the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Forest-floor greenhouse gas fluxes in a subalpine spruce forest: Continuous multi-year measurements, drivers, and budgets
Susanne Burri
Iris Feigenwinter
Mana Gharun
Philip Meier
Nina Buchmann
Abstract. Forest ecosystems play an important role in the global carbon (C) budget by sequestering a large fraction of anthropogenic carbon dioxide (CO2) emissions and by acting as important methane (CH4) sinks. The forest-floor greenhouse gas (GHG; CO2, CH4 and nitrous oxide N2O) flux, i.e., from soil and understory vegetation, is one of the major components to consider when determining the C budget of forests. Although winter fluxes are essential to determine the annual C budget, only very few studies have examined long-term, year-round forest-floor GHG fluxes. Thus, we aimed to i) quantify the seasonal and annual variations of forest-floor GHG fluxes; ii) evaluate their drivers, including the effects of snow cover, timing, and amount of snow melt, and iii) calculate annual budgets of forest-floor GHG fluxes for a subalpine spruce forest in Switzerland. We measured GHG fluxes year-round during four years with four automatic large chambers at the ICOS Class 1 Ecosystem station Davos (CH-Dav). We applied random forest models to investigate environmental drivers and to gap-fill the flux time series. Annual and seasonal forest-floor CO2 emissions responded most strongly to soil temperature and snow depth (2.34±0.20 kg CO2 m-2 yr-1). No response of forest-floor CO2 emissions to leaf area index or photosynthetic photon flux density was observed, suggesting a strong direct control of environmental factors and a weak or even lacking indirect control of canopy biology. Furthermore, the forest-floor was a consistent CH4 sink (-19.1±1.8 g CO2-eq m-2 yr-1), with annual fluxes driven mainly by snow depth. Fluxes during winter were less important for the CO2 budget (6.0–7.3 %), while they contributed substantially to the annual CH4 budget (14.4–18.4 %). N2O fluxes were very low, negligible for the forest-floor GHG budget at our site. In 2022, the warmest year on record with also below-average precipitation at the Davos site, we observed a substantial increase in forest-floor CO2 emissions compared to other years. The mean forest-floor GHG budget indicated emissions of 2317±200 g CO2-eq m-2 yr-1 (mean±standard deviation over four years), with CO2 fluxes dominating and CH4 offsetting a small proportion (0.8 %) of the GHG budget. Due to the relevance of snow cover, we recommend year-round measurements of GHG fluxes with high temporal resolution. In a future with increasing temperatures and less snow cover due to climate change, we expect increased forest-floor CO2 emissions even at this subalpine site, with negative effects on its carbon sink behaviour.
- Preprint
(2133 KB) - Metadata XML
- BibTeX
- EndNote
Luana Krebs et al.
Status: open (until 22 Oct 2023)
-
RC1: 'Comment on egusphere-2023-1852', Andreas Schindlbacher, 15 Sep 2023
reply
General comments:
Luana Krebs and co-authors present an impressive soil GHG flux dataset from a subalpine forest. Multi-year datasets from such ecosystems are rather scarce and therefore definitely deserve publication. CO2, CH4 (and N2O) were measured in high temporal resolution year-round during 4 years and annual CO2equ budgets were calculated showing higher net emission during the warmest year 2022. I have some suggestions to improve the manuscript.
It should be considered removing N2O from the manuscript. Obviously N2O data is available only for 2 years and the extremely low fluxes are only shown in the supplements and not included in the budgets or any other calculations. It appears that the 180 sec chamber closure likely were not long enough for serious N2O flux calculations (extremely low 10th percentile R2 in Table.2) – as mentioned in the manuscript, using such a poor fit to calculate fluxes should be avoided. I don’t know the actual cause. It is unlikely that the subalpine soil does not show any, or not measurable N2O emissions throughout the whole year. It seems more likely that there appeared problems with the laser. I don’t operate one by myself but heard from colleagues that the real measurement accuracy in the field is not really the 0.06 ppb for N2O that is suggested. There might be drift or whatever else. How and how often was the laser calibrated? If you do not have 100% trust in the N2O data, I would take them completely out of the manuscript. Currently you state that there was no soil N2O efflux at this site – are you 100% sure?
Some information should be added to the method section about how the snow in the chamber was treated in flux calculations. The volume of the water(ice) of the snow cover must be subtracted from the camber volume during snow cover. How was that done? Was snow porosity measured or assumed somehow? If the snow volume was not subtracted, it is no wonder that CO2 emissions became the lower the more snow was in the chamber. Or wa sonly the volume above the snow surface used for calculation? (In this case it would be flux from the snow surface, not the forest floor). Just of interest, what had happened in spring? Typically opaque chambers warm up faster and snow melts much earlier in and around them.
It is often stated that high temporal resolution measurements have the advantage to capture hot-moments in GHG fluxes. Well – have such hot moments been observed? Currently it does not really seem so. Did freeze-thaw periods occur? What happened during these periods with CH4 fluxes? It would be important to zoom out some such hot moments from the long-term datasets and show them separately. Even if there was freeze-thaw and no peak in CH4 flux occurred – this could be shown. Were there any CH4 emissions at all, at least short-term? We for instance once observed a small CH4 and huge N2O peak during freeze-thaw in a deciduous forest (Schindlbacher Biogeochemistry 2022) but very similar CH4 pattern with just lower uptake during winter as in your study (Heinzle AgrForMet 2023) in a spruce mountain forest. Another possibility for hot moments is after rain post drought periods. Did you observe any flux peaks of any GHG during such periods? Seems there were some very short term peaks and some longer-term positive CH4 fluxes in the “observed fluxes” in Fig. A.3.
There appeared very high CO2 fluxes during a period in summer 2022 – any idea why? If possible the advantages of the high resolution GHG dataset should be worked out – but probably the advantage is not as great as can be seen from the fact that the simple Q10 driven model produced more or less the same CO2 budgets as the random forest model with a lot more input parameters than soil temperature.
Line Comments:
Title: “Continous” is a bit confusing since there were no measurements done in 2018 and 2019
Intro:
P2 50-60: I would rather not discuss soil warming experiments here. The current study is no soil warming experiment and has no connection. You might better refer to generally increasing soil temperatures (Lembrechts, J. J., (2022) Global Change Biology, 28(9), 3110-3144.) and to the global trend of soil respiration under warming (Jian, Jinshi, et al. "A restructured and updated global soil respiration database (SRDB-V5)." Earth System Science Data 13.2 (2021): 255-267.; Bond-Lamberty, Ben, and Allison Thomson. "Temperature-associated increases in the global soil respiration record." Nature 464.7288 (2010): 579-582.).
Methods:
Calibration of laser?
Tab1 – for snow depth usually rather the maximal depth is provided than the mean depth
P6 140: it is mentioned that 2% of the data were discarded after step 1 but not how many data were discarded after step 2 and 3. Please add.
Discussion:
230: I’d rather write “after snowmelt” instead of late winter
Fig 1 and similar cases: It took me a while until I figured out that the record is not consecutive and that 1.1.2020 follows the 31.12.2017. Please indicate somehow that there is a 2 year gap in the dataset (eg. by a gap between the panels)
300-305: When discussing the budgets terms such as “higher-than-usual” “considerably higher” etc. should be avoided. If you want to make a solid statement that the 2022 fluxes were higher than average, then it would be necessary to apply statistics to the budget data. Otherwise no scientifically valid conclusion can be drawn.
305: delete “exceptionally”
315: There is a bulk of literature that shows that CO2 emissions are depressed under realy dry conditions – please rephrase here (the WFPS in the manuscript are generally rather low, but this is likely a matter of the very low bulk densities, which might have been underestimated a bit?). Anyway, was the soil in summer 2022 really so dry?
Table 4 can be moved into the supplements, or also the results = measured fluxes from all the studies are shown in the table for comparison with the current ones.
Citation: https://doi.org/10.5194/egusphere-2023-1852-RC1 -
RC2: 'Comment on egusphere-2023-1852', Anonymous Referee #2, 29 Sep 2023
reply
General comments
This ms reports high-resolution GHG fluxes from a forest floor in a subalpine coniferous forest using four automated chambers. Such automatic measuring systems are of great scientific interest, because events that occur for a short time can be recorded with them. The GHG measurements are integrated in a network for long-term observations of ecosystem fluxes. Three objectives were defined, but no hypotheses or research questions. The measurement technique of CO2 and CH4 fluxes seems very robust, while the measurement technique for N2O fluxes is obviously critical. Many N2O measurements indicate negative values, a net N2O uptake by the forest floor. Few net N2O uptakes have been reported, but mostly in dry soils during the summer months. I can only speculate that the measurement duration of 180 seconds is too short for the large chamber volume (281 L) or for the height of the chambers (50 cm) at low N2O fluxes. Own measurements in a spruce forest with a different laser technique and a different chamber system showed that the measurement time often required more than 20 min before a significant increase of the N2O concentration could be determined. In this respect, I propose to remove the N2O measurements completely from the manuscript and focus on CO2 and CH4 fluxes. Another problem with respect to the calculation of the GHG budget is the contribution of ground vegetation to CO2 fluxes. Due to the opaque chambers, only the respiration of the vegetation is measured, as it naturally occurs only at night. Thus, CO2 fluxes were overestimated during daylight hours. For a correct GHG budget, however, the CO2 fixation of plants would also have to be recorded. An estimation of the contribution of aboveground plant organs to the CO2 flux would be interesting. Calculating the GHG budget for the forest does not seem justified to me. Overall, a thorough revision of the manuscript is needed. As is usual in scientific papers, clearly formulated research questions or hypotheses, e.g. on the effect of the snowpack, would improve the quality of the ms.
Specific comments
Title: please change the title if N2O fluxes are omitted. 'multi-year' is a bit exaggerated when the fluxes were only measured for 3-4 years.
L 16: please present only means of the annual fluxes.
L 19: provide here the mean CH4 flux, not the CO2 equivalent.
- 19-20: ‘driven mainly by snow depth’ – do you mean that increasing snow depth reduced CH4 uptake? Is the relation between CH4 flux and snow depth significant?
L27-28: ‘with negative effects on its carbon sink behavior` the data don’t show this, please omit the statement.
L54- 58: Experimental soil warming was not investigated in this study, but annual variation of gas fluxes. A more general view at temperature influence would better fit this study.
L 92 (Table 1): provide some data of the forest floor and mineral soil: horizons, thickness, texture, stocks. Does bulk density (5 cm) refer to the mineral soil or forest floor? (see comment below)
L 113: 180 s measuring time - why where chambers closed for 10 min? When where concentrations measured during the 10 min? Please provide the length of the tubing between chamber and detector and the flow rate or pump rate.
L119: Were the chambers closed 16 times per day = 160 min or 11% of daytime? Does this mean that 11% of annual precipitation was also excluded and the forest floor was drier than outside the chambers?
L 155: The installation depth was 5 cm for the SWC sensors. The low bulk density indicates that the sensors were installed in the organic horizon or in the transition from the organic horizon to the mineral A horizon. This is critical because the EC-5 sensors have only a standard calibration, which is often not suitable for many forest soil horizons with high root density or stone fraction. Where the sensors calibrated with the soil from 5 cm depth? While the sensors show nicely the dynamics of the water content, the absolute value is often incorrect. When bulk density changes due to shrinkage and swelling of the forest floor, further uncertainty is added to WFPS. Overall, the WFPS is very low (Fig. 1b), especially after snowmelt where much higher values should be reached.
L 271: this result could be better presented, perhaps by linear/non-linear relationship (decrease in CH4 uptake/cm snow depth)
Discussion
N2O fluxes are not discussed at all.
L 370: how many ‘hot moments’ were identified in this study. One message of this study could be that the effort with automatic measurement systems for these forest types is very large and weekly or bi-weekly measurements with many chambers yield more robust flux rates on a larger spatial scale.
Table 1: Temperature, WFPS and snow cover are presented in Fig. 1. If needed, annual means can be described in the text.
Table 4: Were the fluxes in these studies measured exclusively from forest floors where vegetation had not been removed? If present, the above-ground soil vegetation is very often removed by clipping to measure soil respiration. There are many more long-term studies where GHG fluxes were published in different papers from the same forest site. A table without CO2 and CH4 flow rates is redundant anyway.
Citation: https://doi.org/10.5194/egusphere-2023-1852-RC2
Luana Krebs et al.
Luana Krebs et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 193 | 62 | 11 | 266 | 5 | 5 |
- HTML: 193
- PDF: 62
- XML: 11
- Total: 266
- BibTeX: 5
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1