Mountain biodiversity and ecosystem functions: interplay between geology and contemporary environments

It has long been a core goal in ecology to disentangle the mechanisms underlying spatial and temporal distributions of biodiversity and, in turn, ecosystem functions. Biodiversity is not the sole or even the primary driver of ecosystem functions; rather, both biodiversity and ecosystem functioning are commonly driven by contemporary environmental factors such as climate and local biotic and abiotic attributes. Biodiversity can also be shaped by long-term drivers, including geological processes that leave lasting legacies on contemporary environments. This link may be especially strong in mountain regions, where recent work has shown effects of erosion and soil heterogeneity on terrestrial tetrapod diversity. However, effects of geological processes on microbial communities and ecosystem functions, and their interactions with contemporary environments, have rarely been assessed. To better understand underlying mechanisms, the authors propose an integrative framework that combines contemporary environments (e.g., climate, local soil conditions, biotic attributes) with long-term geological processes (e.g., parent rock type, weathering) along mountainsides. Mountain ranges created and modified by tectonics have boundaries such as sutures and faults and exhibit predictable elevational trends in climate and abiotic conditions that drive plant and animal zonation. The Himalayan-Tibetan orogeny, formed by the collision of Indo-Malaysia and Eastern-Asia plates, contains many major faults, spans most major ecosystem types, and supports high biodiversity and ecosystem functioning (e.g., the South-East Tibet biodiversity hotspot). Elevational gradients thus provide natural laboratories to test hypotheses about biodiversity patterns and their links to ecosystem functions. This study asks how biodiversity and ecosystem functions vary along elevation, what the main drivers are (contemporary environments versus geological processes), and how geological processes directly and indirectly affect biodiversity and ecosystem functions. Along a 700–3760 m gradient on Galongla Mountain (South-East Tibetan Plateau), spanning multiple faults and contrasting rock formations, the authors quantified plant and soil bacterial communities and 38 ecosystem functions across 180 plots at 18 sites representing tropical monsoon rainforest to frigid shrub meadow. They examined geological variables (parent rock composition, weathering indices) and contemporary environments (climate, soil pH and moisture, biotic attributes), characterized biodiversity (relative abundance, richness, community composition) and ecosystem functioning (individual functions, ecosystem multifunctionality, composition of functions), and evaluated the consistency of elevational biodiversity patterns by comparison to reported bacterial communities on Gongga Mountain. The hypotheses were that contemporary environments (e.g., soil pH, temperature) strongly affect biodiversity and functioning, and that long-term processes (parent rock, weathering) have direct effects and also indirect influences via interactions with contemporary environments and biodiversity.

Study area and design: The study was conducted along a 3000-m elevational gradient (700–3760 m; 29°28′–29°75′ N, 95°20′–95°71′ E) on Galongla Mountain in Medog County, South-East Tibetan Plateau, China. The gradient crosses the Indus-Yalu suture zone fault and includes contrasting rock formations from the Himalaya and Gangdese terranes. Vegetation zones span tropical monsoon rainforest (<1100 m), subtropical evergreen broadleaved forests (1100–2000 m), subtropical evergreen and semi-evergreen broadleaved forests (2000–2500 m), temperate mixed conifer-broadleaf forests (2500–3000 m), frigid-temperate coniferous forests (3000–3700 m), and frigid shrub meadows (>3700 m). Sampling design: In July–August 2012, 18 elevational sites were established, each with ten plots. Tree plots were 10 m × 10 m; shrub plots were 5 m × 5 m. Across all sites, this yielded 180 plots. Plant surveys recorded taxa (species to family level), individuals, density, coverage, and height. Plant importance values were computed as the average of relative density, coverage, and height. Plants were categorized as trees (fir, hardwood, softwood), shrubs, and herbs. Plant biomass was estimated using published allometric equations: fir biomass = (0.4642×V + 47.4990)/100; hardwood biomass = (0.6573×V^1.0502)/100; softwood biomass = (2.1529×V^0.6085)/100; shrub biomass = (0.0398×height×100 – 0.3326)×coverage/25; herb biomass = (0.0175×height×100 – 0.2888)×coverage, with V = 100×height×coverage. Soil sampling: At each plot, 25 soil cores (0–10 cm depth, 5 cm diameter) were randomly collected and composited into one sample. This yielded ten composite soil samples per site and 180 composite samples overall. Subsamples were sieved (2 mm) and stored at 4 °C for physicochemical and enzyme analyses or at −80 °C for organic chemical and molecular analyses. Climate and soil variables: Mean annual temperature was predicted using a linear model with latitude, longitude, and elevation based on local meteorological stations. Mean annual precipitation was obtained by integrating CHIRPS data with local station data via least-squares polynomial fitting with elevation. Soil variables measured included soil moisture, pH, total nitrogen (TN), total phosphorus (TP), total organic carbon (TOC), water-soluble soil organic carbon (WSOC), water-soluble soil organic nitrogen (WSON), nitrate (NO3−-N), and ammonium (NH4+-N). Geological variables: Parent rock mineralogy (quartz, plagioclase, K-feldspar, amphibole, muscovite, chlorite) was identified and summarized using PCA; the first two PCA axes (explaining 61.4% variance) represented parent rock composition. Metal elements (Ca, Fe, Mg, Al, K, Na, Mn, Ti) were measured and summarized via PCA; the first two axes (explaining 80.3% variance) indicated geochemical factors. Weathering indices included the Chemical Index of Alteration (CIA) and elemental ratios (Ti/Fe, Ti/Al, Mg/Al, Ca/Al). Measurement details are in Supplementary Information. Microbial biomass and lipids: Phospholipid fatty acids (PLFAs) were extracted and grouped as fungi, bacteria, actinomycetes, and protozoa to estimate microbial biomass and composition. Glycerol dialkyl glycerol tetraethers (GDGTs) were extracted and quantified, including isoprenoid GDGTs (iGDGTs, from Archaea) and branched GDGTs (bGDGTs, from bacteria). Soil enzyme activities measured included β-glucosidase, amylase, invertase, phenol oxidase, and cellulase. Bacterial community profiling: DNA was extracted from 0.5 g soil using FastDNA SPIN Kit for Soil. The 16S rRNA V4 region was amplified with primers 515F/806R and sequenced (Illumina MiSeq, 2×250 bp). Sequences were processed via an in-house Galaxy platform (IEG pipeline). To standardize sampling depth, samples were rarefied to 10,000 sequences for diversity analyses. Biodiversity metrics: For plants, relative abundance was computed for plant types and species present in 5–95% of samples. Species richness (all plants and by type) was calculated. Community composition was assessed using detrended correspondence analysis (DCA); the first two axes represented composition for all plants and by type. For bacteria, relative abundance at phylum level was computed; species richness (OTU counts) was used, supported by strong correlations with Chao1 and phylogenetic diversity (R2 = 0.97 and 0.96). Community composition was represented by DCA axes for the whole bacterial community and for 18 dominant phyla (including Proteobacteria split into classes). A multidiversity (MD) index was computed by averaging standardized species richness of plants, the whole bacterial community, and dominant bacterial phyla. Ecosystem functions: Thirty-eight variables spanning five functional groups were quantified: plant biomass (biomass, individuals, height, coverage for trees/shrubs/herbs), microbial biomass (PLFA groups; bGDGT, iGDGT, GDGT-0, Crenarchaeol), photosynthetic bacteria (relative abundances of Cyanobacteria, Rhodospirillales, Rhodocyclales, Chlorobi), soil nutrients (TOC, TN, TP, WSOC, WSON, NH4+-N, NO3−-N), and enzyme activities. Ecosystem multifunctionality (EMF) was calculated as the average Z-score across functions; EMF was also computed for each functional group. Sensitivity to the number of functions was evaluated via all combinations from 10 to 38 functions with 1000 permutations (R package multifunc). The composition of ecosystem functions was analyzed by DCA (first two axes) for all functions and for subsets (PLFAs, GDGTs, enzymes). Statistical analyses: Elevational relationships were visualized with LOESS. Elevational breakpoints (abrupt changes) for explanatory and response variables were tested using piecewise linear regression (R package Sizer), with bootstrap confidence intervals. Given the Indus-Yalu suture location at 2293–2438 m, candidate breakpoint bands were constrained to 1800–3000 m. Additional analyses included visualization of separation among sites/elevations using gradients of fitted contours and linking DCA ordinations with elevation. Contemporary explanatory variables included climate (temperature, precipitation), local soil properties (pH, moisture), and biotic attributes (diversity and DCA scores of plants and bacteria). Geological variables included parent rock mineralogy and PCA axes, weathering indices and elemental PCA axes. Response variables encompassed plant and bacterial communities and ecosystem functions across the three analytical facets (relative abundance/richness/composition for biota; individual functions/EMF/function composition).

• Biological communities (plants and soil bacteria) and ecosystem functions exhibited consistent elevational breakpoints between 2000 and 2800 m, coinciding with the Indus–Yalu suture zone fault; similar breakpoints were observed for soil bacteria on another mountain range (Gongga) ~1000 km away.
• Primary contemporary determinants of biodiversity and ecosystem functions were mean annual temperature, soil pH, and soil moisture.
• Incorporating geological processes (parent rock, weathering) substantially increased explained variance compared to models without geological variables: +67.9% for plant communities, +35.9% for bacterial communities, and +27.6% for ecosystem functions.
• Geological processes provided additional, distinct links to ecosystem properties: parent rock and weathering had considerable direct effects on biodiversity, while their effects on ecosystem functions were largely indirect, mediated through interactions with biodiversity and contemporary environments.
• The results support an integrated framework in which both contemporary environmental gradients and long-term geological processes structure mountain biodiversity and ecosystem functioning, with tectonic boundaries aligning with biotic and functional discontinuities along elevation.

The study directly addresses how biodiversity and ecosystem functions vary with elevation and disentangles the roles of contemporary versus geological drivers. The detection of consistent elevational breakpoints at 2000–2800 m, aligned with a major tectonic suture, indicates that long-term geological structures can coincide with abrupt biotic and functional transitions. This pattern’s replication for soil bacteria on a distant mountain (Gongga) suggests broader regional consistency, strengthening inference about geological influences. Contemporary environmental factors—temperature, soil pH, and moisture—emerged as dominant immediate controls on both biodiversity and ecosystem functions, aligning with established ecological theory. However, the marked increases in explained variance when geological factors are included demonstrate that parent rock composition and weathering history impose additional constraints and opportunities on biotic assembly and ecosystem functioning. The divergence in pathways—direct effects of geology on biodiversity versus indirect effects on ecosystem functions via biodiversity and contemporary conditions—highlights the need to consider cross-scale interactions. Specifically, geological substrates shape soil chemistry and texture, which filter plant and microbial communities; these communities, in turn, modulate nutrient cycling and other functions, with the resulting functional outcomes also contingent on current climate and soil moisture/pH. Overall, integrating geological processes with contemporary environmental gradients refines our understanding of mountain biodiversity patterns and strengthens predictive capacity for ecosystem functioning across climatic zones, especially near tectonic boundaries where sharp transitions occur.

This study integrates geology with contemporary environmental gradients to explain mountain biodiversity and ecosystem functions along a 3000-m elevational transect on the Tibetan Plateau. It identifies consistent elevational breakpoints in plant and bacterial communities and ecosystem functions that align with a major tectonic suture. Contemporary factors (temperature, soil pH, moisture) are primary determinants, but incorporating geological variables (parent rock, weathering) substantially improves explanatory power. Geological processes directly shape biodiversity and indirectly influence ecosystem functioning through interactions with biodiversity and current environments. Main contributions include: (1) empirical demonstration that tectonic and geological context adds significant explanatory power for biotic and functional patterns; (2) evidence for consistent elevational breakpoints associated with tectonic structures; (3) an integrated analytical framework linking geology, environment, biodiversity, and ecosystem functioning. Future research should: (a) expand functional measurements to include additional key processes (e.g., plant-available phosphorus, carbon fixation, nitrogen fixation); (b) evaluate alternative sequencing data normalization frameworks beyond rarefaction; (c) test generality across more mountain systems and biomes; and (d) develop mechanistic models incorporating parent material and weathering alongside climate and soil chemistry to predict biodiversity and multifunctionality under environmental change.

• Functional coverage: Despite 38 functions measured, some important functions were not included (e.g., plant-available phosphorus, carbon fixation, nitrogen fixation), potentially limiting the comprehensiveness of ecosystem functioning assessment.
• Microbial sequencing normalization: Bacterial diversity analyses relied on rarefaction to 10,000 sequences; while common, alternative approaches (e.g., mixture models) could provide complementary insights.
• Spatial scope: Primary analyses were conducted on a single elevational gradient (Galongla Mountain), with cross-system support limited to comparison with bacterial data from one additional mountain (Gongga). Broader generalization would benefit from more regions and taxa.
• Inference limits: While geological variables improved explained variance, causal pathways among geology, contemporary environments, biodiversity, and functions are complex; some relationships are inferred via statistical associations rather than experimental manipulation.