Water Regulation Ecosystem Services Following Gap Formation in Fir-Beech Forests in the Dinaric Karst
Round 1
Reviewer 1 Report
The manuscript entitled “Variation in water regulation ecosystem services following gap creation in Dinaric Karst forests” written by Urša Vilhar, investigates water regulation ecosystem service in gaps and closed stands of differently managed beech-silver fir forests. I think this paper constitutes valuable contribution given the fact importance of forest ecosystems in providing pristine water resources in Slovenia karst regions. Indeed, water retention is of main concern due to large human-induced land-use changes in Europe. Nonetheless, this kind of research is worthy to be further developed and the article may work as an interesting contribution to promote more similar and more extended research. While the employed data are of interest, the MS suffers from some inaccuracies that reduce the potential value of this study. The author was avoiding some topics which can increase the clarity of a paper, therefore I tried to address some of my concerns. Furthermore, I am concerned about the study design and choice of plots. The first section of my review includes larger conceptual points, and that is followed by detailed line-by-line comments.
Major comments
- i) Author is mentioning that the water balance is not possible to measure in karst area (specifically drainage), and the model is the only possibility. Not sufficient literature review in the introduction is provided about the problematic, as for me was not clear why the model is the only way. In specific there is no info about sap flow or remotely sensed methods (eddy covariance), in my opinion, this should be reviewed. What is the novelty of the presented research? The obtained results are somewhat expected (see below comments on Study design). Few sentences describing the silviculture systems used in karst regions in Slovenia and their influence on water balance should be provided. On my view the biggest issue that structural characteristics are not sufficiently described, both of forest stands and gaps.
- ii) I found problematic the study design, as there are huge differences in structural characteristics of two forests, specifically tree volume (no other structural characteristics are provided). No info about the managed forest was given to justify this. I think the diameter structure should be provided. The studied gap in the managed forest is 3 times bigger than that one in old-growth (but also the shape of the gap could influence (see Table 1) evapotranspiration and regeneration patterns). I would expect these results even without running the model. What silviculture system is used in the managed forest, and was the managed forest uneven-aged? There is no sufficient explanation of the model itself, even the entire MS is based on it. The author is refereeing to other studies, however, I think it should be better explained in the text. Is it precipitation the only input parameter in the model? Calibration of the model should be better explained.
iii) In discussion part I couldn’t find any clue about the gap size that is usually used in beech stands on karst, or in beech stands on drier substrates in Central Europe. According to me, this should be reviewed especially because in the “Conclusion” part “the continuous cover forestry” is suggested. What are beech requirements in terms of light for successful natural regeneration? The gap size might have a different influence on beech and fir regeneration? Is it future goal to create monodominant beech forest or mix-forest with fir? Moreover, what are the implications of monocultures and mix-forest on water balance? Some parts of discussion seem like repetitive, same story explained in different ways (see line-by-line comments). Even the last paragraph of Discussion I liked, more of it should be included.
On my view, this MS might be considered for publication after thorough editing according to the suggested changes. My few concerns and suggestions to improve the MS are detailed below.
Specific line-by-line comments:
Line 26: What type of forest could be introduced earlier in abstract. We got info about beech species at the end of the abstract.
Line 81-82: what was the proportion of each species? In addition, why fir is not mentioned again? Is the goal of management monodominant beech forest?
Line 90: Were there smaller gaps? From this sentence, I get the impression that there were more, why this specifically was selected and not in addition a smaller one (see major comments)? Were the gaps created for the purpose of natural regeneration in the managed forest or some specific investigative reason? Was the managed forest even aged?
Line 94-95: Introduce these in table 1! How far were forests?
Line 96-99: What about silver fir, it is mentioned in the first sentence of M&M that the forest consists of these two species. The managed forest also was beech dominated?
Line 107: What about runoff especially in the gaps? Is it accounted in DF?
Line 113: SWC…. How come? Please explain.
Line 123: Throughfall? Precipitation measured under the canopy? Please explain what parameters are required by the model as inputs? Monthly throughfall was measured and daily SWC data? Is it enough monthly throughfall to check model accuracy? In which forest climatic data were measured, in both?
Line 125: The parameter 125 values were….delete this (same as the previous sentence).
Line 156: Figure caption is without precipitation!
Line 252: This sentence is the same as at the beginning of the following paragraph line 281-283. Overall those two paragraphs are saying the same thing.
Line 327: This figure is not referenced within the text. Given that this MS has just two figures, i suggest to be introduced in the main text and referenced in the results part.
Author Response
Concerning Manuscript ID: forests-1026163
Title: Variation in water regulation ecosystem services following gap creation in Dinaric Karst forests
Author: Urša Vilhar, PhD;
Submitted to Section: Forest Ecology and Management
Special Issue: Forest Structural Characteristics and Their Relationship with Ecosystem Services and Functions
Dear Editors,
Thank you very much for your decision. We understand the concerns of the Reviewer #1. We have followed their useful suggestions and made changes in the manuscript, accordingly, as listed below. We thank the anonymous reviewers for their helpful comments on the manuscript.
Below we reply to the Reviewers comments.
Best regards,
Urša Vilhar (PhD)
Slovenian Forestry Institute
Večna pot 2
1000 Ljubljana
Slovenia
+386 1 200 78 46
Author Response File: 
Author Response.docx
Reviewer 2 Report
Review of manuscript Forests-1026163 ‘Variation in water regulation ecosystem services following gap creation in Dinaric Karst forests’ by Urša Vilhar.
General comments
The subject of the manuscript is very relevant for Forests and the manuscript clearly reflects the topic.
In general, the manuscript is very well written.
With respect to the correlation analyses, I wonder about the meaning of this analysis. As the simulations use the same meteorological driver and a lot of other common parameter values, a strong coherence of the time series should be self-evident. It isn’t described, which temporal aggregation (daily, monthly, annual) was used for the correlation analyses.
Most importantly, I cannot follow the data given for the Old-growth gap. The values reported for DF, SE, TRAN, and I don’t add up to 100% as suggested by eq. (1). If mean values of water fluxes in mm given in Fig. 2 are summed up, it is with 1372 mm substantially lower than for the other plots with 1768 to 1781 mm despite the same input data have been used. It should be impossible to assume ΔSWC to be responsible for the difference.
As I cannot exclude a miscalculation, a judgment of the conclusions isn’t possible and should be made after clarification of this item.
Given these concerns, I can only recommend the acceptance of the manuscript after a major revision. A potential revision should include a recalculation of the water budget or clarification of a possible misunderstanding. A more detailed description of the methods and the hydrologic simulation is recommended.
Specific comments
Page 5, row 124-125 Roots may be present also in the weathered bedrock beneath 40 cm depth, especially in karstic terrain.
Page 7, row 182 consider ‘… significant trend.’ -> ‘… significant temporal trend.’.
Page 7, row 194-196 Was this a relative increase of I in relation to other fluxes or an absolute increase of I?
Page 8, row 218-219 According to eq (1), DF, SE, TRAN, and I should add up to P if ΔSWC is neglected. The sum for P for Old-growth gap significantly deviates from other plots. Please explain.
Page 9, row 221-222 According to eq (1), DF, SE, TRAN, and I should add up to 100% if DSWC is neglected. Data for Old-growth gap significantly deviate from 100%. Please explain.
Page 10, row 257-259 What is meant by ‘water holding capacity’? I doubt that the water holding capacity of soils or vegetation is reduced when annual precipitation is high.
Page 10-11, row 291-293 What is meant by ‘water holding capacity of forests’? The model simulation uses the same soil parametrization for all plots.
Author Response
Concerning Manuscript ID: forests-1026163
Title: Variation in water regulation ecosystem services following gap creation in Dinaric Karst forests
Author: Urša Vilhar, PhD;
Submitted to Section: Forest Ecology and Management
Special Issue: Forest Structural Characteristics and Their Relationship with Ecosystem Services and Functions
Dear Editors,
Thank you very much for your decision. We understand the concerns of the Reviewer #2. We have followed their useful suggestions and made changes in the manuscript, accordingly, as listed below. We thank the anonymous reviewers for their helpful comments on the manuscript.
Below we reply to the Reviewers comments.
Best regards,
Urša Vilhar (PhD)
Slovenian Forestry Institute
Večna pot 2
1000 Ljubljana
Slovenia
+386 1 200 78 46
Author Response File: Author Response.docx
Reviewer 3 Report
The subject of the manuscript is very relevant and necessary in Forest.
It is a very well prepared material that gives a lot of information about the water balance and the penetration of rainfall through tree crowns.
The data is from many years, so there is no randomness here.
Tables and charts needed and clearly presented.
Author Response
Concerning Manuscript ID: forests-1026163
Title: Variation in water regulation ecosystem services following gap creation in Dinaric Karst forests
Author: Urša Vilhar, PhD;
Submitted to Section: Forest Ecology and Management
Special Issue: Forest Structural Characteristics and Their Relationship with Ecosystem Services and Functions
Dear Editors,
Thank you very much for your decision. We thank the anonymous reviewers for their helpful comments on the manuscript.
Below we reply to the Reviewers comments.
Best regards,
Urša Vilhar (PhD)
Slovenian Forestry Institute
Večna pot 2
1000 Ljubljana
Slovenia
+386 1 200 78 46
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Too many keywords? Choose the most important 5-6. No need for all water balance components.
Even i don't feel qualified to judge the English language, seems to me that the first version had better English? I recommend being checked by a native speaker.
The author, in the response letter, mentioned sap flow and eddy covariance but a i was unable to find?
Line 74: I guess this project should be mentioned only once? Use the acronym. Line 152: " as part mentioned project "acronym".
Author Response
Concerning Manuscript ID: forests-1026163
Title: Water regulation ecosystem services following gap formation in fir-beech forests in Dinaric Karst
Author: Urša Vilhar, PhD;
Submitted to Section: Forest Ecology and Management
Special Issue: Forest Structural Characteristics and Their Relationship with Ecosystem Services and Functions
Dear Editors,
Thank you very much for your decision. We understand the concerns of the Reviewer 1 and Reviewer #2. We have followed their useful suggestions and made changes in the manuscript, accordingly, as listed below. In addition, English proofreading was repeated, and a certificate is attached to this letter. All changes in the manuscript are marked with yellow colour. We thank the anonymous reviewers for their helpful comments on the manuscript.
Below we reply to the Reviewers comments.
Best regards,
Urša Vilhar (PhD)
Slovenian Forestry Institute
Večna pot 2
1000 Ljubljana
Slovenia
+386 1 200 78 46
Review #1 Round 2
Comments and Suggestions for Authors
Too many keywords? Choose the most important 5-6. No need for all water balance components.
- Agreed and corrected. Now it reads:
Keywords: water retention capacity; drainage flux; evapotranspiration; Brook90 hydrological model; experimental canopy gap, old-growth forest
Even i don't feel qualified to judge the English language, seems to me that the first version had better English? I recommend being checked by a native speaker.
- Agreed and corrected. English proofreading was repeated, and a certificate is attached to this letter.
The author, in the response letter, mentioned sap flow and eddy covariance but a i was unable to find?
- Agreed and corrected. Now it reads:
L 94: Soil water fluxes in karst areas are therefore very complex [1], and hydrological fluxes are often estimated by eddy covariance [2-4], tree sap flow measurements [5-8] or simulated by hydrological models [9-16].
Line 74: I guess this project should be mentioned only once? Use the acronym.
- Agreed and corrected. Now it reads:
L72: In 2001 several experimental canopy gaps of different sizes (small-medium-large) were established within the EU FP5 project Nat-Man.
Line 152: " as part mentioned project "acronym".
- Agreed and corrected. Now it reads:
L150: …, as part of the EU FP5 project Nat-Man.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
General comments
Thank You very much for the thorough revision of the manuscript.
The subject of the manuscript is very relevant for Forests and the manuscript clearly reflects the topic.
In general, the manuscript is very well written.
The introduction gives a comprehensive overview of hydrological processes in Dinaric forest ecosystems and the regulating ecosystem services they provide with emphasis on created or natural gaps.
The materials and methods section provides extensive information on the study sites and the observations. The number of replications for the throughfall measurements seem very low with respect to the heterogeneity of the stands. Apparently, no throughfall measurements took place at the gaps. As the simulated canopy interception at the old-growth gap is substantial, it would have been beneficial to have some observations at hand. Additionally, observations of stemflow are missing. Here, it should be specified, how the parametrization of stemflow and throughfall at the gap plots was derived.
The results are only referred to as relative values with respect to precipitation. I would suggest including some absolute values as some dependencies seem to be opposed for relative or absolute values (e.g. ETP in wet and dry years). This applies also to the discussion.
The references are mostly up to date and from international journals.
I recommend the acceptance of the manuscript after a minor revision. A potential revision should consider to above and below-given annotations.
Specific comments
Page 3, row 93-104 However, it may be mentioned that Brook90 is only capable to deal with a dual permeability system. A threefold permeability system can’t be represented by Brook90.
Page 6, row 189 Nine collectors seem to be too few to represent the spatial heterogeneity of throughfall (Clarke et al. 2016 https://www.icp-forests.org/pdf/manual/2016/ICP_Manual_2016_01_part14.pdf). Please specify the potential sampling error of the throughfall measurements due to spatial variability. Please specify, if throughfall sampling was conducted in the old-growth gap (at least in 2013)? Please explain, why stemflow wasn’t assessed at the plots. Stemflow may significantly contribute to stand precipitation in beech forests (Clarke et al. 2016).
Page 6, row 199 Dirksen (1999) is missing in references.
Page 8, row 256-261 Please give a unit for RSME values.
Page 9, row 282-283 From Fig. 2 and regression parameters (row 283) it seems that SE is decreasing in the old-growth gap. Please explain.
Author Response
Concerning Manuscript ID: forests-1026163
Title: Water regulation ecosystem services following gap formation in fir-beech forests in Dinaric Karst
Author: Urša Vilhar, PhD;
Submitted to Section: Forest Ecology and Management
Special Issue: Forest Structural Characteristics and Their Relationship with Ecosystem Services and Functions
Dear Editors,
Thank you very much for your decision. We understand the concerns of the Reviewer 1 and Reviewer #2. We have followed their useful suggestions and made changes in the manuscript, accordingly, as listed below. In addition, English proofreading was repeated, and a certificate is attached to this letter. All changes in the manuscript are marked with yellow colour. We thank the anonymous reviewers for their helpful comments on the manuscript.
Below we reply to the Reviewers comments.
Best regards,
Urša Vilhar (PhD)
Slovenian Forestry Institute
Večna pot 2
1000 Ljubljana
Slovenia
+386 1 200 78 46
Review #2 Round 2
General comments
Thank You very much for the thorough revision of the manuscript.
The subject of the manuscript is very relevant for Forests and the manuscript clearly reflects the topic.
In general, the manuscript is very well written.
The introduction gives a comprehensive overview of hydrological processes in Dinaric forest ecosystems and the regulating ecosystem services they provide with emphasis on created or natural gaps.
The materials and methods section provides extensive information on the study sites and the observations. The number of replications for the throughfall measurements seem very low with respect to the heterogeneity of the stands. Apparently, no throughfall measurements took place at the gaps. As the simulated canopy interception at the old-growth gap is substantial, it would have been beneficial to have some observations at hand. Additionally, observations of stemflow are missing. Here, it should be specified, how the parametrization of stemflow and throughfall at the gap plots was derived.
- Throughfall was recorded in forests and gaps monthly or biweekly using a series of nine funnel collectors (240 cm2 each) arranged along a regular grid with 5×5 m spacing. In the experimental gap monthly throughfall values were comparable to incidental precipitation, recorded in an open field near the study sites. In addition, in the experimental gap and above tree crowns automated weather stations (Vantage Pro® wireless, Davis Instruments, Hayward, California, USA) were recording hourly precipitation. Stemflow was recorded monthly on 6 trees with different diameters in each managed and old-growth forest. Results of this study were published in: Vilhar U (2010) Padavinski režim v vrzelih in sestojih dinarskega jelovo-bukovega gozda. Precipitation regime in gaps and mature stands of Dinaric silver fir-beech forests. Zbornik gozdarstva in lesarstva / Acta Silva et Ligni 91. 3-10
- However, stemflow is not simulated by the Brook90 model so it was not used for model fitting or testing.
- Similar to the managed and old-growth forest, also for old-growth gap monthly throughfall was used for model fitting and testing. Accordingly, Results section “3.1. Model fitting and testing” was corrected and now reads:
L 252: Table S1 summarizes the input parameters used in the Brook90 model simulations. Monthly throughfall (Figure S2) and daily soil water contents (Figure S3) were well simulated in both time and magnitude for all sites. For throughfall fitting, the average D was 0.772 and average RMSE was 60.2 mm month-1. For model testing, the average D was 0.772 and average RMSE was 60.2 mm month-1. Throughfall measurements were limited for financial reasons to monthly or biweekly recordings using a series of nine funnel collectors (240 cm2 each) arranged along a regular grid with 5×5 m spacing. Mean deviation of recorded throughfall quantities was 5% in the managed forest and ranged between -6% and 21%. In the old-growth forest mean deviation was 10% and ranged between -12% and 22%. In the old-growth gap, mean deviation of recorded throughfall quantities was 6% and ranged between -16% and 17%, respectively. In the experimental gap, mean deviation of recorded throughfall quantities was only 2% and ranged between -4% and 19%. To cover throughfall variability and assure spatial representativeness, higher number of throughfall collectors is recommended [88], e. g. 35 to 40 collectors for heterogeneous uneven aged single species and mixed forest of two or more tree species, with small canopy gaps or even more than 40 collectors in case of big canopy gaps. For soil water content fitting, the average D was 0.865 and the average RMSE was 13.0 mm day-1. For model testing, the average D was 0.750 and the average RMSE was 15.6 mm day-1 (Figure S4).Figure S2 was corrected and now reads: Monthly throughfall (mm) - measured values and simulated with the Brook90 model for a) the experimental gap, b) old-growth gap, c) managed forest and d) old-growth forest
- Table S4 was corrected and data about throughfall in the old-growth gap were added. Now it reads: Linear regression (y = a + b*x) coefficients, linear correlation coefficient (r), index of agreement (D), root mean square error (RMSE), and sample size (n) describing the goodness-of-fit between Brook90 model simulated (y) and measured (x) daily values for the soil water contents (0-40 cm layer) and monthly throughfall (mm) for two calibration periods.
The results are only referred to as relative values with respect to precipitation. I would suggest including some absolute values as some dependencies seem to be opposed for relative or absolute values (e.g. ETP in wet and dry years). This applies also to the discussion.
- Agreed and corrected. Results and Discussion sections were rewritten accordingly.
The references are mostly up to date and from international journals.
I recommend the acceptance of the manuscript after a minor revision. A potential revision should consider to above and below-given annotations.
Specific comments
Page 3, row 93-104 However, it may be mentioned that Brook90 is only capable to deal with a dual permeability system. A threefold permeability system can’t be represented by Brook90.
- Agreed and corrected. Now it reads:
L 204 However, the model is capable to deal with a dual permeability system [1], whereas a threefold permeability system cannot be represented by this model.
Page 6, row 189 Nine collectors seem to be too few to represent the spatial heterogeneity of throughfall (Clarke et al. 2016 https://www.icp-forests.org/pdf/manual/2016/ICP_Manual_2016_01_part14.pdf). Please specify the potential sampling error of the throughfall measurements due to spatial variability.
- Agreed and corrected. Now it reads:
L 257: Throughfall measurements were limited for financial reasons to monthly or biweekly recordings using a series of nine funnel collectors (240 cm2 each) arranged along a regular grid with 5×5 m spacing. Mean deviation of recorded throughfall quantities was 5% in managed forest and ranged between -6% and 21%. In old-growth forest mean deviation was 10% and ranged between -12% and 22%. In the old-growth gap, mean deviation of recorded throughfall quantities was 6% and ranged between -16% and 17%, respectively. To cover throughfall variability and assure spatial representativeness, higher number of throughfall collectors is recommended [88], e. g. 35 to 40 collectors for heterogeneous uneven aged single species and mixed forest of two or more tree species, with small canopy gaps or even more than 40 collectors in case of big canopy gaps.
Please specify, if throughfall sampling was conducted in the old-growth gap (at least in 2013)?
- Throughfall was recorded in all sites. L 180-185 reads: From 2001 to 2007, monthly or biweekly throughfall (the amount of precipitation that is not intercepted by vegetation and reaches the forest floor) was recorded in forests and gaps at a height of 1.3 m using a series of nine funnel collectors (240 cm2 each) arranged along a regular grid with 5×5 m spacing. Incidental precipitation was recorded monthly from 2001 to 2007 in an open field near the study sites using the same type of funnel collectors. Data sets from 2003 to 2007 were used for the old-growth gap, that occurred naturally during the winter of 2002-2003.
- Figure S2 was corrected and now reads: Monthly throughfall (mm) - measured values and simulated with the Brook90 model for a) the managed forest, b) the old-growth forest and c) old-growth gap;
- Table S4 was corrected and data about throughfall in the old-growth gap were added. Now it reads: Linear regression (y = a + b*x) coefficients, linear correlation coefficient (r), index of agreement (D), root mean square error (RMSE), and sample size (n) describing the goodness-of-fit between Brook90 model simulated (y) and measured (x) daily values for the soil water contents (0-40 cm layer) and monthly throughfall (mm) for two calibration periods.
Please explain, why stemflow wasn’t assessed at the plots. Stemflow may significantly contribute to stand precipitation in beech forests (Clarke et al. 2016).
- Stemflow was recorded monthly on 6 trees with different diameters in each managed and old-growth forest. Results of this study were published in: Vilhar U (2010) Padavinski režim v vrzelih in sestojih dinarskega jelovo-bukovega gozda. Precipitation regime in gaps and mature stands of Dinaric silver fir-beech forests. Zbornik gozdarstva in lesarstva / Acta Silva et Ligni 91. 3-10.
- However, stemflow is not simulated by the Brook90 model so it was not used for model fitting or testing.
Page 6, row 199 Dirksen (1999) is missing in references.
- Agreed and corrected.
Page 8, row 256-261 Please give a unit for RSME values.
- Agreed and corrected. Now it reads:
L 255: For throughfall fitting, the average D was 0.772 and average RMSE was 60.2 mm month-1. For model testing, the average D was 0.772 and average RMSE was 60.2 mm month-1 (Figure S4).
L267: For soil water content fitting, the average D was 0.865 and the average RMSE was 13.0 mm day-1. For model testing, the average D was 0.750 and the average RMSE was 15.6 mm day-1 (Figure S4).
Supplementary material Table S4: Linear regression (y = a + b*x) coefficients, linear correlation coefficient (r), index of agreement (D), root mean square error (RMSE), and sample size (n) describing the goodness-of-fit between Brook90 model simulated (y) and measured (x) daily values for the daily soil water contents (mm day-1; 0-40 cm layer) and monthly throughfall (mm month-1) for two calibration periods.
Page 9, row 282-283 From Fig. 2 and regression parameters (row 283) it seems that SE is decreasing in the old-growth gap. Please explain.
- Agreed and corrected. Now it reads:
L 291: For the old-growth gap a statistically significant decrease in SE was shown during the simulation period (SE = -6.33 * year + 12743, N = 11, R2 = 0.645) due to shading of progressive natural tree regeneration.
