Biological change of western Saudi Arabia: Alien plants diversity and their relationship with edaphic variables

2.1 Terminology

In the current study, the recorded alien plant species were defined and classified following Kowarik and Pyšek (2012), Richardson et al. (2000), and Vilà et al. (1999) into three categories as follows: 1) casual species indicating those plant species that may flourish or reproduce occasionally in an area, but do not form self-replacing populations, and rely on repeated introductions for their persistence, 2) naturalized species indicating plant species that reproduce consistently and sustain populations over many life cycles without direct intervention by humans; they often recruit offspring freely, usually close to adult plants, and do not necessarily invade natural, semi-natural or human-made ecosystems, and 3) invasive species indicating naturalized plant species that produce productive offspring, often in large numbers, at considerable distances from parent plants, and thus have the potential to spread over a considerable area.

2.2 Study area

The current study was performed in the western region of Saudi Arabia and included the Tihama plain and Sarawat Mountains with different habitats (i.e., mountains, protected areas, wadis, farmlands, ruderal areas, and coastal areas) along the coastal region of the Red Sea from Rabigh in the north (N24°38′38″ E39°93′18″) to Jeddah in the south (N24°26′59.5″ E39°14′41.1″) and from Al Figrah mountains in the north (N24°25′32″ E39°23′36″) at elevation of 2100 m above sea level (a.s.l.) to Raidah Sanctuary in the south (N18°20′95″ E42°40′96″) at elevation of 1900 m a.s.l. (Fig. 1). The western region of Saudi Arabia is characterized by sedimentary rocks with the presence of several sand dunes and salt marshes sloping from the mountains towards the Red Sea. The study was performed over a whole year starting from winter 2018 to fall 2019. A total of 58 stands in the different habitats along the western region of Saudi Arabia were examined.

Close

Fig. 1

A map of Saudi Arabia (left bottom corner) showing the study area inside the black box with a zoomed-in map showing the distribution of the 58 stands (red dots) in the western region of Saudi Arabia.

Export to PPT

2.3 Identification of alien plant species

A list of all the plants found in the study area was prepared. For the purpose of the current study, the alien species was defined according to Richardson et al. (2000) as “the alien species (exotic plants, non-native plants, non-indigenous plants) are plant taxa in a given area whose presence is due to intentional or accidental introduction as a result of human activity.” Identification of plants and their chorotypes and life forms, as well as the intentional and accidental introduction, were determined according to the flora references of Saudi Arabia (Collenette, 1985, 1998, 1999, Chaudhary, 1999-2001). The identification of alien species was revised in the King Saud Herbarium (KSUP). The habitats of each alien species were recorded during the different field trips.

2.4 Density and cover estimation

Quadrat method was applied to estimate the density of each recorded plant in the different habitats. In each selected sampling stand, 5 quadrats were applied and the number of individuals per each plant species in each quadrat was counted. The quadrat area varied according to the plants' sizes with a minimum of 5 × 5 m and a maximum of 50 × 50 m. The average quadrat area was 25 × 25 m. The density of each plant species was calculated following Ellenberg and Mueller-Dombois (1974). The density values were converted to number of individuals per 100 m² and then this value was converted to number of individuals per hectare.

Pearson’s correlation analysis between the densities of native and alien plant species recorded in all the studied quadrats was performed using IBM SPSS Statistics 20 (IBM Corp., Armonk, NY).

The plant cover and relative cover were estimated according to Kent (2012).

2.5 Diversity indices

Several diversity indices were calculated (Pielou, 1975, Magurran, 1988, 2003). Briefly, α-diversity (species richness) was calculated as the average number of species per stand. Shannon-Wiener index was used to calculate species relative evenness as follows: $$\widehat{H}={\sum }_{i=1}^s{p}_i\log p_i$$

Species evenness was calculated using the following equation: $$J^{\prime }=\frac{\widehat{H}}{{\widehat{H}}_{\mathit{max}}}$$

where $\widehat{H}$ is the number derived from the Shannon diversity index and ${\widehat{H}}_{\mathit{max}}$ is the maximum possible value of $\widehat{H}$ (if every species was equally likely).

Simpson index was used to measure the relative concentration of dominance as follows: $$C={\sum }_{i=1}^s{p_i}^2$$

where (s) is the number of species and (p_i) is the relative cover value of the i-^th species.

Seeds of the identified plant species were collected. Moreover, the dispersal types for most of the collected diaspores from the recorded species were examined using the scheme proposed by Dansereau and Lems (1957).

2.6 Soil analysis

Soil samples were collected using a soil corer to the depth of 20 cm with tri-replicates for each stand. After air-drying of the soil samples, soil solution (1:5 w:v) was prepared to measure pH and electrical conductivity (EC). Soil texture was measured using Bouyoucos hydrometer (Allen, 1989). Organic matter (OM) content was measured using loss-on-ignition method (Allen, 1989). Atomic absorption spectroscopy was used to determine N and P contents in the soil. Contents of other nutrient elements (K, Ca, Mg, Na, F, and Mn) in the soil were analyzed using Inductively Coupled Plasma Mass Spectrometry (NexION^TM 300D ICP-MS, PerkinElmer, Inc., MA, USA) method (Allen, 1989).

2.7 Data analysis

The data matrix of species cover data was created with 58 stands, species cover values for the identified alien species and soil variables. Canonical correspondence analysis (CCA) was applied to examine the correlation between the soil variables and distribution of the identified alien species in the study area using the CANOCO software (Ter Braak and Smilauer, 2002). One-way ANOVA was performed to examine the variance in all diversity indices and soil parameters across different studied habitats via IBM SPSS Statistics 20 (IBM Corp., Armonk, NY). Means were compared using Duncan’s multiple range test (p < 0.05).

3.1 Floristic diversity and ecological analysis of the alien plant species

A list of all the 146 plant species (either alien or native) recorded in the study area along with their families and life forms is provided (Table S1). The results of the current study revealed the presence of 52 alien plant species (4 casual, 45 naturalized, and 3 invasive) in the western region of Saudi Arabia belonging to 25 different plant families with Asteraceae being the most represented family with 9 species followed by Gramineae (8 species). Several families were represented by only 1 species while Aizoaceae, Chenopodiaceae, Malvaceae, and Solanaceae were represented by 3 species each. Amaranthaceae, Lamiaceae, Leguminosae, and Zygophyllaceae were represented by 2 species each (Table 1). Interestingly, the identified casual and invasive species (except P. juliflora) belonged to families that were represented by one species only.

The dominant life forms observed in the identified alien species were annual herbs (11 species) followed by subshrubs and shrubs (8 and 7 species, respectively). The three identified invasive species belonged to annual herbs, shrubs, and trees; while the casual identified species were 2 subshrubs, a shrub, and a tree.

The majority of the recorded alien species in the western region of Saudi Arabia belonged to the American and the Saharo-Arabian regions (8 species each). The Somali Masai region and the biregional Saharo-Arabian – Somali Masai are represented by 5 species each, while the tropical region was represented by 4 species. The Tropical African region and the biregional Saharo-Arabian – Irano Turanian were represented by 3 species each. It is worth noting that all the identified invasive alien species and two (half) of the identified casual species belonged to the American region.

Pearson’s correlation analysis revealed a negative correlation (r² = -0.586, p = 0.001) between the density of native and alien species recorded in the study area. Moreover, the ecological analysis of the identified alien species revealed that 9 species showed high frequencies along the study area namely A. mexicana, Conocarpus erectus L., Datura innoxia Mill., Heliotropium curassavicum L., Lantana camara L., Nicotiana glauca Graham, O. dillenii, P. juliflora, and Xanthium spinosum L. The other 5 species (Amaranthus hybridus L., Cupressus sempervirens L., Rosmarinus officinalis Spenn., Sesuvium portulacastrum (L.) L., and T. portulacastrum) showed lower frequencies. Therefore, further analyses regarding the distribution of the high-frequent species along different habitats were performed.

3.2 Density of alien species in different habitats

In mountain habitats, O. dillenii showed the highest density (1722.2 individual hectare^-1) followed by A. mexicana (955.6 individual hectare^-1) and D. innoxia (11.1 individual hectare^-1; Fig. 2). Similarly, O. dillenii dominated the protected area habitats with 1516.7 individual hectare^-1 followed by N. glauca (483.3 individual hectare^-1). N. glauca plants were not observed in any other habitat. Moreover, A. mexicana and C. erectus were found in protected area habitats but with lower densities (66.7 and 16.7 individual hectare^-1). In wadi (valley) habitats, P. juliflora showed the highest frequency (590.8 individual hectare^-1) followed by O. dillenii (227.7 individual hectare^-1). C. erectus and H. curassavicum were found in wadi habitats with 12.31 individual hectare^-1 (Fig. 2), while A. mexicana and L. camara showed 6.2 individual hectare^-1. D. innoxia had the lowest frequency in wadi habitats (6.12 individual hectare^-1). The highest number of unique alien species was observed in the farmland habitats as 8 out of the 9 high-frequent species were observed in farmlands. Similar to mountain and protected area habitats, O. dillenii showed the highest frequency (800 individual hectare^-1) in the farmlands (Fig. 2). P. juliflora and C. erectus showed 181.82 and 121.2 individual hectare^-1, respectively. The lowest frequency in farmlands was observed for H. curassavicum (12.12 individual hectare^-1). Interestingly, X. spinosum plant was observed in farmlands with a frequency of 36.4 individual hectare^-1. This plant was not recorded in any other habitat (Fig. 2). In ruderal habitats, only P. juliflora was recorded with a frequency of 1600 individual hectare^-1. Furthermore, P. juliflora and C. erectus dominated the coastal area habitats with 1200 and 66.7 individual hectare^-1, respectively (Fig. 2).

Close

Fig. 2

Density of each identified alien plant species in different habitats measured as average number of individuals per hectare of land area.

Export to PPT

3.3 Distribution of high-frequent alien species in different habitats

A. mexicana plants were mainly observed in mountain habitats (955.6 individual hectare^-1). This species was also recorded in wadi, protected area, and farmland habitats (Fig. 3). The highest frequency of C. erectus was recorded in farmland habitats (181.82 individual hectare^-1) followed by coastal area, protected area, and wadi habitats. Nevertheless, C. erectus was not recorded in mountain or ruderal habitats. D. innoxia was recorded in 3 different habitats (mountains, wadis, and farmlands). D. innoxia showed lower frequencies as compared to other alien species in these habitats. H. curassavicum and L. camara were found in wadis and farmlands. N. glauca was found in protected areas only with a frequency of 483.3 individual hectare^-1. O. dillenii was recorded in 4 different habitats (mountains, protected areas, wadis, and farmlands). Interestingly, O. dillenii showed high frequencies in all habitats where it was found. P. juliflora was also found in 4 different habitats (wadis, farmlands, ruderal, and coastal areas). The highest frequency of P. juliflora was found in ruderal habitats where it was the only recorded alien plant species. X. spinosum was recorded only in farmlands with relatively low frequency (36.4 individual hectare^-1).

Close

Fig. 3

The most frequently identified alien plant species in different habitats measured in individuals per hectare.

Export to PPT

3.4 Variation in diversity indices and soil variables across different habitats

The results of the current study revealed that all the diversity indices (i.e., number of species, species richness, species evenness, Shannon index, and Simpson index) except species cover had significant variation across the studied habitats (Table 2). The highest number of species, species richness, and species evenness (11.56, 2.76, and 0.81, respectively) were recorded in farmland habitats. Furthermore, farmlands showed the highest Shannon index (0.85) while coastal areas showed the highest Simpson index (0.57). Unsurprisingly, the soil of farmlands was characterized by high OM contents. Contents of N and P didn’t show variation among the studied habitats. Nevertheless, K, Mg, Mn, and Na showed significant variation across different habitats with coastal areas having the highest Mg and K values and the second highest Na values after ruderal areas.

3.5 Relationships between alien plant species, diversity indices, and soil variables

The relationships between alien plants distribution along the ordination axes and the soil variables were examined via CCA (Fig. 4). The results revealed that the distribution of A. hybridus, H. curassavicum, and O. dillenii was correlated with soil pH and OM content. Similarly, the distribution of A. mexicana, N. glauca, and X. spinosum was correlated with silt content in the soil on the AX1 positive axis. On the other hand, the distribution of C. erectus and D. innoxia showed correlation with Mn content, EC, and sand content in the soil. The distribution of P. juliflora and T. portulacastrum showed correlation with the soil contents of K, Ca, and Mn on the AX1 negative axis. Nevertheless, the distribution of H. curassavicum showed correlation with soil contents of N and P (Fig. 4).

Close

Fig. 4

A biplot showing the correlation between distribution of the alien species and the significant soil variables based on the canonical correspondence analysis (CCA) ordination. Plant scientific names are shown as the first 4 letters of genus and the first 3 letters of the species.

Export to PPT