Effect of soil water content on soil respiration sensitivity to temperature (Q10) in a temperate beech forest: overview of data processing from four years of observation with automatic chambers

Soil respiration (R_S) is the primary source of atmospheric carbon dioxide from terrestrial ecosystems. R_S has been shown to respond exponentially to temperature, a relationship summarized by the Q₁₀ parameter, which quantifies the increase in R_S with a 10°C rise in temperature. Q₁₀ depends on both temperature and soil water content, but the latter’s effect on R_S and Q₁₀ remains unclear, especially in temperate forests. Unlike soil temperature, whose influence on R_S is widely accepted, soil water content’s effect is more site-due to multiple factors such as land use, soil and climate characteristics. Continuous, high-frequency field data are needed to improve understanding, but such data are challenging to collect in forest environments.

Here, we question the effect of soil water content on Q₁₀ along the annual cycle in a temperate deciduous beech forest. Do soil water content levels have an impact on Q₁₀ values? Is the relationship between soil water content and Q₁₀ the same throughout the annual cycle? The experimental site is located in the Châtillonnais National Forest Park in the North-Eastern part of France. There, R_S and environmental parameters are measured hourly using 4 automated chambers (LI-8100, LI-COR) since 2020 (4-year dataset). The forest has been protected from harvesting for over 30 years and is classified as an integral biological reserve.

The dataset includes over 145,000 R_S measurements with about 6.1% missing data. After quality control and outlier removal, 92.7% of the data are used for analysis. For each hour, the mean R_S is calculated by fitting exponential and linear models, with the best model selected based on AIC (ResChamberProc package in R developed by Wutzler T.). R_S values are considered reliable when the coefficient of determination exceeds 0.9. Soil water content, measured near each chamber, shows high temporal consistency but high magnitude spread due to heterogenous soil conditions. To standardize, values are normalized by dividing each by the maximum recorded value, creating bins from 0 (dry) to 1 (wet).

We found that Q₁₀ can only be calculated for spring and autumn (2.71 and 3.12 respectively) in our study site. During winter, low temperatures prevent meaningful Q₁₀ calculation, while in summer, dry soil conditions limit results. A threshold analysis revealed that soil water content positively affects Q₁₀ when it exceeds 40% of the maximum value. This indicates that R_S is much more temperature-sensitive in wet than dry soils.