATERIALS AND METHODS We collected sediment from a temporary pond in Doñana National Park, in SW Spain (Supplementary Table 1). Temporary ponds are common in this area (Díaz-Paniagua et al. 2010), and their vegetation has been described in detail elsewhere (Fernández-Zamudio et al. 2016). We gathered dry sediment from the upper soil layer (top 1–10 cm) from different areas of pond basin. We homogenized the sediment and then deposited it into pots (6.5 × 6.5 × 7 cm). We used 39 aquariums, each one (22 × 22 × 37 cm) contained two three-pot rows with sediments. The aquariums were then exposed to different experimental treatments beginning in October 2011. They were placed within a climatic chamber in which environmental temperatures and light conditions mimicked those in the field (Fig. 1). We programmed temperature in the chamber taking into account the weekly average of air and water temperature variation recorded in the study area. Light was provided by two lamps model OSRAM 58w 21–840 over each aquarium. We filled the aquariums with 17 L of filtered, dechlorinated tap water. Aquariums were randomly assigned to one of three treatments, which differed in the date on which the aquariums were filled with water. Each treatment comprised 13 replicates. In the autumn filling treatment (hereafter autumn treatment), aquariums were filled on 25 October 2011. In the winter filling treatment (hereafter winter treatment), aquariums were on 24 January 2012. Finally, in the spring filling treatment (hereafter spring treatment), aquariums were filled on 27 March 2012. Figure 1: View largeDownload slide total number of seedlings that emerged during the experiment; numbers of seedlings surviving to identification are also represented. The horizontal bars indicate the hydroperiod length of each treatment. Weekly maximum (day) and minimum (night) temperatures and daily hours of light and darkness are also shown. Figure 1: View largeDownload slide total number of seedlings that emerged during the experiment; numbers of seedlings surviving to identification are also represented. The horizontal bars indicate the hydroperiod length of each treatment. Weekly maximum (day) and minimum (night) temperatures and daily hours of light and darkness are also shown. Once an aquarium was inundated, we monitored it on a weekly basis. We counted the number of seedlings (including spores and their germlings) in the first three-pot row, located at the front of the aquarium. We let the seedlings grow until they could be identified; they were then removed from the aquarium. We followed the botanic nomenclature found in Valdés et al. (1987) and Cirujano (2008). We counted the total number of seedlings of each plant species that emerged in each aquarium. Some seedlings died before they could be identified, so we distinguished between seedlings that survived to identification and dead seedlings, which were simply classified as either monocots or dicots. We were able to identify most seedlings to species level. All the monocot seedlings that died before they could be identified belonged to the family Poaceae, a taxon commonly found in the pond from which sediments were taken. In the second three-pot row, located at the back of the aquarium, we let the seedlings grow until the end of the experiment, on 4 June 2012. At this time, we harvested all the plants, clipping them at ground level and separating out the different species. For each aquarium, we measured the fresh biomass (±0.01 g) of each dicot species and that of the monocots as a group (because of their small size). Before the measurements were made, any excess water was removed with a manual centrifuge. We also recorded phenological data: presence of flowers, fruits or spores. We estimated the total number of plant species (i.e. plant richness) per aquarium by counting all the different species observed throughout the experiment: those recorded during the weekly observations, those detected as seedlings that were removed after their identification and those left to grow to assess biomass at the end of the experiment. We also calculated species diversity using species-specific seedling abundance in each aquarium. Statistical analyses We calculated Shannon diversity indices using the Vegan package in R (version 3.1; R Development Core Team 2014). To assess treatment-related differences in species richness, we used generalized linear models with a Poisson error distribution and a logit link function. To compare the number of seedlings, a quasipoisson error distribution was used instead (except in the case of Myriophyllum alterniflorum DC and Nitella translucens (personal communication) C. Agardh, for which a Poisson error distribution was a better fit). For species diversity, a Gaussian error distribution was employed. We also conducted Tukey’s HSD tests for post hoc comparisons. Plant biomass had a non-normal error distribution, so we used Kruskal–Wallis tests as non-parametric alternative. With the Survival plugin in R-commander (Fox and Carvalho 2012), we constructed seedling emergences curves for each plant species and compared them among treatments. Then, we used Cox proportional hazards models to assess the effects of the different variables (time post-filling, daily hours of light and environmental temperature) on seedling emergences of the most abundant species during the experiment. For among-treatment comparisons, we only used data for the first 10 weeks following pond filling to have comparable curves (i.e. for similar germination periods). RESULTS Plant richness and diversity Plant species richness did not differ among treatments (χ2 = −3.120; P = 0.210). There was, on average, between 6.6 and 8.5 species per aquarium; the maximum was 10 species (Table 1). Some species were very common and appeared in all aquariums in all three treatments, notably: Callitriche brutia Petagna, M. alterniflorum, Juncus heterophyllus Dufour, Elatine macropoda Guss and Poaceae. In contrast, some species were less common, such as Baldellia ranunculoides (L.) Parl. and Ranunculus peltatus Schrank, or were rare, such as Chara connivens Salzm. ex A. Braun, which only grew in autumn treatment (Table 1). Species diversity differed among treatments (F = 72.386; P < 0.001), with the winter treatment displaying significantly lower values (Tukey post hoc P < 0.05; Table 1). Table 1: number of aquariums per treatment in which each species was recorded, as well as the mean (±standard error) and range (minimum–maximum) of total species richness and the Shannon diversity index Species Autumn Winter Spring Callitriche brutia 13 13 13 Myriophyllum alterniflorum 13 13 13 Juncus heterophyllus 13 13 13 Elatine macropoda 13 13 13 Nitella translucens 13 11 9 Ranunculus peltatus 5 3 2 Chara connivens 6 0 0 Baldellia ranunculoides 7 1 4 Isolepis fluitans 13 11 11 Poaceae 13 13 8 Mean total richness/aquarium (minimum–maximum) 8.46 ± 0.32 (7–10) 7.2 ± 0.27 (6–9) 6.62 ± 0.29 (5–9) Mean diversity/aquarium (minimum–maximum) 1.43 ± 0.03 (1.26–1.58) 1.27 ± 0.03 (1.00–1.39) 1.39 ± 0.03 (1.17–1.60) Species Autumn Winter Spring Callitriche brutia 13 13 13 Myriophyllum alterniflorum 13 13 13 Juncus heterophyllus 13 13 13 Elatine macropoda 13 13 13 Nitella translucens 13 11 9 Ranunculus peltatus 5 3 2 Chara connivens 6 0 0 Baldellia ranunculoides 7 1 4 Isolepis fluitans 13 11 11 Poaceae 13 13 8 Mean total richness/aquarium (minimum–maximum) 8.46 ± 0.32 (7–10) 7.2 ± 0.27 (6–9) 6.62 ± 0.29 (5–9) Mean diversity/aquarium (minimum–maximum) 1.43 ± 0.03 (1.26–1.58) 1.27 ± 0.03 (1.00–1.39) 1.39 ± 0.03 (1.17–1.60) View Large Seedling emergence A total of 4256 seedlings emerged in the first pot row. There were significant differences among treatments (χ2 = −63.96; P = 0.005). There were also significant differences in seedling mortality among treatments (χ2 = 32.58; P < 0.005): a higher proportion of seedlings died in the winter (41.0%) than in autumn (14.3%) and spring (15.0%) treatments (Fig. 1). Except for eight dicots in the winter treatment, all the dead seedlings were Poaceae (24.9% of total seedlings). As for the seedlings that survived to identification, there were 1159 individuals emerged in the autumn treatment, 979 individuals in the winter treatment and 1058 individuals in the spring treatment (Fig. 1). Juncus heterophyllus, Isolepis fluitans (L.) R. Br. and C. brutia were the most abundant. There were no significant differences in species-specific germination among treatments, except in the case of C. brutia, which was less abundant in the spring treatment, and of I. fluitans, which was much more abundant in the spring treatment (Table 2; Fig. 2). Table 2: Mean ( ± standard error) number of seedlings per aquarium in each treatment and the results of the statistical comparisons between treatments (χ2 and P) for each species Species Autumn Winter Spring χ2; P values Callitriche brutia 27.61 ± 2.65 26.61 ± 1.74 19.62 ± 1.5 20.829; P = 0.007 Myriophyllum alterniflorum 5.15 ± 0.48 6.84 ± 0.77 5.00 ± 0.39 4.6751; P = 0.097 Elatine macropoda 7.00 ± 1.01 5.46 ± 0.85 4.38 ± 0.49 7.9636; P = 0.064 Ranunculus peltatus 0.23 ± 0.12 0.15 ± 0.10 0 Juncus heterophyllus 35.38 ± 3.11 32.77 ± 2.26 26.23 ± 2.84 18.797; P = 0.053 Isolepis fluitans 10.62 ± 1.81 2.23 ± 0.87 29.38 ± 6.30 3.398; P < 0.0005 Baldellia ranunculoides 0.85 ± 0.30 0 0.38 ± 0.21 Nitella translucens 1.92 ± 0.43 1.15 ± 0.22 0.85 ± 0.25 5.9512; P = 0.051 Chara connivens 0.15 ± 0.10 0 0 Dead Poaceae seedlings 14.85 ± 2.42 52.30 ± 4.02 14.38 ± 4.32 39.861; P = 0.073 Total seedlings surviving to identification 89.2 ± 6.74 75.3 ± 368 81.38 ± 9.0 63.96; P = 0.005 Total seedlings 104 ± 8.17 127.6 ± 6.32 95.8 ± 6.61 Species Autumn Winter Spring χ2; P values Callitriche brutia 27.61 ± 2.65 26.61 ± 1.74 19.62 ± 1.5 20.829; P = 0.007 Myriophyllum alterniflorum 5.15 ± 0.48 6.84 ± 0.77 5.00 ± 0.39 4.6751; P = 0.097 Elatine macropoda 7.00 ± 1.01 5.46 ± 0.85 4.38 ± 0.49 7.9636; P = 0.064 Ranunculus peltatus 0.23 ± 0.12 0.15 ± 0.10 0 Juncus heterophyllus 35.38 ± 3.11 32.77 ± 2.26 26.23 ± 2.84 18.797; P = 0.053 Isolepis fluitans 10.62 ± 1.81 2.23 ± 0.87 29.38 ± 6.30 3.398; P < 0.0005 Baldellia ranunculoides 0.85 ± 0.30 0 0.38 ± 0.21 Nitella translucens 1.92 ± 0.43 1.15 ± 0.22 0.85 ± 0.25 5.9512; P = 0.051 Chara connivens 0.15 ± 0.10 0 0 Dead Poaceae seedlings 14.85 ± 2.42 52.30 ± 4.02 14.38 ± 4.32 39.861; P = 0.073 Total seedlings surviving to identification 89.2 ± 6.74 75.3 ± 368 81.38 ± 9.0 63.96; P = 0.005 Total seedlings 104 ± 8.17 127.6 ± 6.32 95.8 ± 6.61 View Large Figure 2: View largeDownload slide cumulative number of seedlings in the three pond filling treatments (autumn, winter and spring) over the course of the experiment. Figure 2: View largeDownload slide cumulative number of seedlings in the three pond filling treatments (autumn, winter and spring) over the course of the experiment. The timing of seedling emergence significantly differed among treatments (χ2 = 122.0; df = 2; P < 0.0005). In all three treatments, the first seedlings appeared the first week after filling, but they did so at a faster rate in the spring treatment. Emergence peaked from the second to the fifth week in the spring treatment; it was highest from the third to the fifth week in the winter treatment and from the fourth to the seventh week in the autumn treatment. After 10 weeks, in both the winter and autumn treatments, seedling emergence continued at very low levels until the end of the experiment (Fig. 1). In each treatment, we observed two main periods of emergence. Callitriche brutia, J. heterophyllus and M. alterniflorum appeared during the first. Elatine macropoda, I. fluitans and N. translucens appeared during the second. In the spring treatment, the emergence peaks for all species were closer together; N. translucens and I. fluitans emerged slightly later than the others. In the autumn treatment, seedling mortality was low and emergence continued after the seventh week of the experiment. In the winter treatment, in contrast, seedling mortality was high and was greatest around the time of the second seedling-emergence peak (weeks 7–13). In the spring treatment, the seedlings that died had emerged as early as the second week (Fig. 1). In species with sufficiently large sample sizes, we found significant differences in the timing of emergence, except in the case of N. translucens (Table 3). Furthermore, except in N. translucens and M. alterniflorum, time to emergence was significantly related to time post-filling; the effects of temperature and daily hours of light were not significant (Table 3). Table 3: for different species, mean (± standard error) and range (in brackets) of the number of days post-filling, environmental temperature (°C) and daily hours of light during seedling emergence Species Variables Autumn Winter Spring χ2 Days post-filling* 46.8 ± 2.5 (10–217) 30.2 ± 1.1 (8–106) 26.9 ± 1.5 (6–72) 24.9*** Callitriche brutia Temperature 14.2 ± 0.1 (11.3–25.8) 15.0 ± 0.1 (12.6–25.3) 19.9 ± 0.23 (17.6–27.3) Light 607.9 ± 2.5 (576–872) 671.4 ± 2.6 (621–841) 808.6 ± 2.9 (766–879) Days post-filling 80.3 ± 5.8 (28–197) 56.9 ± 3.4 (28–135) 31.5 ± 1.9 (6–63) 60.5*** Elatine macropoda Temperature 13.7 ± 0.28 (11.3–21.8) 18.5 ± 0.5 (14.5–32.3) 20.5 ± 0.3 (17.6–25.8) Light 628.9 ± 8.6 (576–841) 732.1 ± 7.5 (666–879) 823.8 ± 3.5 (766–872) Days post-filling** 57.6 ± 1.8 (14–226) 40.3 ± 1.4 (14–135) 28.0 ± 2.7 (10–69) 137.0*** Juncus heterophyllus Temperature 13.1 ± 0.1 (11.3–27.3) 16.2 ± 0.2 (13.2–32.3) 20.0 ± 0.4 (18.0–27.3) Light 600.4 ± 2.3 (576–878) 692.6 ± 3.2 (635–879) 818.7 ± 4.7 (624–879) Days post-filling 39 ± 1.5 (28–50) Baldellia ranunculoides Temperature 12.8 ± 0.13 (12.1–13.7) Light 580.8 ± 1.1 (576–590) Days post-filling*** 68.8 ± 4.8 (36–217) 56.6 ± 8.0 (28–111) 34.5 ± 2.7 (13–69) 117.0*** Isolepis fluitans Temperature 12.3 ± 0.23 (11.3–25.8) 18.4 ± 1.1 (14.5–26.7) 20.8 ± 0.5 (12.1–27.3) Light 593.6 ± 6.4 (576–872) 732.2 ± 18.4 (666–853) 828.9 ± 4.3 (576–879) Days post-filling 40.6 ± 5.3 (28–197) 3.7 ± 1.6 (8–106) 20.1 ± 1.1 (6–55) 27.6*** Myriophyllum alterniflorum Temperature 14.3 ± 0.3 (11.3–21.8) 14.3 ± 0.2 (12.6–25.3) 19.0 ± 0.1 (17.6–24.3) Light 605.8 ± 7.4 (576–841) 656.9 ± 2.7 (621–841) 802.5 ± 3.03 (766–864) Days post-filling 81.7 ± 16.1 (10–209) 60.3 ± 6.8 (36–118) 35.3 ± 3.21 (20–4) 5.1 Nitella translucens Temperature 15.1 ± 0.9 (11.3–24.3) 19.1 ± 1.0 (15.4–28.4) 20.7 ± 0.5 (18.7–21.8) Light 656.1 ± 27.8 (576–864) 740.0 ± 15.0 (683–864) 830.7 ± 4.4 (813–841) Species Variables Autumn Winter Spring χ2 Days post-filling* 46.8 ± 2.5 (10–217) 30.2 ± 1.1 (8–106) 26.9 ± 1.5 (6–72) 24.9*** Callitriche brutia Temperature 14.2 ± 0.1 (11.3–25.8) 15.0 ± 0.1 (12.6–25.3) 19.9 ± 0.23 (17.6–27.3) Light 607.9 ± 2.5 (576–872) 671.4 ± 2.6 (621–841) 808.6 ± 2.9 (766–879) Days post-filling 80.3 ± 5.8 (28–197) 56.9 ± 3.4 (28–135) 31.5 ± 1.9 (6–63) 60.5*** Elatine macropoda Temperature 13.7 ± 0.28 (11.3–21.8) 18.5 ± 0.5 (14.5–32.3) 20.5 ± 0.3 (17.6–25.8) Light 628.9 ± 8.6 (576–841) 732.1 ± 7.5 (666–879) 823.8 ± 3.5 (766–872) Days post-filling** 57.6 ± 1.8 (14–226) 40.3 ± 1.4 (14–135) 28.0 ± 2.7 (10–69) 137.0*** Juncus heterophyllus Temperature 13.1 ± 0.1 (11.3–27.3) 16.2 ± 0.2 (13.2–32.3) 20.0 ± 0.4 (18.0–27.3) Light 600.4 ± 2.3 (576–878) 692.6 ± 3.2 (635–879) 818.7 ± 4.7 (624–879) Days post-filling 39 ± 1.5 (28–50) Baldellia ranunculoides Temperature 12.8 ± 0.13 (12.1–13.7) Light 580.8 ± 1.1 (576–590) Days post-filling*** 68.8 ± 4.8 (36–217) 56.6 ± 8.0 (28–111) 34.5 ± 2.7 (13–69) 117.0*** Isolepis fluitans Temperature 12.3 ± 0.23 (11.3–25.8) 18.4 ± 1.1 (14.5–26.7) 20.8 ± 0.5 (12.1–27.3) Light 593.6 ± 6.4 (576–872) 732.2 ± 18.4 (666–853) 828.9 ± 4.3 (576–879) Days post-filling 40.6 ± 5.3 (28–197) 3.7 ± 1.6 (8–106) 20.1 ± 1.1 (6–55) 27.6*** Myriophyllum alterniflorum Temperature 14.3 ± 0.3 (11.3–21.8) 14.3 ± 0.2 (12.6–25.3) 19.0 ± 0.1 (17.6–24.3) Light 605.8 ± 7.4 (576–841) 656.9 ± 2.7 (621–841) 802.5 ± 3.03 (766–864) Days post-filling 81.7 ± 16.1 (10–209) 60.3 ± 6.8 (36–118) 35.3 ± 3.21 (20–4) 5.1 Nitella translucens Temperature 15.1 ± 0.9 (11.3–24.3) 19.1 ± 1.0 (15.4–28.4) 20.7 ± 0.5 (18.7–21.8) Light 656.1 ± 27.8 (576–864) 740.0 ± 15.0 (683–864) 830.7 ± 4.4 (813–841) The results of the statistical comparisons among seedling emergences curves (χ2 and P) are provided. For the model variables, only the significance of the regression coefficients are shown (***P < 0.0005; **P < 0.005; *P < 0.05). View Large Phenology: flower, fruit and spore production From February onwards in the autumn treatment, we observed flowers and fruits in C. brutia, R. peltatus and E. macropoda, as well as spores in N. translucens. The same group of species produced fruits in the winter treatment, but over a shorter period. In the spring treatment, only C. brutia produced fruits (Fig. 3). We did not observe flower production by M. alterniflorum in any treatment, which was probably related to the quality of artificial light in the climatic chambers. Figure 3: View largeDownload slide periods during which different species were observed with flowers, fruits or spores in the three pond filling treatments (autumn, winter and spring). Only species that produced reproductive structures are shown. Figure 3: View largeDownload slide periods during which different species were observed with flowers, fruits or spores in the three pond filling treatments (autumn, winter and spring). Only species that produced reproductive structures are shown. Biomass Plant biomass significantly differed among treatments (χ2 = 16.66; df = 2; P < 0.0005), mainly because of the greater amount of biomass in the winter treatment (Fig. 4). In all three treatments, M. alterniflorum was the predominant biomass producer: it generated more than 65% of the total plant biomass observed across all the treatments and had significantly greatest biomass than the other species within the winter treatment (χ2 = 9.18; df = 2; P = 0.010). Callitriche brutia had the second highest level of overall biomass. It produced an average of 15% of the total plant biomass across all three treatments, although its biomass levels were similar to those of other species (i.e. E. macropoda and N. translucens) within the autumn treatment (Fig. 4). Figure 4: View largeDownload slide plant biomass (g), expressed as log (x + 1) at the end of the experiment in the three pond filling treatments (autumn, winter and spring). M alt: Myriophyllum alterniflorum, C bru: Callitriche brutia, N tra: Nitella translucens, I flu: Isolepis fluitans, E mac: Elatine macropoda, R pel: Ranunculus peltatus. Figure 4: View largeDownload slide plant biomass (g), expressed as log (x + 1) at the end of the experiment in the three pond filling treatments (autumn, winter and spring). M alt: Myriophyllum alterniflorum, C bru: Callitriche brutia, N tra: Nitella translucens, I flu: Isolepis fluitans, E mac: Elatine macropoda, R pel: Ranunculus peltatus. DISCUSSION Depending on the year, Mediterranean temporary ponds may fill in the autumn, winter or spring, which means there is wide interannual variability in the initial environmental conditions of the wet phase (Díaz-Paniagua et al. 2010). However, our study has revealed that this variability does not have a major impact on the composition of aquatic plant assemblages. Our three treatments, which simulated pond filling during three different seasons, had similar levels of plant richness with the same species dominating in each treatment. This finding suggests these aquatic plant communities are likely adapted to dealing with the unpredictable conditions found in Mediterranean temporary ponds. Our results contrast with those of other studies, which have found major differences in species composition related to filling season. In temporary wetlands in Australia, Warwick and Brock (2003) found species richness was higher when inundation occurred in summer than in the autumn. In California vernal pools, the composition of their plant communities was sensitive to differences in the timing of first rain, revealing the influence of temperature on this system (Bliss and Zedler 1997). The difference between our study and those cited above could be related to the number of plant biotypes taken into account; those communities included not only aquatic species, which were predominant in our study, but also terrestrial generalists that are less dependent on inundation for germination (Bliss and Zedler 1997). In our experiment, we observed all the common aquatic species that occur in the pond from which the sediment was collected. The longer hydroperiod associated with the autumn treatment allowed a larger number of seeds of different species to germinate, thus favoring higher species diversity. One species was observed exclusively in the autumn treatment—Chara connivens—suggesting that it may require longer inundation times than other species to germinate. In a previous study, charophyte species were found to germinate at low levels in Doñana marshes, even though its oospores were common in the seed bank (Grillas et al. 1993). Time after filling was the main factor that triggered seedling emergence in our temporary ponds. The plants’ rapid response to inundation represents a strategy: plants can thus synchronize their annual life cycles with the unpredictable timing of the wet phase of temporary ponds, thus allowing them to make the most of the growing season. Consequently, from the beginning of the wet phase, temporary ponds acquire a complex structure of macrophytes that provide food and refuge for associated aquatic fauna (Arribas et al. 2015; Díaz-Paniagua 1987; Nilsson 1996; Williams 2006). Mediterranean temporary ponds commonly experience severe summer droughts, and the resulting long desiccation periods and high temperatures are often fatal to vegetative propagules; indeed, aquatic plant communities depend on their seed banks to re-establish themselves (Aponte et al. 2010; Brock et al. 2003; Grillas et al. 1993). Seeds of aquatic plants may last for very long periods of time (Deil 2005), and several studies have demonstrated that seed banks are not depleted in successive inundation periods, making ponds resilient to the environmental fluctuations (Brock and Rogers 1998; Brock et al. 2003; Bonis et al. 1995) that are common in such temporary habitats. Seeds go dormant to survive over the long term but then require exposure to certain conditions to break out of that state and germinate (Carta 2016). Pond desiccation is one of the most important mediating factors (Bonis et al. 1995; Baskin and Baskin 1998; Cronk and Fenessy 2001). Thus, when dormancy is disrupted during a dry summer, seeds are ready to germinate at the beginning of the next wet phase (Baskin and Baskin 1998; Carta 2016). In Mediterranean temporary ponds, seed germination mostly depends on cool temperatures (<15 °C) (Carta 2016). However, temperatures below this optimum range are common in Doñana ponds when they are flooded in the autumn or winter. Yet aquatic plant seeds nonetheless start to germinate. Overall, in our experiment, seedlings emerged over the full range of available temperatures, including values below 15 °C. The ability of these species to germinate at below-optimum temperatures is an important factor that favors their synchronization with the wet phase of temporary ponds. For several species, seedling number did not differ among treatments, suggesting that temperature did not constrain seed germination. However, it may have some influence on the length of the germination period, accelerating seedling emergence in years of spring filling or slightly delaying it in years of autumn filling, during the period when daily temperatures are decreasing. We observed a second major germination peak in the winter treatment, which coincided with monocot seedling mortality. The seedlings came from the seeds of the terrestrial monocots that usually colonize the dry basin of the pond from which the sediment had been taken. Aquatic, terrestrial and amphibious plants may be found within the basin of Mediterranean temporary ponds (see e.g. Fernández-Zamudio et al. 2016; Pinto-Cruz et al. 2009), and species composition may differ among years based on pond inundation and desiccation regimes (Brock and Rogers 1998; Warwick and Brock 2003). Based on the results of our study, the proportion of terrestrial species should be greater in years with spring filling and shorter hydroperiods, while the proportion of aquatic species should be greater in years with autumn or winter filling and longer hydroperiods. When inundation conditions persist, terrestrial seedlings die. However, they can survive in dry years or in years when pond desiccation occurs in the early spring. Therefore, if terrestrial seedlings emerge at the end of the wet phase, ponds may be invaded by terrestrials, as we have observed during dry periods in the basins of many temporary ponds in our study area. Different species made varying contributions to total plant biomass across the three treatments. The group of species with the greatest numbers of seedlings (J. heterophyllus, I. fluitans and C. brutia) did not include the largest producer of biomass. Myriophyllum alterniflorum—which had low levels of seedling emergence. In nature, M. alterniflorum forms thick mats within ponds. The rapid and great increase in I. fluitans seedlings in the spring treatment suggests that this species is able to survive in the ponds when they are close to drying up and may even reproduce during the dry phase. This species appears as an aquatic plant in temporary ponds of our study area (Fernández-Zamudio et al. 2016), but it has also been classified as an amphibious (Pinto-Cruz et al. 2009) or occasionally terrestrial (Cirujano et al. 2014). The winter treatment had the lowest seedling survival but the greatest plant biomass production, probably because temperatures were increasing during the wet phase, favoring plant growth. In the autumn treatment, plants had 2–3 months longer to grow, but temperatures were initially decreasing and thus unfavorable to rapid growth. Although M. alterniflorum also predominated in this treatment, other species increased their contribution to the total biomass in relation to the other treatments, namely N. translucens and E. macropoda. It is likely that, during the colder initial growth conditions, M. alterniflorum could not exert competitive pressure on the other species, which allowed them to produce more biomass than they could in the other treatments. Although the autumn treatment had a longer hydroperiod, and thus growth period, than did the winter treatment, its biomass levels were lower. This pattern could be explained by the death or deterioration of older plants. We harvested plants simultaneously across all the treatments, and it is possible that this moment did not coincide with the peak of plant production in the autumn treatment. Therefore, we cannot exclude the possibility that, at other times during the experiment, plant biomass in the autumn treatment was similar or even higher than that in the winter treatment. However, the fact that biomass was the lowest in the spring treatment is clearly related to the shorter growth period. Pond filling season also influenced species phenology. Clearly, the most notable advantage of the autumn filling was that most species were able to generate a greater number of propagules: earlier germination meant more time was available for propagule production. In contrast, only one species (C. brutia) produced fruits in the spring treatment, suggesting that aquatic plants make limited contributions to the seed bank in years when inundation is delayed and hydroperiods are short. A succession of years in which hydroperiods are short may have severe consequences for the conservation of plant richness in temporary ponds. Seedlings emerge, depleting the seed bank, but there is little to no seed production. In addition, during these short cycles, more terrestrial monocots emerge and survive, covering the pond basin if it dries out. Although aquatic seed banks can persist for long periods of time (Deil 2005), are not depleted by the annual germination, and can produce plants for several successive years without being replenished (Brock and Rogers 1998; Bonis et al. 1995), its capacity to restore aquatic communities could be reduced if the frequency of dry years or short-hydroperiod years increases. It is common for Mediterranean habitats to experience an alternation of wet and dry years; such dynamics are normal dynamics for Mediterranean temporary ponds. However, natural conditions may be altered by different causes, such as global climate change or human exploitation of water resources. Some alterations in the natural dynamics of our study area have been observed: there is greater pond desiccation and hydroperiods are shortening (Díaz-Paniagua and Aragonés 2015; Gómez-Rodríguez et al. 2010; Serrano and Zunzunegui 2008). The water table is thus deeper, which means rainfall levels must be higher for ponds to fill, increasing the likelihood that they will be filled later rather than earlier. With regards to aquatic plant species, our study suggested that an increase in the frequency of dry years, as well as of years in which ponds fill late and thus have shorter hydroperiods, may be detrimental for the conservation of pond vegetation, resulting in limited replenishment of the seed bank and favoring pond colonization by terrestrial species. Doñana Ponds form a natural network of aquatic habitats, with a rich and diverse plant community that includes rare, threatened and endangered species (Fernández-Zamudio et al. 2016). To promote their conservation, it will be important to reduce the exploitation of aquatic resources and thus preserve the natural dynamics of pond inundation and desiccation cycles. SUPPLEMENTARY DATA Supplementary material is available online at Journal of Plant Ecology online. FUNDING This study was funded by the Spanish Ministry of Agriculture, Food and Environment (Project 158/2010). Conflict of interest statement. None declared. REFERENCES Aponte C Kazakis G Ghosn Det al. ( 2010) Characteristics of the soil seed bank in Mediterranean temporary ponds and its role in ecosystem dynamics. Wetlands Ecol Manag 18: 243– 53. Google Scholar CrossRef Search ADS Arribas R Díaz-Paniagua C Caut Set al. ( 2015) Stable isotopes reveal trophic partitioning and trophic plasticity of a larval amphibian guild. PLOS ONE 10: e0130897. Google Scholar CrossRef Search ADS PubMed Bagella S Caria MC( 2012) Diversity and ecological characteristics of vascular flora in Mediterranean temporary pools. C R Biol 335: 69– 76. Google Scholar CrossRef Search ADS PubMed Baskin CC Baskin JM( 1998) Seeds: Ecology, Biogeography and Evolution of Dormancy and Germination . San Diego: Academic Press. Bliss SA Zedler PH( 1997) The germination process in vernal pools: sensitivity to environmental conditions and effects on community structure. Oecologia 113: 67– 73. Google Scholar CrossRef Search ADS PubMed Bonis A Lepart J Grillas P( 1995) Seed bank dynamics and coexistence of annual macrophytes in a temporary and variable habitat. Oikos 74: 81– 92. Google Scholar CrossRef Search ADS Brock MA Rogers KH( 1998) The regeneration potential pf the seed bank of an ephemeral floodplain in South Africa. Aquat Bot 61: 123– 35. Google Scholar CrossRef Search ADS Brock MA Nielsen DL Shiel SJet al. ( 2003) Drought and aquatic community resilience: the role of eggs and seeds in sediments of temporary wetlands. Freshwater Biol 48: 1207– 18. Google Scholar CrossRef Search ADS Carta A( 2016) Seed regeneration in Mediterranean temporary ponds: germination ecophysiology and vegetation processes. Hydrobiologia 782: 23– 35. Google Scholar CrossRef Search ADS Casanova MT Brock MA( 1990) Charophyte germination and establishment from the seed bank of an Australian temporary lake. Aquat Bot 36: 247– 54. Google Scholar CrossRef Search ADS Cherry JA Gough L( 2006) Temporary floating island formation maintains wetland plant species richness: the role of the seed bank. Aquat Bot 85: 29– 36. Google Scholar CrossRef Search ADS Cirujano S( 2008) Flora Ibérica. Algas continentales . Carófitos ( Characeae ). Madrid: Real Jardin Botánico, CSIC. Cirujano S Meco A García–Murillo P( 2014) Flora Acuática Española. Hidrófitos Vasculares . Madrid: Real Jardin Botánico, CSIC. Cronk JK Fenessy MS( 2001) Wetland Plants: Biology and Ecology . Boca Raton, FL: Lewis Publishers. Google Scholar CrossRef Search ADS De Bélair G( 2005) Dynamique de la végétation de mares temporaires en Afrique du Nord (Numidie orientale, NE Algérie). Ecologia Mediterranea 31: 83– 100. Deil U( 2005) A review on habitats, plant traits and vegetation of ephemeral wetlands – a global perspective. Phytocoenologia 35: 533– 706. Google Scholar CrossRef Search ADS Díaz-Paniagua C( 1987) Tadpole distribution in relation to vegetal heterogeneity in temporary ponds. Herpetol J 1: 167– 9. Díaz-Paniagua C Aragonés D( 2015) Permanent and temporary ponds in Doñana National Park (SW Spain) are threatened by desiccation. Limnetica 34: 407– 24. Díaz-Paniagua C Fernández-Zamudio R Florencio Met al. ( 2010) Temporary ponds from Doñana National Park: a system of natural habitats for the preservation of aquatic flora and fauna. Limnetica 29: 41– 58. Ferchichi-Ben Jamaa H Muller SD Daoud-Bouattour Aet al. ( 2010) Vegetation structures and conservation of Mediterranean temporary wetlands: Mogods region (northern Tunisia). C R Biol 333: 265– 79. Google Scholar CrossRef Search ADS PubMed Fernández-Aláez C Fernández-Aláez M Bécares E( 1999) Influence of water level fluctuation on the structure and composition of the macrophyte vegetation in two small temporary lakes in the northwest of Spain. Hydrobiologia 415: 155– 62. Google Scholar CrossRef Search ADS Fernández-Zamudio R García-Murillo P Díaz-Paniagua C( 2016) Physical and chemical features and water permanence determine aquatic plant distribution in a temporary pond network (Doñana National Park). Hydrobiologia 774: 123– 35. Google Scholar CrossRef Search ADS Fox J Carvalho MS( 2012) The RcmdrPlugin.survival Package: extending the R commander interface to survival analysis. J Stat Softw 49: 1– 32. Google Scholar CrossRef Search ADS Fraga P( 2008) Vascular flora associated to Mediterranean temporary ponds on the island of Minorca. Anales del Jardín Botánico de Madrid 65: 393– 414. Gómez-Rodríguez C Bustamante J Díaz-Paniagua C( 2010) Evidence of hydroperiod shortening in a preserved system of temporary ponds. Remote Sens 2: 1439– 62. Google Scholar CrossRef Search ADS Grillas P Garcia-Murillo P Geertz-Hansen Oet al. ( 1993) Submerged macrophyte seed bank in a Mediterranean temporary marsh: abundance and relationship with established vegetation. Oecologia 94: 1– 6. Google Scholar CrossRef Search ADS PubMed Médail F( 2004) Plant species. In Grillas P Gauthier P Yavercovski N,et al. (eds). Mediterranean Temporary Pools, Vol 1. Issues Relating to Conservation, Functioning and Management . Arles: Station Biologique de la Tour du Valat, 20– 6. Nilsson AN( 1996) Aquatic Insects of North Europe. A Taxonomic Handbook . Stenstrup: Apolo Books. Pinto-Cruz C Molina JA Barbour Met al. ( 2009) Plant communities as a tool in temporary ponds conservation in SW Portugal. Hydrobiologia 634: 11– 24. Google Scholar CrossRef Search ADS R Development Core Team( 2014) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Vienna, Austria. http://www.R-project.org/. Rhazi L Grillas P Rhazi Met al. ( 2009) Ten-year dynamics of vegetation in a Mediterranean temporary pool in western Morocco. Hydrobiologia 634: 185– 94. Google Scholar CrossRef Search ADS Rhazi L Grillas P Saber ERet al. ( 2012) Vegetation of Mediterranean temporary pools: a fading jewel? Hydrobiologia 689: 23– 36. Google Scholar CrossRef Search ADS Rouissi M Boix D Muller SDet al. ( 2014) Spatio-temporal variability of faunal and floral assemblages in Mediterranean temporary wetlands. Comptes Rendus Biologies 337: 695– 708. Google Scholar CrossRef Search ADS PubMed Serrano L Zunzunegui M( 2008) The relevance of preserving temporary ponds during drought: hydrological and vegetation changes over a 16-year period in the Doñana National Park (south-west Spain). Aquat Conserv 18: 261– 79. Google Scholar CrossRef Search ADS Valdés B Talavera S Galiano EF( 1987) Flora Vascular de Andalucía Occidental . Barcelona: Ed. Ketres. Warwick NWM Brock MA( 2003) Plant reproduction in temporary wetlands: the effects of seasonal timing, depth, and duration of flooding. Aquat Bot 77: 153– 67. Google Scholar CrossRef Search ADS Williams DD( 2006) The Biology of Temporary Waters . Oxford: Oxford University Press. © The Author(s) 2017. Published by Oxford University Press on behalf of the Institute of Botany, Chinese Academy of Sciences and the Botanical Society of China. All rights reserved. For permissions, please email: firstname.lastname@example.org
Journal of Plant Ecology – Oxford University Press
Published: Jun 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera