Unexpected water uptake under drought conditions and thinning treatments in young and overstocked lodgepole pine (<em>Pinus contorta</em>) forests

Ellis, Emory; Guy, Robert; Wei, Xiaohua A.

doi:https://doi.org/10.5194/egusphere-2024-88

Preprints

https://doi.org/10.5194/egusphere-2024-88

Preprints

25 Jan 2024

| 25 Jan 2024

Unexpected water uptake under drought conditions and thinning treatments in young and overstocked lodgepole pine (Pinus contorta) forests

Emory Ellis, Robert Guy, and Xiaohua A. Wei

Abstract. As drought and prolonged water stress become more prevalent in dry regions under climate change, understanding and preserving water resources has become the focal point of many conversations. Forest regeneration after deforestation or disturbance can lead to over-populated juvenile stands with high water demands and low water use efficiency. Forest thinning improves tree health, carbon storage, and water use while decreasing stand demands in arid and semi-arid regions. However, little is known about the impacts of over-population on seasonal variation in depth to water uptake nor the magnitude of the effect of growing season drought conditions on water availability, and existing reports are highly variable by climatic region, species, and thinning intensity. In this study, stable isotope ratios of hydrogen (δ²H) and oxygen (δ¹⁸O) in water collected from soil varying depths and from twigs of lodgepole pine (Pinus contorta) under different degrees of thinning (control: 27,000 stems per ha; moderately thinned: 4,500 stems per ha; heavily thinned: 1,100 stems per ha) over the growing season and analyzed using the MixSIAR Bayesian mixing model to calculate the relative contributions of different water sources in the Okanagan Valley in the interior of British Columbia, Canada. We found that lodgepole pine trees shift their depth to water uptake depending on water availability under drought conditions and rely more heavily on older precipitation events that percolate through the soil profile when shallow soil water becomes less accessible. Interestingly, forest thinning did not cause a significant change in depth to water uptake. Our results support other findings by indicating that although lodgepole pines are drought tolerant and have dimorphic root systems, they cannot shift from deep water sources when shallow water becomes more available at the end of the growing season.

Received: 10 Jan 2024 – Discussion started: 25 Jan 2024

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 907 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (907 KB)

Supplement (130 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

25 Oct 2024

Seasonal shifts in depth-to-water uptake by young thinned and overstocked lodgepole pine (Pinus contorta) forests under drought conditions in the Okanagan Valley, British Columbia, Canada

Emory C. Ellis, Robert D. Guy, and Xiaohua A. Wei

Hydrol. Earth Syst. Sci., 28, 4667–4684, https://doi.org/10.5194/hess-28-4667-2024,https://doi.org/10.5194/hess-28-4667-2024, 2024

Short summary

Emory Ellis, Robert Guy, and Xiaohua A. Wei

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2024-88', Anonymous Referee #1, 04 Apr 2024

I think that the data presented in this paper may allow for a more nuanced message regarding the impact of the thinning treatments on the water use pattern and uptake depths of lodgepole pine in overstocked stands during drought. The data presented in Figure 7 appear to indicate a faster and earlier depletion of topsoil water storage (5-35 cm depth) by trees in the overstocked control stands during late spring and early summer, as well as an earlier shift to deeper water sources utilization (>35 cm depth). In contrast, trees in the thinned stands (T1, T2) continue using a much larger proportion of topsoil water further into the summer, and shift to uptake of deeper water sources later in the summer season. This pattern appears to be particularly evident for the data collected in July, when trees in the unthinned control treatment were already obtaining more than 60% of their water from deeper sources (>35 cm), whereas trees in the thinned T1 and T2 treatments we still obtaining 55-70% of their total water uptake from shallower soil layers (5-35 cm depth). This pattern strongly suggests to me that the thinning treatments were rather effective at delaying the consumption and depletion of the available topsoil (5-35 cm) water pool by the pines during the early stages of the summer drought, compared to the overstocked control stands. In other words, trees in the unthinned control stands were forced to shift to deeper soil water sources earlier than trees in the thinned stands, probably due to faster depletion of the topsoil water pool by the overstocked control stands. I agree with the authors that the water uptake depth of the lodgepole pines was less affected by the thinning treatments later into the summer drought period, although in August the trees in the thinned stands were still using a somewhat greater proportion of topsoil water (5-35 cm) than the unthinned control stands. So I think that the main message highlighted in the summary and conclusions sections ("forest thinning did not cause a significant change in depth to water uptake") is questionable and should be reconsidered and/or significantly modulated after careful re-analysis and re-interpretation of the data.
Also, I would like to see a much more explicit and detailed description of the impacts of the thinning treatments on the continuously measured soil moisture content data (L311-313). I would even recommend including a new figure to better highlight any differences encountered in soil moisture contents by depth among the different treatments/stand densities.
The data presented in Table 1 appear to indicate greater rainfall interception by the tree canopy during late summer/early fall in the overstocked control stands than in the thinned stands, as suggested by the large differences in topsoil water contents (5 cm) at the end of the growing season. Also, please clarify whether this table reports data measured in October, and please note that the standard deviations mentioned in the figure caption are actually missing from all the columns, so please include them. The data presented in this table appear to suggest that mean soil water content across the full soil profile (5-100 cm depth) was slightly higher in the T2 treatment than in the unthinned control treatment at the end of the growing season, which would be a relevant finding if confirmed by appropriate statistical analyses.
In figure 5, it is unclear why data are missing for the upper soil layers in some of the treatments, or why the scale of the Y axis is not uniform across panels. In Figure 7, it is unclear why the values in the Y axis range between 0-1 instead of 0-100, as the axis title refers to percentages. It is also unclear why some of the columns in the central panel (T2) are shorter than all the other columns in the figure.
L453-455: Several studies have shown that trees that have access to relatively shallow groundwater (6.5 m deep at the study site) can support the integrity and functionality of their shallow fine roots and associated mycorrhizal fungi in dry topsoil layers during prolonged drought through internal hydraulic redistribution (Bauerle et al 2016; Querejeta et al 2007), so the assumption made in L454-455 and L508 is perhaps speculative and questionable, in the absence of any direct measurement of root function. Are there any other plausible explanations for the apparent inability of the lodgepole pines to use recent rainwater during the late growing season? Perhaps those late season rainfall events were of insufficient magnitude to recharge the topsoil layer in a physiologically meaningful way? In L510, do you mean that pines are unable to access water made available by late-season rainfalls during the rewetting period?
I think that the strong assertions made in L520-522 and 526-528 of the Conclusions section (and similar statements in the Abstract) should be reconsidered and rewritten after careful examination of the issues raised above.
Some of the references mentioned in the text appear to refer to wet riparian habitats and tree species, and I thus wonder whether they are really directly relevant to this particular study which has a strong focus on dry interior forests of the semiarid Okanagan Valley (e.g. Gibson &Edwards 2002; Liu et al 2015; Maier et al 2019; Sanchez-Perez et al 2008). It might be more appropriate to replace these references by others referring to semiarid pine forests more closely resembling your study system.

L30-31 in Abstract: They cannot shift from deep water sources when shallow water becomes more available at the end of the growing season? Or perhaps they don't need to shift to shallow water because deep water is plentiful and sufficient to meet their transpiration demand (which is presumably lower in October than in midsummer)? Or perhaps these late-season rain events are insufficient to recharge the topsoil layer meaningfully (i.e. the soil matric potential in upper layers may still remain lower than that in deeper layers in October despite these rain events)?

BAUERLE, T.L., RICHARDS, J.H., SMART, D.R. and EISSENSTAT, D.M. (2008), Importance of internal hydraulic redistribution for prolonging the lifespan of roots in dry soil. Plant, Cell & Environment, 31: 177-186.
Querejeta JI, Egerton-Warburton LM, Allen MF (2007) Hydraulic lift may buffer rhizosphere hyphae against the negative effects of severe soil drying in a California Oak savanna. Soil Biology and Biochemistry, 39, 409–417.

Citation: https://doi.org/10.5194/egusphere-2024-88-RC1
- AC1: 'Reply on RC1', Emory Ellis, 27 Apr 2024
  
  Thank you for your response.
  Partition: I agree that the results shown in Figure 7 indicate that the control trees show less dependence on shallow soil water, specifically when compared to the heavily thinned treatment, and this trend is most pronounced in July. I will reexamine the mixing model results and see if any significant pairwise trends can be discussed to strengthen this argument further in our revisions. As for the trends noted in September, I agree that trees in the thinned treatments could use soil water from below the evaporative front more so than the control trees, which had most of their water uptake from deep soil water (> 35 cm) and groundwater. Thank you for your feedback. We will carefully consider these comments when reevaluating the statistical analysis to see if there is a more compelling and nuanced conclusion to present in this paper.
  Soil Moisture: We continuously measured soil moisture at two depths across the treatment plots in block 1. The other values of percent soil water content (for deeper soil samples) were measured using gravimetric methods.
  We would be happy to include a plot of the soil moisture measurements.
  Interception: The control has higher canopy interception compared to the treatments for low-intensity precipitation events (just from visual observations).
  The error bars are only available for samples for which we ran multiple cryogenic distillations. Removing them altogether could be more effective because there are no standard error calculations for all measurements.
  I will conduct and add the appropriate statistical tests comparing soil water content at each depth and add those results.
  I did not notice the missing data in Figure 5 on the initial submission, but I will correct it and add that updated figure.
  L 510 – We will clarify that the late growing season precipitation was likely insufficient for shallow root uptake because of the low precipitation volume and previously dry soils rather than imply that the trees were unable to experience a shift in water uptake from some other limitation or do not need to shift (as you stated in the abstract comment).
  As for the citations, thank you for recommending references more relevant to arid ecosystems. We will review them and consider adding them when appropriate rather than research on riparian ecosystems.
  
  Citation: https://doi.org/10.5194/egusphere-2024-88-AC1
RC2:
'Comment on egusphere-2024-88', Anonymous Referee #2, 23 Apr 2024

The study by Ellis investigates the impact of thinning and seasonal water availability on root water uptake of lodgepole pine trees in an experimental forest in Canada using water stable isotopes. The authors found that the root water uptake of pine trees responded to the seasonal shift in water availability but not to thinning treatment. Overall, the findings are timely and important for similar research in the field, as well as nicely presented and discussed. My comments probably lead only to minor modifications of the manuscript.
Title: The title of the manuscript could be a bit more specific, so that it matches the story of abstract, and the main findings of the manuscript. For instance, it’s not 100% clear what “unexpected” means throughout the manuscript and whether it’s related to the effect of soil water availability or thinning. Maybe add “Canada” to the title.
Line 158-160: Font and/or font size changed. Please check.
Line 177-183: Indicate briefly how close the soil pits were to the “sample” trees? What was the exact sampling period for the vertical 1 m pit samples. Where the soil sample taken directly after digging the pits?
Line 196-198: The process of sampling handling of soil and plant samples between sampling in the field and before CVD extraction is not clear. For instance, were the samples transferred into a new tube or glass vial before CVD? What is a test tube?
Line 196-208: CVD extraction can affect isotope values of the extracted water (Chen et al, 2020, PNAS, Barbeta et al 2022, New Phytologist). Given the growing number of papers stating a bias for d2H values of woody material, I was wondering whether such a bias could affect the conclusion of the manuscript. This is because d2H values are used in the MixSIAR model to estimate root water uptake. Please clarify. Please also briefly give a statement on the extracted water amount (Diao et al. 2022, HESS).
Line 249: But samples were taken every 20 cm, not every 10 cm, right? See Line 181-182.
Line 293-301 and Figure 3: Some results are not well described. For example, lines 293-295: which samples exactly were used to determine the slope and intercept? Additionally, there is a lack of description regarding branch, soil, ground, and stem water. While OMWL is a bit unusual, it is acceptable. What does LEL represent? Furthermore, distinguishing between blue and green values is difficult in the small figure. Could one 'zoom in' or alter the colors slightly to ensure differentiation, such as between branch and soil water isotope values?
Table 1: Standard deviation not shown as stated in the caption. What does SMC mean?
Line 367 and throughout: d2H and d18O can be higher or lower, but not depleted or enriched. However, e.g., soil water can be 18O-depleted or 2H-enriched.
Line 369 and throughout: Please consider placing “isotope” before the word “fractionation” to make more explicitly what is actually fractionated.
Figure 7: In contrast to 5 and 35 cm soil samples, “deep” soil water has been sampled only once per growing season. Therefore, the deep soil samples lack a temporal component, right? Is the potential variation in deep soil water isotope values with time irrelevant for the study conclusion? If yes, can the authors back this up for the experimental site? Would it make sense to consider precipitation (e.g. modelled precipitation) as an additional source or do the authors assume that the 5 cm soil samples reflect isotope variation in precipitation?
Discussion point 4.2: Does the soil evaporative effect, regardless of transpiration, increase with increasing thinning?
Line 471: Sentence not clear, please rephrase.

Citation: https://doi.org/10.5194/egusphere-2024-88-RC2
- AC2: 'Reply on RC2', Emory Ellis, 27 Apr 2024
  
  Thank you for your invaluable feedback and meticulous attention to detail. Your insights are instrumental in fortifying this manuscript and enhancing the clarity of our message.
  Working new title: Seasonal changes in water uptake under drought conditions and thinning treatments of young and overstocked lodgepole pine (Pinus contorta) forests in British Columbia, Canada
  Methods:
  Soil pits were randomly selected across the plots to avoid sampling disturbed soils and were not persistently located near sample trees.
  1 m pit samples were taken in mid-July
  10 mL glass vials sealed with parafilm, foil, and tape and frozen until CDV. Branches remained in the tubes through CDV to capture any liquid water lost from the branches between field collection and frozen storage. – We will add in more detail
  L 196-208: We are unfamiliar with these articles, but we did focus on potential bias in the soil water samples and d2H fractionation in photosynthetically active materials and grasses. We will review these articles and make appropriate adjustments.
  L 249: Correct every 20 cm – will revise and correct this interval in the methods
  L 295: The local evaporative line was created from all of the collected samples; we will clarify that and elaborate on its significance. This figure needs to be reworked to be more readable. Your feedback on this aspect is particularly crucial for the overall understanding of our research.
  Table 1: Soil moisture content (SMC) – will clean up the table and caption
  Line 367 and throughout – agreed. This is a misuse of terminology that will be clarified
  Figure 7: We assume that the soil water at 5 cm reflects precipitation after evaporative fractionation. We collected precipitation samples when possible but did not consider precipitation as a potential input because meteorological drivers of evaporation are so strong over the growing season.
  There is no temporal component to deep soil water (> 35 cm) because the literature indicates that these depths are below the evaporative front. It was challenging to repeat sampling at this depth due to the soil structure, and there were only precipitation events that indicated deep infiltration in September (which can be reflected by precipitation and continuous soil moisture plots that we will add). The isotopic composition of deep soil water was isotopically similar to snowfall collected at the beginning of the sampling period, which supports our assumptions. However, this study has some limitations. We may need a section of our discussion highlighting this limitation. After high-intensity precipitation events in September and October, the isotopic composition of deep soil water was no longer similar to that measured in July.
  Discussion point 4.2: Figure 4 shows that the d18O values of soil water at depth 5 cm are more positive in treatments 1 and 2, which is likely to indicate a more substantial evaporative effect with increased thinning, but this conclusion could be clearer in lines 437-440
  
  Citation: https://doi.org/10.5194/egusphere-2024-88-AC2
RC3:
'Comment on egusphere-2024-88', Anonymous Referee #3, 26 Apr 2024

Review EGUsphere „Unexpected water uptake under drought conditions and thinning treatments in young and overstocked lodgepole pine forests”
The manuscript submitted by Ellis et al. approaches a timely and very interesting topic especially during intensified climate change conditions. However, I feel it does not qualify to be published at this stage. And I do say this with regret because I know the first author is an ECS and this work the result of a master thesis. Therefore, my critique is primarily directed towards the more senior authors. It's crucial for senior researchers to ensure the readiness of their work before submission, and this feedback is not intended to deter the young researcher's enthusiasm or efforts, because this work has been a great effort as a master thesis.
Water stable isotopes are wonderful tools for supporting inferences about flux origins and necessary to support research in forest hydrology or physiology. However, they often cannot stand alone without additional information about the fluxes of interest, and this is, in my opinion one of these cases. The authors monitor soil and branch water stable isotopes throughout an entire season without adding information about precipitation amount and transpiration fluxes, but because 2021 seemed to be an exceptionally dry year in Canada, this would have been key. Inferring water uptake depths without knowing how much water left the system in which direction (i.e. as transpiration or percolation) is not meaningful and results in the discussion turning around speculations. I also read the publication by Wang et al. 2019, which the authors mention, and it seems that the differences in transpiration fluxes are significant especially in the drier year of 2017. This information would greatly support parts of the current manuscript but is not provided. If transpiration in the control stand is as significantly reduced as shown by Wang et al. for 2017, then how can you be sure that there are significant amounts of transpiration in the even drier 2021, especially considering that you describe scorching in the crown? Subsequently the conclusion you draw that forest thinning does not significantly influence the change in water uptake depth is misleading and at least incomplete because likely the transpiration in the control plot is so low that no significant amounts of water do leave the system.
In case you do decide to move forward in finding a way to provide this information or re-writing the manuscript with a different research question, I do have some more specific comments that can help moving forward:
INTRODUCTION:
L 111-113: why would they be limited in depth? Would not the horizontal extend be limited in an overstocked forest plot?
METHODS:
L 123: course = coarse?
LL 149-152: This study should already be mentioned in the introduction as it leads to the formulation of hypothesis 1.
General comment: this section would benefit from a tabular overview of the sampling campaigns for all the compartments. It is very difficult to follow when you sampled what and a table could help solve this issue.
L 180: “in the middle of the growing season…” when is that exactly, please provide the dates to when you did things.
LL 184-193: “Precipitation samples were collected when available during field collection days.” What does this mean? Is this then a cumulated precipitation sample? “Groundwater and stream samples were collected at the end of the growing season as stream beds were dry and groundwater was inaccessible during the dry period.” When was this? Please provide the dates. Also see above, fill a table with this information so that the reader knows when you have which data available. From how this reads you have three snow/rain samples and samples groundwater once?
L 220: give more info on the LEL
RESULTS:
Figure 3: Please change the colours to more distinct and enlarge the text and axis descriptions (in R at least size=16)
Table 1: please add the SD’s to the table
Figure 5: why is the uppermost soil layer missing for some treatments? Please make sure you also mention this in the text.
L369: could also be cryogenic extraction bias? See (Allen and Kirchner 2022)
Figure 6: What is the M treatment? Also the figure caption is incomplete. What do the boxes show? Again also enlarge axis and figure text.
Figure 7: incorrect y-axis description. Please correct
LL419-420: is that not a contradiction of what the MixSiar model shows in fig. 7? What does that mean?
DISCUSSION
I think the discussion is too speculative and will be automatically improve once you can either provide flux data or direct the findings toward a different research question.
LL 429- 430: how do you then explain the differences in soil water signatures shown in figure 4C in 07 and 09, where the blue (T2 treatment heavy thinning) boxes show a clear enrichment at the 5cm depth?
LL442-443 because this is the biggest seasonal water influx?? Relate to overall precip and summer precip?
CONCLUSION:
This is a bit of a stretch and largely based on speculation about the fluxes in the discussion.

References:
Allen, Scott T., and James W. Kirchner. 2022. ‘Potential Effects of Cryogenic Extraction Biases on Plant Water Source Partitioning Inferred from Xylem-Water Isotope Ratios’. Hydrological Processes 36 (2): e14483. https://doi.org/10.1002/hyp.14483.

Citation: https://doi.org/10.5194/egusphere-2024-88-RC3
- AC3: 'Reply on RC3', Emory Ellis, 27 Apr 2024
  
  Thank you for your comment and detailed feedback. Firstly, not including the summer precipitation was an oversight in our first submission. We have meteorological data under the canopy for the control and thinning treatments and continuous soil moisture measurements at two depths (5 and 35 cm) from mid-July through October that would help explain inputs and some of the subsurface infiltration patterns. However, we do not have data on transpiration fluxes due to complications in the field during the data collection period. The addition of transpiration would be a complementary component that would strengthen our results for the drought period compared to the study conducted in 2017. I agree with your recommendation to change the manuscript's narrative to not focus on water use and fluxes but instead only focus on the location of water uptake, given the limited information about deep soil infiltration and tree water use. Further statistical analysis is necessary to elaborate on the conclusion that thinning/ tree density does not change seasonal water uptake depths as there are differences in the timing and relative contribution of different sources. I agree that a likely solution to the lack of difference in depth to water uptake by treatment is likely to be explained by the difference in stand transpiration volume, but that is only speculation.
  Response to the methods general comment: I agree, and other reviewers have mentioned the ambiguity of the sampling methods, specifically the timing of samples and precipitation sample collection methods – this will be clarified in revisions.
  L220: We will provide more information on the local evaporative line (LEL)
  Results: These figures and tables will be cleaned and more readable (enlarged axis and points and complete figure descriptions.
  M = mature stands for comparison. We collected branch samples but no soil water samples.
  Discussion:
  We agreed that adding meteorological data will provide insight that will strengthen the discussion. However, as previously recommended and mentioned, it may be better to change the discussion and conclusions only to discuss the depth of lodgepole pine water uptake rather than the volume of water uptake.
  Line 429-430: We need to revisit our statistical analysis of the isotopic composition of soil water between treatments because, as you suggest, Figure 4C shows differences in the isotopic composition driven by evaporation.
  L442-443: We will more clearly relate this to overall and summer precipitation using the continuous soil moisture measurements and precipitation.
  Conclusion: Hopefully, the inclusion of meteorological data will better inform this, but we will also revisit these claims and take a more conservative approach to discussing our findings relative to depth to water uptake and not water use.
  
  Citation: https://doi.org/10.5194/egusphere-2024-88-AC3

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2024-88', Anonymous Referee #1, 04 Apr 2024

I think that the data presented in this paper may allow for a more nuanced message regarding the impact of the thinning treatments on the water use pattern and uptake depths of lodgepole pine in overstocked stands during drought. The data presented in Figure 7 appear to indicate a faster and earlier depletion of topsoil water storage (5-35 cm depth) by trees in the overstocked control stands during late spring and early summer, as well as an earlier shift to deeper water sources utilization (>35 cm depth). In contrast, trees in the thinned stands (T1, T2) continue using a much larger proportion of topsoil water further into the summer, and shift to uptake of deeper water sources later in the summer season. This pattern appears to be particularly evident for the data collected in July, when trees in the unthinned control treatment were already obtaining more than 60% of their water from deeper sources (>35 cm), whereas trees in the thinned T1 and T2 treatments we still obtaining 55-70% of their total water uptake from shallower soil layers (5-35 cm depth). This pattern strongly suggests to me that the thinning treatments were rather effective at delaying the consumption and depletion of the available topsoil (5-35 cm) water pool by the pines during the early stages of the summer drought, compared to the overstocked control stands. In other words, trees in the unthinned control stands were forced to shift to deeper soil water sources earlier than trees in the thinned stands, probably due to faster depletion of the topsoil water pool by the overstocked control stands. I agree with the authors that the water uptake depth of the lodgepole pines was less affected by the thinning treatments later into the summer drought period, although in August the trees in the thinned stands were still using a somewhat greater proportion of topsoil water (5-35 cm) than the unthinned control stands. So I think that the main message highlighted in the summary and conclusions sections ("forest thinning did not cause a significant change in depth to water uptake") is questionable and should be reconsidered and/or significantly modulated after careful re-analysis and re-interpretation of the data.
Also, I would like to see a much more explicit and detailed description of the impacts of the thinning treatments on the continuously measured soil moisture content data (L311-313). I would even recommend including a new figure to better highlight any differences encountered in soil moisture contents by depth among the different treatments/stand densities.
The data presented in Table 1 appear to indicate greater rainfall interception by the tree canopy during late summer/early fall in the overstocked control stands than in the thinned stands, as suggested by the large differences in topsoil water contents (5 cm) at the end of the growing season. Also, please clarify whether this table reports data measured in October, and please note that the standard deviations mentioned in the figure caption are actually missing from all the columns, so please include them. The data presented in this table appear to suggest that mean soil water content across the full soil profile (5-100 cm depth) was slightly higher in the T2 treatment than in the unthinned control treatment at the end of the growing season, which would be a relevant finding if confirmed by appropriate statistical analyses.
In figure 5, it is unclear why data are missing for the upper soil layers in some of the treatments, or why the scale of the Y axis is not uniform across panels. In Figure 7, it is unclear why the values in the Y axis range between 0-1 instead of 0-100, as the axis title refers to percentages. It is also unclear why some of the columns in the central panel (T2) are shorter than all the other columns in the figure.
L453-455: Several studies have shown that trees that have access to relatively shallow groundwater (6.5 m deep at the study site) can support the integrity and functionality of their shallow fine roots and associated mycorrhizal fungi in dry topsoil layers during prolonged drought through internal hydraulic redistribution (Bauerle et al 2016; Querejeta et al 2007), so the assumption made in L454-455 and L508 is perhaps speculative and questionable, in the absence of any direct measurement of root function. Are there any other plausible explanations for the apparent inability of the lodgepole pines to use recent rainwater during the late growing season? Perhaps those late season rainfall events were of insufficient magnitude to recharge the topsoil layer in a physiologically meaningful way? In L510, do you mean that pines are unable to access water made available by late-season rainfalls during the rewetting period?
I think that the strong assertions made in L520-522 and 526-528 of the Conclusions section (and similar statements in the Abstract) should be reconsidered and rewritten after careful examination of the issues raised above.
Some of the references mentioned in the text appear to refer to wet riparian habitats and tree species, and I thus wonder whether they are really directly relevant to this particular study which has a strong focus on dry interior forests of the semiarid Okanagan Valley (e.g. Gibson &Edwards 2002; Liu et al 2015; Maier et al 2019; Sanchez-Perez et al 2008). It might be more appropriate to replace these references by others referring to semiarid pine forests more closely resembling your study system.

L30-31 in Abstract: They cannot shift from deep water sources when shallow water becomes more available at the end of the growing season? Or perhaps they don't need to shift to shallow water because deep water is plentiful and sufficient to meet their transpiration demand (which is presumably lower in October than in midsummer)? Or perhaps these late-season rain events are insufficient to recharge the topsoil layer meaningfully (i.e. the soil matric potential in upper layers may still remain lower than that in deeper layers in October despite these rain events)?

BAUERLE, T.L., RICHARDS, J.H., SMART, D.R. and EISSENSTAT, D.M. (2008), Importance of internal hydraulic redistribution for prolonging the lifespan of roots in dry soil. Plant, Cell & Environment, 31: 177-186.
Querejeta JI, Egerton-Warburton LM, Allen MF (2007) Hydraulic lift may buffer rhizosphere hyphae against the negative effects of severe soil drying in a California Oak savanna. Soil Biology and Biochemistry, 39, 409–417.

Citation: https://doi.org/10.5194/egusphere-2024-88-RC1
- AC1: 'Reply on RC1', Emory Ellis, 27 Apr 2024
  
  Thank you for your response.
  Partition: I agree that the results shown in Figure 7 indicate that the control trees show less dependence on shallow soil water, specifically when compared to the heavily thinned treatment, and this trend is most pronounced in July. I will reexamine the mixing model results and see if any significant pairwise trends can be discussed to strengthen this argument further in our revisions. As for the trends noted in September, I agree that trees in the thinned treatments could use soil water from below the evaporative front more so than the control trees, which had most of their water uptake from deep soil water (> 35 cm) and groundwater. Thank you for your feedback. We will carefully consider these comments when reevaluating the statistical analysis to see if there is a more compelling and nuanced conclusion to present in this paper.
  Soil Moisture: We continuously measured soil moisture at two depths across the treatment plots in block 1. The other values of percent soil water content (for deeper soil samples) were measured using gravimetric methods.
  We would be happy to include a plot of the soil moisture measurements.
  Interception: The control has higher canopy interception compared to the treatments for low-intensity precipitation events (just from visual observations).
  The error bars are only available for samples for which we ran multiple cryogenic distillations. Removing them altogether could be more effective because there are no standard error calculations for all measurements.
  I will conduct and add the appropriate statistical tests comparing soil water content at each depth and add those results.
  I did not notice the missing data in Figure 5 on the initial submission, but I will correct it and add that updated figure.
  L 510 – We will clarify that the late growing season precipitation was likely insufficient for shallow root uptake because of the low precipitation volume and previously dry soils rather than imply that the trees were unable to experience a shift in water uptake from some other limitation or do not need to shift (as you stated in the abstract comment).
  As for the citations, thank you for recommending references more relevant to arid ecosystems. We will review them and consider adding them when appropriate rather than research on riparian ecosystems.
  
  Citation: https://doi.org/10.5194/egusphere-2024-88-AC1
RC2:
'Comment on egusphere-2024-88', Anonymous Referee #2, 23 Apr 2024

The study by Ellis investigates the impact of thinning and seasonal water availability on root water uptake of lodgepole pine trees in an experimental forest in Canada using water stable isotopes. The authors found that the root water uptake of pine trees responded to the seasonal shift in water availability but not to thinning treatment. Overall, the findings are timely and important for similar research in the field, as well as nicely presented and discussed. My comments probably lead only to minor modifications of the manuscript.
Title: The title of the manuscript could be a bit more specific, so that it matches the story of abstract, and the main findings of the manuscript. For instance, it’s not 100% clear what “unexpected” means throughout the manuscript and whether it’s related to the effect of soil water availability or thinning. Maybe add “Canada” to the title.
Line 158-160: Font and/or font size changed. Please check.
Line 177-183: Indicate briefly how close the soil pits were to the “sample” trees? What was the exact sampling period for the vertical 1 m pit samples. Where the soil sample taken directly after digging the pits?
Line 196-198: The process of sampling handling of soil and plant samples between sampling in the field and before CVD extraction is not clear. For instance, were the samples transferred into a new tube or glass vial before CVD? What is a test tube?
Line 196-208: CVD extraction can affect isotope values of the extracted water (Chen et al, 2020, PNAS, Barbeta et al 2022, New Phytologist). Given the growing number of papers stating a bias for d2H values of woody material, I was wondering whether such a bias could affect the conclusion of the manuscript. This is because d2H values are used in the MixSIAR model to estimate root water uptake. Please clarify. Please also briefly give a statement on the extracted water amount (Diao et al. 2022, HESS).
Line 249: But samples were taken every 20 cm, not every 10 cm, right? See Line 181-182.
Line 293-301 and Figure 3: Some results are not well described. For example, lines 293-295: which samples exactly were used to determine the slope and intercept? Additionally, there is a lack of description regarding branch, soil, ground, and stem water. While OMWL is a bit unusual, it is acceptable. What does LEL represent? Furthermore, distinguishing between blue and green values is difficult in the small figure. Could one 'zoom in' or alter the colors slightly to ensure differentiation, such as between branch and soil water isotope values?
Table 1: Standard deviation not shown as stated in the caption. What does SMC mean?
Line 367 and throughout: d2H and d18O can be higher or lower, but not depleted or enriched. However, e.g., soil water can be 18O-depleted or 2H-enriched.
Line 369 and throughout: Please consider placing “isotope” before the word “fractionation” to make more explicitly what is actually fractionated.
Figure 7: In contrast to 5 and 35 cm soil samples, “deep” soil water has been sampled only once per growing season. Therefore, the deep soil samples lack a temporal component, right? Is the potential variation in deep soil water isotope values with time irrelevant for the study conclusion? If yes, can the authors back this up for the experimental site? Would it make sense to consider precipitation (e.g. modelled precipitation) as an additional source or do the authors assume that the 5 cm soil samples reflect isotope variation in precipitation?
Discussion point 4.2: Does the soil evaporative effect, regardless of transpiration, increase with increasing thinning?
Line 471: Sentence not clear, please rephrase.

Citation: https://doi.org/10.5194/egusphere-2024-88-RC2
- AC2: 'Reply on RC2', Emory Ellis, 27 Apr 2024
  
  Thank you for your invaluable feedback and meticulous attention to detail. Your insights are instrumental in fortifying this manuscript and enhancing the clarity of our message.
  Working new title: Seasonal changes in water uptake under drought conditions and thinning treatments of young and overstocked lodgepole pine (Pinus contorta) forests in British Columbia, Canada
  Methods:
  Soil pits were randomly selected across the plots to avoid sampling disturbed soils and were not persistently located near sample trees.
  1 m pit samples were taken in mid-July
  10 mL glass vials sealed with parafilm, foil, and tape and frozen until CDV. Branches remained in the tubes through CDV to capture any liquid water lost from the branches between field collection and frozen storage. – We will add in more detail
  L 196-208: We are unfamiliar with these articles, but we did focus on potential bias in the soil water samples and d2H fractionation in photosynthetically active materials and grasses. We will review these articles and make appropriate adjustments.
  L 249: Correct every 20 cm – will revise and correct this interval in the methods
  L 295: The local evaporative line was created from all of the collected samples; we will clarify that and elaborate on its significance. This figure needs to be reworked to be more readable. Your feedback on this aspect is particularly crucial for the overall understanding of our research.
  Table 1: Soil moisture content (SMC) – will clean up the table and caption
  Line 367 and throughout – agreed. This is a misuse of terminology that will be clarified
  Figure 7: We assume that the soil water at 5 cm reflects precipitation after evaporative fractionation. We collected precipitation samples when possible but did not consider precipitation as a potential input because meteorological drivers of evaporation are so strong over the growing season.
  There is no temporal component to deep soil water (> 35 cm) because the literature indicates that these depths are below the evaporative front. It was challenging to repeat sampling at this depth due to the soil structure, and there were only precipitation events that indicated deep infiltration in September (which can be reflected by precipitation and continuous soil moisture plots that we will add). The isotopic composition of deep soil water was isotopically similar to snowfall collected at the beginning of the sampling period, which supports our assumptions. However, this study has some limitations. We may need a section of our discussion highlighting this limitation. After high-intensity precipitation events in September and October, the isotopic composition of deep soil water was no longer similar to that measured in July.
  Discussion point 4.2: Figure 4 shows that the d18O values of soil water at depth 5 cm are more positive in treatments 1 and 2, which is likely to indicate a more substantial evaporative effect with increased thinning, but this conclusion could be clearer in lines 437-440
  
  Citation: https://doi.org/10.5194/egusphere-2024-88-AC2
RC3:
'Comment on egusphere-2024-88', Anonymous Referee #3, 26 Apr 2024

Review EGUsphere „Unexpected water uptake under drought conditions and thinning treatments in young and overstocked lodgepole pine forests”
The manuscript submitted by Ellis et al. approaches a timely and very interesting topic especially during intensified climate change conditions. However, I feel it does not qualify to be published at this stage. And I do say this with regret because I know the first author is an ECS and this work the result of a master thesis. Therefore, my critique is primarily directed towards the more senior authors. It's crucial for senior researchers to ensure the readiness of their work before submission, and this feedback is not intended to deter the young researcher's enthusiasm or efforts, because this work has been a great effort as a master thesis.
Water stable isotopes are wonderful tools for supporting inferences about flux origins and necessary to support research in forest hydrology or physiology. However, they often cannot stand alone without additional information about the fluxes of interest, and this is, in my opinion one of these cases. The authors monitor soil and branch water stable isotopes throughout an entire season without adding information about precipitation amount and transpiration fluxes, but because 2021 seemed to be an exceptionally dry year in Canada, this would have been key. Inferring water uptake depths without knowing how much water left the system in which direction (i.e. as transpiration or percolation) is not meaningful and results in the discussion turning around speculations. I also read the publication by Wang et al. 2019, which the authors mention, and it seems that the differences in transpiration fluxes are significant especially in the drier year of 2017. This information would greatly support parts of the current manuscript but is not provided. If transpiration in the control stand is as significantly reduced as shown by Wang et al. for 2017, then how can you be sure that there are significant amounts of transpiration in the even drier 2021, especially considering that you describe scorching in the crown? Subsequently the conclusion you draw that forest thinning does not significantly influence the change in water uptake depth is misleading and at least incomplete because likely the transpiration in the control plot is so low that no significant amounts of water do leave the system.
In case you do decide to move forward in finding a way to provide this information or re-writing the manuscript with a different research question, I do have some more specific comments that can help moving forward:
INTRODUCTION:
L 111-113: why would they be limited in depth? Would not the horizontal extend be limited in an overstocked forest plot?
METHODS:
L 123: course = coarse?
LL 149-152: This study should already be mentioned in the introduction as it leads to the formulation of hypothesis 1.
General comment: this section would benefit from a tabular overview of the sampling campaigns for all the compartments. It is very difficult to follow when you sampled what and a table could help solve this issue.
L 180: “in the middle of the growing season…” when is that exactly, please provide the dates to when you did things.
LL 184-193: “Precipitation samples were collected when available during field collection days.” What does this mean? Is this then a cumulated precipitation sample? “Groundwater and stream samples were collected at the end of the growing season as stream beds were dry and groundwater was inaccessible during the dry period.” When was this? Please provide the dates. Also see above, fill a table with this information so that the reader knows when you have which data available. From how this reads you have three snow/rain samples and samples groundwater once?
L 220: give more info on the LEL
RESULTS:
Figure 3: Please change the colours to more distinct and enlarge the text and axis descriptions (in R at least size=16)
Table 1: please add the SD’s to the table
Figure 5: why is the uppermost soil layer missing for some treatments? Please make sure you also mention this in the text.
L369: could also be cryogenic extraction bias? See (Allen and Kirchner 2022)
Figure 6: What is the M treatment? Also the figure caption is incomplete. What do the boxes show? Again also enlarge axis and figure text.
Figure 7: incorrect y-axis description. Please correct
LL419-420: is that not a contradiction of what the MixSiar model shows in fig. 7? What does that mean?
DISCUSSION
I think the discussion is too speculative and will be automatically improve once you can either provide flux data or direct the findings toward a different research question.
LL 429- 430: how do you then explain the differences in soil water signatures shown in figure 4C in 07 and 09, where the blue (T2 treatment heavy thinning) boxes show a clear enrichment at the 5cm depth?
LL442-443 because this is the biggest seasonal water influx?? Relate to overall precip and summer precip?
CONCLUSION:
This is a bit of a stretch and largely based on speculation about the fluxes in the discussion.

References:
Allen, Scott T., and James W. Kirchner. 2022. ‘Potential Effects of Cryogenic Extraction Biases on Plant Water Source Partitioning Inferred from Xylem-Water Isotope Ratios’. Hydrological Processes 36 (2): e14483. https://doi.org/10.1002/hyp.14483.

Citation: https://doi.org/10.5194/egusphere-2024-88-RC3
- AC3: 'Reply on RC3', Emory Ellis, 27 Apr 2024
  
  Thank you for your comment and detailed feedback. Firstly, not including the summer precipitation was an oversight in our first submission. We have meteorological data under the canopy for the control and thinning treatments and continuous soil moisture measurements at two depths (5 and 35 cm) from mid-July through October that would help explain inputs and some of the subsurface infiltration patterns. However, we do not have data on transpiration fluxes due to complications in the field during the data collection period. The addition of transpiration would be a complementary component that would strengthen our results for the drought period compared to the study conducted in 2017. I agree with your recommendation to change the manuscript's narrative to not focus on water use and fluxes but instead only focus on the location of water uptake, given the limited information about deep soil infiltration and tree water use. Further statistical analysis is necessary to elaborate on the conclusion that thinning/ tree density does not change seasonal water uptake depths as there are differences in the timing and relative contribution of different sources. I agree that a likely solution to the lack of difference in depth to water uptake by treatment is likely to be explained by the difference in stand transpiration volume, but that is only speculation.
  Response to the methods general comment: I agree, and other reviewers have mentioned the ambiguity of the sampling methods, specifically the timing of samples and precipitation sample collection methods – this will be clarified in revisions.
  L220: We will provide more information on the local evaporative line (LEL)
  Results: These figures and tables will be cleaned and more readable (enlarged axis and points and complete figure descriptions.
  M = mature stands for comparison. We collected branch samples but no soil water samples.
  Discussion:
  We agreed that adding meteorological data will provide insight that will strengthen the discussion. However, as previously recommended and mentioned, it may be better to change the discussion and conclusions only to discuss the depth of lodgepole pine water uptake rather than the volume of water uptake.
  Line 429-430: We need to revisit our statistical analysis of the isotopic composition of soil water between treatments because, as you suggest, Figure 4C shows differences in the isotopic composition driven by evaporation.
  L442-443: We will more clearly relate this to overall and summer precipitation using the continuous soil moisture measurements and precipitation.
  Conclusion: Hopefully, the inclusion of meteorological data will better inform this, but we will also revisit these claims and take a more conservative approach to discussing our findings relative to depth to water uptake and not water use.
  
  Citation: https://doi.org/10.5194/egusphere-2024-88-AC3

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Publish subject to revisions (further review by editor and referees) (08 May 2024) by Nadia Ursino

AR by Emory Ellis on behalf of the Authors (20 Jul 2024) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (24 Jul 2024) by Nadia Ursino

RR by Anonymous Referee #2 (19 Aug 2024)

Suggestions for revision or reasons for rejection

I have now reviewed the revised version of the manuscript of Ellis et al. and acknowledge the significant improvements made by the authors. However, I believe there are still some minor points that should be addressed before publication.

Line 183: What does n = 3333 reflect? Maybe not related to this study?

Line 236-238: Please clarify the definition of d2h “biases” and explain why these d2h biases are similar to those reported by Allen & Kirchner (2022).

Line 489-491: Please rephrase to clarify that this statement is based on isotope analysis.

Line 495: Change "enrichment of d18O" to "enrichment in 18O."

Line 461-466: The logic in this sentence is not entirely clear. I believe the authors intend to explain that Scenario 6 is the most realistic and statistically sound scenario. However, the text seems to "jump" between Scenarios 1, 2, and 6, as well as between Scenarios 4 and 6, without clear transitions. Additionally, Scenarios 3 and 5 are not mentioned, leaving gaps in the explanation. Could you please clarify the reasoning and ensure all relevant scenarios are addressed in a logical sequence?

Line 618-621: The current statement is not entirely clear. If thinning (or a decrease in stem density) increases soil water use below 35 cm during prolonged drought periods, then soil water below 35 cm becomes more relevant as the occurrence and duration of drought spells increase in the future (assuming this holds true for UPC). Therefore, lodgepole pine dependency on deep soil or groundwater would increase, not decrease as currently stated. Is this correct?

Figures/Tables in general: Please explain the meaning of the lines in point plots as well as the boxes, whiskers, and symbols in boxplots. Additionally, provide more details on the location, species, and other relevant context. Ensure that all abbreviations are clearly defined. In general, each figure and table should be understandable on its own, without requiring the reader to refer to the entire manuscript.

Figure 9: The authors also collected isotope data from mature trees, but the rationale for including these data is not clearly expressed in the manuscript. I assume the mature trees are intended as a control to clarify the representativeness of the young tree water uptake patterns, correct? Additionally, I suggest adding the statistical results mentioned in Lines 453 and 455 to Figure 9 to make these key findings more explicit. Currently, these important results are somewhat "hidden" in the text, and highlighting them in the figure would enhance their visibility and impact.

Figure 10: This is one of the most important figures in the paper. Could the authors please add more details to ensure the figure can be fully understood without needing to refer to the main text? In particular, providing detailed information about Scenario 6 is crucial. Although the d2h biases are now mentioned in the Methods and Results sections, it remains unclear how these biases are considered in the MixSIAR model results. Please clarify how this bias is accounted for and its impact on the interpretation of the data.

Table S1: This table is not cited in the main text, or I may have missed the reference.

Hide

RR by Anonymous Referee #1 (21 Aug 2024)

ED: Publish subject to minor revisions (review by editor) (22 Aug 2024) by Nadia Ursino

AR by Emory Ellis on behalf of the Authors (01 Sep 2024) Author's response Author's tracked changes Manuscript

ED: Publish as is (03 Sep 2024) by Nadia Ursino

AR by Emory Ellis on behalf of the Authors (07 Sep 2024)

Journal article(s) based on this preprint

25 Oct 2024

Seasonal shifts in depth-to-water uptake by young thinned and overstocked lodgepole pine (Pinus contorta) forests under drought conditions in the Okanagan Valley, British Columbia, Canada

Emory C. Ellis, Robert D. Guy, and Xiaohua A. Wei

Hydrol. Earth Syst. Sci., 28, 4667–4684, https://doi.org/10.5194/hess-28-4667-2024,https://doi.org/10.5194/hess-28-4667-2024, 2024

Short summary

Emory Ellis, Robert Guy, and Xiaohua A. Wei

Supplement

https://doi.org/10.5194/egusphere-2024-88-supplement

Emory Ellis, Robert Guy, and Xiaohua A. Wei

Viewed

Total article views: 444 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
326	84	34	444	53	18	24

HTML: 326
PDF: 84
XML: 34
Total: 444
Supplement: 53
BibTeX: 18
EndNote: 24

Views and downloads (calculated since 25 Jan 2024)

Month	HTML	PDF	XML	Total
Jan 2024	62	13	2	77
Feb 2024	48	8	1	57
Mar 2024	20	8	2	30
Apr 2024	81	16	15	112
May 2024	29	8	3	40
Jun 2024	47	9	4	60
Jul 2024	18	8	3	29
Aug 2024	8	6	4	18
Sep 2024	8	4	0	12
Oct 2024	5	4	0	9

Cumulative views and downloads (calculated since 25 Jan 2024)

Month	HTML	PDF	XML	Total
Jan 2024	62	13	2	77
Feb 2024	48	8	1	57
Mar 2024	20	8	2	30
Apr 2024	81	16	15	112
May 2024	29	8	3	40
Jun 2024	47	9	4	60
Jul 2024	18	8	3	29
Aug 2024	8	6	4	18
Sep 2024	8	4	0	12
Oct 2024	5	4	0	9

Viewed (geographical distribution)

Total article views: 429 (including HTML, PDF, and XML) Thereof 429 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 25 Oct 2024

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (907 KB)
Metadata XML

Short summary

This study analyzes the water-stable isotope composition by analyzing the impact of forest thinning on lodgepole pine depth-to-water uptake and water use strategies. The lodgepole pine's primary source is spring snowmelt and shifts to rely on deeper soil water to maintain water uptake. There was no effect of decreased stand density on depth-to-water uptake. It will become more critical that we know how much water forests are using and what strategies trees use to sustain their water supply.


Total:	0
HTML:	0
PDF:	0
XML:	0