| 16 Dec 2025
Strong toluene and p-cymene emission from waterlogged hotspots of a temperate mixed-forest soil
Abstract. Ecosystem processes taking place at hotspots, i.e., at areas of small spatial extent, but at high intensity, might potentially affect the whole system. Due to their great importance for atmospheric chemistry, we studied the exchange of volatile organic compounds (VOCs) from the forest floor of the ECOSENSE forest, a temperate mixed forest, at selected waterlogged sites. These sites were mainly located along skid trails, hence, characterized by soil compaction and strongly reduced drainage of rainwater. In addition, we monitored VOC exchange from undisturbed, well-drained forest floor during summer and the transition from summer to autumn/winter when strong changes in soil moisture occurred. We complemented these field measurements by experiments under controlled conditions in the laboratory, in which soil cores collected from well-drained and waterlogged sites in the forest were exposed to anoxia. Anoxic conditions enormously stimulated emissions of aromatic compounds, particularly of toluene and the aromatic monoterpene p-cymene 50-fold and 20-fold, respectively. Moreover, anoxia also strongly enhanced the emission of the monoterpenes camphene, limonene, δ-3-carene and α-pinene. In the field, the forest floor changed from a sink of toluene and p-cymene to a strong source of these compounds in response to waterlogging. We observed enormously increased toluene and p-cymene emissions from the waterlogged forest floor along skid trails, amounting to over 5000 ng m-2 h-1 and 3000 ng m-2 h-1, respectively. Considering that the share of skid trails of managed forests can be considerable, we further investigated a potential impact of these hotspots of toluene emission at the ecosystem scale. Indeed, ambient atmospheric mixing ratios of toluene in the ECOSENSE forest peaked in autumn when soil moisture levels were increased. The work therefore suggests that the soil of temperate forests might be a yet underestimated source of toluene and p-cymene for the atmosphere. Still, elevated ambient toluene levels co-occurred with litterfall and senescence of European beech leaves which might be an additional source of atmospheric toluene. At periods with high ambient toluene mixing ratios, the well-drained forest floor acted as a significant sink of toluene. Our results underline the importance of the forest floor’s spatial heterogeneity to act as a concurrent sink (main area) and a source (from waterlogged hotspots) of aromatic VOCs.
This manuscript presents a comprehensive field and laboratory study examining how waterlogging-induced oxygen limitation alters soil–atmosphere exchange of volatile organic compounds in a temperate mixed forest. By combining seasonal forest-floor chamber measurements with controlled anoxic incubations of soil cores, the authors show that waterlogged/anoxic soil patches, mainly associated with skid trails, can switch from being a sink to a strong source of aromatic VOCs, particularly toluene and p-cymene. The observed emission rates are unexpectedly high for a temperate forest system and are supported consistently across in situ observations and laboratory experiments. Overall, the manuscript addresses an underexplored source of biogenic aromatic VOCs and provides novel insights that are relevant for both forest biogeochemistry and atmospheric chemistry.
Overall, I would support the publication of the manuscript, but would like the authors to address or note the following comments/concerns:
Line 45-47, BVOC/AVOC ratio needs to be updated.
Line 72-73, ref missing
Line 91, could you please be more specific with examples? What microbial activities?
Line 127, spell out OL and OF
Line 151, did you use any base collars?
Line 156, I am a bit concerned about the large volume of air samples by the adsorption tubes, which may have exceeded the breakthrough volume of the adsorption materials. Check here for the potential breakthrough volume of the compounds you are interested in – https://www.ingenieria-analitica.com/attachment/pdf/print/td20-confirming-sorbent-tube-retention-volumes-and-checking-for-analyte-breakthrough-pdf-417
Section 2.4 I am also curious how many potential VOCs you have detected from the temperate forest soils. Were these compounds you focused on the most emitted ones?
I noticed that the observed toluene emission rates were substantially higher from the waterlogged skid trail soils than from the undisturbed forest floor soils. Could longer-term accumulation of toluene in skid trail soils, for example through repeated deposition from vehicle exhaust, potentially contribute to the elevated emissions observed at these locations, in addition to the proposed biogeochemical mechanisms?
Figures 3 and 4: It may be clearer to present the emission increases on the y-axis as fold changes rather than percentages, as this would make the values easier to read and compare.
In the anoxic incubation experiments, nitrogen was used to establish anoxic conditions. I was wondering how quickly the soils became anoxic after the onset of nitrogen flushing, and whether this transition can be considered effectively instantaneous. If the establishment of anoxia was rapid, I find it somewhat surprising that VOC emission fluxes could increase by several tens of times over such a short timescale, as microbial processes are not generally expected to respond that quickly.
Section 3.3 Ambient toluene mixing ratios show a pronounced seasonal increase, while soil exchange rates indicate net uptake during the same period. How would you reconcile this decoupling between atmospheric concentrations and net soil fluxes? To what extent might changes in canopy structure, phenology, or boundary-layer dynamics during the seasonal transition influence the observed ambient VOC mixing ratios? Elevated ambient toluene concentrations kind of coincide with the period of leaf senescence. Could there be any potential contribution of senescing leaves or fresh litter to the observed atmospheric signal?
Section 3.3 The chamber-based exchange rates integrate fluxes over a two-hour period, whereas ambient mixing ratios reflect cumulative atmospheric conditions. Could differences in temporal integration affect the interpretation of the soil–atmosphere coupling?
Section 3.3 Although waterlogging is proposed as a key driver of enhanced toluene emissions, ambient toluene mixing ratios do not show a clear positive correlation with measured soil moisture. Could this discrepancy be briefly discussed?
For Section 4.3, I found parts of the mechanistic discussion to be somewhat speculative, although valuable and well grounded in the existing literature. Given that the proposed biochemical pathways are inferred rather than directly demonstrated in this study, it may help to more clearly distinguish between interpretations that are directly supported by the data and those that remain more hypothetical.
The discussion suggests that emissions from waterlogged hotspots may influence ecosystem-scale toluene budgets. Given that such hotspots likely occupy only a small fraction of the forest area, might it be appropriate to slightly tone down the strength of this inference, or to more clearly state its limitations?