The study was conducted in Raudat Altinhat (26.0807° N, 46.5072° E) in the northeast of Riyadh region, central Saudi Arabia (Figure 1). The area is belonging to the King Abdulaziz Royal Reserve. The site is a low-lying plain covering approximately 32 km², bordered by the sand dunes of Al-Dahna to the east. Due to numerous wadi/valley canals, including Wadi Al-Shwaki, Wadi Al-Tayri, Wadi Al-Atiq, and Wadi Al-Wadi, the Raudat has superior soil and vegetation compared to neighboring areas. The vegetation is primarily composed of Vachellia and Ziziphus trees, along with various seasonal wildflowers, which flourish particularly during springtime (Saudipedia, 2024). The average mean temperature and average annual rainfall are 25.22°C and 119.82 mm, respectively, for the years 1991–2020 (World Bank, 2024). The soil in the selected site is primarily composed of loamy sand and sandy soil (Saudipedia, 2024). The selection of sampling sites within the study area was conducted to ensure a comprehensive representation of vegetation characteristics, including diversity and intensity. Additionally, the topography features of the entire Raudhat ecosystem were considered to facilitate a systematic assessment of the ecosystem’s spatial heterogeneity. These features encompass water streams, water accumulation areas, land aspects, depressions, and sandy plains.

Fourteen plots were selected to cover the whole area of the Raudhat Altinhat. Within each plot, three quadrats were selected for the vegetation analysis. The minimal area for each quadrat was determined to be 10 m × 10 m in areas for herbaceous vegetation, and 20 m × 20 m in areas with shrub vegetation. In total, 42 quadrats were performed. The species in each quadrat were identified and named in accordance with the protocols of Collenette (1999), Chaudhary (1999), Chaudhary (2000), and Chaudhary (2001), and were subsequently reconfirmed by the Royal Botanical Garden, Kew (POWO, 2023). These species were observed at a variety of sites during the summer-fall and winter-spring seasons of 2023 and 2024. Plant density was calculated in accordance with Bonham (2013), and plant cover was determined using the Braun-Blanquet scale (Mueller-Dombois and Ellenberg, 1974). Before calculating the species importance value index (IVI) to ascertain the species dominance, the density, cover, and frequency data were converted to their respective relative values using the following equations: Relative density of species  A = 100 × A b s o u l t e   d e n s i t y   o f   s p e c i e s   A ∑ a b s o l u t e   d e n s i t i e s   o f   a l l   s p e c i e s Relative cover of species  A = 100 × A b s o u l t e   c o v e r   o f   s p e c i e s   A ∑ a b s o l u t e   c o v e r   o f   a l l   s p e c i e s Relative frequency of species  A = 100 × N u m b e r   o f   p l o t s   h a v e   s p e c i e s   A T o t a l   n u m b e r   o f   a l l   p l o t s Importance value index = R e l a t i v e   d e s n i t y + R e l a t i v e   c o v e r + R e l a t i v e   f r e q u e n c y

Various published materials and online resources, including https://flora.org.il/en/plants/, http://www.theplantlist.org/1/, and https://powo.science.kew.org/, were used to examine the chorotype of each species, and the species life forms were identified according to Raunkiaer (1937). The life forms and chorotypes of all identified plants were examined in accordance with Raunkiaer (1937). The species richness (Simpson index) and species evenness (Shannon evenness) of all sites were ascertained using the subsequent equations: S i m p s o n   i n d e x   S = ∑ i n i × n i − 1 N × N − 1 S h a n n o n − W i e n e r   i n d e x   H = ∑ i = 1 s P i ⁡ ln P i S h a n n o n − E v e n n e s s   i n d e x   E = H ⁡ ln s where Pi = ni/N = proportional abundance of species, “i” in a habitat made up of “S” is the number of species, “ni” is the number of sites contai006Eing species i, and N = S ni. The species richness and species evenness were calculated according to the following equations: S p e c i e s   R i c h n e s s = S − 1 L o g N S p e c i e s   E v e n n e s s = H ′ L o g S

Plastic bags were used to collect a central soil sample with a volume of 4,000 cm³ (20 cm width × 20 cm length × 10 cm depth) in each quadrat where the vegetation survey was conducted. A total of 42 soil seed bank samples were collected, labeled, and transferred to the Range Science Lab, College of Food Science and Agriculture at King Saud University in Riyadh, Saudi Arabia, for further preparation and analysis. The seedling emergence techniques are employed to evaluate the seed density of the seed bank (Baker, 1989). Approximately 60% of the previous studies have employed direct greenhouse germination to evaluate the densities and species composition of soil seed banks (Padonou et al., 2022). This method is a common method for studying the soil seed bank in the desert ecosystem (Jiang et al., 2013; Loydi and Collins, 2021; Luo et al., 2021; Gomaa et al., 2023; Al-Huqail et al., 2025). The soil samples were first passed through a 4 mm mesh sieve to remove any detritus, including plant fragments, roots, and coarse stones, before being stored at 2 °C for a month to break seed dormancy (Niu et al., 2022). This was done prior to the initiation of the germination trial. A manual examination was conducted on the excluded material to detect the presence of seeds. The rectangular perforated plastic trays (45 × 45 × 7 cm) were lined with sterilised sand substrate that was less than 2 cm thick at the bottom. The sieved soil samples were spread to a thickness of approximately 2 cm. The trays were positioned on benches in a greenhouse that was adjusted to maintain a temperature of 25 °C ± 2 during the trial period. Water was introduced daily to ensure that the soil remained moist. All germinated seedling species were identified and their numbers were recorded. Subsequently, they were removed at weekly intervals. The germination trials were terminated when no new seedlings emerged. The plants were identified in accordance with the methods of Collenette (1999), Chaudhary (1999), and Chaudhary (2001). Additionally, the relative density of each species was determined using the following equation, and the number of species in each tray was counted: R e l a t i v e   d e n s i t y   % = 100 × N u m b e r   o f   i n d i v i d u a l   o f   o n e   s p e c i e s T o t a l   n u m b e r   o f   a l l   i n d i v i d u a l s   o f   a l l   s p e c i e s

Soil samples were collected from each quadrat under investigation at a depth of 5–25 cm in polythene bags for soil chemical and physical analyses. The soil samples that were collected were labeled and transported to the laboratory for additional processing and preparation. The soil samples were immediately distributed over paper sheets to air-dry at room temperature (25°C ± 3°C) upon arrival in the laboratory. Subsequently, they were sieved through a 2 mm sieve to remove any extraneous materials, including gravel and plant elements. Mualem’s approach was employed to evaluate soil porosity (Mualem, 1976), while Bouyoucos’s method was employed to determine soil texture (Bouyoucos, 1962). The method of Keen and Raczkowski (1921) was performed to ascertain the water holding capacity (WHC). The graduated cylinder method (Blake and Hartge, 1986) was used to ascertain the particle density of the soil. The pH and electrical conductivity (EC) were estimated using soil-water extracts (1:5) (Rowell, 1994). The calcimeter method, as outlined by Page (1982), was employed to ascertain the calcium carbonate content. The total nitrogen (N) was determined using the Kjeldahl method, as outlined by Bremner and Mulvaney (1982), and the total phosphorus (P) was determined spectrophotometrically in accordance with Nelson and Sommers (Nelson and Sommers, 1982). The soil organic matter content was determined by the loss on ignition method at 550°C according to (Jones, 2001). The exchangeable sodium (Na) and potassium (K) were determined using the flame photometer method (PHF 80 B Biologie Spectrophotometer), as described by Rhoades (1982). The atomic absorption spectrometer (A Perkin-Elmer, Model 2,380, United States) was employed to estimate the calcium (Ca) and magnesium (Mg) as described by Allen et al. (1974). The sulphuric acid method was used to measure the carbonate (CO₃) and bicarbonate (HCO₃) in the soil samples through titration (Page, 1982). Finally, AgNO₃ was used to determine the chloride (Cl) content through a titration technique (Jackson, 1962).

Hierarchical cluster analysis was conducted using Community Analysis Package, version 1.2 (Pisces Conservation Ltd. IRC House, Pennington, Lymington, UK), where the importance value index (IVI) of all identified plant species dataset was subjected to an agglomerate hierarchical clustering via Ward’s linkage to identify the various plant communities within each season. The datasets of the IVI and the soil variables of all examined sites were prepared and subjected to ordination analysis using canonical correspondence analysis (CCA) in order to evaluate the relationships between the soil variables and vegetation composition. The MVSP software program, version 3.22 (Kovach Computing Services, United Kingdom), was employed to generate the CCA. A one-way ANOVA was conducted using CoStat software (v. 6.311, CoHort Software, CA, United States) on the raw data of all soil parameters of the plant communities of the winter-spring and summer-fall seasons to compare variances across the identified plant communities and their soil characteristics. Duncan’s multiple range test was then performed. Correlation analysis was conducted on a dataset that comprises the density percentage of all identified plant species in the soil seed bank composition and the aboveground vegetation via the JMP^® Pro 16.0.0 software program. In addition, the JMP^® Pro was used for hierarchical cluster analysis, the constellation plot, and the heatmap between the aboveground vegetation and soil seed bank composition were generated to demonstrate a distinct and representative correlation among the investigated sites.

The vegetation survey of the different sites of Raudhat Altinhat showed the presence of 86 plant species (Table 1). These plant species were 38 (44.2%) perennial plants and 48 (55.8%) annuals. These plant species belong to 27 families, where the major families are Asteraceae, Poaceae, Boraginaceae, and Fabaceae, which attained 47.7% of the total identified species (Supplementary Figure S1A). The plant species of these families are usually reported in such arid regions of Saudi Arabia (Assaeed and Al-Doss, 2002; Abd-ElGawad et al., 2021; Al-Huqail et al., 2025), and this could be attributed to the adaptability of the species of these plants belonging to these families for such arid environmental conditions (Waheed and Arshad, 2024). Among the life forms of the identified plant species, therophytes/annuals were the most represented (47.7%), followed by chamaephytes (32.6%), while the geophyte life form was the lowest represented (Supplementary Figure S1B). The predominance of annual species can be ascribed to stressful environments, where the seeds of these species germinate during the favorable/rainy season and complete their life cycle before the start of the arid season (Gutterman, 2000). In addition, these annual plant species developed various survival strategies to tolerate the harsh conditions, such as phenotypic plasticity and powerful flowering and seed dispersal tactics (Venable et al., 2008; Stotz et al., 2021).

The chorological analysis of the identified plant species showed that 55.8% are mono-regional (found in one biogeographical region), while 40.7% of the recorded species are categorized as bi-regional and 3.5% as pluri-regional species. The Saharo-Arabian element was the most representative, where it was determined with 37.2% as mono-regional and 29.1% as bi-regional. In this context, the Irano-Turanian chorotype was also found in a high percentage (30.2%), which is 3.5% mono-regional, 23.3% bi-regional, and 3.5 pluri-regional (Supplementary Figure S1C). This structure of the chorotype in the present study is in agreement with other studies in similar environmental conditions in Saudi Arabia (Abd-ElGawad et al., 2021; Elkordy et al., 2022; Al-Huqail et al., 2025). Among the studied species, 17.4% of species have been reported as spiny plant species that showed adaptability to desert or drought conditions (Batanouny, 2000). Also, six plant species were reported as leaf or stem succulent plants.

A dataset of the relative densities of all identified plant species of the aboveground vegetation as well as that determined in the soil seed banks is prepared and subjected to the Pearson correlation coefficient (r) to show the correlation between the aboveground (current) vegetation and the soil seed bank of various studied plots (Figure 5). As an average of all studied plots, a low negative correlation was observed between the aboveground vegetation and soil seed bank composition either during the winter-spring season (r = −0.060) or the summer-fall season (r = −0.036).

The correlation analysis between the aboveground vegetation and soil seed bank during the winter-spring season showed a low negative correlation for all studied plots, except for plot 3, where it showed a low positive correlation (r = 0.141). However, during the summer-fall season, all plots revealed a negative correlation between the aboveground vegetation and the soil seed banks, except for plot 5 and plot 7, where they showed a positive correlation (Figure 5; Supplementary Table S3). The negative correlation between the aboveground vegetation and soil seed bank could be ascribed to the degradation of the sites due to overgrazing, camping, and logging. The Raudhat Altinhat site was not protected before, but after the establishment of the King Abdulaziz Royal Reserve in 2018, the area become protected and it is expected that the vegetation of the site will be recovered over time. The difference between the aboveground vegetation and the composition of the soil seed bank increased with decreasing disturbance intensity (Dölle and Schmidt, 2009).

In a similar study, Savadogo et al. (2016) reported a low correlation between the aboveground vegetation and soil seed bank species in the degraded areas of the National Park transboundary biosphere reserve, West Africa. In this context, the study of the soil seed bank of Raudhat Alkhafs, northeast of Riyadh City, Saudi Arabia, also showed low correlation in the northern part between the vegetation and soil seed bank that was attributed to the overgrazing and anthropogenic activities (Al-Huqail et al., 2025). The ecosystem of Raudhats in Saudi Arabia was accessible to various human activities like overgrazing, camping, cutting of wood for firing, and harvesting medicinal plants (Al-Qarawi, 2011). These anthropogenic activities, coupled with the environmental stresses and harsh environmental conditions, disrupt the relationships between the aboveground vegetation and the soil seed bank in this ecosystem (Al-Qarawi, 2011; Al-Huqail et al., 2025). Various studies have reported that anthropogenic activity leading to disruption of the soil seed bank as well as the vegetation, and in consequence, negatively impacts their relationships and harms the vitality of the natural ecosystems (Amiaud and Touzard, 2004; Amrein et al., 2005; dos Santos et al., 2018; Shinoda and Akasaka, 2019; Luo et al., 2023). Among the various anthropogenic activities, the overgrazing in Saudi Arabia is the most common (Mirreh and Al Diran, 2014; Al-Rowaily et al., 2015), and this overgrazing can alter the soil seed bank by decreasing the production of seeds and affecting the seed bank composition due to the selection of certain and more preferable plants by animals (Tessema et al., 2012; van Langevelde et al., 2016; Gonzalez and Ghermandi, 2021).

The species diversity indexes showed that aboveground vegetation during the winter-spring season attained higher species richness (25.1) and evenness (1.9), compared to the soil seed bank, which showed species richness of 5.7 and species evenness of 1.1 (Figure 6). This observed variation between the aboveground vegetation and SSB during the winter season could be attributed to the rainfall, where most annual species seeds germinated and flourished after the rainy season, affecting the SSB pool (Gomaa, 2014; Shi et al., 2022; Al-Huqail et al., 2025).

In this context, the vegetation in the desert habitat usually produces more dormant seeds for most taxa compared to less stressful habitats (Shi et al., 2022), which could explain the low species richness and evenness of the soil seed bank in the present study. This dormancy is considered a good strategy by the desert plant to sustain a long-term seed bank within unpredictable environments, like deserts (Kleemann and Gill, 2013). On the other hand, during the summer-fall season, the species richness of the aboveground vegetation and soil seed bank was consistent, while the species evenness was higher for the soil seed bank (1.8) compared to the aboveground vegetation (1.3). This clear decline in the species richness and evenness during the summer-fall season is attributed to the harsh and dry conditions, where perennial species some few summer annuals that adapted to the drought conditions, can survive (Abd-ElGawad et al., 2021). Also, the considerable species richness and evenness of the SSB during summer could be ascribed to the accumulation of seeds in the soil after the rainy season (Gomaa et al., 2023). Our results showed that the Raudhat Altinhat site needed rehabilitation and strict protection against all unmanaged activities, such as overgrazing and logging. Based on our survey, the Z. nummularia community showed a high degradation level as well as very low regeneration, and no seed germination in the soil seed bank. Therefore, this issue needs further studies to determine specifically the causes for the degradation of Z. nummularia community and the best way to rehabilitate and recruit these very important shrubs on the site. Also, it is worth mentioning that in our study, the seedling emergence method was used, which provides an underestimate for the total number of seeds in the soil due to seed dormancy. Therefore, a further study to compare using seedling emergence method with othermethods such as seed extraction or using chemical treatments for estimating the soil seed bank in desert ecosystem with comparision to the aboveground vegetation.