the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Feb 2026
Drought responses of a Norway spruce forest on drained peat soil: combining sap-flow sensors, eddy-covariance, soil and UAV data
Abstract. The summer of 2021 brought a severe drought to southern Finland. We explored how a drained boreal peatland forest responds to drought by combining a wide range of in situ monitoring and Unmanned Aerial Vehicle (UAV) remote sensing. A dataset combining eddy-covariance (EC) fluxes, sap-flow sensors, UAV mapping, soil and weather data was collected. Spruce stand reaction to drought is examined at sub-daily to seasonal time scales, and its temporal and spatial features are identified separately for a recently thinned Continuous Cover Forestry (CCF) block and a control block. Sap flow data showed that the CCF-harvested block shows greater tree resilience to high vapour pressure deficit (VPD) than the control block, likely due to the higher soil water availability during the rainless and hot period. At the same time, both the control and CCF harvest blocks had notably reduced net ecosystem exchange (NEE) and increased Bowen ratio, particularly on high VPD days. The UAV surveys indicated that trees in the CCF block tend to have higher canopy temperatures and lower Normalized Difference Vegetation Index (NDVI) than in the control block, implying their possible susceptibility to more extreme drought.
- Preprint
(6330 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-568', Anonymous Referee #1, 11 Mar 2026
- RC2: 'Comment on egusphere-2026-568', Prajwal Khanal, 02 Apr 2026
-
RC3: 'Comment on egusphere-2026-568', Anonymous Referee #3, 24 Apr 2026
General comments:
This study has used sapflow data, eddy covariance and UAV measurements to assess the response of a Norway spruce forest to drought over two years (2020 and 2021). In the meantime, the forest was affected by thinning. Throughout the paper it is not clear what exactly the research questions are (effect of drought on transpiration or carbon fluxes or both? Effect of thinning on carbon fluxes? How thinning changes response to drought? What is the question?). Because of the poor structure of the paper, it is not clear what new insights are gained here which we did not already have, based on considerably longer timeseries and a greater number of sites. The paper is written very poorly (results are provided often with a qualitative language) and basic statistical analysis is missing. Thus, what the authors present as “response” to drought might not hold at all. Also, I have reservations with how the authors defined drought (see details below). Overall, this was a frustrating experience reviewing this paper. I do not believe that the study is mature enough for publication in Biogeosciences. I have listed my reasoning below.
Title: Rather than listing the methods, the title should indicate what “response” of the forest is the central point of this paper.
The Abstract is written very poorly. It is not clear what exactly the research questions are (not clear response to drought is assessed based on what function?) and it is a summary of methods and some results without an appropriate link made across the methods. Only the results from each of the observation methods are reported. Judging by the Abstract alone, there is no novel finding or mechanistic understanding provided by the study. It is not clear to me what this paper adds to the existing body of literature.
Unfortunately, still by the end of the Introduction the objectives are not clear (is it management effect? Why is not the relevant literature assessed for this topic?) The Introduction is a collection of points without a clear link to form the overall objective of the study and the motivation of the study is lacking.
Not enough details on EC flux calculation are given: e.g., what frequency are the high-resolution data, raw data filtering approaches and checks, coordinate rotation, time-lag detection etc.
I understand that the study employs multiple different observation methods but rather than focusing on the method would be good to focus on the ecosystem properties these methods were used to assess. E.g., in Line 22-25: combining monitoring for what factor? Dataset of what?
Is the research question about soil (highlighted in the abstract) effect on drought response? Or on management (thinning) effect on drought response? What is the question here? Not clear. The Introduction starts with the topic of management so maybe this is the key question? We don’t know.
There is also a major issue with how “drought” in the soil is extracted. The authors have used absolute soil moisture content as a threshold for drought, and not even based on the local observations but based on another study! Soil water limitation depends on water potential, NOT water content and this link is highly dependent on soil texture. The studies that are used to determine the soil water content threshold for drought are on mineral soil and not relevant for this forest.
Results are presented with qualitative language: ample, remarkable. They don’t provide sufficient comparison and a statistical test to show if the responses are at all significantly different is completely missing from the paper. Bare minimum standard writing and analysis for a scientific paper is lacking.
Specific comments:
Abstract should already mention the timeframe of the study.
Line 26. “It’s temporal and spatial features are identified „ temporal features in what? What is meant by spatial features of the forest stand?
Line 32: Are you talking about an increase in temperature and decline in NDVI due to drought, which was higher in CCF plots? Not clear how this comparison reflects the effect of drought on the two blocks.
Line 37-39: the two statements are not linked.
Line 41: What exactly is the example reference supporting?
Line 46: given that low resilience is confirmed for Norway spruce, what is your study adding?
Line 51: the topic of drought identification is not linked to the previous statement.
Line 55: “atmospheric drought”, not air drought
Line 61 and 62 could have been merged
Line 77: provides a buffer rather than protection
Line 76 so the non-dominant trees and understory are growing in the dark?!
Line 85: and? What does this “less competition” mean for their response to drought? What is the link?
Line 87: What is even the point of this sentence here that UAV and RS is helpful or research??
Line 91: this is not true. There are numerous studies using UAV to detect drought effect on a range of structural and physiological functional traits in forests. Also, is this the place to justify a methodology? I am struggling to find the research questions. Yes, there are some background information given here and there (that the resilience of Norway spruce is tested before, effect of tree response on stand response) but what is the exact question in this study that addressed all these functions in a manner that has not been tested before?
Line 98: if the objective was to compare for the effect of management, this topic should be the focus and properly addressed and related literature reviewed for it, and provide the basis for the hypothesis that is put forward.
Line 104: what causes this difficulty? Reading this sentence, I would conclude that you have developed methods that specifically address the challenge of detecting drought response? (a remote sensing-focused analysis). How does using multiple methods that are each very demanding and not available easily, provide a solution to the challenges of studying boreal forests? So again: What is the research question exactly?
Line 107: OK so finally there is a mention what is studied here: carbon, water and energy? This should have been made clear in the Abstract. Sapflow is clear but UAV or EC don’t indicate what entity has been measured and what is the focus of this paper.
Line 23: this is about density. What about the contribution to stand basal area by each species? If inventory data was collected DBH should be available for the trees.
Figure 1: footprint area around the tower should have been shown to link the sapflow measurements with the fluxes
Line 150: Standard flux terminology is ignored. Calculating fluxes is processing, gapfilling and partitioning is referred to as post-processing.
Line 193: nearby means how many m away? Ground level means on the ground!? What variables were measured on the ground?
Section 2.2.2 The location of the GW measurements and soil profile should also have been shown on the map.
Line 220-226: why was soil drought defined based on thresholds of SWC from soil that is not even organic? This undermines the definition of soil drought selected for this analysis.
Line 274: stem growth sensors
Line 284-282 what is the reference that validates this approach? Same as for equation 10.
Figure 2: a map without coordinates is just a shape figure.
Line 387: What supports this statement? When can the reader get this information?
Line 417: How much is “ample”?
Figure 4: ET (or LE) should have been plotted too
Line 455: seasonality in what? Light? Phenology? What is this statement describing and how is it connected to the previous statement
Figure 5 The difference between the lines (mean for each category plus the uncertainty around them) should have been tested to see if the differences are significantly large enough
Line 468: but we don’t know if it was significantly lower?
Line 476: what is meant by mild weather?
Figure 6: why is the daytime 30-min fluxes called flux density?
Figure 6 lower panel: this is describing the drier conditions and should have been mentioned earlier
Figure 7 which of these differences are significant and which not?
Line 506: if it is usual, then references should have been given to support this.
Figure 8 seems like some panels (b, c) are cropped at top (y-axis limit should be increased)
How was resilience defined (the abstract mentions higher resilience of thinned stands to high VPD. Also how do the authors know that this range of VPD is “too high”?
The manuscript should have been checked for language (many grammar mistake that I won’t list here)
Citation: https://doi.org/10.5194/egusphere-2026-568-RC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 197 | 90 | 27 | 314 | 47 | 49 |
- HTML: 197
- PDF: 90
- XML: 27
- Total: 314
- BibTeX: 47
- EndNote: 49
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments
Overall, the paper is suitable for publication in Biogeosciences, as it provides an important overview of the efficacy of various methodological approaches for investigating the responses of boreal forests to drought. However, given the many concepts and datasets presented in this paper, the overall structure needs improvement to provide more clarity for the reader. Some of the ideas presented are not well connected, or require clarification to contextualize them within the paper's premise and the scale of the results. Moreover, because the figures can be very data-heavy, captions and/or legends need to be improved so they can stand alone without requiring embedded supporting text in the manuscript. Nonetheless, the conclusions are clearly drawn and can support further research on the usability of the proposed methods to understand the onset and effects of drought on boreal forest productivity.
Specific comments
The introduction needs some light reorganization to improve the flow of ideas. For example, in lines 51 to 56, the introduction of the Standardized Precipitation Evaporation Index (SPEI) feels disconnected from the ideas presented in the previous paragraph. I understand that this index is used later in the paper to explore regional drought characterization, but it is not well integrated here, nor does it support a strong segue to the next paragraph. A suggestion is to combine the paragraphs from lines 45 to 56 into one and make the connection between Norway spruce responses to VPD clearer in preparation for the next paragraph.
In the introduction, the paper's main goal is first presented: to investigate the drought responses of two adjacent stands in a boreal forest in Finland (lines 98 and 99). However, in line 104, the statement about the difficulty of detecting drought responses in boreal forests serves as the main premise. Given the importance of this premise in the introduction and abstract, I suggest reframing the goals to clarify the paper's main question. If the paper aims to identify the utility of different methodologies to characterize drought and then explain how two adjacent stands behaved differently under drought, a reframing could benefit the reader.
In the methodology section, specifically in “meteorological and soil environmental conditions” (Section 2.2.2), there is mention of datasets that were collected but not incorporated into the discussion. I suggest removing mention of any datasets that are not presented in the results and discussion.
I suggest renaming section 2.2.3, “meteorological drought assessment”, to better represent both the atmospheric and soil drought definitions. Moreover, a clearer distinction between these two definitions is needed, as their use in the results and discussion sections was often confusing (see other comments below). Moreover, this section does not seem to belong in the “field setup” description and would be more appropriate in another section describing derived indices and drought characterization. The same would be more appropriate for Section 2.2.4 on drought proxies. In this section, acronyms need to be more clearly defined and used consistently throughout the paper, and providing more context on why these proxies were chosen would probably help the reader.
The novel approach to measuring tree stomatal conductance (section 2.2.7) also does not appear to be well suited to the “field setup” section. Moreover, this seems like a pivotal contribution of this paper to the literature. It may be best to include it in another section of the methods for physiological stress characterization, for example, or within section 2.2.6 (e.g., 2.2.6.1).
In the results section, I am unclear what is the goal of the monthly SPEI distribution for the entire country (Figure 2c). It is tied to national rainfall patterns, but this is not investigated enough to provide sufficient contextualization for the study site-specific results. I suggest this be reframed for greater clarity or removed entirely. Moreover, Figure 2b needs more clarity: the soil drought points correspond to both atmospheric and soil drought conditions (line 395). The figure should be comprehensible on its own, and the current caption and legend are not sufficiently self-explanatory. Please include a legend for Figure 2a. I assumed that the colors and line patterns corresponded to different blocks and years, as in other figures, but that should be clearer here. Lastly, I understand that Figure 2b is representative of the harvest block, likely due to the placement of the eddy-covariance tower. That is not well described in section 2.2.2, nor in the caption or the text. Make this clearer to the reader in one of these places.
Results for the Bowen Ratio in lines 425-428 are unclear because there is no frame of reference for the Bowen Ratio in the methods section. Please include contextualization for what it means when the Bowen ratio increases or decreases. Contextualization may also be needed for the Canopy Water Stress Index, since the averaged CWSI during drought periods is not presented relative to other values.
In Figure 6a-6d, it is unclear why and how the VPD values were binned. What was the criterion for binning the values? Could a line have displayed that with either shaded standard deviations or a shaded confidence interval? Moreover, the legend is small and hard to read. Please increase the point size of the legend.
Please consider reorganizing section 3.2 for greater clarity. This is a very data-heavy section and can be confusing for the reader. It might be beneficial to discuss ecosystem-level data first, as currently presented in Figures 4 and 6, and then use the block distinctions in Figure 6 to discuss differences between blocks, including Figure 5. The main issue identified here is the discussion of NEE that opens and closes this section, which leads to some repetition; perhaps bringing the two together could facilitate the reader’s comprehension of the data. Moreover, a reorganization may help with the weather contextualization of this section (e.g., drought is often associated with more sunlight, weather conditions for EC measurements, etc.).
Throughout the results section, please consider being clearer about the atmospheric and soil drought. Currently, the term “drought” is used in both contexts with much explanation, which has hindered easier comprehension of the results. Perhaps some context can be introduced in the relevant methods section and reinforced here.
In Figure 7, similar to Figure 6a-6d, it is unclear how the VPD values were binned. Please provide context for the specific criteria used to bin these values.
In lines 511-512, it is noted that the harvest-block trees show higher sap flow than the control trees. Is that the result on average? Is it per range of tree diameter? How are these values contextualized?
In the UAV survey results (section 3.4), inconsistent use of abbreviations and symbols may hinder the reader’s ability to understand the results. Please consider using abbreviations and symbols more consistently, such as when discussing thermal stomatal resistance and stomatal conductance.
The paragraph in lines 585-595 compares the median of the distributions from the tree stomatal conduction to the eddy covariance values, which is not immediately clear from the figure. An indication of the tree medians could be included to facilitate comparison with the EC lines. Moreover, rewording this paragraph may benefit the reader; lines 593-595 were confusing.
In Figure 12, tree height and NDVI values were binned. As with previous figures showing binned VPD values, it is unclear here why or how these variables were binned. Please provide some context.
Again, in the discussion section, there’s confusion about which type of drought the argument concerns (atmospheric or soil). In the first paragraph of this section, for example, it is not immediately clear that changes in vegetation functioning with drought refer to the atmospheric drought. Improving clarity would greatly benefit the dissemination of this paper.
The ideas in the paragraph on lines 654-657 should be distributed across the previous paragraph and the following paragraphs for greater clarity.
In lines 661-665, it is unclear why bring up a previous study finding greater reductions in boreal forest productivity under low soil moisture contents if the last sentence is about the effects of atmospheric drought in the absence of soil drought.
In the results section, it states that soil moisture was measured at 20 cm in addition to 5 cm. Why is that not shown in the paper, and why was it not used to explore additional stress from soil drought further? Moreover, how might the water table depth mediate the soil response to drought stress? How can the decoupling between soil moisture and water table depth help explain the effects of soil drought on boreal forest productivity?
The discussion section follows a strong structure, from ecosystem-level responses to drought and block effects (thinning) to individual tree responses. It ends with a comparison of the EC data to the UAV approach, which is incredibly valuable to the community.
Technical corrections
Line 234: include BR abbreviation in parentheses or brackets (consistent with other abbreviations).
Line 239: include symbol and units for aerodynamic resistance to water vapour transport in parentheses or brackets (consistent with other symbols).
Line 253: include symbol and units for surface conductance in parentheses or brackets (consistent with other symbols).
Line 254: include Canopy Water Stress Index abbreviation in parentheses or brackets (consistent with other abbreviations).
Line 263: include WUE abbreviation in parentheses or brackets (consistent with other abbreviations).
Line 268: include LUE abbreviation in parentheses or brackets (consistent with other abbreviations).
Line 435: Figure 4 needs a title before the description of the specific panels.
Line 467: add a comma after PAR for greater clarity.
Line 474: unsure whether there’s a typo or if the sentence needs more clarification. Unsure what “at no drought on the c shows” means.
Line 481: Figure 6 needs a title before the description of the specific panels.
Line 529: Figure 9 needs a title before the description of the specific panels.
Line 578: Figure 10 needs a title before the description of the specific panels.
Line 669: correct Mirabel et al. (2023) citation.
Line 709: Is Figure 12 the correct figure here? Shouldn’t it be figure 9?