Dynamic stability and canopy structure drive spatio-temporal variability of greenhouse gas concentrations in the sub-canopy of a temperate spruce forest

Titelangaben

Abdoli, Mohammad ; Lapo, Karl ; Thomas, Christoph:
Dynamic stability and canopy structure drive spatio-temporal variability of greenhouse gas concentrations in the sub-canopy of a temperate spruce forest.
In: Agricultural and Forest Meteorology. Bd. 383 (2026) . - 111129.
ISSN 0168-1923
DOI: https://doi.org/10.1016/j.agrformet.2026.111129

Volltext

Link zum Volltext (externe URL):

Angaben zu Projekten

Abstract

We investigated the micrometeorological controls on spatio-temporal variability of CO2, CH4, and H2O mixing ratios in the horizontal and vertical dimensions within a temperate forest canopy using high-resolution measurements from sensor networks and machine learning techniques. The main objective was to identify the governing factors controlling scalar gas variability within the sub-canopy and to assess how static stability and thermal stratification influence scalar mixing processes under varying wind regimes. Scalar variability was characterized in both the vertical and horizontal across weak- and strong-wind regimes, defined using sub- and above-canopy turbulence statistics. A Random Forest model was employed to identify key drivers of scalar variability showing that in the sub-canopy it is primarily controlled by shear-induced turbulence during strong winds and by buoyancy during weak winds. Under strong wind conditions, CO2 variability is mainly influenced by TKE and dynamic stability, whereas air temperature becomes the key driver under weak-wind conditions. For CH4, temperature remains an important factor across all conditions, with wind shear playing a role during strong winds and horizontal advection influencing its variability under weak winds. H2O variability is closely tied to temperature in all wind regimes, reflecting evapotranspiration dynamics and energy distribution. Additionally, vertical profiles of potential temperature from Fiber-Optic Distributed Sensing (FODS) were clustered using k-means to classify stratification regimes, showing that unstable temperature profiles promote mixing and reduce variability, whereas stable stratification suppresses mixing and enhances scalar variability. Horizontal and vertical CO2 and CH4 spatial variability could be explained by thermal stratification and horizontal wind, integrating effects of turbulence and radiation, while H2O variability was primarily governed by evapotranspiration dynamics and energy availability, particularly during early morning and late afternoon transition periods. The diurnal course of carbon dioxide flux across the forest revealed a pronounced contrast between above-canopy and sub-canopy with regime-dependent differences in vertical coupling and “flushing” events that enhance subcanopy CO₂ transport under weakly turbulent conditions. Our results emphasize the critical role of the non-linear interplay of turbulence, static stability, canopy structure, and enthalpie inputs in shaping fine-scale greenhouse gas mixing and transport. They also demonstrate the effectiveness of dense sensor networks combined with machine learning methods to enhance process-level understanding of gas exchange dynamics within forest sub-canopies.