Plant Species Classification and Diversity of the Understory Vegetation in Oak Forests of Swat, Pakistan

Abstract

The forest ecosystem has understory vegetation that plays a vital role in sustaining diversity, providing nutrients, and forming a useful association for developing a balanced ecosystem. The current study provides detailed insights into the plant biodiversity and species classification of the understory vegetation of Swat, Pakistan. The floral diversity of the area was comprised of 58 plant species belonging to 32 families. The physiognomy of the studied area was dominated by herbaceous growth form with 47 species. The dominant life-form class was hemicryptophytes with 19 species (33%), followed by nanophanerophytes with 15 species (26%) and therophytes with 13 species (22%). Of the 58 species, 43 plant species were associated with group III clustered by applying Ward’s agglomerative clustering that indicated wide sociability of the species in the studied oak-dominated forests. Group III had higher species richness (10.3), α-diversity (2.74) and β-diversity (9.85), and Margalef index values (3.95). While the group I had maximum Pielous and Simpson index values of 0.97 and 7.13, respectively. Redundancy analysis revealed that seven variables (i.e., latitude, elevation, clay, wilting point, bulk density, saturation, and electric conductivity) were significantly influential concerning the understory vegetation of oak-dominated forests. The understory vegetation of these forests plays an important role in the forest ecosystem of the region. The present study reveals floral divergence and physiognomic scenario of the unexplored study area, which could be an important reference for future ethnobotanical, phytosociological, and conservational endeavors. Moreover, this information is important to the success of efforts intended to prevent the loss of species diversity in these forests by destroying their natural habitats.

1. Introduction

The most useful source of botanical information of a particular area is its floristic checklists [1,2]. Since floristic composition is a decent floristic marker, any changing floristic composition in various endogenous milieus highlights the presence of various environmental variables, prompting entomb and intra-particular variations [3]. The floristic structure of a specific territory provides fundamental. However, information about the plant dissemination and motherland show diversity in landscapes, topographies, and biological/ecological zones, bolstering the distinctive floristic composition [4]. To contemplate the vegetation of a territory, it is essential to oversee the plant’s life-form, which is an indicator for both micro- and macroclimate [5]. As indicated by Raunkiaer [6], the biological spectrum is divided into five major classes (i.e., phanerophytes, hemicryptophytes, cryptophytes, chamaephytes, and therophytes).

The plant species are also categorized based on leaf size classes which are also extremely helpful for association mapping of vegetation; it may also be helpful for comprehension of the plant communities and their physiological processes [7,8]. Leaf sizes and shapes (physiognomy) connect unequivocally with moisture and temperature from worldwide to neighborhood scales [9]. Leaf characteristics assume an especially imperative part in carbon absorption, water relations, and vitality adjust (energy balance) [9]. Furthermore, a life-form and leaf size spectra provide a clear picture of an area. Therefore, climatic conditions of a certain region may accurately be referred from such observations [10]. These physiognomic characters have been widely used in vegetation investigations in anthropogenic and naturally vulnerable areas [5,10].

The understory vegetation plays a major role in sustaining the structure and function of forest ecosystems [11,12], promoting energy transfer and nitrogen cycles, and shaping the canopy succession of forest ecosystems [13,14]. The understory makes comparatively little contribution to the overall biomass of forest plants [13,15]; it promotes the greatest proportion in floristic diversity [16]. In addition, diversified understory vegetation boosts the structural complexity of forests and provides other biotic groups with habitats and food, increasing their diversity [17]. For forest regeneration, understory vegetation is extremely crucial, as it can influence the germination, survival, and development of tree seedlings by competing for light, water, and nutrients [11,15,18]. Therefore, increasing attention is being paid to forest understory vegetation [16]. It is essential to know about the most influential variables affecting their distribution for ecological protection, conservation, and forest management [12,16,19].

Numerous studies have shown the effect on the species composition and diversity of understory flora. These include flora canopy species and structure [20], stand management [21], soil disturbances [22], light resources [23], litter properties [20,24], and soil nutrients and pH [23,25]. Topography can dramatically change microclimates and the availability of energy under the tree canopy [23,26] and, in turn, affect the composition and diversity of understory species [12,27]. In coniferous forests, hardwood forests, and mixed-wood forests, understory vegetation has been well studied [25,28]. The latter two types of forests are generally agreed to create more favorable conditions for their biodiversity conservation and restoration than coniferous forests [21]. In general, broad-leaved evergreen forest and oak forest have been ignored in understory vegetation and its composition. Beg and Mirza [29] studied oak-dominated forests in the Hindukush-mountain region. In the study, four oak species have been reported from the region, i.e., Quercus lanata Sm., Quercus robur L., Quercus baloot Griff., and Quercus semecarpifolia Sm. The research work also reported thirty-six understory species in the oak-dominated forest. So far, no comprehensive study has investigated the details of phytosociological attributes and environmental variables in association with those understories, aiming to understand its diversity. Forest ecosystems’ floristic composition, geographic distribution, and ecological conditions are essential for a rational management plan. In this study, the understory species composition is considered to differ significantly among the major groups concerning species richness and diversity. Additionally, attempts were made to determine the extent to which physiographic and edaphic variables could explain variation in the understory species composition and major distinct groups of oak-dominated forests.

2. Materials and Methods

2.1. Study Site

Swat district is located in northwest of Pakistan in Khyber Pakhtunkhwa (KP) province (Figure 1). It is a well-known summer resort in the country for national and international tourists and visitors. It lies between 34°0′34″ and 35°0′55″ north latitudes and 72°0′08″ and 72°0′50″ east longitudes. The total area of the district is 5337 Km². Most of the area consists of mountainous regions belonging to the Hinduraj series of Hindukush mountains. The climate of Swat is relatively warm in the lower parts but cooler and refreshing in the upper parts of the Valley. Summers are short and moderate, temperatures seldom rise above 37 °C. Annual rainfall ranges from 800 to 900 mm, and snowfall during winter is the constant feature of upper Swat [30].

2.2. Field Investigation

Extensive field surveys were carried out during the flowering and fruiting stages. Timings for the fieldwork were carefully selected according to the plant’s growth phases and collection season. A handheld (Garmin manufactured in Germany) Geographical Positioning System (GPS) recorded the study area coordinates. A total of 300 plots having an area of 20 m × 20 m were established in thirty stands (10 plots/stand) of the oak forest at an elevation ranging from 1000 to 2900 m. The understory vegetation is defined as shrubs and herbaceous plants growing on the forest floor. Each 20 m × 20 m plot was divided into four 10 m × 10 m quadrats, three of which were chosen at random to survey shrub species. Five 1 m × 1 m quadrats, one in the middle and four at the corners of the 20 m × 20 m plot, were used to study herbaceous plants. Phytosociological parameters, i.e., height and density, with the percent cover’s visual calculations, were measured [31,32]. The life-form classes and leaf spectra of all plant species were determined and classified according to Raunkiaer [7]. Plants reported and collected were identified by expert plant taxonomists and were then rechecked with the Royal Botanical Garden Kew online web (http://www.plantsoftheworldonline.org/ accessed on 14 November 2021) and the Flora of Pakistan [33,34]. The Herbarium specimens were submitted to the Department of Botany, University of Malakand.

2.3. Soil Analysis

A stainless-steel cylindrical soil sampler with a diameter of 5 cm was used to collect five topsoil samples (0–10 cm depth) at random sites from each plot. The samples were then carefully combined to create a composite sample for analysis. The hybrid soil samples were air-dried and sieved to 0.2 mm for soil organic carbon (SOC), and total nitrogen (TN) tests and 2 mm for soil pH tests prior to analysis. The updated Mebius approach was used to examine SOC [35]. The Kjeldahl procedure was used to evaluate TN [36]. In a 1:2.5 soil to water suspension, the pH of the soil was determined [37]. The volumetric ring procedure was used to determine the bulk density of the soil [38]. Soil textural properties were determined using Bouyoucos hydrometer [39], while electrical conductivity was calculated using a conductometer (Model CON 5). Important nutrients like potassium, phosphorus, and lime were determined by following standard procedures adopted by [38].

2.4. Data Analysis

The relative important value (IV) and importance value index (IVI) of each species in the understory plant community was calculated as follows:

IV (x) = F₃ (x) + D₃ (x) + C₃ (x)

(1)

$$\mathrm{IVI} \left(\mathrm{x}\right)=\frac{\mathrm{IV} \left(\mathrm{x}\right)}{300}\times 100$$

(2)

where F₃ (x), relative frequency; D₃ (x), relative density; and C₃ (x), relative cover.

IVI (x) represents the computed importance value index of understory species and seedling sapling of tree species, while IV (x) represents the computed importance value of the associated understory species. The species richness and α-diversity (Shannon index, H; Evenness index, E; Margalef’s index, M and Simpson’s index, 1/D) were used to describe plot diversity and β-diversity for changes in community structure across sites within contrasting forest types.

$${\mathrm{H}}^{\prime }={\sum }_{\mathrm{i}=1}^{\mathrm{S}}pi \mathrm{In}pi$$

(3)

$$\mathrm{E}=\frac{{\mathrm{H}}^{\prime }}{\mathrm{InS}}$$

(4)

M = S − 1/In N

(5)

1/D = 1/Σ (pi²)

(6)

β = St/a

(7)

where pi = proportion of the species (i) to total number of species, In = natural logarithm, S = species richness, N = total number of species, St = number of species in all plots, and a = total number of plots

Ward’s agglomerative cluster analysis was used for classification of the understory vegetation stands as the species distribution varies across the elevation gradient; therefore, cluster analysis was preferred in comparison with other technique available in Pc-ord version 6 in which Euclidean distance was opted for measuring distance using Ward’s linkage method [40]. Redundancy analysis (RDA) evaluated the effects of different environmental and understory parameters using PC-ord version 6. The suitability of RDA analysis was assessed by using published literature in which first DCA-ordination was performed [41] to elucidate whether unimodal [42,43] or linear [44] response curve should be used in ordination analysis. The gradient length on DCA-axes 1 was 12, which is more than 4.1, and therefore the use of Canonical Correspondence analysis (CCA) or RDA was tested. In the choice between CCA and RDA-ordination, we prefer the RDA as the % age variance was (30%) and stands distribution was uniform in biplot compared to CCA ordination having variance of 19%. Ms-Excel 2010 was used for data tabulation and graphic presentation, while SPSS version 22 was used for statistical analysis. The significance of the variables was tested at p < 0.05. In addition, post hoc honestly significant differences (HSD) test was used for variation between the groups.

3. Results

3.1. Understory Species Composition

A total number of 58 understory plant species belonging to 32 families was identified at 30 different locations in the Hindukush ranges of Swat. The physiognomy was dominated by herbaceous growth form with 47 plant species (81%), followed by shrubs with 10 (17%), and trees with a single plant species (2%). Herbaceous dominancy may be due to altitudinal and geographical variations, indicating the herbaceous flora’s climatic factors. Among the families, Fabaceae was the leading family with eight plant species, followed by Asteraceae and Lamiaceae with seven species each. Further, seven families (i.e., Amaranthaceae, Malvaceae, Poaceae, Plantaginaceae, Polygonaceae, Pteridaceae, and Rosaceae) were recorded with two plant species each. The remaining 22 families are represented by individual species (

Table 1. List of the plant species with the family names, growth form/habit, chorotype, life-form, and leaf size classes that were recorded in understory vegetation of oak-dominated forests of Swat, Pakistan.

Taxon	Family	Habit	Chorotype	Life-Form	Leaf Size
Achyranthes aspera L.	Amaranthaceae	Herb	Plurireg	Th	N
Adiantum venustum D. Don	Pteridaceae	Herb	Cosm	G	N
Ajuga bracteosa Wall. ex Benth.	Lamiaceae	Herb	Unireg	H	Mi
Amaranthus viridis L.	Amaranthaceae	Herb	Cosm	Th	Mi
Arabidopsis thaliana (L.) Heynh.	Brassicaceae	Herb	Cosm	Th	L
Arenaria serpyllifolia L.	Caryophyllaceae	Herb	Cosm	Th	N
Artemisia vulgaris L.	Asteraceae	Herb	Plurireg	Th	N
Asplenium trichomanes L.	Aspleniaceae	Herb	Bireg	H	N
Astragalus grahamianus Benth.	Fabaceae	Herb	Unireg	Ch	L
Atropa acuminata Royle ex Lindl.	Solanaceae	Herb	Plurireg	G	Mi
Berberis lycium Royle	Berberidaceae	Shrub	Unireg	NPh	L
Bidens cernua L.	Asteraceae	Herb	Bireg	H	Mi
Buddleja asiatica Lour.	Scrophulariaceae	Shrub	Cosm	NPh	Mi
Calamintha vulgaris (L.) H. Karst.	Lamiaceae	Herb	Plurireg	Th	N
Chenopodium album L.	Chenopodiaceae	Herb	Cosm	H	N
Convolvulus arvensis L.	Convolvulaceae	Herb	Cosm	H	N
Erigeron canadensis L.	Asteraceae	Herb	Cosm	Th	L
Cynodon dactylon (L.) Pers.	Poaceae	Herb	Plurireg	H	L
Daphne mucronata Royle	Thymelaeaceae	Shrub	Bireg	NPh	N
Dioscorea deltoidea Wall. ex Griseb.	Dioscoreaceae	Herb	Unireg	H	Mi
Dodonaea viscosa Jacq.	Sapindaceae	Shrub	Cosm	NPh	Mi
Dryopteris stewartii Fraser-Jenk.	Dryopteridaceae	Herb	Bireg	G	Me
Elaeagnus angustifolia L.	Elaeagnaceae	Shrub	Bireg	NPh	Mi
Festuca gigantea (L.) Vill.	Poaceae	Herb	Bireg	G	L
Fragaria indica Andrews	Rosaceae	Herb	Plurireg	H	N
Galium asperifolium Wall.	Rubiaceae	Herb	Unireg	Th	L
Hibiscus syriacus L.	Malvaceae	Shrub	Plurireg	NPh	Mi
Indigofera gerardiana Graham ex Baker	Fabaceae	Herb	Unireg	NPh	N
Indigofera heterantha Wall. ex Brandis	Fabaceae	Herb	Unireg	NPh	L
Isodon rugosus (Wall. ex Benth.) Codd.	Lamiaceae	Herb	Unireg	NPh	Mi
Justicia adathoda L.	Lamiaceae	Herb	Plurireg	NPh	Me
Lamium album L.	Lamiaceae	Herb	Unireg	Th	Mi
Lespedeza juncea (L. f.) Pers.	Fabaceae	Herb	Unireg	H	N
Lotus corniculatus L.	Fabaceae	Herb	Bireg	H	Mi
Malva neglecta Wallr.	Malvaceae	Herb	Unireg	H	Mi
Medicago lupulina L.	Fabaceae	Herb	Plurireg	Ch	N
Myrsine africana L.	Primulaceae	Shrub	Bireg	NPh	L
Origanum vulgare L.	Lamiaceae	Herb	Unireg	Th	N
Otostegia limbata (Benth.) Boiss.	Lamiaceae	Herb	Unireg	NPh	L
Oxytropis mollis Royle ex Benth.	Fabaceae	Herb	Unireg	H	L
Plantago lanceolata L.	Plantiganaceae	Herb	Plurireg	H	Mi
Plantago major L.	Plantaginaceae	Herb	Plurireg	H	Mi
Polygonatum verticillatum (L.) All.	Asparagaceae	Herb	Plurireg	Th	N
Pteris cretica L.	Pteridaceae	Herb	Plurireg	G	Mi
Ranunculus laetus Salisb.	Ranunculaceae	Herb	Unireg	H	Me
Rosa webbiana Wall. ex Royle	Rosaceae	Herb	Plurireg	NPh	N
Rumex denticulatus K. Koch.	Polygonaceae	Herb	Bireg	H	Me
Rumex nepalensis Spreng.	Polygonaceae	Herb	Bireg	H	Me
Sarcococca saligna (D.Don) Müll.Arg.	Buxaceae	Shrub	Bireg	NPh	Mi
Smilax lanceolata L.	Smilacaceae	Herb	Bireg	L	Mi
Solidago virgaurea L.	Asteraceae	Herb	Unireg	H	Mi
Tagetes minuta L.	Asteraceae	Herb	Bireg	Th	Mi
Taraxacumofficinale (L.) Weber ex F.H.Wigg.	Asteraceae	Herb	Cosm	H	Mi
Trifolium repens L.	Fabaceae	Herb	Cosm	G	N
Viburnum grandiflorum Wall. ex DC.	Viburnaceae	Shrub	Unireg	NPh	Ma
Violaserpens Wall. ex Ging.	Violaceae	Shrub	Plurireg	G	Mi
Youngia japonica (L.) DC	Asteraceae	Herb	Cosm	Th	Mi
Zanthoxylum armatum DC.	Rutaceae	Tree	Cosm	MPh	Mi

Plurireg (Pluriregional); Unireg (Uniregional); Cosm (Cosmopolitan); Bireg (Biregional); Th (Therophyte); G (Geophytes); NPh (Nanophanerophyte); H (Hemicryptophyt); MPh (Microphanerophytes); Ch (Chamaephyte); Mi (Microphyllous); Ma (Megaphyllous); N (Nanophyllous); L (Leptophyllous); Acronyms were assigned by following Shehata [45] except Raunkiaer life-form.

).

The present study identified seven different life-form classes in the study area (Figure 2). The most dominant life-form was hemicryptophytes with 19 species (33%), followed by nanophanerophytes with 15 species (26%), therophytes with 13 species (22%), and geophytes with 7 species (12%)., Two species of chamaephytes (3%) and one species (2%) of nanophanerophytes and liana were also recorded (Table 1).

3.2. Leaf Size Spectra

Results of leaf size spectra trait revealed that species with microphyllous leaves dominated the area with 24 species (41%) followed by nanophyll 17 species (29%), leptophyll 11 species (19%), and mesophyll 5 species (9%) (Figure 3). Single species (2%) with megaphyllous leaf size was also recorded. A high rate of microphylls may be identified within the cool environment of the sub-elevated and snowcapped areas. Here, the top layer was less settled; it contained a slim sheet that may deny them the entrance of roots. The microphyllous and nanophyllous leaves species were inexhaustible because of the biological variety for this dry condition.

The chorotype of the species varies from uniregional to cosmopolitan, i.e., 22% were cosmopolitan, 26% pluriregional, 23% bioregional, and 29% were uniregional.

3.3. Understory Classification

A total of 58 plant species were recorded as associates in the understory vegetation and clustered into three major groups by applying Ward’s agglomerative clustering (Figure 4 and

Table 2. Understory flora associated with oak-dominated forest recorded in the three clusters separated by Ward’s agglomerative clustering procedure.

Species Name	Code	Cluster I	Cluster II	Cluster III
		M ± SE	M ± SE	M ± SE
Achyranthes aspera L.	As	2.41 ± 2.41	-	-
Adiantum venustum D. Don	Av	-	-	0.77 ± 0.77
Ajuga bracteosa Wall. ex Benth.	Ab	3.87 ± 2.65	-	-
Amaranthus viridis L.	Am	5.10 ± 3.23	-	2.41 ± 1.44
Arabidopsis thaliana (L.) Heynh.	At	-	-	1.16 ± 0.79
Arenaria serpyllifolia L.	Am	-	-	0.53 ± 0.53
Artimisia vulgaris L.	Av	7.2 ± 4.71	-	-
Asplenium trichomanes L.	As	9.14 ± 4.71	4.23 ± 4.23	6.6 ± 2.6
Astragalus grahamianus Benth.	Ag	-	-	0.53 ± 0.53
Atropa acuminata Royle ex Lindl.	Aa	3.4 ± 3.46	-	2.42 ± 2.43
Berberis lycium Royle	B	9.86 ± 5.49	7.66 ± 4.91	7.82 ± 3.54
Bidens cernua L.	Bc	-	1.71 ± 1.71	-
Buddleja asiatica Lour.	Ba	0.93 ± 0.93	-	1.69 ± 1.17
Calamintha vulgaris (L.) H. Karst.	Cv	-	11.5 ± 4.11	3.46 ± 1.66
Chenopodium album L.	Ca	-	4.4 ± 4.42	-
Convolvulus arvensis L.	Ca	2.5 ± 2.52	4.2 ± 2.74	6.87 ± 1.76
Erigeron canadensis L.	Co	0.85 ± 0.85	-	-
Cynodon dactylon (L.) Pers.	Cd	7 ± 5	6.17 ± 6.17	4.88 ± 3.32
Daphne mucronata Royle	Dm	3.83 ± 3.83	-	1.94 ± 1.34
Dioscorea deltoidea Wall. ex Griseb.	Dd	1.83 ± 1.83	-	-
Dodonaea viscosa Jacq.	Do	1.42 ± 1.42	-	-
Dryopteris stewartii Fraser-Jenk.	Ds	-	9.86 ± 9.86	-
Elaeagnus angustifolia L.	Ea	2.3 ± 2.3	-	-
Festuca gigantea (L.) Vill.	Fg	-	-	0.46 ± 0.46
Fragaria indica Andrews	Fi	-	2.71 ± 2.71	5.53 ± 2.22
Galium asperifolium Wall.	Ga	4.36 ± 4.36	-	1.34 ± 1.34
Hibiscus syriacus L.	Hs	-	-	0.6 ± 0.62
Indigofera gerardiana Graham ex Baker	Ig	-	-	3.65 ± 2.86
Indigofera heterantha Wall. ex Brandis	Ih	8.93 ± 8.93	-	-
Isodon rugosus (Wall. ex Benth.) Codd	Ir	-	8.2 ± 8.2	-
justica adathoda L.	Ja	-	-	1.96 ± 1.45
Lamium album L.	La	-	-	0.53 ± 0.53
Lespedeza juncea (L. f.) Pers.	Lj	5.02 ± 5.02	-	2.91 ± 2.91
Lotus corniculatus L.	Lc	2.5 ± 2.5	-	2.52 ± 1.57
Malva neglecta Wallr.	Mn	-	-	1.06 ± 1.06
Medicago lupulina L.	Ml	-	-	1.21 ± 1.21
Myrsine africana L.	Ma	4.5 ± 3.7	1.85 ± 1.85	5.96 ± 2.53
Origanum vulgare L.	Ov	-	2.6 ± 2.68	-
Otostegia limbata (Benth.) Boiss	Ol	-	-	1.60 ± 1.13
Oxytropis mollis Royle ex Benth.	Om	3.63 ± 3.63	1.71 ± 1.71	3.36 ± 1.97
Plantago lanceolata L.	Pl	2.3 ± 2.3	9.34 ± 6.15	6.05 ± 3.29
Plantago major L.	Pm	-	5.9 ± 4	0.46 ± 0.46
Polygonatum verticillatum (L.) All.	Pv	-	1.71 ± 1.71	-
Pteris cretica L.	Pc	-	-	1.67 ± 1.18
Ranunculus laetus Salisb.	Rl	-	-	0.93 ± 0.93
Rosa webbiana Wall. ex Royle	Rw	-	-	-
Rumex denticulatus K. Koch	Ru	-	-	4.49 ± 2.57
Rumex nepalensis Spreng.	Rn	-	3.57 ± 3.57	0.53 ± 0.53
Sarcococca saligna (D.Don) Müll.Arg.	Ss	-	-	0.53 ± 0.53
Smilax lanceolata L.	Sl	1 ± 12	-	3.08 ± 1.79
Solidago virgaurea L.	Sv	-	-	0.53 ± 0.53
Tagetes minuta L.	Tg	5.02 ± 5.02	-	1.21 ± 1.21
Taraxacumofficinale (L.) Weber ex F.H.Wigg.	To	-	-	0.6 ± 0.61
Trifolium repens L.	Tr	-	-	1.36 ± 0.95
Viburnum grandiflorum Wall. ex DC.	Vg	-	-	2.28 ± 2.28
Violaserpens Wall. ex Ging.	Vs	-	-	1.35 ± 0.94
Youngia japonica (L.) DC.	Yj	-	2.7 ± 2.72	-
Zanthoxylum armatum DC.	Za	-	-	1.06 ± 1.06

Note: - Represent absence of specie in particular group.

). Among 58 species, 43 plant species were found in association with group III, clearly indicating the species’ wide sociability in the studied oak-dominated forests. Moreover, the analysis reveals that IVI of Berberis lycium Royle (7.82), Convolvulus arvensis L. (6.87), and Asplenium trichomanes L. (6.6) were the most abundant species in group III. Group II was recorded intermediately diverse in understory species having 25 plant species with Calamintha vulgaris (L.) H. Karst., Dryopteris stewartii Fraser-Jenk., and Plantago lanceolata L. as the dominant associates having IVI of 11.5, 9.34, and 9.86, respectively. The results revealed group I as the less diverse in terms of species (18 species) and B. lycium having IVI of 9.86, A. trichomanes (9.14), and Indigofera heterantha Wall. ex Brandis. (8.93) were the most abundant species.

3.4. Understory Species Diversity and Richness

Significant differences in species richness (p < 0.015), Shannon–Wiener, Margalef (p < 0.00076), Simpson (p < 0.040) and Beta indexes were detected among three major groups of the understory layer of oak-dominated forests (

Table 3. Species richness and diversity indexes of oak-dominated forests of Swat, Pakistan.

Indexes	Group I	Group II	Group III	F	p
	Mean ± SE	Mean ± SE	Mean ± SE
Total number of Species (S)	5.87 ± 0.66 b	4 ± 0.37 a	10.32 ± 1.5 b	4.85	0.015
Shannon–Wiener Index (H)	1.84 ± 0.19 a	1.3 ± 0.11 a	2.74 ± 0.20 b	13.93	6.97 × 10−5
Pielous Index (J)	0.97 ± 0.01	0.95 ± 0.01	0.93 ± 0.01	2.09	0.14
Margalef Index (M)	2.9 ± 0.33 a	1.4 ± 0.21 a	3.95 ± 0.42 b	9.48	0.00076
Simpson Index (1/D)	7.13 ± 1.10 a	3.68 ± 0.42 b	6 ± 0.61 a	3.61	0.04
Beta diversity (B = S/a)	6.41 ± 1.27 a	3.68 ± 0.43 a	9.85 ± 0.54 b	16.45	2.13 × 10−5

Note: Different superscript letters mean significant variation at p < 0.5 tested by post hoc Tukey HSD; insignificant variations have no superscript.

). Nonetheless, Pielous evenness was found non-significantly different (p > 0.05) among the three major groups of understory vegetation. Group III had the highest species richness (10.3), α-diversity (2.74), and β-diversity (9.85) and the highest value of Margalef index (3.95). On the contrary, the group I had the highest Pielous and Simpson index values with 0.97 and 7.13, respectively (Table 3).

3.5. Influence of Physiographic and Edaphic Variables on the Understory Species Composition

In the ordination analysis, the permutation test revealed that the eigenvalues for the first axis and those for all canonical axes were significant (p < 0.01) and indicated that understory species composition was influenced by physiographic and edaphic variables (Figure 5). All three axes explained 65% of the cumulative variance and 30.8% of the variance in the relationship among understory species composition (

Table 4. Ordination analysis of the understory vegetation concerning physiographic and edaphic variables.

Axis	Axis 1	Axis 2	Axis 3
Eigenvalue	33.27	25.12	23.96
Variance in species data
% of variance explained	12.42	9.41	9
Cumulative % explained	12.41	21.81	30.82
Pearson Correlation Response-Pred. *	1.00	0.98	1
Kendall Correlation Response-Pred.	0.96	0.81	0.99

* Symbol of steric show signidicant of the data.

). Forward selection of the RDA ordination revealed that seven variables (i.e., latitude, elevation, clay, wilting point, bulk density, saturation, and electric conductivity) were the significant influential among all 23 variables in relation to the understory vegetation of oak-dominated forests (

Table 5. Average values (Mean ± stand error) of the environmental variables, i.e., topographic, edaphic and soil in the three community types (vegetation groups) separated by Ward’s agglomerative cluster analysis. The three groups were Group I (Quercus baloot–Pinus roxburghii), Group II (Quercus semecarpifolia–Quercus robur), and Group III (Quercus lanata–Olea ferruginea).

Environmental Variables	Code	Group I	Group II	Group III	F-Value	p-Value
		Mean ± SE	Mean ± SE	Mean ± SE
Latitude (°)	Lat	35.0 ± 0.12 a	34.9694 ± 0.19 b	34.58 ± 0.050 b	5.57	0.009
Longitude (°)	Long	72.28 ± 0.03	72.29 ± 0.023	72.25 ± 0.016	1.72	0.19
Elevation (m)	Elev	2017.75 ± 205.23 a	2289.5 ± 240.22 b	1659.4 ± 64.48 a	4.87	0.015
Slope (°)	Slope	41.8 ± 1.81	41.14 ± 2.55	38.66 ± 1.14	1.15	0.33
Clay (%)	CLY	16.37 ± 1.17 a	17.5 ± 2.2 a	12.11 ± 1.02 b	4.65	0.01
Silt (%)	SLT	24.44 ± 2.20	26 ± 3.6	29.03 ± 2.74	0.66	0.52
Sand (%)	SND	59.18 ± 2.66	56.38 ± 3.42	58.85 ± 3.08	0.17	0.84
pH (1:5)	pH	6.4 ± 0.12	6.6 ± 0.17	6.26 ± 0.13	1.25	0.30
Organic matter (%)	OM	2.21 ± 0.78	4.62 ± 0.76	2.61 ± 0.54	2.92	0.07
Lime (%)	L	3.16 ± 0.52	4.06 ± 0.61	3.61 ± 0.38	0.66	0.52
Nitrogen (mg/Kg)	N	0.12 ± 0.04	0.18 ± 0.03	0.16 ± 0.03	0.503	0.60
Phosphorus (mg/Kg)	P	8.24 ± 0.63	7.44 ± 1.12	12.94 ± 4.46	0.61	0.54
Potassium (mg/Kg)	K	146.25 ± 27.07	233.42 ± 37.52	186.33 ± 26.33	1.55	0.22
Wilting point (mL/gm)	WP	0.114 ± 0.42	0.11 ± 0.09	0.09 ± 0.004	4.36	0.02
Field capacity (mL/gm)	FC	0.22 ± 0.61	0.23 ± 0.09	0.21 ± 0.07	1.31	0.28
Bulk density (g/cm3)	BD	1.4 ± 0.11 a	1.46 ± 0.02 a	1.52 ± 0.01 ab	3.43	0.04
Saturation Point (0 kPa)	SP	0.44 ± 0.52 a	0.44 ± 0.008 a	0.42 ± 0.06 ab	3.41	0.04
Electrical Conductivity (mS/m)	EC	12.41 ± 1.63 a	12.8 ± 0.03 a	23.46 ± 2.80 b	5.44	0.01
Available water (%)	AW	0.10 ± 0.38	0.11 ± 0.005	0.11 ± 0.04	0.46	0.63
Temperature minimum (°C)	Tpmi	14.30 ± 1.92	11.88 ± 0.37	11.89 ± 0.17	2.13	0.14
Temperature maximum (°F)	Tpma	28.98 ± 8.75	27.89 ± 5.33	36.46 ± 5.75	2.11	0.14
Precipitation (mm)	Preci	49.84 ± 1.21	49.16 ± 1.86	51.84 ± 0.97	1.32	0.28
Relative humidity (%)	RH	49.84 ± 1.22	49.16 ± 1.87	51.84 ± 0.98	1.32	0.28
Water Holding capacity (%)	WHC	52.12 ± 1.82	54.82 ± 1.98	53.47 ± 2.89	0.18	0.83

Note: Different superscript letters mean significant variation at p < 0.5 tested by post hoc Tukey HSD; insignificant variations have no superscript.

). Group I (Quercus baloot–Pinus roxburghii community), which lies at an intermediate elevation among the three communities, has lower quantities of organic matter, Lime, Nitrogen and Potassium. In contrast, group II (Quercus semecarpifolia–Quercus rubur), which lies at a higher elevation, has higher quantities of these nutrients. Moreover, in the ordination axes summery, the first axis was significantly associated with total importance value index (r = 0.639), potassium (r = 0.363), and elevation (r = 0.239). The second axis was found in close relation with QIVI (r = −0.414), wilting point (r = −0.374), and clay (r = −0.369). While the third axis was found in close correlation with silt (r = 0.381), lime (r = 0.395), and available water (r = −0.377) (

Table 6. Results of the RDA showed correlation coefficients between the physiographic and edaphic variables and the RDA axes and biplot scores of each variable on all the three axes.

S. No	Variables	Correlation			Biplot Scores
		Axis 1	Axis 2	Axis 3	Axis 1	Axis 2	Axis 3
1	Lat	−0.06	−0.49 **	−0.02	−0.55	−3.94	−0.155
2	Long	−0.19	0.18	−0.22	−1.81	1.48	−1.74
3	E	0.23	−0.39 *	0.06	2.18	−3.09	0.518
4	ASP	0.08	−0.09	0.16	0.74	−0.74	1.25
5	CLY	0.19	−0.36 *	−0.19	1.77	−2.92	−1.48
6	SLT	−0.20	0.29	0.38 *	−1.83	2.28	2.95
7	SND	0.09	−0.09	−0.26	0.85	−0.72	−2.03
8	pH	0.22	−0.22	−0.21	1.97	−1.72	−1.63
9	OM	0.09	0.16	−0.003	0.88	1.26	−0.02
10	L	0.21	0.17	−0.39 **	1.96	1.32	−3.05
11	N	0.22	0.09	−0.02	1.82	0.75	−0.18
12	P	0.007	−0.13	0.20	0.07	−1.06	1.56
13	K	0.36	−0.04	−0.12	3.31	−0.31	−0.96
14	WP	0.18	−0.37	−0.18	1.66	−2.96	−1.39
15	FC	0.03	−0.13	0.09	0.27	−1.05	0.72
16	BD	−0.10	0.30	0.05	−0.93	2.37	0.42
17	SP	0.104	−0.31	−0.05	0.95	−2.38	−0.37
18	EC	−0.121	0.43 **	0.13	−1.11	3.41	1.004
19	AW	−0.176	0.25	0.38	−1.60	1.97	2.92
20	TIVI	−0.63 **	−0.47 **	0.02	−5.83	−3.79	0.17
21	QIVI	0.162	−0.41 **	0.13	1.48	−3.28	0.97
22	BA	−0.025	−0.11	0.22	−0.23	−0.88	1.55
23	D/ha	0.112	−0.22	−0.17	1.02	−1.78	−1.28
24	TpMi	−0.246	−0.09	−0.04	−2.24	−0.77	−0.34
25	TpMa	−0.185	0.01	0.32	−1.68	0.11	2.48
26	Preci	0.259	−0.09	0.03	2.36	−0.71	0.24
27	RH	0.102	−0.05	−0.04	0.935	−0.37	−0.35
28	WHC	−0.273	0.005	0.079	−2.49	0.04	0.61

Note: * (p < 0.05); ** (p < 0.01).

).

4. Discussion

Vegetation is a group of plants growing together in a particular locality [43], and it may also be defined as a unit that retains its unique structure and physiognomic characteristics in an adequately great and sufficient way to permit their distinction from other units [46]. The understory plant species in the oak-dominated forests consist of 58 species belonging to 32 families at 30 different locations in the Hindukush ranges of Swat. The physiognomy of the studied area was dominated by herbaceous growth form with 47 plant species (81%), followed by shrubs with 10 plant species (17%), and trees with individual plant species (2%). Herbaceous dominancy and less tree percentage might be due to altitudinal and geographical variations, which indicates that the climatic factors favor the herbaceous flora. Our findings are associated with researchers in allied, neighboring, and national regions in which the species reported were mostly herbaceous [47,48]. Among all families, Fabaceae was recorded as the leading family with eight plant species, followed by Asteraceae and Lamiaceae with seven species each. For instance, several researchers [5,32,33,34,38,46], for instance [31], have documented Asteraceae and Fabaceae as the two most important families. While other researchers [38] cited Lamiaceae and Rosaceae, and [48] cited Lamiaceae, Moraceae, and Asteraceae, as dominant plant families from Darra Adam Khel, KP, Pakistan.

Plant life-forms and vegetation are indicators of the climate [6]. According to Meher-Homji [49], life-forms are reflective of the bioclimates of an area. Raunkiaer [6] designed three major phytoclimates based on life-form spectra on the earth. It includes phanerophytic climate for the tropics, therophytic for the xeric environments, and hemicryptophytic for the cold temperate region. The present study identified seven different life-form classes in the study area. The most dominant life-form was hemicryptophytes with 19 species (33%), followed by nanophanerophytes with 15 species (26%) and therophytes with 13 species (22%). Hemicryptophytes usually prevail in open physiognomies, while phanerophytes are mainly in closed ones [50,51]. Biological spectra are important in comparing geography and habitats and might change due to biotic influences, viz., grazing, human activities, and climatic changes [52,53,54]. Hemicryptophytes lose their aboveground parts during most summer and winter months while therophytes remain seeded to avoid summer, drought, and cold winter stresses.

Furthermore, the flora of the study area is under anthropogenic pressure in the form of overgrazing and deforestation by nomadic and native people. Nasir and Sultan [55] reported a similar finding from District Chakwal, where therophytes were the dominant biological spectra. A community that therophytes dominate can be a characteristic feature of a highly disturbed area under anthropogenic pressure [48,54]. In addition, [56] reported therophytes as the dominant plant species of the rangeland, district Tank, Pakistan, and [57] assessed that therophytes is the leading life-form of Lahor, District Swabi, Pakistan, showing that the results of the current study are in compliance with the findings of several other studies.

Results of leaf size spectra trait revealed that species with microphyllous leaves dominated the area (with 24 species, 41%), followed by nanophylls (17 species, 29%), leptophylls (11 species, 19%), and mesophyll (5 species, 9%). Nonetheless, 2% with megaphyllous leaf size was also recorded. Microphyllous leaf size class is normally the characteristic feature of meadow plant species, while leptophylls and nanophylls are illustrative of hot deserts [7,54]. Comparable outcomes, where microphylls and nanophylls overwhelmed the vegetation and were reported in the work of researchers [7] and [45], where the authors linked these two leaf size classes with the physiognomies of mild ranges. A high rate of microphylls may be identified with cool atmospheres of the sub-elevated and snowcapped areas. Here, as well, the top layer was less settled, containing a slim sheet that is not conclusive about the entrance of roots. Microphyllous and nanophyllous leaves species were inexhaustible because of the biological variety for this dry condition. The present results concur with [58], who detailed that microphylls and nanophylls predominance in the dry, mild atmosphere of District Quetta. Our inferences additionally uncovered that the high extent in leaf size class changes with increasing altitude, and the rate of microphylls was emphatically connected with this as well [59,60].

Among the 58 species documented, 43 plant species were found in association with group III clustered by applying Ward’s agglomerative clustering, which indicates wide sociability of the species in the studied oak-dominated forests. Moreover, the analysis reveals that B. lycium (7.82), C. arvensis (6.87), and A. trichomanes (6.6) were the most abundant species in group III. Group II was less diverse in terms of understory species numbering in total of 25 plant species. In this group, Calamintha vulgaris (L.) H. Karst., Dryopteris stewartii Fraser-Jenk. and Plantago lanceolata L. were dominant (11.5, 9.34, and 9.86 respectively). Plant species in the understory with varying degrees of resistance to these factors prefer a specific forest type. Since certain plant species are limited to a specific forest type, the extinction of that forest could result in the extinction of certain understory plants [27,61].

According to a recent analysis, the richness and diversity of understory species in oak forests differed greatly depending on canopy dominants. Herbaceous species abundance and α-diversity were highest in the oaks forest, but there was less heterogeneity (β-diversity) across sites and major groups. These findings highlight the importance of oak forests in protecting habitat and plant composition in the understory vegetation. An herb layer with higher species abundance and diversity in oak forests may be due to higher solar radiation on south-facing slopes, which is better for herbaceous species [62,63,64].

Environmental variables such as geographic parameters and physiochemical soil characteristics are important in determining the community structure and composition [65]. In particular, latitude and altitude were found to vary significantly in the understory groups. Clay particles in soil texture and bulk density, saturation point, and electrical conductivity varied significantly in the understory vegetation. Similarly, in the same region [66], the soil parameters in understory vegetation of Pinus wallichiana A.B. Jacks. dominated forest have been studied, revealing that elevation, slope, and phosphorus are vital factors in determining the understory vegetation. An oak species, Q. rubur L., was also reported as a major associated species. Furthermore, many other researchers from other parts of the world have reported that edaphic factors can significantly affect the species composition and species diversity in plant communities [10,66,67,68].

Forward selection of the RDA ordination revealed that latitude, elevation, clay, wilting point, bulk density, saturation, and electric conductivity were significantly influential in determining the understory vegetation of oak-dominated forests. These results follow the findings of many other previous studies [12,59,69,70]. Ali and Begum [62] also investigated the vegetation of Swat and determined congruent results by stating a strong relationship between vegetation and edaphic factors. Many other researchers worldwide have recorded that soil characteristics matter more than canopy organisms in deciding understory vegetation [62,63,64,70]. The study of species composition analysis in the oak-dominated forest revealed a general pattern of variation in the community diversity and species composition. A similar pattern of species composition was also reported by [65] in P. wallachiana understory from the same region. The species diversity was found to increase with time as Beg and Mirza [29] reported only 36 understory species due to the increase in anthropogenic activities.

5. Conclusions

The current study reveals that Swat oak forests have rich floristic diversity with dominance of the therophytic life-form and microphyllous leaf size class, indicating sub-tropical and moist temperate type climates. The species diversity was observed to increase during the spring and summer seasons and decreased later in the autumn. As the winter season approaches, the decline in diversity was observed, associated with dry environmental conditions, slow growth rate, and other climatic factors. Nonetheless, critical impacts of seasonal variation on life-forms and overall species diversity were evident. Among 58 species, 43 plant species were associated with group III clustered by applying Ward’s agglomerative clustering, indicating the wide sociability of the species in the studied oak-dominated forests. The understory vegetation of these forests plays an important role in the forest ecology of the region.

Moreover, the environmental and soil variables, i.e., latitude, elevation, clay percentage, wilting point, electrical conductivity, and potassium (mg/Kg), affect the understory vegetation. In addition, important overstory variables like Quercus IVI and Total IVI of overstory were also found to affect the understory vegetation in oak-dominated forests. Further, it is concluded that the area is vulnerable due to the pressure from the local inhabitants in the form of overgrazing and deforestation, which may significantly affect the understory vegetation. This information is particularly important for the success of efforts intended to prevent the loss of genetic diversity of species within these forests by destroying their natural habitats. Moreover, it is imperative to conserve the area’s biodiversity and provide alternative means of livelihood for the local communities that may allow sustainable utilization and conservation of the invaluable biodiversity of this area for future generations.