This study was conducted in the riparian forest of Maroon River in Behbahan within Iran’ s southern Khuzestan Province. The study site was located at 32° 38'53"-30° 39'30"N and 50° 09'30"-50° 10'25"E with an average elevation of 250-300 m a.s.l. (Fig. 1). The average annual precipitation and annual mean temperature of the site were 350 mm and 24° , respectively. This site was typically dominated by P. euphratica and T. arceuthoides with a floral coverage by other species like Lyciumsp. and Vitex pseudonegundo to a lesser degree.

The riparian forest along the width direction was divided into three zones with a 200-meter wide zone in between. The locations studied were named Distance I (riverbank), Distance II (intermediate), and Distance III (furthest from riverbank). The reason behind this division was that sand mining and truck movements took place within Distance II. To differentiate the distance effect from the disturbance effect, we have left 50 meters as a buffering zone between each of the distances studied. On the other hand, we have left 50 meters before and after sand mining locations as a buffering zone (Fig. 1). Since the dominant species in this riparian forest were P. euphratica and T. arceuthoides, we randomly selected 10 trees from each species per studied distance (in total 60 samples, 30 samples for each species). Rhizosphere samples were taken from a depth of 15 cm after litter removal in spring (Moradi et al., 2015). These soil samples were used for spore extraction and the determination of soil physical-chemical properties.

Fig. 1

	Figure Option ViewDownloadNew Window
	Fig. 1 Location of study site and sampling design

Spores were extracted from the soil by wet sieving and decanting (Gerdemann and Nicolson, 1963), followed by centrifugation in water. Sieve sizes ranged from 38 to 500 µ m. Five grams of the rhizosphere soil of the plant species studied were used in extracting spores. All extracted spores were mounted on slides stained with PVLG (polyvinyl alcohol lactic acid glycerol), as well as PVLG with Melzer reagent. The identification procedure was based on the observation of morphological features like spore walls, colors and sizes using a stereomicroscope (Olympus CH-2) and in accordance with Schenck and Perez (1988).

All soil samples were air-dried and passed through a 2-mm sieve. Soil phosphorus (P) (Olsen et al., 1954), exchangeable potassium (K) (Morwin and Peach, 1951), nitrogen (N) (Bremner and Mulvaney, 1982), organic carbon (OC) (Walkley and Black, 1934), pH (deionized water suspension of 1:2.5) and textures (hydrometric) were determined. Soil moisture content was determined using oven drying at 105° C for 24 h.

AMF diversities were evaluated using the Shannon-Weiner indices of diversity (H’ ), evenness, Simpson’ s index of dominance, and Margalef’ s richness index (Eqs. 1-4).

$H'=-\sum{{{P}_{i}}}\ln {{P}_{i}}$, (1)

where P_i is the relative abundance of each identified species per sampling site.

Evenness (E) was calculated by Equation 2:

$E=\frac{H'}{{{H}_{\max }}}$, (2)

where H_max=log₂(S), and S is the total number of species.

Simpson’ s index of dominance (D) was calculated by Equation 3:

$D=\sum{\left[ {{n}_{i}}\left( {{n}_{i}}-1 \right)/N\left( N-1 \right) \right]}$, (3)

where n_i is the number of spores of a species and N is the total number of identified spore samples.

Margalef’ s richness index (D_m) was calculated by Equation 4:

${{D}_{m}}=\frac{S-1}{\ln (N)}, $ (4)

where S is the number of species.

All data were subjected to normality tests and analyzed using one-way ANOVA to determine any differences between diversity indices and physical-chemical soil properties within the denoted distances. If data were significantly different, a pairwise comparison was conducted using the Duncan’ s post hoc analysis. Moreover, the Pearson’ s correlation coefficient was used to determine any significant correlation between the soil parameters and AMF diversity indices. All the analyses mentioned were performed by SPSS 16.0 software. Furthermore, principal component analysis (PCA) was performed to obtain the most important factors on AMF diversity. To evaluate the effects of soil physical-chemical properties, we applied redundancy analysis (RDA) with Monte Carlo (unrestricted 999 permutations) to verify the relationship between AMF diversity indices and soil physical-chemical properties. This analysis was performed using the CANOCO for Windows version 4.5. Also, PC-ORD for Windows (V.5) (McCune and Mefford, 1999) was used to determine the keystone species in the study field.

The results indicate that there were 21 AMF species belonging to 6 genera and 6 families in the riparian forest (Table 1). Thirteen species were observed in the T. arceuthoides rhizosphere and 19 in the P. euphratica rhizosphere. As shown in Table 1, Glomus segmentatum, G. geosporum, G. rubiforme, G. nanolumen, G. spinuliferum, Claroideoglomus drummondii, Gigaspora giganteaandAcaulospora paulinae appeared only in the P. euphratica rhizosphere. Meanwhile, G. multiforumandC. claroideum occurred in Tamarixsp. rhizosphere (Table 1). Glomus was the most frequent genus observed in stands of T. arceuthoides and P. euphratica, being consistent with the finding of Bö rstler et al. (2008). It should be noted that some AMF species are plant species-specific for T. arceuthoides and P. euphratica. The reason for this may be related to health of host plants and/or soil conditions that may affect AMF diversity (Chaudhary et al., 2008). Another reason for this may be due to genetic differences among different AMF species, since such a difference could affect the survivals or functionalities of AMF and their host plants (Colard et al., 2011).

The T. arceuthoides rhizosphere had 9, 6, and 11 AMF species detected in Distances I, II, and III, respectively, while they were respectively 11, 9, and 13 AMF species in the P. euphratica rhizosphere, indicating that Distance III (i.e., the farthest research zone from the riverside) had the highest AMF, while Distance II or the zone severely disturbed by sand mining had the least AMF. Moreover, G. geosporumandG. rubiformewere AMF species that were only detected in the P. euphratica rhizosphere within the boundaries of Distance II. It means that the tolerance of these AMF to disturbance may be useful for later restoration activities (Lara-Pé rez et al., 2014). We also found that F. badium, F. constrictum, and C. etunicatum were among AMF whose frequency remained unaffected by the distance from the riverside and the location of sand mining. This result may indicate that these fungi have adapted well to these anthropogenic disturbances. In contrast, the distance from both the riverside and the sand mining had negative effects on the frequencies of F. mossea, G. gibbosum, G. nanolumen, G. spinuliferum, C. drummondii, A. trappei, and P. fransiscana (Table 1). These AMF may well be the indicators of land degradation (Kennedy and Papendick, 1995). Based on this study, a list of AMF species can be proposed both for disturbed and for non-disturbed distances. This kind of list can be important because it may be helpful for site restoration (Symanczik et al., 2014).

Table 1

Table 1

Table 1 AMF and their frequencies (%) within the designated distances Family Species Distance I Distance II Distance III P. euphratica Tamarisk P. euphratica Tamarisk P. euphratica Tamarisk Glomeraceae Funneliformis badium 100 100 100 100 100 100 Funneliformis constrictum 100 100 100 100 100 100 Funneliformis mossea 10 20 10 Glomus deserticola 20 10 Glomus arenarium 100 100 80 100 100 80 Glomus segmentatum 10 20 10 Glomus geosporum 10 Glomus caesaris 10 Glomus rubiforme 10 Glomus gibbosum 10 50 10 10 50 10 Glomus multiforum 10 Glomus nanolumen 10 Glomus spinuliferum 10 Claroideoglomeraceae Claroideoglomus drummondii 10 Claroideoglomus etunicatum 100 100 100 100 100 100 Claroideoglomus claroideum 20 Gigasporaceae Gigaspora margarita 10 10 Gigaspora gigantean 10 Archaeosporaceae Archaeospora trappei 50 60 10 40 60 60 Acaulosporaceae Acaulospora paulinae 10 Pacisporaceae Pacispora fransiscana 20 10 20

Table 1 AMF and their frequencies (%) within the designated distances

F. badium and G. gibbosum were identified as the keystone indicator species at Distance III, while no keystone species were found at the other two distances in the P. euphraticastand (Table 2). It is well documented that AMF are keystone soil micro-organisms for the conservation of biodiversity (O’ Neill et al., 1991; Power and Mills, 1995; Piraino et al., 2002). The lowest frequency of G. gibbosum at Distance II (i.e., the severely disturbed distance) indicated that the negative effects of sand mining on keystone species are rather pronounced. Our results are important in that they provide a list of resistant AMF species that could be used in the conservation of biodiversity.

In the T. arceuthoides stand, G. gibbosum and C. etunicatumwere considered as keystone species in Distance I and Distance II, respectively (Table 3). This finding is consistent with the findings of Piraino et al. (2002), who provided evidence that C. etunicatum occurred in a highly disturbed environment. Different AMF keystone species found in P. euphratica and T. arceuthoides stands reflect not only the differences in responses to plant species but also the AMF symbiosis to anthropogenic activity.

The changes of AMF species by anthropogenic regimes suggest that AMF is a good indicator of soil quality changes, being consistent with the findings of Verbruggen et al. (2012).

The PCA results for T. arceuthoides indicated that the first and second axes explained 48.80% and 23.46% of total variance, respectively (Fig. 2). Furthermore, the results of Monte Carlo permutation tests on redundancy analysis (RDA) showed that soil sand, silt, moisture, clay, and phosphorus (P) are the most important factors affecting AMF diversity inT. arceuthoides(Table 4). The PCA result for P. euphratica revealed that the first and second axes explained 69.27% and 19.71% of the total variance, respectively (Fig. 3). The Monte Carlo permutation tests on RDA indicated that the most significant affecting factors in the physical-chemical properties of soil on AMF diversity in P. euphratica stands are soil moisture, clay, and pH (Table 5). Our results are consistent with Deepika and Kothamasi (2015). These results imply that soil-property changes caused by anthropogenic activity can result in significant AMF changes. Furthermore, it implies that sand mining disturbance affects AMF directly and indirectly by changing either soil conditions or the status of host plant roots.

Table 2

Table 2

Table 2 Keystone indicator species for P. euphratica stand AMF species Distance Importance value index Pvalue Glomus arenarium I 39.3 0.3997 Pacispora fransiscana 17.1 0.5381 Glomus caesaris 10.0 1.0000 Gigaspora margarita 10.0 1.0000 Acaulospora paulinae 10.0 1.0000 Gigaspora gigantean 10.0 1.0000 Claroideoglomus etunicatum II 35.9 0.7269 Glomus geosporum 10.0 1.0000 Glomus rubiforme 10.0 1.0000 Funneliformis badium III 53.3 0.0116 Glomus gibbosum 44.6 0.0270 Funneliformis constrictum 40.9 0.1576 Archaeospora trappei 36.0 0.1526 Funneliformis mossea 20.0 0.3069 Glomus deserticola 20.0 0.3143 Glomus nanolumen 10.0 1.0000 Glomus spinuliferum 10.0 1.0000 Claroideoglomus drummondii 10.0 1.0000 Glomus segmentatum 7.5 0.7487

Table 2 Keystone indicator species for P. euphratica stand

Table 3

Table 3

Table 3 Keystone indicator species for T. arceuthoides stand AMF species Distance Importance value index P value Glomus gibbosum I 39.6 0.0176 Funneliformis badium 36.6 0.8344 Gigaspora margarita 10.0 1.0000 Glomus multiforum 10.0 1.0000 Claroideoglomus etunicatum II 41.6 0.0236 Glomus arenarium 38.5 0.4445 Funneliformis constrictum III 39.9 0.2180 Archaeospora trappei 24.0 0.7704 Claroideoglomus claroideum 20.0 0.3137 Pacispora fransiscana 20.0 0.3001 Glomus segmentatum 10.0 1.0000 Glomus deserticola 10.0 1.0000 Funneliformis mossea 7.0 1.0000

Table 3 Keystone indicator species for T. arceuthoides stand

Fig. 2

	Figure Option ViewDownloadNew Window
	Fig. 2 Principal component analysis of AMF diversity indices and soil physical-chemical properties in T. arceuthoides rhizosphere. SD, spore density; D, Simpson’ s index of dominance; Ric, richness; Dm, Margalef’ s richness index; Col, colonization; H’ , Shannon-Weiner indices of diversity; N, nitrogen; OC, organic carbon; K, potassium; P, phosphorus; Moi, moisture; Eve, evenness.

Table 4

Table 4

Table 4 Monte Carlo permutation test on redundancy analysis for soil physical-chemical properties of T. arceuthoides Soil physical-chemical property F value P value Sand 15.572 0.002 Silt 15.413 0.005 Moisture 14.062 0.003 Clay 10.360 0.008 Phosphorus 4.715 0.040 Organic carbon 4.393 0.060 Nitrogen 4.342 0.070 Potassium 1.749 0.200 pH 0.007 0.980

Table 4 Monte Carlo permutation test on redundancy analysis for soil physical-chemical properties of T. arceuthoides

The SD associated with the T. arceuthoidesrhizosphere in Distance II, where sand mining and truck movements occurred, had the highest significant value (Table 6). Nevertheless, no significant difference was observed in SD values with regard to P. euphratica (Table 7). The higher SD in Distance II may be explained by soil compaction that reduced root growth (Wallace, 1987; Trejo et al., 2016). In the present study, P. euphratica and T. arceuthoides showed different SD and root colonization rates, reflecting the fact that different plant species respond differently to these two factors (Li et al., 2007).

Furthermore, a high level of symbioses was observed between AMF and P. euphratica, being in accordance with the findings of Wang et al. (2010). Yet, our results are not in accordance with those of Yang et al. (2013) who reported a significantly lower SD in P. euphratica rhizospheres. The reason for a significantly higher level of SD in the current study may be attributable to a higher AMF diversity compared to Yang et al. (2013). The minimum and maximum colonization rates for T. arceuthoides and P. euphratica were observed at Distances II and III, respectively (Table 6). A clear indication of the negative effect of sand mining and truck movement was found on root colonization rates at Distance II. Changes in SD and root colonization may be induced by soil physical-chemical changes caused by sand mining activity. This study provides a much improved understanding of the impact of sand mining on AMF diversity, colonization, and SD.

An increasing trend in the root colonization rate with increasing distance from the river was observed in the present study, which probably indicates that soil conditions were improved for AMF symbiosis farther from the river. High level of root colonization in this study is in agreement with the findings of Wang et al. (2010).

Fig. 3

	Figure Option ViewDownloadNew Window
	Fig. 3 Principal component analysis of AMF diversity indices and soil physical-chemical properties in P. euphratica rhizosphere. SD, spore density; D, Simpson’ s index of dominance; Ric, richness; Dm, Margalef’ s richness; Col, colonization; H’ , Shannon-Weiner indices of diversity; E, Evenness; N, nitrogen; OC, organic carbon; K, Potassium; P, phosphorus; Moi, moisture.

Table 5

Table 5

Table 5 Monte Carlo permutation test on redundancy analysis for soil physical-chemical properties of P. euphratica Soil physical-chemical property F value P value Moisture 5.671 0.03 Clay 4.786 0.04 pH 4.082 0.04 Sand 3.549 0.08 Organic carbon 3.247 0.09 Silt 2.998 0.11 Nitrogen 2.768 0.13 Potassium 2.628 0.13 Phosphorus 0.913 0.37

Table 5 Monte Carlo permutation test on redundancy analysis for soil physical-chemical properties of P. euphratica

T. arceuthoides stand had the least AMF diversity at Distance II (Table 6) and P. euphratica stand had the highest AMF diversity at Distanced III (Table 7). Although, Shannon-Weiner indices of diversity (Hʹ ) and Simpson’ s index of dominance (D) displayed no significant differences between the triple designated Distances, the least significant D was observed in Distance II (Tables 6 and 7). The reason for this is that D is affected by species richness (Magurran, 2004) and higher species richness results in a higher D value (Mahecha-Vá squez et al., 2017). As the least species richness was observed in Distance II and then the least D was also observed in this zone (Stü rmer and Siqueira, 2011). These results suggest that the impact of sand mining was severe enough to have negatively affected AMF diversity in riparian forest. Sand mining resulted in changes in AMF diversity and composition, and this knowledge is crucial to the understanding of mycorrhizal function in these ecosystems (Soka and Ritchie, 2014a, b).

In addition, distance from the local river also affected AMF diversity, while Distance III (the farthest distance from the river) hosted the highest AMF diversity. This may be due to the least negative effects of natural phenomena among the three distances.

Table 6

Table 6

Table 6 AMF diversity for T. arceuthoides stand Distance I Distance II Distance III Root colonization 82.4± 1.98a 73.7± 2.89b 84.4± 2.54a SD 189± 19.60b 254± 19.54a 189± 19.23b Richness 5.4± 0.37a 4.4± 0.16b 5.2± 0.38ab H’ 0.84± 0.07a 0.75± 0.08a 0.95± 0.05a D 0.59± 0.04a 0.62± 0.05a 0.51± 0.04a E 0.45± 0.04a 0.50± 0.04a 0.53± 0.05a Dm 0.85± 0.06a 0.62± 0.03b 0.81± 0.07a Note: SD, Spore density per 5 grams of soil; H’ , Shannon-Weiner indices of diversity; D, Simpson’ s index of dominance; E, Evenness; Dm, Margalef’ s richness index. Different lowercase letters indicate significance among three distances at P< 0.05 level. Abbreviations are the same as in Table 7.

Table 6 AMF diversity for T. arceuthoides stand

Table 7

Table 7

Table 7 AMF diversity for P. euphratica stand Distance I Distance II Distance III Root colonization 85.5± 1.52a 65.3± 4.24b 86.4± 1.56a SD 174± 22.71a 177± 24.63a 186± 15.96a Richness 5.2± 0.29b 4.3± 0.26c 6.1± 0.34a H’ 0.91± 0.07a 0.84± 0.08a 1.06± 0.09a D 0.54± 0.04a 0.57± 0.05a 0.48± 0.05a E 0.50± 0.04a 0.55± 0.03a 0.49± 0.04a Dm 0.83± 0.05a 0.65± 0.05b 0.97± 0.06a

Table 7 AMF diversity for P. euphratica stand

Pearson’ s correlation between soil factors and AMF diversity indices at the P. euphraticastand revealed that AMF richness had a positive correlation with soil P and a negative correlation with pH. However, evenness had a significant negative correlation with richness, Simpson’ s index of dominance (D) and Margalef’ s richness index (D_m). Furthermore, soil moisture had significant negative correlations with root colonization and AMF richness, while, had a negative significant correlation with spore density (SD) (Table 8). However, for the T. arceuthoides stand, AMF richness had a significantly positive correlation with root colonization (RC). Soil phosphorus (P) had a significantly positive correlation with D and a significantly negative correlation with Shannon-Weiner indices of diversity (H’ ). Yet, Margalef’ s richness index (D_m) was significantly correlated with root colonization. Also, soil moisture had a significantly positive correlation with root colonization (RC) and a significantly negative correlation with SD (Table 9). These results indicate the importance of better soil conditions for higher AMF diversity (Mirzaei and Moradi, 2017). Also, in this study, Distance III with better soil conditions and less anthropogenic activities had higher AMF diversity.

Table 8

Table 8

Table 8 Pearson’ s correlation between soil parameters and AMF diversity indices at P. euphratica stand N OC P K Clay Silt Sand pH Moi RC SD R D H’ E Dm N 1.00 OC 0.96* * 1.00 P 0.36 0.26 1.00 K -0.01 -0.05 0.48 1.00 Clay 0.20 0.20 0.66* * 0.16 1.00 Silt 0.16 0.17 0.51 0.15 0.95* * 1.00 Sand -0.17 -0.18 -0.56* -0.16 -0.97* * -0.99* * 1.00 pH -0.26 -0.21 -0.26 0.36 -0.52* -0.47 0.49 1.00 Moi 0.28 0.30 -0.23 -0.48 0.04 0.15 0.13 -0.53 1.00 RC 0.36 0.37 0.24 -0.04 0.37 0.35 -0.37 -0.12 0.72* * 1.00 SD -0.14 -0.18 0.12 -0.37 0.11 0.04 -.062 -0.11 -0.69* * -0.16 1.00 R 0.36 0.27 0.57* 0.17 0.48 0.42 -0.44 -0.58* 0.61* 0.06 0.17 1.00 D 0.06 0.21 -0.24 -0.26 -0.01 -0.06 0.05 0.18 -0.48 0.30 0.16 -0.27 1.00 H’ 0.04 -0.12 0.34 0.29 0.12 0.15 -0.14 -0.29 0.46 -0.25 -0.14 0.44 -0.97* * 1.00 E -0.36 -0.39 -0.25 0.09 -0.33 -0.20 0.26 0.30 0.47 -0.31 -0.27 -0.57* -0.62* 0.47 1.00 Dm 0.39 0.32 0.49 0.25 0.41 0.37 -0.39 -0.57* -0.52* 0.01 -0.03 0.96* * -0.30 0.47 -0.52* 1.00 Note: N, nitrogen; OC, organic carbon; P, soil phosphorus; K, soil potassium; Moi, moisture; RC, root colonization; SD, spore density; R, richness; D, Simpson’ s index of dominance; H’ , Shannon-Weiner indices of diversity; E, evenness; Dm, Margalef’ s richness index; * * , correlation is significant at P< 0.01 level. * , correlation is significant at P< 0.05 level. Abbreviations are the same as in Table 9.

Table 8 Pearson’ s correlation between soil parameters and AMF diversity indices at P. euphratica stand

Table 9

Table 9

Table 9 Pearson’ s correlation between soil parameters and AMF diversity indices at T. arceuthoides stand N OC P K Clay Silt Sand pH Moi RC SD R D Hʹ E Dm N 1.00 OC 0.99* * 1.00 P 0.44 0.44 1.00 K 0.52* 0.52* 0.02 1.00 Clay 0.33 0.34 0.28 0.56* 1.00 Silt 0.31 0.32 0.29 0.64* 0.73* * 1.00 Sand -0.33 -0.34 -0.26 -0.68* * -0.80* * -0.99* * 1.00 pH 0.40 0.39 -0.18 0.34 -0.24 -0.04 0.05 1.00 Moi 0.003 0.01 -0.21 0.14 0.33 0.59* -0.55* -0.06 1.00 RC 0.15 0.14 -0.22 0.41 -0.14 0.14 -0.14 0.24 0.78* * 1.00 SD -0.27 -0.26 0.27 -0.15 0.38 0.13 -0.16 -0.36 -0.71* * -0.31 1.00 R 0.17 0.18 -0.10 0.39 0.15 0.16 -0.20 0.20 0.24 0.65* * -0.19 1.00 D 0.08 0.08 0.66* * -0.18 0.11 0.07 -0.05 -0.41 -0.29 -0.32 0.40 -0.25 1.00 Hʹ -0.03 -0.02 -0.66* * 0.26 -0.07 -0.04 0.01 0.44 0.33 0.42 -0.43 0.41 -0.98* * 1.00 E -0.13 -0.12 -0.53* 0.04 -0.14 -0.10 0.10 0.32 -0.07 -0.04 -0.28 -0.33 -0.82* * 0.71* * 1.00 Dm 0.17 0.18 -0.19 0.40 0.03 0.09 -0.13 0.32 0.39 0.68* * -0.38 0.97* * -0.37 0.53* -0.18 1.00

Table 9 Pearson’ s correlation between soil parameters and AMF diversity indices at T. arceuthoides stand

Several factors were reported to have affected AMF diversity, such as environmental (Moradi et al., 2015) and soil conditions (Yang et al., 2011). But in the present study, sand mining and truck movement were documented to have resulted in AMF variations. Also, soil P and pH affect AMF diversity, being in line with the findings of other studies (Treseder, 2004; Yoshimura et al., 2013). Being similar to previous studies highlighting the use of AMF species as soil bio-indicators (Fokom et al., 2012; Vasconcellos et al., 2016), the present study has also demonstrated that AMF diversity could be a key factor for understanding soil quality. Furthermore, it should be reiterated that because fungi are able to adapt to local habitats and also because they are important for biodiversity conservation, AMF species should be carefully identified for each specific region (Symanczik et al., 2014).

Unlike other studies reporting that soil compaction did not affect AMF (Miransari, 2013; Thorne et al., 2013), the present study demonstrated that sand mining and truck movement did negatively affect AMF diversity and SD through reducing fine roots (Waltert et al., 2002). As shown earlier, sand mining caused approximately 24.4% and 10.8% reductions in SD in P. euphratica and T. arceuthoides, respectively. These results are in accordance with Jasper et al. (1991).