TY - JOUR
AU1 - Zogas,, Alexander
AU2 - Kosman,, Evsey
AU3 - Sternberg,, Marcelo
AB - Abstract Aims Climate change in the eastern Mediterranean region will have a strong impact on ecosystem functioning and plant community dynamics due to a reduction in annual rainfall and increased variability. We aim to understand the role of seed banks as potential buffers against climatic uncertainty determined by climate change. Methods We examined germination strategies of 18 common species present along an aridity gradient. Data were obtained from soil seed banks germinated during nine consecutive years from arid, semi-arid, Mediterranean and mesic Mediterranean ecosystems. At the semi-arid and Mediterranean sites, rainfall manipulations simulating 30% drought and 30% rainfall increase were applied. Germination strategies were tested under optimal irrigation conditions during three consecutive germination seasons to determine overall seed germinability in each soil sample. Changes in germination strategy were examined using a novel statistical approach that considers the climatic and biotic factors that may affect seed germinability. Important Findings The results showed that dominant species controlled their germination fractions by producing seeds with a different yearly germination fraction probability. The amount of rainfall under which the seeds were produced led to two major seed types with respect to germinability: high germinability, seeds leading to transient seed banks, and low germinability, seeds leading to persistent seed banks. We conclude that differential seed production among wet and dry years of both seed types creates a stable balance along the aridity gradient, enabling the soil seed bank to serve as a stabilizing mechanism buffering against rainfall unpredictability. Additionally, we present a general model of germination strategies of dominant annual species in Mediterranean and arid ecosystems that strengthens the notion of soil seed banks as buffers against climatic uncertainty induced by climate change in the region. 摘要 由于年降雨量减少和变异性的增加，地中海东部地区的气候变化将对生态系统功能和植物群落动态产生重大影响。我们旨在了解种子库作为应对气候变化所导致的气候不确定性的潜在缓冲作用。我们研究了沿干旱梯度出现的18种常见物种的萌发策略。数据由干旱、 半干旱、地中海和中等地中海生态系统连续九年内萌发的土壤种子库获得。在半干旱和地中海地区，采用了模拟30%干旱和30%降雨增加的降雨处理方法。在连续三个萌发季的最佳灌溉条件下检测了萌发策略，以确定每种土壤样品的总体种子萌发能力。使用一种新颖的统计方法研究了萌发策略的变化，该方法考虑了可能影响种子发芽性的气候和生物因素。研究结果表明，优势种通过产生具有不同年度发芽率概率的种子来控制其发芽率。可产生种子的降雨量导致了关于可萌发性的两种主要种子类型：高萌发性(可形成短暂种子库的种子)和低萌发性(可形成持久种子库的种子)。我们得出的结论是，两种类型的干湿年之间种子产生的差异沿干旱梯度建立了一个稳定的平衡，使土壤种子库可以充当稳定机制，以防止降雨的不可预测性。此外，我们提出了地中海和干旱生态系统中占主导地位的一年生物种萌发策略的一般模型，该模型加强了土壤种子库可以作为应对该地区气候变化引起的气候不确定性的缓冲剂的概念。 climate change, desert, drought, Mediterranean, rainfall manipulations, seed bank 气候变化, 沙漠, 干旱, 地中海, 降雨控制, 种子 INTRODUCTION In recent decades, significant changes in climatic conditions have been noted in the eastern Mediterranean basin (Kafle and Bruins 2009). Observations and new models have revealed a continuing tendency of decreasing annual rainfall and increasing frequency of extreme events (Samuels et al. 2018). Current models show that by 2050 the amount of annual rainfall will decrease by 20% combined with an increase in winter temperatures (Shohami et al. 2011; Ziv et al. 2014). These changes will probably increase rainfall variability and seasonality with the consequent effect of increasing the water deficit in the region and increasing the duration of drought periods (Alpert et al. 2008; Black 2009; Evans 2009; García-Ruiz et al. 2011). Additionally, expected changes in rainfall distribution and intensity will affect the overall water balance of the eastern Mediterranean region (Samuels et al. 2018). The greatest magnitude of this change is apparent in arid and semi-arid areas, where natural ecosystems are mostly dependent on annual rainfall (Mahlstein et al. 2013). Soil moisture is one of the main filters determining vegetation structure around the Mediterranean basin by shaping seed germination processes, plant growth, survival and reproduction (Estrelles et al. 2015; Peralta et al. 2016). The expected new climatic conditions will have a direct impact on vegetation productivity and ecosystem processes (Beier et al. 2012; Tuffa and Treydte 2017). Additionally, it is expected that the combined increase in annual rainfall variability and temperature with prolonged periods of soil moisture deficit could have important implications for biodiversity and biogeochemical cycles in dry terrestrial ecosystems (Austin et al. 2004; Huxman et al. 2004; Schwinning and Sala 2004), and particularly along the aridity gradient in Israel (Golodets et al. 2013). This is particularly important considering that changes in the soil moisture balance are critical for soil seed bank continuity, which plays an important role in plant community dynamics, especially with respect to the conservation of plant populations and communities (Harel et al. 2011; Lebrija-Trejos et al. 2011). Some studies have shown that rainfall reduction does not significantly affect plant productivity in arid and semi-arid ecosystems (Miranda et al. 2011). In a notable contrast, positive effects of rare rainfall events on productivity were found in other semi-arid ecosystems (Heisler-White et al. 2008, 2009), whereas negative effects on productivity were reported for Mediterranean shrublands (Alon and Sternberg 2019), and more mesic ecosystems (Heisler-White et al. 2009; Knapp et al. 2002). The importance of the structure and dynamics of soil seeds bank (SSB) is widely recognized in ecological studies, but knowledge is still lacking with respect to the processes governing the dynamics of SSBs, particularly under climate change (del Cacho et al. 2012; Peralta et al. 2016). A range of studies has described the structure of the soil seed bank and species composition of target communities in the eastern Mediterranean region (Harel et al. 2011; Lebrija-Trejos et al. 2011; Sternberg et al. 2011; Tielbörger et al. 2014). But our understanding about the processes associated with the formation and dynamics of soil seed banks in both optimal and stressful conditions (water stress), as well as their relationship with vegetation resilience in the eastern Mediterranean basin, is still limited (Peralta et al. 2016). The transition from seed to seedling is one of the most drastic developmental transition periods for plant species, because abiotic and biotic factors may determine the dynamics of the SSB (Donohue et al. 2010; Kigel 1995). Plants may respond to rapidly changing environmental conditions through adjustment of phenotypes to maximize their performance, or by adopting conservative strategies to avoid risks (Gremer and Venable 2014). Most studies support the classical theories of long-term survival strategies of annual plants, namely: (i) delayed germination (Gremer and Venable 2014), (ii) the trade-off between dispersal and dormancy (Siewert and Tielbörger 2010), (iii) phenotypic plasticity as a major mode of plant adaptation (Matesanz and Valladares 2014) and (iv) maternal effects on offspring quality and habitat selection (Galloway 2005, Germain and Gilbert 2014). The most common theory of long-term survival is the dormancy–dispersal trade-off, but experimental tests have revealed a discrepancy between theory and empirical data (Lebrija-Trejos et al. 2011; Siewert and Tielbörger 2010). It has been shown that each plant species must possess a wide range of possible mechanisms to deal with environmental fluctuations (Rysavy et al. 2016; Siewert and Tielbörger 2010). Thus, understanding the strategies of SSB dynamics is of great importance for understanding the evolution and survival of plant communities (Pakeman et al. 2008), and particularly for understanding the process of germination and the formation of soil seed banks. In this study, we evaluated the impact of changes in rainfall conditions on the germination strategies of the dominant species present along a rainfall gradient and modelled them within the framework of climate change. We also studied processes of SSB formation in order to determine changes in the density of dominant annual species. We addressed the following questions: (a) How do changes in rainfall conditions along an aridity gradient affect germination dynamics? (b) Are there differential germination strategies along the aridity gradient? (c) Under which conditions of rainfall amount and variability change may we expect a shift in germination strategy? (d) What is the role of the SSB as a stabilizing mechanism under climate change? MATERIALS AND METHODS Study sites Studies were conducted at four research sites located along a steep rainfall gradient in Israel (length: ~245 km). All sites shared the same calcareous bedrock and were positioned on south-facing slopes (Table 1). Ecosystems represent arid, semi-arid, Mediterranean and mesic Mediterranean conditions characterized by mild, rainy winters and prolonged warm, dry summers. The growing season for herbaceous vegetation commences soon after the first rains, during October–December each year and generally ends in March–April with seed production. Similarly, to other eastern Mediterranean ecosystems, the study sites do not usually experience any rainfall events between May and October. Table 1: Climatic, physical and biotic characteristics of the study sites along the rainfall gradient (adapted from Sternberg et al. 2011) Ecosystem type . Rainfall (mm/CV) . Mean temperature (°C) . Elevation (a.s.l.) . Soil type . Vegetation characteristics . Arid (N 30°52′; E 34°46′) 90/51 19.1 470 m Desert lithosol Vegetation dominated by small shrubs such as Zygophyllum dumosum, Artemisia sieberi and Hammada scoparia with sparsely growing annuals and geophytes. Semi-arid (N 31°23′; E 34°54′) 300/37 18.4 590 m Light brown rendzina Dwarf-shrubs of Sarcopoterium spinosum and Coridothymus capitatus associated with herbaceous (mainly annual) plant species. Mediterranean—Matta LTER (N 31°42′; E 35°3′) 540/30 17.7 620 m Terra rossa Dwarf-shrubland dominated by S. spinosum and high richness of herbaceous (mostly annual) plant species. Mesic Mediterranean (N 33°0′; E 35°14′) 780/22 18.1 500 m Montmorillonitic terra rossa Oak maquis (Quercus calliprinos) and open garrigue formations dominated by shrubs (e.g. Calicotome villosa, S. spinosum and Cistus spp.) and associated herbaceous plants. Ecosystem type . Rainfall (mm/CV) . Mean temperature (°C) . Elevation (a.s.l.) . Soil type . Vegetation characteristics . Arid (N 30°52′; E 34°46′) 90/51 19.1 470 m Desert lithosol Vegetation dominated by small shrubs such as Zygophyllum dumosum, Artemisia sieberi and Hammada scoparia with sparsely growing annuals and geophytes. Semi-arid (N 31°23′; E 34°54′) 300/37 18.4 590 m Light brown rendzina Dwarf-shrubs of Sarcopoterium spinosum and Coridothymus capitatus associated with herbaceous (mainly annual) plant species. Mediterranean—Matta LTER (N 31°42′; E 35°3′) 540/30 17.7 620 m Terra rossa Dwarf-shrubland dominated by S. spinosum and high richness of herbaceous (mostly annual) plant species. Mesic Mediterranean (N 33°0′; E 35°14′) 780/22 18.1 500 m Montmorillonitic terra rossa Oak maquis (Quercus calliprinos) and open garrigue formations dominated by shrubs (e.g. Calicotome villosa, S. spinosum and Cistus spp.) and associated herbaceous plants. Temperature refers to annual means. CV refers to coefficient of rainfall variation (%). Open in new tab Table 1: Climatic, physical and biotic characteristics of the study sites along the rainfall gradient (adapted from Sternberg et al. 2011) Ecosystem type . Rainfall (mm/CV) . Mean temperature (°C) . Elevation (a.s.l.) . Soil type . Vegetation characteristics . Arid (N 30°52′; E 34°46′) 90/51 19.1 470 m Desert lithosol Vegetation dominated by small shrubs such as Zygophyllum dumosum, Artemisia sieberi and Hammada scoparia with sparsely growing annuals and geophytes. Semi-arid (N 31°23′; E 34°54′) 300/37 18.4 590 m Light brown rendzina Dwarf-shrubs of Sarcopoterium spinosum and Coridothymus capitatus associated with herbaceous (mainly annual) plant species. Mediterranean—Matta LTER (N 31°42′; E 35°3′) 540/30 17.7 620 m Terra rossa Dwarf-shrubland dominated by S. spinosum and high richness of herbaceous (mostly annual) plant species. Mesic Mediterranean (N 33°0′; E 35°14′) 780/22 18.1 500 m Montmorillonitic terra rossa Oak maquis (Quercus calliprinos) and open garrigue formations dominated by shrubs (e.g. Calicotome villosa, S. spinosum and Cistus spp.) and associated herbaceous plants. Ecosystem type . Rainfall (mm/CV) . Mean temperature (°C) . Elevation (a.s.l.) . Soil type . Vegetation characteristics . Arid (N 30°52′; E 34°46′) 90/51 19.1 470 m Desert lithosol Vegetation dominated by small shrubs such as Zygophyllum dumosum, Artemisia sieberi and Hammada scoparia with sparsely growing annuals and geophytes. Semi-arid (N 31°23′; E 34°54′) 300/37 18.4 590 m Light brown rendzina Dwarf-shrubs of Sarcopoterium spinosum and Coridothymus capitatus associated with herbaceous (mainly annual) plant species. Mediterranean—Matta LTER (N 31°42′; E 35°3′) 540/30 17.7 620 m Terra rossa Dwarf-shrubland dominated by S. spinosum and high richness of herbaceous (mostly annual) plant species. Mesic Mediterranean (N 33°0′; E 35°14′) 780/22 18.1 500 m Montmorillonitic terra rossa Oak maquis (Quercus calliprinos) and open garrigue formations dominated by shrubs (e.g. Calicotome villosa, S. spinosum and Cistus spp.) and associated herbaceous plants. Temperature refers to annual means. CV refers to coefficient of rainfall variation (%). Open in new tab Soil seed bank (SSB) sampling and determining a ‘core’ set of species SSB samples were collected in autumn before the onset of the rainy season, for nine consecutive years (late September 2001–09). The resulting SSB database included incidence and abundance of 347 plant species obtained during the entire research period. The late collection date ensured that seeds present in these soil samples were exposed for at least 5 months to local natural climatic conditions after seed set and shedding (March/April). This period of field exposure may be important for breaking seed dormancy in some species. Moreover, since major losses due to granivory (mainly by ants and rodents) also occur during this period, sampling at this late date ensures that the number of seeds present in the soil samples represents a more accurate representation of potential seed germination at each station (Lebrija-Trejos et al. 2011). Soil seed banks at each study site were sampled in open areas between shrubs dominated by herbaceous vegetation as representing the dominant habitat. In each study site, we randomly collected a total of 50 SSB samples. Each sample measured 5 cm × 5 cm to a depth of 5 cm, and included seeds on the surface and plant litter. Each soil sample was thoroughly mixed, and stones and coarse roots were removed. These samples were later spread in plastic trays (12 cm × 14 cm, 6.5 cm depth, with drainage holes) on a gauze sheet placed on top of a 3-cm-thick layer of perlite. The thickness of the soil layer varied between 0.75 and 1 cm. The trays were irrigated during winter (beginning early October) in a net-house at the Botanical Garden of Tel Aviv University. Emerging seedlings were identified once a week, counted and continuously removed until no further emergence was observed (mid-March). Individuals plants that were not possible to identify at the seedling stage were grown in the trays until morphological traits of adult plants allowed to do so. The overall seed germinability in each soil sample was determined by replicating the above procedure for each tray during three consecutive germination seasons (winters). During summer, seed bank trays were naturally dried in the net-house to mimic typical hot, dry field conditions during this period. At the end of the third germination year, soils were passed through 5- and 0.30-mm sieves, to retrieve non-germinated seeds that were then counted under a microscope (80× magnification). As the counting of retrieved seeds per sample was very time-consuming and their numbers were very low (<1% of total number of emerged seedlings), this fraction was omitted from the analysis. Seeds smaller than 0.30 mm were not considered due to workload issues. According to our knowledge of the seed bank flora (unpublished database), the number of species with seeds smaller than 0.30 mm that remained non-germinated following three consecutive germination events was not important, as described by Harel et al. (2011). Considering optimal germination conditions (i.e. high watering) we assumed that seed viabilities remained constant during the three consecutive germination years. Fungi development that may affect seed viability was not evident at the germination trays. A ‘core’ set of species (18 species, Table 2) was determined for effective modelling. The species were selected based on the following criteria: (i) each species was detected during at least four of the 9 years of the experiment (2001–09), of which 2 years were control and 7 years experienced rainfall manipulations; (ii) each species appeared in at least two consecutive years; (iii) the amount of seeds produced by each species allowed reasonable statistical analyses (minimum of 600 seeds per m2). Only species from the core set are further considered. Table 2: List of dominant species considered by germination strategies Species . Family . Strategies . Regression . R2 . Filago desertorum Compositae A 0.0033x + 0.9277a 0.14 Mercurialis annua Euphorbiaceae A 0.0052x + 0.8164a 0.16 Picris longirostris Compositae A 0.0044x + 0.9333b 0.28 Anagallis arvensis Primulaceae B −0.0140x + 0.9083a 0.79 Catapodium rigidum Gramineae B −0.0029x + 0.9338a 0.21 Crithopsis delileana Gramineae B −0.0036x + 0.9546a 0.06 Linum corymbulosum Linaceae B −0.0055x + 1.0194a 0.29 Lolium rigidum Gramineae B −0.0020x + 0.9550a 0.09 Plantago afra Plantaginaceae B −0.0023x + 0.8843a 0.07 Stachys neurocalycina Labiatae B −0.0096x + 0.9545a 0.19 Stipa capensis Gramineae B −0.0026x + 0.9741a 0.09 Trifolium scabrum Papilionaceae B −0.0042x + 0.9466a 0.05 Trisetaria macrochaeta Gramineae B −0.0125x + 1.0037a 0.82 Valantia hispida Rubiaceae B −0.0029x + 0.9683a 0.18 Brachypodium distachyon Gramineae R −7E−06x + 0.9810c <<0.01 Convolvulus siculus Convolvulaceae R −0.0003x + 0.9380c <<0.01 Scorpiurus muricatus Papilionaceae R −0.0003x + 0.9324c <<0.01 Urospermum picroides Compositae R −0.0004x + 0.9412c <<0.01 Species . Family . Strategies . Regression . R2 . Filago desertorum Compositae A 0.0033x + 0.9277a 0.14 Mercurialis annua Euphorbiaceae A 0.0052x + 0.8164a 0.16 Picris longirostris Compositae A 0.0044x + 0.9333b 0.28 Anagallis arvensis Primulaceae B −0.0140x + 0.9083a 0.79 Catapodium rigidum Gramineae B −0.0029x + 0.9338a 0.21 Crithopsis delileana Gramineae B −0.0036x + 0.9546a 0.06 Linum corymbulosum Linaceae B −0.0055x + 1.0194a 0.29 Lolium rigidum Gramineae B −0.0020x + 0.9550a 0.09 Plantago afra Plantaginaceae B −0.0023x + 0.8843a 0.07 Stachys neurocalycina Labiatae B −0.0096x + 0.9545a 0.19 Stipa capensis Gramineae B −0.0026x + 0.9741a 0.09 Trifolium scabrum Papilionaceae B −0.0042x + 0.9466a 0.05 Trisetaria macrochaeta Gramineae B −0.0125x + 1.0037a 0.82 Valantia hispida Rubiaceae B −0.0029x + 0.9683a 0.18 Brachypodium distachyon Gramineae R −7E−06x + 0.9810c <<0.01 Convolvulus siculus Convolvulaceae R −0.0003x + 0.9380c <<0.01 Scorpiurus muricatus Papilionaceae R −0.0003x + 0.9324c <<0.01 Urospermum picroides Compositae R −0.0004x + 0.9412c <<0.01 Strategy A: increasing germination fraction in the first year with increasing rainfall; strategy B: decreasing germination fraction with increasing rainfall; strategy C: random, independent of rainfall germination fraction. aP-value <0.001; bP-value <0.01; cP-value >0.05 (no rainfall correlation). Open in new tab Table 2: List of dominant species considered by germination strategies Species . Family . Strategies . Regression . R2 . Filago desertorum Compositae A 0.0033x + 0.9277a 0.14 Mercurialis annua Euphorbiaceae A 0.0052x + 0.8164a 0.16 Picris longirostris Compositae A 0.0044x + 0.9333b 0.28 Anagallis arvensis Primulaceae B −0.0140x + 0.9083a 0.79 Catapodium rigidum Gramineae B −0.0029x + 0.9338a 0.21 Crithopsis delileana Gramineae B −0.0036x + 0.9546a 0.06 Linum corymbulosum Linaceae B −0.0055x + 1.0194a 0.29 Lolium rigidum Gramineae B −0.0020x + 0.9550a 0.09 Plantago afra Plantaginaceae B −0.0023x + 0.8843a 0.07 Stachys neurocalycina Labiatae B −0.0096x + 0.9545a 0.19 Stipa capensis Gramineae B −0.0026x + 0.9741a 0.09 Trifolium scabrum Papilionaceae B −0.0042x + 0.9466a 0.05 Trisetaria macrochaeta Gramineae B −0.0125x + 1.0037a 0.82 Valantia hispida Rubiaceae B −0.0029x + 0.9683a 0.18 Brachypodium distachyon Gramineae R −7E−06x + 0.9810c <<0.01 Convolvulus siculus Convolvulaceae R −0.0003x + 0.9380c <<0.01 Scorpiurus muricatus Papilionaceae R −0.0003x + 0.9324c <<0.01 Urospermum picroides Compositae R −0.0004x + 0.9412c <<0.01 Species . Family . Strategies . Regression . R2 . Filago desertorum Compositae A 0.0033x + 0.9277a 0.14 Mercurialis annua Euphorbiaceae A 0.0052x + 0.8164a 0.16 Picris longirostris Compositae A 0.0044x + 0.9333b 0.28 Anagallis arvensis Primulaceae B −0.0140x + 0.9083a 0.79 Catapodium rigidum Gramineae B −0.0029x + 0.9338a 0.21 Crithopsis delileana Gramineae B −0.0036x + 0.9546a 0.06 Linum corymbulosum Linaceae B −0.0055x + 1.0194a 0.29 Lolium rigidum Gramineae B −0.0020x + 0.9550a 0.09 Plantago afra Plantaginaceae B −0.0023x + 0.8843a 0.07 Stachys neurocalycina Labiatae B −0.0096x + 0.9545a 0.19 Stipa capensis Gramineae B −0.0026x + 0.9741a 0.09 Trifolium scabrum Papilionaceae B −0.0042x + 0.9466a 0.05 Trisetaria macrochaeta Gramineae B −0.0125x + 1.0037a 0.82 Valantia hispida Rubiaceae B −0.0029x + 0.9683a 0.18 Brachypodium distachyon Gramineae R −7E−06x + 0.9810c <<0.01 Convolvulus siculus Convolvulaceae R −0.0003x + 0.9380c <<0.01 Scorpiurus muricatus Papilionaceae R −0.0003x + 0.9324c <<0.01 Urospermum picroides Compositae R −0.0004x + 0.9412c <<0.01 Strategy A: increasing germination fraction in the first year with increasing rainfall; strategy B: decreasing germination fraction with increasing rainfall; strategy C: random, independent of rainfall germination fraction. aP-value <0.001; bP-value <0.01; cP-value >0.05 (no rainfall correlation). Open in new tab We classified these species into two categories based on their germination fractions along the three consecutive germination seasons: species with high germination fractions (HG—mean germination fraction in the first year is above 75%); and low germination fractions (LG—mean germination fraction in the second and third year combined is above 75%). This means that species with LG fraction has LG capacity in the first germination year, but higher in the next two germination years. Climate change scenarios Two climate change scenarios were tested using rainfall manipulations: drought and increased rainfall. These contrasting scenarios were originally based on spatially variable changes in observed rainfall patterns in the region and were confirmed by recent observations and climate change models (Black 2009; Kafle and Bruins 2009; Samuels et al. 2018). Commencing in winter 2002–03, rainfall manipulations were applied at the two central study sites along the rainfall gradient, the Mediterranean and the semi-arid site. These sites are located in the two intermediate locations along the climatic gradient and they represent the transition from mesic to arid conditions: the Mediterranean and the semi-arid desert regions. The rationale of climatic manipulations at these sites is based on predicted potential climate change scenarios in which the strongest changes occur in the transition zone between mesic Mediterranean and arid desert areas. The mesic Mediterranean and the arid desert stations at the ends of the gradient are kept under natural climatic conditions and serve as controls for the climate-manipulated areas. This experimental design is in alignment with the Space-For-Time (SFT) substitution approach (Fukami and Wardle 2005; Pickett 1989). The idea of SFT is to identify natural gradients, therefore ‘the space’, representing contrasting conditions that indicate current and future scenarios, therefore ‘the time’ (Pickett 1989). Our SFT approach to study the effects of rainfall changes used the natural north–south climatic gradient present in Israel. Rainfall treatments mimic the natural timing, frequency and intervals of rainfall events at the sites (see Sternberg et al. 2011, for full experimental design). The rainfall manipulation period covered by this study range from the 2002–03 rainfall year until 2008–09 rainfall season. The 2001–02 rainfall season was considered as a baseline year (zero year) before the rainfall treatments were applied. Rainout shelters (drought treatment) similar to the ones described by Yahdjian and Sala (2002) were constructed to reduce rainfall by 30% at each rainfall event. The fixed shelters utilized V-shaped bands of transparent greenhouse plastic, supported by a frame of galvanized aluminium (mean height = 2.5 m) and covered an area of 10 m × 25 m. Intercepted rain simulating drought conditions was collected in gutters (plastic strips) covering 30% of the area and drained outside the study site. The sides of the shelters were open to allow for air movement and minimize temperature and humidity differences between sheltered and unsheltered areas. Solar radiation on vegetation was minimally affected as the rainout shelters were positioned east–west, and any potential solar dimming changed along the day due to the earth’s rotation. Irrigation systems (increased rainfall treatment) were established to increase the long-term mean annual precipitation by 30%. Irrigation with tap water was applied at the end of each rainfall event exceeding 5 mm by means of drizzle sprinklers. Each treatment was applied to five plots (10 m × 25 m), while five additional plots were left untreated (control treatment). At the mesic Mediterranean and arid sites, five control plots were established, and no rainfall manipulations were conducted. In a previous analysis, we showed that the effects of rainfall manipulation treatments on community biomass and vegetation diversity in both the Mediterranean and semi-arid sites were inconsistent or negligible (Tielbörger et al. 2014). Therefore, in the present analysis we pooled together the plots of the control and rainfall manipulation treatments in each site. This also included the arid and mesic Mediterranean stations where no rainfall manipulations were carried out. With this procedure we obtained a much wider range of conditions within sites, allowing us to examine spatial processes related to soil seed bank germination strategies along with the whole rainfall range. Modelling germination strategies along the aridity gradient Rainfall model Our new approach focuses on changing the resolution of observed rainfall data using the following climate factors: rainfall regime and its variability during two fixed periods of the rainy season (December–January and February–March). Annual plant life cycles are short, and plants generally complete most of their life cycle, from germination to seed production, during the rainy season. Critical periods in the life cycle for Mediterranean and desert annual vegetation are: (i) establishment during December–January (in this period there is very strong seedling competition both between and within species) and (ii) flowering and seed production during February–March. The daily rainfall curve at the studied sites was characterized as a broken line that does not allow for a clear understanding of the rainfall trend during the period considered in this study. In order to smooth out daily fluctuations and to better describe longer-term trends of rainfall amounts, we considered running averages for time intervals of 120 and 60 days (Meehl et al. 2011). We empirically determined three periods along the rainy season: (i) a long period, December–March—almost the entire growing season and (ii) two short periods, December–January—germination and establishment, and February–March—flowering and seed set. Note, e.g., that the running average with an interval of 60 days calculated for December 1 is the mean value of rainfall amounts recorded from December 1 to January 29. Thus, for the running average in December–January, rainfall amounts are considered until the end of March. We calculated Mean Period Precipitation (MPP) for each of the 2-month periods, and for the December–March period, and described climate change scenarios in terms of MPP. Data analysis of species by germination fractions Our approach was to test the effect of rainfall amounts on seed germination fractions during the first germination year for each core species. The first germination year was selected as most germination occurred during this period under the optimal germination conditions (maximum watering) imposed in this study. The following algorithm was developed and determined for every relevant species, and included: Total rainfall amount recorded in every plot where a given species was recorded, independent of site, treatment (control, drought or increased rainfall) and year of the experiment (2001–09). Then, a precipitation presence range (minimum, min, and maximum, max, values) (i.e. rainfall niche) was determined for each core species based on the observations across all samples (around 18 000 in total) collected during the entire experiment (different plots, sites, treatments and years). Thus, for each species a rainfall niche was determined based on its presence along the entire rainfall gradient considered. Additionally, a range [min, max] for each species was determined for each of the three considered periods (December–January, February–March and December–March). The entire precipitation range [min, max] was divided into k = 2n equal intervals [minj, maxj] for j = 1, 2, …, k (binary division; min1 = min, minj = maxj−1, maxk = max), where n depends on a particular species to achieve a resolution of subdivisions of up to 25 mm rainfall (i.e. the fixed precipitation range of each of k intervals does not exceed 25 mm for a given species, maxj − minj ≤ 25, and may vary between 12.5 and 25 mm for different species). The total number of seeds germinated in the first year was counted for each sample, where a given species was observed. Then for sample i, the proportion, fiS ⁠, of seeds germinated in the first year was calculated. A range (minimum, fminP=min{fiS:i∈P} ⁠, and maximum, fmaxP=max{fiS:i∈P} ⁠, values) and average, fP, of the sample estimates of the seed germination fraction in the first year were determined in each plot, P, where a given species was observed. All plots (and samples, as a result) were divided into k groups, Gj, with rainfall amounts within the precipitation interval [minj, maxj], and an average amount of precipitations, aprj was calculated based on the rainfall data for all plots from group Gj (j = 1, 2, …, k = 2n). For each group, Gj, the range (minimum and maximum values) and average of the seed germination fraction in the first year were determined based on the corresponding estimates for all samples (independent of the plots and sites they came from) and plots from that group. Note that the estimate of the average germination fraction, (⁠ f¯S(Gj) ⁠), was based on the sample data, (⁠ fiS ⁠, i∈Gj ⁠), and that (⁠ f¯P(Gj) ⁠) for the plot data, (fP, P∈Gj ⁠), is usually different, while the corresponding minimum and maximum values are identical (⁠ fmin(Gj)=min{fiS:i∈Gj}=min{fminP:P∈Gj} and fmax(Gj)=max{fiS:i∈Gj}=max{fmaxP:P∈Gj} ⁠). A linear regression model was applied to analyse whether there is any correlation between the seed germination fraction in the first year and amount of rainfall in the corresponding sample/plot. The following estimates (dependences) versus the average precipitation, aprj, for samples/plots in group Gj (j = 1, 2, …, k) were tested: (i) fiS ⁠, where sample i∈Gj ⁠; (ii) f¯S(Gj) ⁠; (iii) f¯P(Gj) ⁠; (iv) fmin(Gj); and (v) fmax(Gj). The distribution of the seed germination fraction for the samples in the first year was empirically tested with histograms within each group, Gj (j = 1, 2, …, k). Based on observations of samples germinated during three consecutive rainfall seasons, it is assumed that no viable seeds remained in the germination trays, meaning that all seeds present in the soil seed bank germinated. Data analyses were carried out using IBM SPSS Statistics for Windows, version 24 (IBM Corp., Armonk, NY, USA, R software package 2013). RESULTS For each species, the range of rainfall amounts for its presence as well as the range of seed germination in the first year (proportion of HG fraction) was established (Table 2, Figs 1 and 2; Supplementary Figs S1a, S2a and S3a). We found that the germination strategies of the examined species depended on the amount of rainfall occurring along the rainfall range. These could be divided into three categories (i.e. types) according to the patterns of change in the proportion of the HG fraction observed with increasing rainfall. Figure 1: Open in new tabDownload slide Schematic representation of probabilities of germination strategies by species: (A) correlation between germination probability in first year and rainfall, strategy A. An increase in rainfall increases the germination probability of HG seed types; (B) correlation between germination probability in first year and rainfall, strategy B. An increase in rainfall increases the germination probability of LG seed types; (R) correlation between germination probability in first year and rainfall, strategy R. An increase in rainfall does not affect the germination probabilities of seed type LG or HG. Figure 1: Open in new tabDownload slide Schematic representation of probabilities of germination strategies by species: (A) correlation between germination probability in first year and rainfall, strategy A. An increase in rainfall increases the germination probability of HG seed types; (B) correlation between germination probability in first year and rainfall, strategy B. An increase in rainfall increases the germination probability of LG seed types; (R) correlation between germination probability in first year and rainfall, strategy R. An increase in rainfall does not affect the germination probabilities of seed type LG or HG. Figure 2: Open in new tabDownload slide Germination fractions of Anagallis arvensis showing lower germination fractions with increasing rainfall in the first year, representing strategy B. HG represent high germination fractions and LG represent low germination. Black squares indicate germination fraction of the first year. Red triangles indicate germination fraction of the second year and white circles represent germination fractions of the third year. Values of the linear regression of germination fraction of the first year are indicated and dotted regression line presented. Average germination fraction was pooled from all germination years and sites along the gradient. Figure 2: Open in new tabDownload slide Germination fractions of Anagallis arvensis showing lower germination fractions with increasing rainfall in the first year, representing strategy B. HG represent high germination fractions and LG represent low germination. Black squares indicate germination fraction of the first year. Red triangles indicate germination fraction of the second year and white circles represent germination fractions of the third year. Values of the linear regression of germination fraction of the first year are indicated and dotted regression line presented. Average germination fraction was pooled from all germination years and sites along the gradient. Type A: Increasing rainfall amounts along the rainfall gradient lead to an increase in the fraction of HG seeds on average (probability of germination in the first year increases). In other words, the relative number of samples characterized by a larger amount of seeds germinating in the first year increases along the rainfall gradient (Fig. 1, line A, Supplementary Fig. S1a–c). Three species (Filago desertorum, Mercurialis annua and Picris longirostris) out of 18 showed this type of strategy (Table 2). For a given species, specific ranges exist for both rainfall and the HG fraction. The proportion of the HG seed fraction increases from the low rainfall range reaches a natural balance with the LG fraction (average of the HG fraction interval) around the point of average along the range, and continues to increase with increasing rainfall. Note that the rainfall ranges and HG fraction intervals are usually not the same for different species. Type B: Increasing rainfall amounts along the rainfall gradient lead to an increase in the fraction of LG seeds on average. The probability of germination in the first year decreases with increasing rainfall along the range. In other words, the relative number of samples characterized by a larger amount of seeds germinating in the first year decreases along the rainfall gradient (Figs 1, line B and 2). The proportion of the HG seed fraction decreases from the low rainfall range, reaches a natural balance with the LG fraction (average of the HG fraction interval) around the point of average rainfall along the range and continues to decrease with increasing rainfall. This was the most common strategy found (above 60%) among the 18 species considered (Table 2). Anagallis arvensis provide a clear representation of Type B (Fig. 2). Germination fractions during the first germination year of A. arvensis decreased with increasing rainfall along the gradient. Germination fractions of the second and third year showed an opposite trend to that shown for the first year, however statistically non-significant (Fig. 2). Type R: Seeds germinate randomly either in the first or subsequent years independent of rainfall mount. No consistent pattern in the proportion of the HG fraction was found (Fig. 1, line R, Supplementary Fig. S3a–d). Only four species (Brachypodium distachyon, Convolvulus siculus, Scorpiurus muricatus and Urospermum picroides) out of 18 showed this type of strategy (Table 2). DISCUSSION When considering an aridity gradient with decreasing rainfall and increasing rainfall uncertainty, such as the one considered in this study, we may expect increasing seed dormancy based on bet-hedging strategies (Claus and Venable 2000; Cohen 1967). However, our results only partially support this idea, as only three species followed these expectations (Table 2, Type A strategy). A long evolutionary process has led the studied species to develop high adaptability to the rainfall fluctuations and variability found along the studied aridity gradient (Harel et al. 2011; Tielbörger et al. 2014). An unambiguous mechanistic response to changes in rainfall was not found, despite important plant trait differences such as seed mass and germination fractions (Harel et al. 2011). The lack of response to changes in rainfall in most species present in the seed bank is also explained by the high spatial heterogeneity of soil moisture along the aridity gradient (i.e. geodiversity, see Yizhaq et al. 2017) combined with additional environmental constraints such as soil depth in shallow soils typical of the region, presence of biological soil crusts and granivory (Harel et al. 2011; Lebrija-Trejos et al. 2011; Sternberg et al. 2011). In the analysis considered here, species were grouped according to their germination fraction envelop along the rainfall gradient. This type of approach is rather different from those considered in other studies. For each species, the interval of rainfall was found for which the species can exist. As a result, it was found that most species can be present in drier conditions (annual rainfall less than 100 mm). This finding led to our understanding of the presence of a mechanistic germination response to rainfall variability that ensures the stability of the soil seed bank (Kigel 1995). We associate this mechanism to the high spatial and temporal heterogeneity typical of eastern Mediterranean ecosystems. High soil heterogeneity combined with climatic variability has a significant influence on the dynamics of the species composition and abundance of the soil seed bank (SSB) along gradients, as supported by other studies (Kutiel and Lavee 1999; Ulrich et al. 2014). Dominant plant species and germination strategies Dominant species along the rainfall gradient showed different germination strategies. Species with increasing germination fractions in the first year when rainfall increases, such as F. desertorum and P. longirostris, are species with relatively small seeds (0.03 and 0.15 mg, respectively; Sternberg unpublished data), typical of arid and semi-arid ecosystems in Israel. As noted by Harel et al. (2011), small-seeded species originating from arid sites showed higher germination fractions with increasing rainfall (2001–09). This type of strategy prevails under increasing rainfall unpredictability assuming that it buffers against rainfall uncertainty. In desert environments increasing rainfall leads to higher germination rates that are compatible with species from Type A. The additional species of Type A, M. annua, is a common Mediterranean species present in disturbed areas. Increasing germination fractions with increasing rainfall may boost its chances of establishment in this type of disturbed habitat as site pre-emption is a typical trait of this type of species. The largest group of species found in this study belongs to Type B, meaning decreasing germination fraction with increasing rainfall during the first germination year. These species (Table 2) are mostly typical of semi-arid ecosystems, relatively shallow soils and exhibit short to medium height (not taller than 40 cm). In good rainfall years, in this type of habitat, competition for light may become a limiting factor as the taller species outcompete the shorter ones. Under such conditions (i.e. higher resource availability) and lower rainfall unpredictability, disturbance agents such as fire and grazing may also become key drivers in Mediterranean ecosystems. The relatively lower germination fraction observed with increasing rainfall probably buffers against these common disturbance agents in these systems, while higher dormancy in arid ecosystems acts against climatic unpredictability. This hypothesis was confirmed by our data that demonstrated a relatively lower germination fraction in the first year and higher seed mass of the annual plant communities in more mesic sites (Harel et al. 2011). An important environmental constraint on germination strategies in Mediterranean ecosystems appears to be intense competition by neighbouring plants and disturbance agents. These ecosystems are characterized by higher seed densities and are consequently expected to be highly competitive environments (Turnbull et al. 2004). Species from Type R showed no clear pattern of germination fraction along the rainfall gradient. The species forming this group generally belong to Mediterranean ecosystems from diverse taxonomic groups and have relatively larger seeds than those of the other two types (Table 2). The random strategy (Type R) has important advantages in unpredictable conditions. All species in this group are capable of flowering and producing seeds even in dry years and even if plants are exposed to drought conditions. The lack of a clear pattern of germination fraction allows these species to maintain a relatively stable soil seed bank allowing persistence through time and across a wide range of rainfall conditions along the aridity gradient. The presence of alternative germination types along the rainfall gradient serves as a stabilizing mechanism of the plant communities developing along the gradient, and strengthens the notion of soil seed banks as a buffer against increasing environmental unpredictability arising from climate change (Kiss et al. 2018; Ooi 2012). At the arid site we expect that that year-to-year variability should be mostly related to stochastic variability inherent to this ecosystem type. This stochastic variability typical of desert ecosystems may confer to the seed bank additional resistance to climatic change (Venable 2007). Considering that the soil seed banks were germinated under optimal germination conditions, the different germination types and variability along the rainfall gradient could also be attributed to maternal effects. Although not directly tested in this study, we can assume that the growing conditions of the mother plant may also determine the optimal germination strategy for its progeny. Under more mesic conditions, mother plants could ‘support’ investment in compounds around the seed coat that increase seed dormancy capabilities (Wulff 2017). This strategy prevents competition among seedlings from the same mother plant, and may also buffer against potential exposure to risks such as fire and grazing typical of this type of ecosystem (Gutterman 1994, 2000, 2012). Considering the climate change scenarios for the eastern Mediterranean (Hochman et al. 2018), it is crucial to maintain the natural replenishment of the soil seed bank in order to secure relatively stable plant communities. Intensive and continuous seed depletion through grazing or fire could affect this balance with potential negative feedbacks on the stability of such ecosystems. Furthermore, seed viability loss under increasing temperatures predicted to the region could affect species composition as shown in other dryland ecosystems (Ooi et al. 2009). Adaptive management of these areas should consider the importance of maintaining soil seed banks to overcome increasing uncertainty induced by climate change (Kiss et al. 2018). Supplementary Material Supplementary material is available at Journal of Plant Ecology online. Figure S1: Species showing strategy A characterized by increasing germination fraction with increasing rainfall during the first germination year: a) Filago desertorum, b) Mercurialis annua, c) Picris longirostris. Figure S2: Species showing strategy B characterized by decreasing germination fraction with increasing rainfall during the first germination year: a) Anagallis arvensis, b) Catapodium rigidum, c) Crithopsis delileana, d) Linum corymbulosum, e) Lolium rigidum, f) Plantago afra, g) Stachys neurocalycina, h) Stipa capensis, j) Trifolium scabrum, k) Trisetaria macrochaeta, i) Valantia hispida. Figure S3: Species showing strategy R characterized by random germination fraction during the first year: a) Brachypodium distachyon, b) Convolvulus siculus, c) Scorpiurus muricatus, d) Urospermum picroides. Funding The present study was supported by the GLOWA Jordan River project and funded by the German Federal Ministry of Education and Research (BMBF), in collaboration with the Israeli Ministry of Science and Technology (MOST). Acknowledgements We are most grateful to Claus Holzapfel, Hadas Parag, Danny Harel, Irit Konsens and Danny Wallach for soil seed bank sampling and monitoring along the years of the study. We are also grateful to Carly Golodets for English editing and comments. Conflict of interest statement. The authors declare that they have no conflict of interest. Authors’ Contributions MS originally formulated the idea and gathered the data; AZ and EK developed the methodology and the mathematical model; AZ analysed the data; MS, EK and AZ wrote the manuscript. REFERENCES Alon M , Sternberg M ( 2019 ) Effects of extreme drought on primary production, species composition and species diversity of a Mediterranean annual plant community . J Veg Sci 30 : 1045 – 55 . Google Scholar Crossref Search ADS WorldCat Alpert P , Krichak SO, Shafir H, et al. ( 2008 ) Climatic trends to extremes employing regional modelling and statistical interpretation over the E. Mediterranean . Glob Planet Change 63 : 163 – 70 . Google Scholar Crossref Search ADS WorldCat Austin AT , Yahdjian L, Stark JM, et al. ( 2004 ) Water pulses and biogeochemical cycles in arid and semiarid ecosystems . Oecologia 141 : 221 – 35 . Google Scholar Crossref Search ADS PubMed WorldCat Beier C , Beierkuhnlein C, Wohlgemuth T, et al. ( 2012 ) Precipitation manipulation experiments—challenges and recommendations for the future . Ecol Lett 15 : 899 – 911 . Google Scholar Crossref Search ADS PubMed WorldCat Black E ( 2009 ) The impact of climate change on daily precipitation statistics in Jordan and Israel . Atmos Sci Lett 10 : 192 – 200 . Google Scholar Crossref Search ADS WorldCat Cohen D ( 1967 ) Optimizing reproduction in a randomly varying environment when a correlation may exist between the conditions at the time a choice has to be made & the subsequent outcome . J Theor Biol 16 : 1 – 14 . Google Scholar Crossref Search ADS PubMed WorldCat del Cacho M , Saura-Mas S, Estiarte M, et al. ( 2012 ) Effect of experimentally induced climate change on the seed bank of a Mediterranean shrubland . J Veg Sci 23 : 280 – 91 . Google Scholar Crossref Search ADS WorldCat Donohue K , de Casas RR, Burghardt L, et al. ( 2010 ) Germination, post germination adaptation, and species ecological ranges . Annu Rev Ecol Evol Syst 41 : 293 – 319 . Google Scholar Crossref Search ADS WorldCat Estrelles E , Biondi E, Galiè M, et al. ( 2015 ) Aridity level, rainfall pattern and soil features as key factors in germination strategies in salt-affected plant communities . J Arid Environ 117 : 1 – 9 . Google Scholar Crossref Search ADS WorldCat Evans JP ( 2009 ) 21st century climate change in the Middle East . Clim Change 92 : 417 – 32 . Google Scholar Crossref Search ADS WorldCat Fukami T , Wardle DA ( 2005 ) Long-term ecological dynamics: reciprocal insights from natural and anthropogenic gradients . Proc Biol Sci 272 : 2105 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat Galloway LF ( 2005 ) Maternal effects provide phenotypic adaptation to local environmental conditions . New Phytol 166 : 93 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat García-Ruiz JM , López-Moreno JI, Vicente-Serrano SM, et al. ( 2011 ) Mediterranean water resources in a global change scenario . Earth-Sci Rev 105 : 121 – 39 . Google Scholar Crossref Search ADS WorldCat Germain RM , Gilbert B ( 2014 ) Hidden responses to environmental variation: maternal effects reveal species niche dimensions . Ecol Lett 17 : 662 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Golodets C , Sternberg M, Kigel J, et al. . ( 2013 ) From desert to Mediterranean rangelands: will increasing drought and inter-annual rainfall variability affect herbaceous annual primary productivity? Clim Change 119 : 785 – 98 . Google Scholar Crossref Search ADS WorldCat Gremer JR , Venable DL ( 2014 ) Bet hedging in desert winter annual plants: optimal germination strategies in a variable environment . Ecol Lett 17 : 380 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Gutterman Y ( 1994 ) Strategies of seed dispersal and germination in plants inhabiting deserts . Bot Rev 60 : 373 – 425 . Google Scholar Crossref Search ADS WorldCat Gutterman Y ( 2000 ) Environmental factors and survival strategies of annual plant species in the Negev desert, Israel . Plant Species Biol 15 : 113 – 25 . Google Scholar Crossref Search ADS WorldCat Gutterman Y ( 2012 ) Survival Strategies of Annual Desert Plants. Switzerland : Springer Science & Business Media . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Harel D , Holzapfel C, Sternberg M ( 2011 ) Seed mass and dormancy of annual plant populations and communities decreases with aridity and rainfall predictability . Basic Appl Ecol 12 : 674 – 84 . Google Scholar Crossref Search ADS WorldCat Heisler-White JL , Blair JM, Kelly EF, et al. ( 2009 ) Contingent productivity responses to more extreme rainfall regimes across a grassland biome . Glob Change Biol 15 : 2894 – 904 . Google Scholar Crossref Search ADS WorldCat Heisler-White JL , Knapp AK, Kelly EF ( 2008 ) Increasing precipitation event size increases aboveground net primary productivity in a semi-arid grassland . Oecologia 158 : 129 – 40 . Google Scholar Crossref Search ADS PubMed WorldCat Hochman A , Harpaz T, Saaroni H, et al. ( 2018 ) Synoptic classification in 21st century CMIP5 predictions over the eastern Mediterranean with focus on cyclones . Int J Climatol 38 : 1476 – 83 . Google Scholar Crossref Search ADS WorldCat Huxman TE , Snyder KA, Tissue D, et al. ( 2004 ) Precipitation pulses and carbon fluxes in semiarid and arid ecosystems . Oecologia 141 : 254 – 68 . Google Scholar Crossref Search ADS PubMed WorldCat Kafle HK , Bruins HJ ( 2009 ) Climatic trends in Israel 1970–2002: warmer and increasing aridity inland . Clim Change 96 : 63 – 77 . Google Scholar Crossref Search ADS WorldCat Kigel J ( 1995 ) Seed germination in arid and semi-arid regions . In Seed Development and Germination . New York : Routledge , 645 – 99 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kiss R , Deák B, Török P, et al. ( 2018 ) Grassland seed bank and community resilience in a changing climate . Restor Ecol 26 : 141 – 50 . Google Scholar Crossref Search ADS WorldCat Knapp AK , Fay PA, Blair JM, et al. ( 2002 ) Rainfall variability, carbon cycling, and plant species diversity in a mesic grassland . Science 298 : 2202 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat Kutiel P , Lavee H ( 1999 ) Effect of slope aspect on soil and vegetation properties along an aridity transect . Isr J Plant Sci 47 : 169 – 78 . Google Scholar Crossref Search ADS WorldCat Lebrija-Trejos E , Lobato MC, Sternberg M ( 2011 ) Reproductive traits and seed dynamics at two environmentally contrasting annual plant communities: from fieldwork to theoretical expectations . Isr J Ecol Evol 57 : 73 – 90 . Google Scholar Crossref Search ADS WorldCat Mahlstein I , Daniel JS, Solomon S ( 2013 ) Pace of shifts in climate regions increases with global temperature . Nat Clim Change 3 : 739 – 43 . Google Scholar Crossref Search ADS WorldCat Matesanz S , Valladares F ( 2014 ) Ecological and evolutionary responses of Mediterranean plants to global change . Environ Exp Bot 103 : 53 – 67 . Google Scholar Crossref Search ADS WorldCat Meehl GA , Arblaster JM, Fasullo JT, et al. ( 2011 ) Model-based evidence of deep-ocean heat uptake during surface-temperature hiatus periods . Nat Clim Change 1 : 360 – 4 . Google Scholar Crossref Search ADS WorldCat Miranda JD , Armas C, Padilla FM, et al. ( 2011 ) Climatic change and rainfall patterns: effects on semi-arid plant communities of the Iberian Southeast . J Arid Environ 75 : 1302 – 9 . Google Scholar Crossref Search ADS WorldCat Ooi MKJ ( 2012 ) Seed bank persistence and climate change . Seed Sci Res 22 : 53 – 60 . Google Scholar Crossref Search ADS WorldCat Ooi MKJ , Auld TD, Denham AJ ( 2009 ) Climate change and bet-hedging: interactions between increased soil temperatures and seed bank persistence . Glob Change Biol 15 : 2375 – 86 . Google Scholar Crossref Search ADS WorldCat Pakeman RJ , Garnier E, Lavorel S, et al. ( 2008 ) Impact of abundance weighting on the response of seed traits to climate and land use . J Ecol 96 : 355 – 66 . Google Scholar Crossref Search ADS WorldCat Peralta AM , Sánchez AM, Luzuriaga AL, et al. ( 2016 ) Factors driving species assemblage in Mediterranean soil seed banks: from the large to the fine scale . Ann Bot 117 : 1221 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Pickett STA ( 1989 ) Space-for-time substitution as an alternative to long-term studies . In Likens GE (ed). Long-Term Studies in Ecology: Approaches and Alternatives . New York, NY : Springer , 110 – 35 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC R Core Team ( 2013 ) R: A Language and Environment for Statistical Computing . Vienna, Austria : R Foundation for Statistical Computing . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rysavy A , Seifan M, Sternberg M, et al. ( 2016 ) Neighbour effects on shrub seedling establishment override climate change impacts in a Mediterranean community . J Veg Sci 27 : 227 – 37 . Google Scholar Crossref Search ADS WorldCat Samuels R , Hochman A, Baharad A, et al. ( 2018 ) Evaluation and projection of extreme precipitation indices in the Eastern Mediterranean based on CMIP5 multi-model ensemble . Int J Climatol 38 : 2280 – 97 . Google Scholar Crossref Search ADS WorldCat Schwinning S , Sala OE ( 2004 ) Hierarchy of responses to resource pulses in arid and semi-arid ecosystems . Oecologia 141 : 211 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat Shohami D , Dayan U, Morin E ( 2011 ) Warming and drying of the eastern Mediterranean: additional evidence from trend analysis . J Geophys Res: Atmos 116 : 22 . Google Scholar Crossref Search ADS WorldCat Siewert W , Tielbörger K ( 2010 ) The dispersal dormancy trade-off in annual plants: putting model predictions to the test . Am Nat 176 : 590 – 600 . Google Scholar Crossref Search ADS WorldCat Sternberg M , Fleischer A, Holzapfel C, et al. . ( 2011 ) The Use and Misuse of Climatic Gradients for Evaluating Climate Impact on Dryland Ecosystems—An Example for the Solution of Conceptual Problems . Croatia : INTECH Open Access Publisher . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Tielbörger K , Bilton MC, Metz J, et al. ( 2014 ) Middle-Eastern plant communities tolerate 9 years of drought in a multi-site climate manipulation experiment . Nat Commun 5 : 5102 . Google Scholar Crossref Search ADS PubMed WorldCat Tuffa S , Treydte AC ( 2017 ) Modeling Boran cattle populations under climate change and varying carrying capacity . Ecol Model 352 : 113 – 27 . Google Scholar Crossref Search ADS WorldCat Turnbull LA , Coomes D, Hector A, et al. ( 2004 ) Seed mass and the competition/colonization trade-off: competitive interactions and spatial patterns in a guild of annual plants . J Ecol 92 : 97 – 109 . Google Scholar Crossref Search ADS WorldCat Ulrich W , Soliveres S, Maestre FT, et al. ( 2014 ) Climate and soil attributes determine plant species turnover in global drylands . J Biogeogr 41 : 2307 – 19 . Google Scholar Crossref Search ADS PubMed WorldCat Venable DL ( 2007 ) Bet hedging in a guild of desert annuals . Ecology 88 : 1086 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat Wulff RD ( 2017 ) Environmental maternal effects on seed quality and germination . In Seed Development and Germination . New York : Routledge , 491 – 505 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Yahdjian L , Sala OE ( 2002 ) A rainout shelter design for intercepting different amounts of rainfall . Oecologia , 133 , 95 – 101 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Yizhaq H , Stavi I, Shachak M, et al. ( 2017 ) Geodiversity increases ecosystem durability to prolonged droughts . Ecol Complex 31 : 96 – 103 . Google Scholar Crossref Search ADS WorldCat Ziv B , Saaroni H, Pargament R, et al. ( 2014 ) Trends in rainfall regime over Israel, 1975–2010, and their relationship to large-scale variability . Reg Environ Change 14 : 1751 – 64 . Google Scholar Crossref Search ADS WorldCat © The Author(s) 2020. 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: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Germination strategies under climate change scenarios along an aridity gradient
JF - Journal of Plant Ecology
DO - 10.1093/jpe/rtaa035
DA - 2020-08-01
UR - https://www.deepdyve.com/lp/oxford-university-press/germination-strategies-under-climate-change-scenarios-along-an-aridity-olfPv90d0y
SP - 470
EP - 477
VL - 13
IS - 4
DP - DeepDyve
ER -