Soil carbon fluxes in dominant forest ecosystems along elevation gradient in the Western Carpathians

Undisturbed forest ecosystems serve as crucial natural carbon (C) sinks, where photosynthesis typically exceeds carbon losses from respiration and leaching. However, the ability of these ecosystems to sequester carbon is closely tied to soil C dynamics. Even small changes in forest soil carbon pools can significantly impact atmospheric CO₂ levels. Understanding soil respiration, a major carbon release pathway from terrestrial plant ecosystems, is crucial; however, it remains poorly studied in the Western Carpathians, particularly regarding its response to increasing temperatures projected as the new climate norm. To address these knowledge gaps, we conducted research at four Long-Term Ecological Research (LTER) sites within Slovakia, representing distinct forest ecosystems along an elevation gradient (~300 to 1200 m a.s.l.). These ecosystems include oak ( Quercus cerris and Quercus robur ), beech ( Fagus sylvatica ), mixed (fir-spruce-beech) ( Abies alba – Picea abies – Fagus sylvatica ), and spruce ( Picea abies ) forests. At each site, we established a 10 × 10 m grid across two replicated plots (0.81 ha each), totaling 100 points per replicate. To account for within-plot variability, we stratified each plot into three sections based on Rs autocorrelation, assessed during a preliminary measurement campaign on all 100 points. Measurement points were then assigned proportionally to stratum size, selecting 20 points in total per plot. Additionally, separate trenched plots were established in each stratum, with three measurement points per trenched plot, to isolate the heterotrophic component of Rs by severing roots and preventing regrowth. Rs and its heterotrophic component were measured biweekly during the vegetation period (April–October) in 2023 and 2024 using a custom-built chamber equipped with a Vaisala GMP252 probe (Vaisala Oyj, Vantaa, Finland). Soil respiration was measured alternately between two plots every second week, totaling 29 (20 + 9) measurements per site in one campaign. Additionally, in 2024, we conducted an open-top greenhouse experiment to assess Rs responses to warming. Data were analyzed using generalized linear mixed models (GLMM) implemented in the lme4 package in R to compare respiration rates across sites and conditions. This study aimed to: (a) quantify soil CO₂ efflux and determine the ratio of autotrophic (Ra) to heterotrophic (Rh) respiration across four forest ecosystems during the 2023 growing season; (b) examine the relationship between Rs and soil temperature, soil moisture, and soil organic C content; and (c) assess the response of Rs and its components to experimental warming in 2024. Results from 2023 revealed significantly higher Rs rates in the spruce forest ecosystem (mean growing season Rs = 0.634 gCO₂ m⁻² h⁻¹) compared to other sites (mean growing season Rs = 0.413–0.500 gCO₂ m⁻² h⁻¹). Rs showed a weak correlation with soil temperature and moisture. Interestingly, trenched plots exhibited higher soil moisture levels, introducing confounding effects that complicated the separation of Ra and Rh. The 2024 warming experiment showed varying responses among ecosystems, with the strongest Rs increases observed in oak (92%) and mixed forests (66%), while the spruce forest showed the least response to warming (7%). Additionally, a significant positive correlation (R² = 0.89) was found between Rs and organic C content in the humus layer in 2023. These findings indicate that different forest ecosystems release varying amounts of CO₂ into the atmosphere, with heterotrophic respiration playing a more substantial role in ecosystems with higher organic C content. Furthermore, results highlight the variability in forest ecosystem responses to climate warming, with implications for carbon cycle feedbacks under future climate scenarios. Further research is needed to disentangle the mechanisms driving these responses and assess long-term impacts.