Discussion
Our results showed that among 22 plant families in SSB, the highest number of species belonged to Asteraceae. Previous studies showed that these plants were also widely present in the SSB (e.g. Gomaa et al., 2012). One of the possible reasons for the increase of these plants in SSB is the abundant seed production and morphological characteristics of the seeds. Species of Asteraceae with small seed size and wing shape, light and easy dispersal provide conditions for the presence of their seeds in the SSB (Harper, 1977; Hong et al., 2012). Forbs were the most abundant plants in the SSB composition. Our results showed that number of forb species was higher (33, 27, 24 and 17 species beneath E. stellata , D. mezareum , A. scoparia and control, respectively) than grasses (6, 7, 5 and 5 species beneath E. stellata , D. mezareum , A. scoparia and control, respectively) in the study area. In accordance with the results of our study, Bertiller et al. (2011), Parlak et al. (2011) and Tessema et al. (2012) reported that forbs had the highest number of species in SSB. Higher number of forbs in the AGV might be a reason for increasing the seeds of these species in the SSB. In contrast, woody plant species (trees and shrubs) were scarcely found in the greenhouse. Although, Teketay and Granström (1997) and Chaideftou et al. (2009) attributed the lack of woody species in the SSB to the lack of mature species in the AGV, in our study, this cannot be the reason because woody species in the AGV were frequent. Many factors are involved in reducing the density and richness of woody species in the SSB in an area. These could include the larger size of the seeds, the higher amount of predations and seed dormancy (Esmailzadeh et al., 2011). Some studies have shown that breaking seed dormancy of woody plants requires special conditions and if these conditions are not provided, these plants will eventually be removed from the SSB (Chaideftou et al., 2009).
In general, this study showed that the SSB density and species richness and diversity under the shrubs was higher than control and this differentiation was more pronounced for forbs. Previous studies (e.g. Marone et al., 2004) showed that the seed density of forbs were often higher under woody plants and positively correlated with the cover of woody vegetation, whereas the seed density of grasses were less associated by woody vegetation. Our results are consistent with some previous studies (e.g. Erfanzadeh et al., 2014) and disagree with others (e.g. Mndela et al., 2020). Positive effects of shrubs on SSB are exerted through direct and indirect ways. They increase buried seeds in soil by directly trapping seeds or by indirect mechanisms through an intermediary animal or plant species (Bullock and Moy, 2004; Giladi et al., 2013). Shrubs significantly influence the movement wind or water around their canopy (Hoffman et al., 2013) and thus can trap seeds or act as barrier for movement (Giladi et al., 2013). Shrubs can indirectly facilitate seed arrival by acting as a perching site for seed-carrying birds (Debussche and Isenmann, 1994) or as cache for granivorous rodents (Beck and van der Wall, 2010) and ants (Vergara-Torres et al., 2018). Additionally, shrubs can indirectly increase SSB by facilitating the plants that are able to increase seed production or viability and vigority of produced seeds (Pugnaire and Lázaro, 2000). Shrubs provide a suitable conditions for growing, flowering and seeding of herbaceous plants under their canopies through modifying the physical and chemical properties of soil with litter and root exudation, improving soil micro-relief, decreasing direct sunlight, increasing soil moisture, protecting the surface soil from erosion and adding organic matter into the soil (Ruiz et al., 2008; Barness et al., 2009; Olvera-Carrillo et al., 2009; Li et al., 2011; Sylvain and Wall, 2011; García-Sánchez et al., 2012).
However, different effects of different shrubs on SSB characteristics were observed in increasing SSB density under the canopy of A. scoparia and species richness and diversity under the canopy ofE. stellata. Previous studies showed that the size of a shrub could impact the arrival of seeds (Pugnaire and Lázaro, 2000) because larger shrubs can provide greater facilitative effects. Larger shrubs can intercept more solar radiation (Maestre and Cortina, 2005), have higher soil nutrients (Zhang et al., 2015), or lower evapotranspiration (Kidron and Gutschick, 2013) creating a favourable microclimate for seed production, particularly by annuals (Filazzola et al., 2019). As a result, comparing to the other shrubs, taller and larger canopy inA. scoparia might increase SSB density through higher seed production by plants. The seeds of some annuals were found at strong frequent under A. scoparia , i.e. Bromus tectorum ,Galium aparine and Veronica anagalis . However, procumbent canopy in E. stellata might increase species diversity and richness of SSB. It might be that attached crown cover to the ground inE. stellata physically obstruct more seeds and enhance species diversity and richness in SSB. Our results showed that the seeds of many species were found under E. stellata , while they were absent in control or under other shrubs, e.g. Polygonum dumosum , Poa sinaica , Tragopogon jezdianus and Dianthus orientalis ,Tianthus crinitus . Briefly, indirect effect of A. scopariaon seed production of few plant species (some annuals) and direct effect of E. stellata on seed trapping of many species resulted these significant differences of SSB density, species richness and diversity between shrubs.
Similarity between the seed bank and the AGV was generally low in three shrubs and control. The low similarity between the AGV and the SSB in our and other studies is usually due to the fact that some species were present in the vegetation, while they were absent from the seed bank, and vice versa (e.g. Valkóa et al., 2014; Erfanzadeh et al., 2016). However, the similarity between the AGV and the SSB was lowest in the control. In this area, many species, such as Acantholimon scorpius and Ebenus stellata , were absent from the seed bank while they were present in the AGV. Most of these species were perennial, and these, especially shrubby ones, are well-known for their transient seed bank (Thompson et al., 1997). Moreover some annuals such as Bromus tectorum and Galium aparine were found in the SSB in control and under three shrubs. These species were present in the AGV under the shrubs while absent in the AGV of control. At the sampling time, some annuals in the AGV might be grazed or dried and ended their phenological stages in the control due to higher availability to grazers or solar radiation, temperature and wind speed comparing with under the canopies of shrubs.