Altitudinal gradient and its correlation with plant diversity in Daral Valley, Swat in Pakistan using multivariate analysis

Abstract

Background

Biodiversity is facing direct threats due to climate change and anthropogenic disturbance. Daral Valley, Swat is not an exception and was selected based on its remote location and altitudinal variation, encompassing diverse ecosystems from moist temperate forest to alpine region. Starting at 1400 m a.s.l. (above sea level) through moist temperate region up to the alpine region at 5001 m a.s.l.

Methods

A quadrat sampling method was employed, taking 300 quadrats of varying sizes to record the necessary phytosociological data for herb shrub and trees. Importance value indices (IVI) for vegetation were calculated and subjected to ordination techniques, such as canonical correspondence analysis (CCA).

Results

The flora comprised 381 taxa from 224 genera and 81 families. The most represented families being Asteraceae, Rosaceae, and Lamiaceae with 46, 28, and 22 species, respectively. Therophytes were the dominant life form, followed by hemicryptophytes, and geophytes. Among the leaf spectra classes, microphyll was the dominant leaf form followed by nanophyll and mesophyll, respectively. Using Ward’s agglomerative cluster analysis, we identified three floristically and ecologically distinct associations related with different topographic and edaphic variables, viz., alpine zone (Group I, 3582–5001 m), sub alpine zone (Group II, 2900–3580 m) and moist temperate forest (Group III, 1432–2900 m). The calculated values for evenness and species richness were 0.99 (with a variance of 16.0%). The Pearson’s correlation coefficient was 0.99, indicating a significant portion of the data aligned with the ordination axes.

Conclusion

Based on the data, it is evident that the altitudinal gradient in Daral Valley, Swat significantly influences plant diversity, with varying biotic and abiotic stresses impacting local flora at different elevations. Lower altitudes experience higher anthropogenic pressures such as deforestation, overgrazing, and collection for medicinal purposes, while higher altitudes faces climate challenges like temperature fluctuations and snowfall. These stressors collectively threaten the sustainability of plant habitats across the gradient. To ensure the preservation of plant diversity, it is crucial to implement targeted conservation measures that address the specific challenges at each altitudinal zone. This will help mitigate anthropogenic interruptions and promote the long-term sustainability of the valley’s unique ecosystems.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12862-025-02402-x.

Background

A plant community is a group of plant species that form a vital functional unit as a result of seasonal fluctuations and sampling intervals [1–3]. The species diversity of moist temperate forests, alpine, and subalpine zones is a crucial gauge of community stability [4, 5]. In moist forest ecosystems, community organization, taxonomic distribution patterns, and vegetative function are essential ecological components [1, 6, 7]. Understanding the distribution and composition of plant communities over altitudinal gradients in alpine habitats and wet forest environments is therefore crucial [8–11]. Plant community patterns in moist forests are influenced by environmental gradients, including soil and topographic variables [12–14]. Phytosociological studies reveal that ecological elements such as pedology, anthropogenic load, and aspect and slope of the region have a major impact on vegetation patterns [15]. Differential community patterns are established by the interaction of vegetative features, indicator values (predictable richness), and ecological gradients [12, 16]. Indicator species serve as crucial baselines for management choices by tracking changes in plant associations brought about by a variety of environmental and climatic conditions within an ecosystem [7, 17–19].

Elevation is an intricate environmental gradient that helps to examine species diversity and influence the structure and composition of plant communities [14, 20]. Diverse plant communities exhibit increased variety due to variations in elevation, slopes, and other aspects [21, 22]. These gradients work as natural monitoring sites in tropical rainforests, evaluating the functions of the ecosystem and patterns of community dispersion over limited phytogeographical regions [18, 23, 24]. The close connection between ecological parameters and biodiversity is a major problem in the fields of ecological and environmental studies. High-altitude, humid temperate forests in the Hindukush, Karakoram, and Himalaya are especially susceptible to climate change [25, 26]. Elevation is one of the most important ecological variables for tracking species richness and plant community diversification. Elevation gradients strongly influence species richness and community diversity, as widely studied in plant synecology [10, 27–29]. This suggests that the species richness and distribution patterns of plant taxa that increase in elevation often change at different dimensional scales and along different transects in the flora [30, 31].

Within Pakistan’s large mountain ranges of the Hindukush, Karakoram, and Himalaya, moist temperate forests such as Lalkoo Valley are hotspots for biodiversity [32]. Global reports have been made on several phytosociological research conducted on plant communities in wet temperate forests along elevation gradients. Ahmad & Siddiquie [33, 34] are two ground-breaking papers that examined the vegetation-environment combination of blue pine forests in Pakistan moist temperate zones using multivariate statistical analysis. Shaheen [35] recorded conifer species distribution patterns and community structures in the wet temperate forests of the Kashmir Himalayas [26]. The wet temperate forest habitats of India provide ecosystem provisioning services, according to Dhyani & Dhyani [36]. In Swat district, the study area is one of the climate-prone areas where natural and human factors purposefully alter the composition and structure of the vegetation.

Previous studies on altitudinal gradients and plant diversity in the Himalayan and Hindukush regions, including Swat, have primarily focused on broad-scale patterns of species distribution and diversity. However, several critical gaps i.e., limited focus on plant community dynamics, insufficient integration of environmental drivers, anthropogenic impacts, and localized studies, still remained. Therefore, the current research focused on the research question: how does plant species composition, diversity and anthropogenic activities vary along the altitudinal gradient in Daral Valley, Swat?

To properly address the research question, this study formulated the following objectives.

1. To assess the variation in plant species composition, richness, and diversity along the altitudinal gradient in Daral Valley, Swat;

2. To evaluate the influence of temperature and edaphic factors (soil pH, organic matter, texture, and nutrient content) on plant community dynamics across different altitudinal zones; and.

3. To quantify the impact of anthropogenic activities (deforestation, overgrazing, and medicinal plant collection) on plant diversity and community structure at various elevations. To identify dominant plant species and their ecological roles in the moist temperate forest, subalpine, and alpine zones.

This study aims to contribute a deeper understanding of plant community dynamics in relation to altitudinal gradients and environmental factors in Daral Valley, Swat, Pakistan.

Materials and methods

Study area

Daral Valley, located in the northwest of Swat district within the Hindu Kush Mountain range, spans elevations from 1400 m to its highest peak, Chambargahai, at 5001 m above sea level (a.s.l.) (Fig. 1). The valley is positioned between latitudes 33°19′ to 36°46′N and longitudes 70°10′ to 72°30′E [37]. It comprises ten large villages and twelve small hamlets, housing approximately 15,000 inhabitants [37]. The region experiences minimal to no monsoon influence due to its high peaks, which significantly impact the environment. Precipitation during winter and spring is primarily in the form of snow, supporting alpine, sub-alpine, and moist temperate vegetation [38].

Fig. 1.

Study area, Daral, Swat, Pakistan

Daral Valley’s diverse altitudinal gradient, ranging from 1400 to 5001 m, creates a mosaic of ecosystems, each shaped by distinct climatic conditions, soil properties, and slope aspects. At lower elevations (1400–2500 m), dense forests rich in medicinal plants thrive in loamy soils with moderate organic matter. Mid-elevations (2500–3500 m) feature mixed coniferous and shrub vegetation adapted to cooler climates and rocky, well-drained soils. Higher elevations (3500–5001 m) are dominated by hardy, cold-adapted alpine species growing in shallow, acidic soils with limited nutrients.

The valley’s slope aspects further influence vegetation patterns i.e., south-facing slopes are warmer and drier, supporting xerophytic vegetation, while north-facing slopes are cooler and moister, favoring shade-tolerant species like conifers and mosses. Valley floors, with their fertile alluvial soils, support dense vegetation and agricultural activities. This interplay of natural factors i.e., altitude, climate, soil, and slope make Daral Valley a unique and ecologically significant region, showcasing remarkable biodiversity and highlighting the importance of understanding human impacts on its fragile ecosystems.

The slope aspect of Daral Valley varies across its altitudinal gradient, influencing microclimatic conditions and vegetation patterns:

• South-Facing Slopes:

  • i.
    Receive more sunlight, leading to warmer and drier conditions.
  • ii.
    Support xerophytic vegetation and are more prone to erosion and overgrazing.

• North-Facing Slopes:

  • i.
    Receive less direct sunlight, resulting in cooler and moister conditions.
  • ii.
    Favor the growth of shade-tolerant species, including conifers and mosses.

• Valley floors

Generally, have alluvial soils with higher moisture retention, supporting dense vegetation and agricultural activities.

Systematic floral surveys

Systematic floral surveys were conducted from 2019 to 2021 to analyze the species diversity in the study area. Semi structural questionnaires and interviews were used as a tool to collect the data about the decline of vegetation [39]. The respondents for the semi-structured questionnaires and interviews were primarily local inhabitants of Daral Valley, including farmers, herders, traditional healers, and elderly community members who have extensive knowledge of the local flora and its historical changes [39]. These individuals were selected based on their long-term residency in the area and their direct interaction with the natural environment, making them key informants for understanding vegetation decline and its causes. A purposive sampling technique was employed to select respondents. This method was chosen to ensure that individuals with relevant knowledge and experience were included in the study. The sampling was stratified to cover different altitudinal zones (moist temperate forest, subalpine, and alpine) and various villages within Daral Valley to capture a comprehensive understanding of vegetation changes across the study area [25, 37]. The semi-structured questionnaire was designed to gather both qualitative and quantitative data on vegetation decline, anthropogenic activities, and local perceptions of environmental changes. It included the following sections viz., Demographic Information i.e., age, gender, occupation, and length of residency in the area. Vegetation Knowledge [12, 25] - Local names of plants, their uses (medicinal, fodder, fuelwood, etc.), and observed changes in species abundance over time. Anthropogenic Activities taken questions on deforestation, overgrazing, agricultural expansion, and medicinal plant collection. Environmental Changes i.e., Perceptions of climate change (e.g., temperature, snowfall patterns) and its impact on vegetation. Conservation Practices for sustainable resource use and suggestions for conservation measures [37]. The questionnaire included a mix of closed-ended (e.g., yes/no, multiple-choice) and open-ended questions to allow for detailed responses and insights. Pre-Testing of the questionnaire was pre-tested with a small group of local respondents (n = 10) to ensure clarity, relevance, and comprehensiveness. Feedback from the pre-test was used to refine the wording of questions, remove ambiguities, and adjust the structure to better align with the study objectives [25, 39]. This step was crucial to ensure the reliability and validity of the data collected during the main survey. The respondents were selected purposively to include knowledgeable local inhabitants, and the sampling technique ensured representation across different altitudinal zones and villages [25, 37, 39]. The semi-structured questionnaire was designed to capture detailed information on vegetation decline, anthropogenic activities, and local perceptions, and it was pre-tested to ensure its effectiveness. This approach provided a robust foundation for understanding the drivers of vegetation changes in Daral Valley.

Specimens were collected from the field, and preserved following the techniques outlined by Hussain [39] and were labeled for permanent storage as herbarium collections. The shade dried specimens were deposited to the herbarium at the University of Swat (SWAT), while duplicate specimens were also deposited in other herbaria like Karachi University Herbarium (KUH) and Royal Botanic Gardens, herbarium at KEW (K), UK. These specimens were identified using keys contained in the Flora of Pakistan [40, 41].

The precision and accuracy of the scientific nomenclature was revised using tools developed by medicinal plants naming services (MPNS) KEW, team (MPNS, 2023) which are themselves built upon the botanical reference resources of International Plant Names Index (IPNI 2023) and Plants of the World Online (POWO 2023).

Determination of anthropogenic pressure and diversity indices

The degrees of anthropogenic disturbance induced by medicinal and food purposes, construction accelerated erosion, exploitation by human fuel, deforestation, forest fires, over grazing, seasonal rainfall and snow (climate change), agriculture, traffic density, and landslide were measured using a six-point scale (0–5). A plot that received a score of 0 was regarded as undisturbed, while one that received a score of 5 was considered very disturbed [26, 29]. Thus, 0 indicated no disturbance, 1 indicated that 0–20% of the plot had been disrupted, 2 indicated that 21–40% of the plot had been disturbed, 3 indicated that 41–60% of the plot had been disturbed, 4 indicated that 61–80% of the plot had been disturbed, and 5 indicated that 81–100% of the plot had been disturbed. The degrees of disturbance were graded based on the proportions of the given parameter persisting in disturbed plots, with each kind of disturbance being studied independently in this semiqualitative assessment. The point scale values were derived according to [13] to account for various forms of anthropogenic disturbance.

The species diversity indices were calculated using the following equations following [26, 29].

1

2

3

Where H’ is Shannon–Wiener diversity index; E is Evenness index; 1/D is Simpson’s index; pi is Species proportion; in is Natural logarithm.

Sampling procedure and data collection

The quadrat method of vegetation sampling [39] was employed to record data on the vegetation of the study area. The selection of 300 quadrats across 30 different sites was based on a stratified random sampling strategy to ensure comprehensive coverage of the study area and representation of the three major vegetation zones: moist temperate forest, subalpine, and alpine regions. The study area was divided into three strata based on elevation and vegetation type, with each stratum further subdivided to ensure spatial representation. Within each stratum, 10 sites were randomly selected to minimize bias, and at each site, 10 quadrats were established, totaling 300 quadrats. Sites were chosen to capture variations in slope aspect, soil type, and anthropogenic influence, while also considering accessibility and safety, particularly in the alpine zone’s challenging terrain [37, 39].

The quadrat sizes were chosen based on the growth forms of the plants and established ecological protocols: 1 × 1 m for herbs, 5 × 5 m for shrubs, and 10 × 10 m for trees [41]. These sizes balanced accuracy with fieldwork practicality, ensuring efficient data collection in the steep and rugged terrain of Daral Valley. The stratified random sampling strategy ensured representativeness across all vegetation zones, minimized bias, and provided statistical robustness, enabling meaningful analysis and generalization of findings [39, 41]. This approach ensured robust data collection for analyzing vegetation diversity and structure in Daral Valley.

All the quadrats were taken in different elevations of moist temperate forest (1432–2900 m a.s.l.), subalpine (2900–3400 m a.s.l.), and alpine regions (3400–5001 m a.s.l.) of the study area. Phytosociological studies for each species were calculated from relative frequency, relative density, and relative cover were recorded [39]. The phytosociological attributes were converted into importance value index (IVI) [39]. Moreover, the life form was evaluated use classification of Raunkiaer [42].

F3, D3, and C3 represent the species’ relative frequency, density, and cover.

Soil analysis

Soil samples were collected from a single layer of 1–30 cm depth, as this layer is the most relevant for plant growth and nutrient availability. This approach aligns with standard ecological practices and ensures that the soil data accurately reflects the conditions influencing the vegetation in Daral Valley. This depth was chosen because it represents the root zone for most herbaceous plants, shrubs, and some tree species, making it the most relevant layer for analyzing soil properties that influence plant growth and distribution. The surface Litter was removed, and the samples were mixed before further analysis. Approximately 500 g of soil from each sample were placed in a polyethylene bag, labeled, and transported to the soil laboratory at the Agriculture Research Institute (Mingora North), Swat, Pakistan. In the lab, air-dried soil samples (dried at 25–30 °C) were passed through a 2-mm sieve. The physicochemical characteristics of the soil were then recorded. The pH values of the soil samples were measured in a 1:5 soil-to-water paste using a Dynamic digital pH meter (model Sension, TM 105). Soil organic matter (SOM) was determined using the method described by Jackson [43]. Total nitrogen (N) was measured using the Kjeldahl method as outlined by Bremner [44], and total phosphorus (P) was estimated following the method of Bingham [45].

Multivariate analysis

Ward’s agglomerative clustering was employed to classify the plant community types while canonical correspondence analysis (CCA) was utilized to examine vegetation patterns and species distributions in relation to environmental variables, using PC-ORD software [46]. The importance value index (IVI) was used as species abundance input data [47]. IVI was calculated by summing the relative frequencies, densities and canopy covers of the species. The edaphic and topographic variables used in the CCA analysis included elevation, slope, aspect, clay, silt, sand, pH, organic matter%, lime%, nitrogen%, phosphorus (mg/kg), and potassium (mg/kg).

Data analysis

The data obtained from field work were entered into an Excel spreadsheet (Microsoft Office 2020), for analysis, matrix formulation for classification and ordination data. The data were visualized with means and standard deviation for importance value index (IVI) and environmental variables. The variation between the environmental variables were determined by using one way ANOVA, setting the P-value < 0.5. Floristic cumulative variance percent was used to determine variation between the different communities in terms of floristic composition. The relationship between vegetation and environmental variable were determined using Canonical Correspondence Analysis (CCA), for which two matrices were made one for IVI and other for environmental variables. Semi structural questioner and interview were used to determine the causes of decline vegetation of the study area.

Results

Floristic composition

A total of 381 species represented by 244 genera related to 81 families were documented. Asteraceae is the leading family with 46 taxa, followed by Rosaceae with 28 spp. while Lamiaceae is the third largest family with 22 taxa found in the study area. According to Ward’s agglomerative clustering the vegetation was divided into three groups on the basis of elevation gradient. Group I indicates alpine region; group II indicates subalpine region and group III indicates moist temperate forest. The cumulative variance among the habit groups are herb (28.86%), shrub (81.96%), and tree (173%). The cumulative variance found among the life cycle groups are Perennial (36.97%), Annual (30.04%), and Biennial (34.64%). The recorded cumulative variance of life form among the three group’s viz., chamaephytes (50.68%), hemicryptophyte (10.60%), geophyte (69.17%), therophytes (25.34%), phanerophytes (81.96%), megaphanerophytes (173.20%), hydrophyte (86.60%), and lianas (173.20%) (Fig. 2). The cumulative variance found among leaf spectra of vegetation across the groups i.e., nanophyll (22.56%), microphyll (31.71%), leptophyll (39.62%), mesophyll (62.22%), megaphyll (43.30%) (Table 1).

Fig. 2.

Comparisons of floristic and ecological characters in Group I, II, III

Table 1.

Floristic characters, ecological parameters of group I, II, & III with CV% cumulative variance percentage

S. No	Floristic characters	Ecological parameters	Alpine region, Group I	Subalpine region, Group II	Moist temperate region, Group III	CV%
1	Habit	Herb	115	81	147	28.86
		Shrub	1	9	14	81.96
		Tree	0	0	14	173.20
2	Life cycle	Perennial	78	63	126	36.97
		Annual	36	26	48	30.04
		Biennial	2	1	2	34.64
3	Life Form	Cha	16	5	15	50.68
		HC	35	30	37	10.60
		Geo	15	7	31	69.17
		The	49	37	62	25.34
		Phe	1	9	14	81.96
		MPh	0	0	14	173.20
		Hyd	0	1	1	86.60
		L	0	1	0	173.20
4	Leaf Spectra	NP	30	27	41	22.56
		MI	58	40	77	31.71
		LP	9	8	16	39.62
		MS	18	13	41	62.22
		MG	1	2	1	43.30

Cha Chamaephytes, HC Hemicryptophyte, Geo Geophyte, The Therophyte, Phe Phanerophyte, MPh Megaphanerophyte, Hyd Hydrophyte, L Liyanas, NP Nanophyll, MI Microphyll, LP Leptophyll, MS Mesophyll, MG Megaphyll

Plant diversity and multivariate analysis

Plant community classification clustering describes the unsupervised learning task of partitioning observations into homogenous subsets, or clusters, to uncover subpopulation structure in a dataset. Wards cluster analysis classified 381 plant taxa into three plant associations by using PC-ORD 7 (Fig. 3). These communities concerning associated species of plants with ecological, edaphic and topographic variables.

Fig. 3.

Ward’s agglomerative clustering, classify the plant communities

The vegetation of the study area was classified into three groups on the basis of elevation viz., the elevation of alpine regions was 3400–5001 m, subalpine was 2900–3400 m and the elevation of the moist temperate forest was 1432–2900 m, above sea level (a.s.l.). Furthermore, in each vegetation group, some plants make close association in the given area due to the variation occurring in elevation. These alpine association consists of Aconitum-Anaphalis-Geum community with elevation of 4000 to 5001 m, followed by Bergenia-Androsace-Lomatogonium community, having altitudinal range between 3600 and 4000 m, while Primula-Ranunculus-Campanula community found at an elevation range of 3400–3600 m. Similarly, the Subalpine region consist of Thymus-Parnassia-Epilobium community with altitudinal range between 3200 and 3400 m, Cerastium-Corydalis-Geranium community found at altitudinal range between 3100 and 3200 m, while Juniperus-Sedum-Potentilla community found at the elevation range of 2900 to 3100 m a.s.l. The moist temperate forest association consists of Arisaema-Viburnum-Veronica community found at the elevation range 2300–2900 m, Abies-Pinus-Taxus found at the range of 1700–2300 m, and Juglans-Quercus-Ailanthus community found at an elevation range of 1432–1700 m.

Alpine vegetation analysis (Group I)

This vegetation group comprises 115 species confined to SPIN SAR, study area. This area consists of upper and lower chamber, Daral Dand Lake, Saidgai Lake, Mahnoor Lake and Love Lake. The elevation range of this plant association is between 3200 and 5001 m a.s.l. It occupies a latitudinal 35°19’40"N range between and a longitudinal range between 72°21’59"E. The dominant herb was Bergenia ciliata (Haw.) Sternb. with a high IVI value about (9.66 ± 4.22). The other co-dominant herbaceous species of this group with importance value index IVI were Bergenia stracheyi (Hook.f. & Thomson) Engl. (6.94 ± 5.52) (Fig. 4b), Sibbaldia cuneata Edgew. (6.57 ± 6.49), Sibbaldia procumbens L. (6.06 ± 5.84), Aconitum violaceum Jacquem. ex Stapf (5.63 ± 7.08) (Fig. 4a), Lomatogonium carinthiacum (Wulfen) A.Braun (5.34 ± 4.48), and Gnaphalium stewartii C.B.Clarke ex Hook.f. (5.04 ± 3.17). The single dominant shrub of this group was Rhododendron anthopogon D.Don with (6.22 ± 9.5), IVI. The detail of other species is presented in supplementary materials (Appendix.1). The soils in this association were of sandy-loamy texture with 58.46 ± 4.26% sand, 23.14 ± 1.80% silt and 3.42 ± 0.45% clay. The soils of this association were slightly acidic with a pH of 4.65 ± 0.36. Such soils contained 7.6 ± 0.39% CaCO₃ (lime), 2.6 ± 1.4% organic matter, 4.121 ± 0.45% nitrogen, 0.25 ± 0.04 ppm phosphorus, 7.82 ± 0.52 ppm potassium, 105.8 ± 4.26 ppm and soil electrical conductivity of 0.34 ± 0.05 (Table 2). The alpine communities are subject to Temperature fluctuations, high grazing pressure, collection for medicinal for trade purposes, habitat loss due to climate changes and deforestation pressure.

Fig. 4.

a Aconitum violaceum Jacquem. ex Stapf. b Bergenia stracheyi (Hook.f. & Thomson) Engl

Table 2.

Environmental variables associated with edaphic and topographic factors

Variable	Code	Alpine region	Sub Alpine region	Temperate region	F	P-Value
		Group-1	Group-II	Group-III
Latitude	Lat	35.21 ± 0.04	35.22 ± 0.03	35.21 ± 0.012	0.39	0.67
Longitude	Lon	72.38 ± 0.005	72.37 ± 0.02	72.53 ± 0.019	239.81	1.37×−16
Altitude	Alt	3667 ± 286	3445.9 ± 103	1658.3 ± 170	254.81	6.85×−17
Sand	Sand	58.46 ± 4.26	62.1 ± 2.41	71.72 ± 2.08	2.06	0.148
Silt	Silt	23.14 ± 1.80	21.95 ± 1,31	21.45 ± 2.06	2.06	0.14
Clay	Clay	3.42 ± 0.45	4.19 ± 0.54	4.46 ± 0.45	9.73	0.0008
Organic matter	OM	4.121 ± 0.45	4.655 ± 0.41	3.672 ± 0.61	6.91	0.0042
Nitrogen	N	0.25 ± 0.04	0.223 ± 0.08	0.133 ± 0.03	9.52	0.0009
Phosphorus	P	7.82 ± 0.52	8.52 ± 0.45	6.47 ± 0.41	42.2	1.37×−08
Potassium	K	105.8 ± 4.26	105.9 ± 4.35	90.1 ± 7.07	29.14	3.79×−07
Lime	Lime	7.6 ± 0.39	7.57 ± 0.48	6.43 ± 0.26	29.03	3.92×−07
pH	pH	4.65 ± 0.36	5.02 ± 0.35	5.6 ± 0.51	9.81	0.0007
Electrical conductivity	EC	0.34 ± 0.05	0.33 ± 0.03	0.381 ± 0.03	2.69	0.08

Sub alpine vegetation analysis (Group II)

This association comprised of 104 species confined to sub alpine region of the valley, spread over an altitudinal range of 2900–3400 m. It occupied a range between 35˚11.02’’N of latitude and 72˚25.18’'E of longitude. The association was represented a beautiful Meadows (Atrang) of Union Council kula band linked with the facing slope of spin sar (Union Council), of the study area. The dominant herb in this group with high importance value index IVI values was Thymus linearis Benth. (6.46 ± 9.84), Potentilla subjuga Rydb. (2.71 ± 4.33), and Swertia purpurascens (D.Don) C.B.Clarke (2.62 ± 2.4). While the dominant shrub was Viburnum cotinifolium D.Don (9.8 ± 11.88) and Rosa canina L. (2.78 ± 5.85). The other co-dominant species having with a high IVI were Hylotelephium ewersii (Ledeb.) H.Ohba (8.09 ± 12.98), and Geranium pratense L. (6.54 ± 7.55) (Appendix 1). The soil parameters reveals that the soil type of this association was sandy-loam with 62.1 ± 2.41% sand, 21.95 ± 1.31% silt and 4.19 ± 0.54% clay. The soils of this association were slightly acidic with a pH of 5.02 ± 0.35. Such soils contained 7.57 ± 0.48% CaCO₃ (lime), 2.6 ± 1.4% organic matter, 4.655 ± 0.41% nitrogen, 0.223 ± 0.08 ppm phosphorus, 8.52 ± 0.45 ppm, potassium, 105.9 ± 4.35 ppm and soil electrical conductivity of 0.33 ± 0.03 (Fig. 5a and b; Table 2).

Fig. 5.

a Viburnum cotinifolium D. Don. b Thymus linearis Benth. Sternb

Moist temperate vegetation analysis (Group III)

This association contained 162 species confined to different sub-valleys i.e., Tilba, Daral Dam, Arin, Mali, Oshah, Peyazaki. Shaladar, and Gantar spread over an altitudinal range of 1432–2300 m. Latitudinally too this association occupied northern parts of the higher mountains between 35°13’49.5"N and 72°29’22.0"E. Persicaria hydropiper (L.) Delarbre was the herb with importance value index IVI was (5.38 ± 10.61), followed by Fragaria nubicola (Lindl. ex Hook.f.) Lacaita (2.82 ± 1.39), and Pteridium aquilinum (L.) Kuhn (4 ± 3.26), while the dominant shrubs were Berberis calliobotrys Bien. ex Koehne (5.19 ± 2.66), and Rubus niveus Thunb. (4.48 ± 2.36). The dominant tree in moist temperate forest was Pinus wallichiana A.B.Jacks. (9.03 ± 4.57). The associated species with IVI were Pinus roxburghii Sarg. (8.34 ± 10.32) and Abies pindrow (Royle ex D.Don) Royle (4.44 ± 5.67) (Fig. 6 & Appendix 1). The soils in this association were of loamy texture with 71.72 ± 2.08% sand, 21.45 ± 2.06% silt and 4.46 ± 0.45% clay. The soils of this association were slightly acidic a pH of 5.6 ± 0.51 pH value. Such soils contained 6.43 ± 0.26% CaCO₃ (Lime), 2.6 ± 1.4% organic matter, 3.672 ± 0.61% nitrogen, 0.133 ± 0.03 ppm, phosphorus, 6.47 ± 0.41 ppm, potassium, 90.1 ± 7.07 ppm and soil electrical conductivity of 0.381 ± 0.03 (Table 2).

Fig. 6.

Pinus, Abies and Taxus community in moist temperate forest of study area

Vegetation-Environment correlations

The CCA biplot shows the distribution of thirty phytosociological stands across the environmental gradient. The biplot shows a clear anti-clockwise rotation of stands distribution with demarcation of the stand and environmental variables (Fig. 7).

Fig. 7.

CCA Biplot showing the dominant and codominant species of the site along with its determinants (Environmental variables)

The species-environment correlation explains sufficient cumulative variance i.e., 34.2% where axis contributes highest Eigenvalue (0.99), percentage of variance (16.0) and Pearson’s correlation (0.99), showing bulk of the data on the axis of the ordination axes (Table 3).

Table 3.

Eigen value and percentage of variance of the species-environment correlation data

Factor	Axis 1	Axis 2	Axis 3
Eigen value	0.99	0.77	0.35
% of variance explained	16.0	12.5	5.7
Cumulative % explained	16.0	28.5	34.2
Pearson Correlation, Spp-Envt*	0.99	0.89	0.93

* Show Significance

The first ordination axis has significant positive correlations with longitude (r = 0.968), Sand 72.4% (r = 0.86), Clay 4.4% (r = 0.495), pH 5.9(r = 0.69) and EC 0.458 (r = 0.45), respectively while significant negative correlation was found for altitude (r = −0.977), OM 4.14% (r = −0.529), N 0.207 (r = −0.676), P 9% (r = −0.82) and Lim 7.5% (r = −0.82), respectively, showing its importance in species-environment relation. The second axis explained 12.5% of the floristic variation and can be interpreted as an axis of soil textural and nutrients importance as soil clay and organic matter show significant positive correlation having r-value of 0.52 and 0.42, respectively. In contrast, silt content showed a relatively weak positive relationship (r = 0.29) and potentially favored community establishment in Daral Valley. Canonical Correspondence Analysis Biplot distributing sites based on the effects of Environmental and Soil variables (Table 4).

Table 4.

Correlation and biplot scores of environmental variables

Variable	Correlation			Biplot Scores
	Axis 1	Axis 2	Axis 3	Axis 1	Axis 2	Axis 3
Lat D	−0.090	0.141	0.071	−0.090	0.141	0.071
Lon D	0.968	−0.096	−0.154	0.968	−0.096	−0.154
Alt.	−0.977	−0.075	0.026	−0.977	−0.075	0.026
Sand	0.864	0.226	0.176	0.864	0.226	0.176
Silt	−0.297	−0.294	−0.119	−0.297	−0.294	−0.119
Clay	0.495	0.58	−0.232	0.495	0.58	−0.232
OM	−0.529	0.428	0.082	−0.529	0.428	0.082
N	−0.676	−0.188	0.448	−0.676	−0.188	0.448
P	−0.829	0.354	0.067	−0.829	0.354	0.067
L	−0.827	0.041	−0.07	−0.827	0.041	−0.07
Lime	−0.83	0.005	−0.061	−0.83	0.005	−0.061
pH	0.657	0.312	0.068	0.657	0.312	0.068
EC	0.455	−0.115	0.073	0.455	−0.115	0.073

Anthropogenic pressure on vegetation

The main cause of vegetation loss in the study area were natural and anthropogenic activities. The primary cause were temperature, seasonal rainfall and snow (climate change) 19%. The other causes were collection of plants for medicinal and food purposes (17%). deforestation (16%), over grazing (16%), agriculture (12%), construction (7%), fire activities (5%), soil erosion and land slide (3%), and traffic density (0.24%). Due to these activities the vegetation of the area faces a great natural and anthropogenic pressure. These severe threats were the main cause of decline in flora of the study area (Table 5).

Table 5.

Anthropogenic causes of vegetation decline in the study area

Main causes of vegetation decline site wise
S. No	Factors	Age (30–40)	Age (40–50)	Age (50–60)	(above 70)	%age
1	Medicinal and food purposes	24	15	25	6	17%
2	Construction	6	7	5	9	7%
3	Accelerated erosion	0	3	2	9	3%
4	Exploitation by human Fuel, deforestation	24	18	13	9	16%
5	Fires	7	9	3	1	5%
6	Over grazing	22	16	20	7	16%
7	Seasonal rainfall and snow (climate change)	26	31	16	3	19%
8	Agriculture land	15	16	10	9	12%
9	Traffic density	0	0	1	0	0.24%
10	Landslide	6	5	2	1	3%

The Table 6 provides a comparison of species diversity indices across three groups (Group-I, Group-II, and Group-III), along with statistical results from an analysis of variance (ANOVA). The indices include species richness, the Shannon-Weiner index, Simpson’s index, and the evenness index, each offering insights into different aspects of species diversity.

Table 6.

Diversity indices of the sampling vegetation

Diversity index	Group-I	Group-II	Group-III	F-value	P-value
Species richness	70.2 ± 22.06	62.9 ± 33.92	117.1 ± 47.08	6.73	0.00
Shannon-Weiner index	4.01 ± 0.36	3.69 ± 0.67	4.36 ± 0.61	3.60	0.04
Simpson’s index	49.35 ± 19.54	36.48 ± 20.23	70.73 ± 29.55	5.40	0.01
Evenness Index	0.95 ± 0.02	0.93 ± 0.04	0.94 ± 0.01	2.97	0.07

Species richness, which represents the number of species in each group, is highest in Group-III (117.1 ± 47.08), followed by Group-I (70.2 ± 22.06) and Group-II (62.9 ± 33.92). The significant F-value (6.73) and P-value (0.00) indicate statistically significant differences in species richness among the groups, suggesting that Group-III is notably richer in species compared to the other two groups.

The Shannon-Weiner index, which measures species diversity by combining both species richness and evenness, also shows the highest diversity in Group-III (4.36 ± 0.61), followed by Group-I (4.01 ± 0.36) and Group-II (3.69 ± 0.67). The F-value (3.60) and P-value (0.04) suggest significant differences in diversity among the groups, reinforcing the conclusion that Group-III is the most diverse.

Simpson’s index, which reflects dominance within the community, with higher values indicating greater diversity (lower dominance), further supports this trend. Group-III exhibits the highest diversity (70.73 ± 29.55), followed by Group-I (49.35 ± 19.54) and Group-II (36.48 ± 20.23). The F-value (5.40) and P-value (0.01) indicate significant differences in dominance among the groups, highlighting that Group-III has the lowest dominance and thus the highest diversity.

The evenness index, which measures how evenly individuals are distributed among species, shows that all groups have high evenness values (close to 1), with Group-I being the most even (0.95 ± 0.02). However, the F-value (2.97) and P-value (0.07) suggest no significant differences in evenness among the groups, indicating that the distribution of individuals across species is relatively uniform across all three groups (Table 6).

Moist temperate forest (Group-III), stands out as the most diverse group, with the highest species richness, Shannon-Weiner index, and Simpson’s index values. Group-II, on the other hand, has the lowest diversity across all indices. Group-I shows intermediate diversity but the highest evenness. Significant differences exist in species richness, Shannon-Weiner index, and Simpson’s index among the groups, but no significant differences are observed in evenness. This analysis highlights the varying levels of species diversity and dominance across the three groups, with Group-III being the most diverse and Group-II the least.

Discussion

The vegetation structure, dynamics, and distribution patterns are significantly influenced by environmental factors, with certain plant families showing broad ecological adaptability, particularly in temperate forest microhabitats [10, 29]. Therefore, a multivariate analysis was conducted to examine the distribution and classification of plant communities in the study area. The study area was situated in the junction of three large mountains range, Hindu Kuch, Karakoram Himalayan (HKH) of Pakistan. Key factors influencing this variation include elevation, soil texture, soil organic matter, soil pH, soil electrical conductivity, soil chemical properties, and topography. Among these, elevation is a primary factor affecting community distribution [48, 49]. Systematic classification of plant communities is fundamental in syne ecology and is used for purposes such as conservation, land planning, and data synthesis, as noted by various researchers [9, 50, 51]. However, traditional classification often relies on general species characteristics and personal experience [52, 53]. The methods used in this analysis provide a foundation for future classification procedures.

The study identified 381 vascular plant species, classified into nine community types with distinct diagnostic species in the study area. Variations in dominant species and community composition reflect the influence of environmental and topographic factors. This study aligns with findings from the Himalayas, where Asteraceae and Rosaceae and are dominant families, likely due to their wide distribution across temperate to alpine zone of the northern hemisphere [15, 54, 55].

Species richness and abundance decreased with increasing altitude, a pattern also reported in Kuman, India, and Beer Hills, Haripur, Pakistan [56, 57]. This highlights the negative relationship between altitude, topography, and vegetation diversity, emphasizing the role of environmental gradients in shaping plant communities [56–58]. The decline in species richness at higher altitudes can be attributed to harsher climatic conditions and reduced resource availability. However, the presence of stress-tolerant species thriving at these elevations highlights the role of facilitation in mitigating environmental stress, as emphasized by Callaway et al. (2013). Additionally, species turnover rates across altitudinal gradients suggest that niche differentiation plays a key role in maintaining biodiversity, with species partitioning resources to reduce competition in resource-limited environments [58, 59].

Our findings align with the Stress-Gradient Hypothesis (SGH) Callaway et al., (2013), which posits that facilitative interactions among plants become more dominant as environmental stress increases, such as at higher altitudes [56, 60, 61]. We observed a clear shift from competitive interactions at lower elevations to facilitative interactions at higher altitudes, supporting the idea that harsher conditions, like colder temperatures and reduced resource availability, promote cooperation among species. For example, stress-tolerant species at higher elevations exhibited mutualistic behaviors, enhancing their survival in extreme environments. This pattern underscores the importance of facilitation in structuring plant communities under stress [21, 47].

However, our results also reflect the nuanced perspective introduced by Maestre et al. (2009), who argue that plant interactions are not linear but vary based on species traits, habitat conditions, and ecosystem-specific factors. In our study, mid-elevation zones exhibited the highest species diversity, likely due to a balance between competition and facilitation [17, 59, 62]. This aligns with ecological niche theory, suggesting that species interactions and environmental constraints jointly shape community structure [40]. For instance, moderate stress levels at mid-elevations allow both competitive and facilitative interactions to coexist, creating a more diverse plant community [62].

The importance value index (IVI) was used to measure plant community structure and composition. Species with similar IVI values have comparable population structures [59, 62, 63]. Different IVI values for studied species indicate their varied ecological importance and adaptation to disturbances and local community influences in the study area [14].

Canonical correspondence analysis (CCA) assessed the correlation between plant communities and environmental variables. Influential factors include altitude, clay, silt, sand, pH, organic matter (OM), calcium carbonate (CaCO₃), nitrogen (N), phosphorus (P), potassium (K), and electrical conductivity (EC). Species with high cumulative variance and environmental correlation were notable along axis 1. Key factors included calcium carbonate for the Abies-Pinus-Taxus (APT) community, silt for the Pinus-Verbascum-Veronica (PVV) community, and altitude [26] for the Aconitum-Anaphalis-Geum (AAG) community. Similar outcomes were described by Ahmad et al., [32, 64].

Altitudinal variation significantly impacts plant community structure in mountainous forests, limiting species distribution and community types [19, 65]. Numerous studies have explored species diversity along elevation gradients in mountainous forests [66]. Altitudinal variation significantly impacts plant community structure in mountainous forests, which is a well-established concept in ecological studies [31, 44]. This aligns with the findings of our study in Daral Valley, Swat, but there are also some unique aspects and differences worth discussing. The comparison of how previous studies relate to or differ from the results of our study viz., similarities with previous studies were altitudinal gradient, decrease in species diversity with altitude, and role of environmental factors [62]. Differences with previous studies were localized patterns, anthropogenic influences, phytosociological attributes and soil-plant interactions. This study integrated altitudinal gradients, environmental factors, and anthropogenic influences to provide a holistic understanding of plant community dynamics in Daral Valley.

Indicator species in the study Area Mountains exhibited different altitudinal preferences, such as Aconitum, Bistorta, Geum, Bergenia Lomatogonium, Primula, Campanula were the indicator of alpine regions, followed by Thymus, Parnassia, Epilobium, Cerastium, Juniperus indicating subalpine regions, while Abies, Pinus, Taxus, Quercus, Verbascum and Veronica are the indicator species of moist temperate forest. These species showed significant floristic and structural heterogeneity related to local ecological variables, especially edapadology. Similar elevation ranges for temperate ecosystems have been documented in other studies in the Hindu Kush-Himalayas [22, 40].

The variations in species diversity across altitudinal gradients are well-documented, with many studies reporting a hump-shaped (unimodal) pattern, were diversity peaks at intermediate elevations [62]. This pattern is often attributed to a balance between environmental stress and resource availability at mid-elevations. However, exceptions exist, and the relationship between altitude and diversity can vary depending on geographic and environmental factors [14, 31]. In our study, species richness and diversity followed a unimodal pattern, with mid-elevations supporting the highest diversity due to favorable environmental conditions.

Topographic and environmental factors, such as aspect and habitat type, also significantly influence species distribution and diversity. For example, southern aspects, which receive more sunlight, tend to support higher species richness compared to northern aspects [16, 48, 67]. This is likely due to temperature differences and variations in solar radiation, which affect the soil-plant-atmosphere continuum (SPAC) and create microclimatic conditions that favor certain species. Additionally, habitat heterogeneity plays a crucial role in maintaining biodiversity, as it allows species with different ecological niches to coexist [16, 63].

Conservation implications

The preservation of habitats is critical for maintaining species diversity, particularly for species with narrow habitat preferences, such as those found in riverine or shady, moist habitats. Habitat destruction and fragmentation, often caused by anthropogenic activities, such as by medicinal and food purposes, construction accelerated erosion, exploitation by human fuel extraction, deforestation, forest fires, over grazing, seasonal rainfall and snow (climate change), agriculture, traffic density and landslides can lead to biodiversity loss and the invasion of exotic species [37]. Conservation efforts should focus on protecting these specialized habitats and preventing the spread of invasive species [37, 68].

Our study highlights the importance of altitudinal gradients, topographic factors, and habitat heterogeneity in shaping plant diversity and community structure [26]. These findings provide valuable insights for conservation planning and underscore the need for targeted strategies to protect vulnerable species and ecosystems in the face of climate change. Future research should focus on integrating functional trait analysis, long-term monitoring, and advanced modeling techniques to further understand the ecological dynamics of altitudinal gradients [69, 26, 40].

Conclusion and recommendations

Nine distinct plant communities were identified across the three vegetation zones of Daral Valley: moist temperate forest, subalpine, and alpine. Species richness decreased significantly with altitude, with 162 species in the moist temperate forest (1432–2900 m), 106 in the subalpine zone (2900–3400 m), and 115 in the alpine zone (3400–5001 m). Soil organic matter (SOM) declined from 4.5% in the moist temperate forest to 1.2% in the alpine zone, while soil pH became more acidic, ranging from 6.2 to 5.5. Calcium carbonate content was highest in the subalpine zone (8.5%) and lowest in the alpine zone (2.3%).

Species diversity followed a hump-shaped pattern, peaking in the moist temperate forest and declining sharply in the alpine and subalpine zones, consistent with global altitudinal biodiversity trends. Temperature and precipitation were key climatic drivers, with milder conditions in the moist temperate forest and extreme cold and snowfall in the alpine zone Limiting plant growth. Anthropogenic activities, such as deforestation, overgrazing, and agricultural expansion, were most intense at lower elevations, exacerbating stress on plant communities. Overgrazing and deforestation were particularly severe in the moist temperate forest, with 60% of respondents reporting significant vegetation decline over the past two decades, while soil erosion affected 40% of subalpine sites.

Indicator species identified in the study provide a baseline for monitoring ecological changes and guiding conservation efforts. Restoration plans should focus on rangeland management, reforestation, and soil conservation to mitigate overgrazing and deforestation impacts. Local communities need access to alternative fuel sources, such as solar energy, to reduce reliance on forest resources. Education programs should raise awareness about biodiversity loss and sustainable practices, while policies should prioritize protecting critical habitats, especially in the biodiverse moist temperate forest, which provides essential ecosystem services.

Authors’ contributions

MS and HS conceptualized the study; MS and HA assisted with methodology and project administration; RU and DKYT helped with software, resources and writing—original draft preparation; MY performed validation; MS carried out formal analysis; MFE conducted investigation; KSA curated the data; MY and HA contributed to writing—reviewing and editing; DL took part in visualization; HS and MY participated in supervision and funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

Not applicable.

Data availability

All data generated or analyzed during this study are included in this manuscript.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors approved the manuscript to be published.

Competing interests

The authors declare no competing interests.