Yu et al. (2026) How vertical stand structure shapes transpiration in larch plantations: Implications for the integrated forest-water management

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-01-06
• Authors: Songping Yu, Yanhui Wang, Q. Wang, Zebin Liu, Lihong Xu, Yang Chao, Xin Ma
• DOI: 10.1016/j.agwat.2025.110111

Research Groups

• Key Laboratory of Ecological Environment Evolution and Pollution Control in Mountainous and Rural Areas of Yunnan Province, School of Soil and Water Conservation, Southwest Forestry University, Kunming, China
• Key Laboratory of Forest Ecology and Environment of National Forestry and Grassland Administration, Ecology and Nature Conservation Institute, Chinese Academy of Forestry, Beijing, China
• School of Soil and Water Conservation, Beijing Forestry University, Beijing, China

Short Summary

This study developed and integrated stratified leaf area index (LAI) and transpiration (T) models for larch plantations, revealing how vertical stand structure, site conditions, and environmental factors shape transpiration dynamics. The findings provide a theoretical basis for site- and stand-specific forest-water management, especially under climate change scenarios, by quantifying the required stand density and LAI stratification to maintain target transpiration rates.

Objective

• To elucidate the relationships and driving factors of LAI across tree, shrub, and herb layers.
• To quantify coupled relationships between transpiration and both environmental and vertical stand structural factors across forest layers.
• To identify divergent transpiration responses to stand structural and environmental variations among stands with distinct site/vegetation characteristics.

Study Configuration

• Spatial Scale: Field investigations were conducted in the southern part of the Liupan Mountains, located in the central-western Loess Plateau, northwestern China. 72 temporary plots, each 900 m² (30 m × 30 m), were established across elevations ranging from 2033 m to 2693 m. Three permanent plots, each 900 m² (30 m × 30 m), were established for long-term monitoring.
• Temporal Scale: LAI measurements in temporary plots were conducted between late July and mid-August (when LAI stabilized). Stratified transpiration, meteorological factors, and soil moisture were continuously monitored in permanent plots from May to October in 2021 and 2022 (growing seasons). LAI dynamics in permanent plots were measured at intervals of 7.78 × 10⁵ seconds to 1.21 × 10⁶ seconds (9 to 14 days).

Methodology and Data

• Models used:
  • Stratified LAI models (Eqs. 9-11) based on the Jarvis-Stewart (J-S) model structure, coupling elevation, slope aspect (in radians), stand age, stand density, and upper shading.
  • Modified Transpiration Component (Tc) models (Eqs. 12-14) reflecting the integrated effects of reference evapotranspiration (ETo), relative soil water content (RSWC), and stratified LAI (including shading effects from upper layers).
  • FAO-56 Penman-Monteith equation (Eq. 2) for calculating daily ETo.
• Data sources:
  • Vegetation structure (LAI): LAI-2200c canopy analyser (LI-COR) for tree and shrub layers; LAISmart system for the herb layer.
  • Meteorological factors: ATMOS41 all-in-one sensors (METER) for precipitation, air temperature, relative humidity, solar radiation, and wind speed, logged at 300-second intervals using CR1000X dataloggers.
  • Soil moisture: 5-TE soil moisture, temperature, and electrical conductivity sensors (Decagon) installed at 0.1 m, 0.3 m, and 0.5 m depths within the 0–0.6 m root zone, logged at 300-second intervals using EM50 dataloggers.
  • Transpiration (Sap flow): Two-pin thermal diffusion sap flow probes (Granier) of 0.02 m length were mounted on tree stems (at 1.30 m height) and shrub stems, logged at 300-second intervals using DL2e dataloggers. Sap flux density was converted from mL⋅m⁻²⋅s⁻¹ to m³⋅m⁻²⋅s⁻¹.
  • Transpiration (Herb layer): Pot experiments using dominant herb species (P. aquilinum and C. hancokiana) with daily weight recording (precision ± 0.0001 kg) using a precision electronic balance. Daily herb transpiration was calculated from changes in pot weight (converted from g⋅d⁻¹ to kg⋅s⁻¹).
  • Plot characteristics: Field surveys for stand age, stand density (stems⋅m⁻²), diameter at breast height (DBH, in meters), tree height (in meters), shrub coverage, and herb species biomass (in kg⋅m⁻²).

Main Results

• The developed stratified LAI models effectively captured layer relationships and site-stand influences, showing good simulation accuracy (R² = 0.70–0.78, RMSE = 0.09–0.23).
• Tree layer LAI (LAIt) exhibited a unimodal variation with stand age (peaking at 40 years), a saturating exponential relationship with stand density (stabilizing at 0.1 stems⋅m⁻²), a power-exponential relation with elevation (peaking at 2300 m), and a polynomial relation with slope aspect (minimum at 2.62 rad, maximum at 5.76 rad).
• Shrub layer LAI (LAIs) and herb layer LAI (LAIh) consistently decreased with rising elevation and exponentially with increasing upper shading (LAIt and LAIt+s, respectively).
• Transpiration components (Tt, Ts, Th) displayed unimodal seasonal variation patterns during the growing season, with Tt being the dominant component (65.2%–74.9% of total T), followed by Ts (12.6%–24.5%) and Th (12.5%–21.3%).
• The developed stratified T models, incorporating ETo, RSWC, and stratified LAI (including shading effects), demonstrated high predictive accuracy (R² = 0.73–0.79, NSE = 0.71–0.78, RMSE = 3.47 × 10⁻¹⁰ to 2.08 × 10⁻⁹ m⋅s⁻¹ for components; R² = 0.86, NSE = 0.86, RMSE = 1.39 × 10⁻⁹ m⋅s⁻¹ for total stand T).
• Maintaining a target stand transpiration (e.g., 9.26 × 10⁻⁹ m⋅s⁻¹) required site-specific maximum stand densities (e.g., 0.0706 stems⋅m⁻² at P1, 0.0699 stems⋅m⁻² at P2, and 0.0762 stems⋅m⁻² at P3) due to divergent site and stand attributes.
• Simulations indicated that a longer rotation by managing older forests (beyond 45 years) could save water, as total stand transpiration showed an initial increase of 6.94 × 10⁻¹⁰ m⋅s⁻¹ per 5 years until 45 years, followed by a decrease of 5.79 × 10⁻¹⁰ m⋅s⁻¹ per 5 years.
• Under a climate change scenario (15% ETo rise), maintaining initial transpiration levels required stand density reductions of 31%–32% (current RSWC), 27%–28% (25% reduced RSWC), and 15%–17% (50% reduced RSWC), with optimal LAI stratification varying significantly across plots.

Contributions

• Developed a novel, integrated framework of stratified LAI and transpiration models that mechanistically couples site/vegetation characteristics, LAI, and transpiration across tree, shrub, and herb layers, providing a more realistic representation of vertical stand structure's influence on forest water use.
• Quantified the site-specific requirements for stand density and LAI for water-saving management, highlighting the inadequacy of "one-size-fits-all" approaches and emphasizing the need for localized strategies.
• Provided a robust decision-support tool for integrated forest-water management, particularly for developing adaptive strategies under projected climate change scenarios (e.g., increased ETo and reduced soil water availability).
• Identified optimal stand age for targeted thinning operations (before the 45-year transpiration peak) to reduce excessive water consumption.

Funding

• National Key R & D Program of China (2022YFF0801803 and 2022YFF1300404)
• National Natural Science Foundation of China (42477090 and 32171559)
• Yunnan Fundamental Research Projects (202501AU070071)

Citation

@article{Yu2026How,
  author = {Yu, Songping and Wang, Yanhui and Wang, Q. and Liu, Zebin and Xu, Lihong and Chao, Yang and Ma, Xin},
  title = {How vertical stand structure shapes transpiration in larch plantations: Implications for the integrated forest-water management},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2025.110111},
  url = {https://doi.org/10.1016/j.agwat.2025.110111}
}