Early European Observations of Precipitation Partitioning by Vegetation: A Synthesis and Evaluation of 19th Century Findings
Round 1
Reviewer 1 Report
The manuscript “Early European observations ok precipitation partitioning by vegetation: A synthesis and evaluation of 19th Century literature” summarizes the German language literature about rainfall partitioning in the 19th Century as done before by Ebermayer (1897), present data graphically that have been published before only in different tables, provides location and distribution parameters of throughfall relative to stem distance and rainfall size based on detailed data from Hoppe (1896), and calculated linear relationships between stemflow and rainfall depth with the first stemflow data (Ney, 1893). Indeed it is of interest for the (non-German speaking) community to make this first data on rainfall partitioning available. But the manuscript has some shortcomings that need to be addressed. First the title suggests that all European 19th Century literature was reviewed, but in fact only the German language literature was included in this review. The title should be changed accordingly. The first aim was to describe the methods and study sites, but the description still lacks some information as stated in the specific comments. The second aim states that results are complied, analysed and contextualized. This aim is not satisfying as the reader is not told what the aim of the analysis and contextualization is. What do the authors what to find out? In addition, in the current version of the manuscript it is not always possible to trace the source of numbers provided or the choice of part of the numbers to include and the omission of other numbers. This should be central focus of this paper, and the authors might even consider publishing the numbers as supplementary material, which would be of added value for the non-German speaking community. The third aim was to share the German language study motivations and interpretations with the English speaking community. This is a valid aim, but the authors need to take much more care to report text passages and interpretations correctly (see specific comments). The last aim was to compare the 19th century findings to recent research. While the authors say that most of the old findings are based on only 1 to 3 collectors, they do not satisfactorily explain what implications this has for the comparison with modern studies. In fact, the comparison with modern studies should be improved as stated in the specific comments. Comparisons are only made in the text, but not in figures and tables. And in the text it often says that numbers are in line with recent ranges and some reference is given. It would convince the reader more, if numbers are directly compared in this manuscript graphically or in tables.
To summarize, reworking this old literature has to be made very carefully and accurate as this old references are not available or readable for most of the actual scientific community. Traceability, completeness and correctness of numbers and statements are essential. Therefore I can’t recommend the current version of the manuscript for publication in Geosciences.
Specific comments:
Line 12: “returns water back to the atmosphere” equals evaporation and not interception. Of cause intercepted water also gets evaporated later, but both terms should be mentioned.
Line 13: “to the surface”: the surface of what? Vegetation, soil, litter, forest?
Line 15: “large number”: substitute with the actual number and the reader can judge himself if this is a large number.
Line 17: Delete “incredible”. This is too wordy for a scientific publication.
Line 18: “their data is digitized”: add that is was only partly digitized. I can’t see where the detailed data of the Bavarian, Prussian and Saxon network original data was used for in this manuscript. I could only identify summary data from earlier reviews in this manuscript. If I’m wrong, please clearly indicate for which statements the original digitized data was used. Minor remark: this should be past tense.
Line 18: Add the aim of data analysis and the comparison to recent studies.
Line 20ff: Not all mentioned studies and networks had the strength of detail. Please be more precise. Stand tree species and age should not be part of this rigorous level of detail. Of cause this is essential information in forestry which can’t be omitted. Worth to mention here is indeed the spatial resolution of daily throughfall collection of 1-3 studies.
Line 22: Delete “intra-event sampling” as I could not find any of that in the manuscript.
Line 22: Delete “(just to name a few)”. If there are other rigorous details please mention them in the abstract. In the manuscript I couldn’t find additional examples.
Line 23: Add that this is restricted to German language literature.
Line 25: “in many ways” – be more precise and tell the reader in which ways.
Line 31: As this is a review on 19th Century literature it would be more interesting to get to know what was the vegetation cover in that time than today. Or you could compare it to the current state. Also the global coverage is less of interest compared to the coverage in the area relevant for the review.
Line 36: Delete “often-significant”. What is significant or insignificant?
Line 39: “international attention”: The 19th Century scientist also exchanged themselves on an international level, but in German language. For example, as you cited it, there is the report on the Meeting of the international union of experimental stations in forestry (Ney, 1894). Therefore please correct your statement.
Line 41-43: “rarely cited” again, I assume you mean the literature in English language. Please correct. Furthermore, it is also not correct that this literature has not been cited by English language reviews (e.g. Zon, 1927).
Line 50: If land use change was the essential motivation, than it should be described here as far as possible.
Line 51-74: The description of motivation is part of aim (iii) and should be reported after, but not in the introduction. Please move this to another section. E.g. chapter 2.
Line 77-78: This should be part of the results.
Line 81-83: This should be part of the conclusion, as it is a conclusion that can only be verified by the reader after having read the whole paper.
Line 83: Is this really planned as a contribution of a special issue?
Figure 1:
It is hard to read the symbols in a black and white printout. Different symbols in addition to different colours could improve this. It would be better suited to have a 19th Century map as background. It is not possible to link the stations on the map with the stations in the list. The stations could be numbered in the list and on the map for easy identification by the reader. There are several Saxon and Bavarian stations printed on the map, but in the list single stations are only listed for Prussia. This should be added to the list as well. Where did you get the coordinate from for the stations on the map. References needed or indication how they were obtained. What makes you think that there were further stations that you could not add to the map?Line 104: It is quite normal to have the knowledge of forest age in forestry. But “rather detailed information” is exaggerated and should be omitted.
Table 1:
Open area rainfall instead of interception should be mentioned as interception was calculated from throughfall and rainfall. The header is repeated in the middle of the table on page 5, but should be repeated only on the next page.Line 171: Which studies? This information should be included in table 1.
Line 195: Which ones are the current estimates? Please add references.
Line 218: Change “patterns” to “variability”.
Line 234f: To calculate this “portion” of rainfall diverted to stems, one needs an arial basis for calculation. Was this done on basis of the canopy area of single canopies, on a field basis, or on a stem basal area basis?
Line 236: “limited number” Please indicate how many observation points were monitored.
Line 261ff: Please add how he calculated it. For better understanding, it would also be good to name at least the most important current methods.
Line 275: Change forest to tree age? It is the same age, but the throughfall was estimated for single trees and not for the forest.
Line 275: Why do you think that the annual estimates are more robust than the monthly ones? Please give details in the text.
Figure 3a: It is impossible for the reader to trace the values included in this figures. They are taken from the review of Ebermayer (1897), but not all values from this old review are plotted in figure 3a. E.g. why did you not include 98% throughfall from 20year old Fagus? The data shown in this figure need to be traceable.
Line 275ff and Figure 3a: The authors included forests with different closure in this table. As the collectors were located at the edges of canopies, it is obvious that this figure will not show a clear difference between needleleaved and broadleaved forests. And 19th Century were well aware of this fact.
Line 279f: This only indicates that forest structure was different but it does not say anything about canopy structure of single trees. This is also what Ebermayer said in 1897.
Figure 3:
3a – what do dots and lines indicate? 3c – Why did you include Fagus values from Page 288 (Ebermayer, 1897), but not Picea values from the same table? They do not fit so well, to make your point. But what is the valid reason to omit them?Line 307f: Why should be the closure higher for older stands? These were managed forests of which part of the trees are harvested in more or less regular intervals. Is this really what Ebermayer says? You should check the text. And Figure 3c only shows the relation of throughfall to age and not what you mentioned in this line. Please correct.
Line 313: The crowns do not increase winter moisture, but they decrease less the incoming rainfall.
Table 4 and line 344: Relative through can’t be collected. Only absolute rainfall and throughfall can be collected.
Line 348f: “rarely reported” – in the 19th Century? Please rewrite. Now it sounds as today (since mid 20th Century) these events were rarely reported and this is not true.
Figure 4: Probably the data was not collected by Ebermayer himself. Therefore please indicate who published the data first and at which station they were collected.
Line 362f: The number of gauges is reported twice. Please delete this in one of the two sentences.
Figure 5:
Add the sampling size per box. Indicate the meaning of the boxes whiskers as there is not one standard for it. Indicate why some boxes don’t have wiskers.Line 380: “permitting the estimation”: Please specify which criteria need to be fulfilled to permit it.
Line 381: typo in (iii)
Line 383f: Please indicate if these ranges of recent studies have been calculated for the same tree species.
Line 390: This is not true for the German language literature.
Table 5:
Change “Class” to an expression with more information (e.g. distance to…). Also name the classes listed in the second line. Mention in the table header that these numbers are about throughfall.Line 401: There was enough publications out to compile a review about SF in 2003 (Levia and Frost, 2003). Your statement is not correct.
Line 405: This is not surprisingly as this reference reports a range of 0 to 34% of stemflow. Every measurement can fall in this range. I’m sure you can find better references to make your point.
Figure 6:
6a: indicate which data points were obtained from Ebermayer and which ones from Riegler. And as Ebermayer was probably citing other studies, please provide the original data source. Always indicate to what the stemflow depth (mm) is related to. Stem basal area or projected canopy area? The stemflow data should also be compared to common measures used in the recent literature as done for throughfall in this manuscript. Please use the funnelling ratio on basis of stem basal area and canopy projection area. Figure caption line 4: this is not from an example stand, but from an example tree, as only one (?) tree was sampled.Since 427f: This is not correct. It has been observed much more often as summarized in recent reviews.
Line 429: Rewrite this sentence as it is too wordy.
Line 439ff: It is not appropriate to repeat your own (long) text here. A lot of other researchers also described this before.
Line 452: “as mentioned earlier” Where? It was only shown in the figure, but I think it was not mentioned in the text.
Line 477ff: This translation is wrong. This is about the needed correction due to stemflow that was not measured. As it is written in the manuscript, it says that the researchers in the end of the 19th Century didn’t know if interception was 20 or 80% and this is of cause not true.
References
Ebermayer, G., 1897. Untersuchungs-Ergebnisse über die Menge und Vertheilung der Niederschläge in den Wäldern. Rieger, München.
Hoppe, E., 1896. Regenmessung unter Baumkronen. Mitteilungen aus dem förstlichen Versuchswesen Österreichs.
Levia, D.F., Frost, E.E., 2003. A review and evaluation of stemflow literature in the hydrologic and biogeochemical cycles of forested and agricultural ecosystems. Journal of Hydrology 274, 1–29.
Ney, K.E., 1894. Ueber die Messung des an den Schäften der Bäume herabfliessenden Regenwassers, in: Bericht Über Die ... Versammlung Des Internationalen Verbandes Forstlicher Versuchsanstalten : Zu Mariabrunn. Frick, Wien.
Ney, K.E., 1893. Der Wald und die Quellen. Pietzker, Tübingen.
Zon, R., 1927. Forests and water in the light of scientific investigation By Raphael Zon. Forest service, United States Department of agriculture. Govt. print.off., Washington.
Author Response
R1
The manuscript “Early European observations ok precipitation partitioning by vegetation: A synthesis and evaluation of 19th Century literature” summarizes the German language literature about rainfall partitioning in the 19th Century as done before by Ebermayer (1897), present data graphically that have been published before only in different tables, provides location and distribution parameters of throughfall relative to stem distance and rainfall size based on detailed data from Hoppe (1896), and calculated linear relationships between stemflow and rainfall depth with the first stemflow data (Ney, 1893). Indeed it is of interest for the (non-German speaking) community to make this first data on rainfall partitioning available. But the manuscript has some shortcomings that need to be addressed. First the title suggests that all European 19th Century literature was reviewed, but in fact only the German language literature was included in this review. The title should be changed accordingly. The first aim was to describe the methods and study sites, but the description still lacks some information as stated in the specific comments. The second aim states that results are complied, analysed and contextualized. This aim is not satisfying as the reader is not told what the aim of the analysis and contextualization is. What do the authors what to find out? In addition, in the current version of the manuscript it is not always possible to trace the source of numbers provided or the choice of part of the numbers to include and the omission of other numbers. This should be central focus of this paper, and the authors might even consider publishing the numbers as supplementary material, which would be of added value for the non-German speaking community. The third aim was to share the German language study motivations and interpretations with the English speaking community. This is a valid aim, but the authors need to take much more care to report text passages and interpretations correctly (see specific comments). The last aim was to compare the 19th century findings to recent research. While the authors say that most of the old findings are based on only 1 to 3 collectors, they do not satisfactorily explain what implications this has for the comparison with modern studies. In fact, the comparison with modern studies should be improved as stated in the specific comments. Comparisons are only made in the text, but not in figures and tables. And in the text it often says that numbers are in line with recent ranges and some reference is given. It would convince the reader more, if numbers are directly compared in this manuscript graphically or in tables.
To summarize, reworking this old literature has to be made very carefully and accurate as this old references are not available or readable for most of the actual scientific community. Traceability, completeness and correctness of numbers and statements are essential. Therefore I can’t recommend the current version of the manuscript for publication in Geosciences.
Reply: Thank you for your detailed review, we actually agree with many of your suggestions. Replies to the specific comments are below.
Specific comments:
Line 12: “returns water back to the atmosphere” equals evaporation and not interception. Of cause intercepted water also gets evaporated later, but both terms should be mentioned.
Reply: We clarified both processes in the text.
Line 13: “to the surface”: the surface of what? Vegetation, soil, litter, forest?
Reply: We changed the wording.
Line 15: “large number”: substitute with the actual number and the reader can judge himself if this is a large number.
Reply: We changed the sentence instead of reporting numbers as the sentence talks about both observatories and experimental stations. As observatories consist of several stations detailed numbers would be too confusing for the abstract.
Line 17: Delete “incredible”. This is too wordy for a scientific publication.
Reply: We changed the wording.
Line 18: “their data is digitized”: add that is was only partly digitized. I can’t see where the detailed data of the Bavarian, Prussian and Saxon network original data was used for in this manuscript. I could only identify summary data from earlier reviews in this manuscript. If I’m wrong, please clearly indicate for which statements the original digitized data was used. Minor remark: this should be past tense.
Reply: We have changed the wording to be more precise. Only results included in the literature synthesis has been digitized (and are now directly referenced in a new supplemental material). As with any literature synthesis effort, most results relevant to current theory are digitized and presented. Digitization of raw data from these studies, although available for a rare few (like Hoppe), was not possible. We also argue that a full raw data digitization effort is not necessary for a high-impact review and evaluation (note that such an effort has not done for other high-impact reviews on this topic: see modern review references in the manuscript).
Line 18: Add the aim of data analysis and the comparison to recent studies.
Reply: The aim is provided at the end of the paragraph: “to reveal the roots of interest in precipitation partitioning processes and represent a generally forgotten piece of history shared by the hydrology, meteorology, forestry and agricultural scientific communities. … we hope modern scientists interested in plant-precipitation interactions will find new inspiration in our synthesis and evaluation of this literature.”
Line 20ff: Not all mentioned studies and networks had the strength of detail. Please be more precise. Stand tree species and age should not be part of this rigorous level of detail. Of cause this is essential information in forestry which can’t be omitted. Worth to mention here is indeed the spatial resolution of daily throughfall collection of 1-3 studies.
Reply: Thank you for your comment. This is definitely true for many forestry studies. Ecohydrology, nowadays, however, is not solely linked to forestry studies, but a large body of literature comes from ecohydrology which is more embedded in hydrology than forestry. Therefore, for many recent studies, forest structural information (like stand age) are not reported. Reasons for this may be that observed forests are not necessarily managed and that resources are lacking to conduct dendrochronological studies of the observed trees. Some recent studies do include stand age and evenchronosequential analyses, however, this is not the norm in all precipitation partitioning studies, especially those that do not stem from the forestry community.
Line 22: Delete “intra-event sampling” as I could not find any of that in the manuscript.
Reply: We removed this statement.
Line 22: Delete “(just to name a few)”. If there are other rigorous details please mention them in the abstract. In the manuscript I couldn’t find additional examples.
Reply: We removed this statement.
Line 23: Add that this is restricted to German language literature.
Reply:We believe that it is apparent that the literature is predominantly in German, but have also added details in the introduction.
Line 25: “in many ways” – be more precise and tell the reader in which ways.
Reply: To keep the abstract brief, we have removed this term and explained these studies’ relevance within the manuscript body.
Line 31: As this is a review on 19th Century literature it would be more interesting to get to know what was the vegetation cover in that time than today. Or you could compare it to the current state. Also the global coverage is less of interest compared to the coverage in the area relevant for the review.
Reply: We agree that some brief introduction and discussion of 19thcentury vegetation patterns compared to today would be useful. However, the purpose of this paper is to show the importance of precipitation partitioning as a field during this time and discuss the knowledge - not necessarily the historic context of land cover change (which could be its own, very long review paper). Thus, in the introduction we include some estimates and a brief discussion. Although this is rather difficult to assess exactly, we have added results from a study that analyzed 110-year land cover changes in Europe (Fuchs et al.). Fuchs et al. do not completely coincide with the time frame reported in our manuscript but do give a good estimate as to land cover in Europe around 1900. We have included the studies in the introduction.
Line 36: Delete “often-significant”. What is significant or insignificant?
Reply: We have rephrased the sentence.
Line 39: “international attention”: The 19th Century scientist also exchanged themselves on an international level, but in German language. For example, as you cited it, there is the report on the Meeting of the international union of experimental stations in forestry (Ney, 1894). Therefore please correct your statement.
Reply: True – we have modified the sentence. Horton 1919 is arguably the most widely cited benchmark paper as a starting point for modern studies on precipitation partitioning (a comparison of citing literature in Google Scholar makes this very clear). That, of course, should not suggest that 19thcentury or even earlier scholars did not engage in international cooperation.
Line 41-43: “rarely cited” again, I assume you mean the literature in English language. Please correct. Furthermore, it is also not correct that this literature has not been cited by English language reviews (e.g. Zon, 1927).
Reply: We have better contextualized the statements and cited Zon (1927).
Line 50: If land use change was the essential motivation, than it should be described here as far as possible.
Reply:Land use change and the increasing deforestation, linked to earlier and ongoing industrialization at the time of the early observations was definitely a motivation for studies to investigate the effect of forests on the water cycle. To put the recent land cover numbers stated earlier into context to the 19thcentury land cover we point to the studies by Fuchs et al who reconstructed European land cover from 1900 to 2010. Although not fully covering the time period that is of relevance to this manuscript it gives an indication of how land cover changed within Europe over the past decades. We modified the text accordingly.
Fuchs, M. Herold, P.H. Verburg, J.G.P.W. Clevers, J. Eberle (2015): Gross changes in reconstructions of historic land cover/use for Europe between 1900-2010, Global Change Biology, doi: 10.1111/gcb.12714
R. Fuchs, M. Herold, P.H. Verburg, J.G.P.W. Clevers (2013): A high-resolution and harmonized model approach for reconstructing and analysing historic land changes in Europe, Biogeosciences, 10(3), 1543–1559, doi:10.5194/bg-10-1543-2013
Line 51-74: The description of motivation is part of aim (iii) and should be reported after, but not in the introduction. Please move this to another section. E.g. chapter 2.
Reply: Moved to section 2.
Line 77-78: This should be part of the results.
Reply: Moved to section 2.
Line 81-83: This should be part of the conclusion, as it is a conclusion that can only be verified by the reader after having read the whole paper.
Reply: Moved.
Line 83: Is this really planned as a contribution of a special issue?
Reply: Yes, to a special issue on rainfall and evaporation partitioning.
Figure 1:
It is hard to read the symbols in a black and white printout. Different symbols in addition to different colours could improve this. It would be better suited to have a 19th Century map as background. It is not possible to link the stations on the map with the stations in the list. The stations could be numbered in the list and on the map for easy identification by the reader. There are several Saxon and Bavarian stations printed on the map, but in the list single stations are only listed for Prussia. This should be added to the list as well. Where did you get the coordinate from for the stations on the map. References needed or indication how they were obtained. What makes you think that there were further stations that you could not add to the map?
Reply: We have modified the map to ensure better distinction of station networks utilizing http://colorbrewer2.org and different symbols. The purpose of the map is to show the spatial extent that were covered by these very early observations, not so much to link each station in table one to a point as the map would get more confusing. The station coordinates are now provided as supplementary material S1.
In addition, some literature states numbers for certain tree species from e.g. Bavarian or Prussian stations without mentioning the exact stations from which the data originated.
Locations were taken from the different literature references (e.g. Müttrich) and where coordinates were missing, the locations were researched using auxiliary information stated in the literature, e.g. vicinity to major landmarks such as hilltops or forest names.
With regard to the comment that there may have been further stations, as mentioned above sometimes networks and not stations are referenced – in addition the networks changed over time in the sense that stations were added or discontinued. Since we do not always have a clear overview of all stations but the list is more a compilation of several sources we chose to mention that there is the possibility of incompleteness. After all, also the raw data from these early observations are mostly not available and have, as Ney reported about his first stemflow observations, been lost.
Line 104: It is quite normal to have the knowledge of forest age in forestry. But “rather detailed information” is exaggerated and should be omitted.
Reply:That is definitely true for forestry. With the growing interdisciplinarity, however, we feel that it is important to mention this and hopefully motivate more colleagues to collect these data – after all they managed to do so more than 100 years ago… In ecohydrology, for example, where scholars also come from hydrological backgrounds, collecting such data is not always standard. At the same time, we strongly agree with the reviewer that such data should be part of every study!
Table 1:
Open area rainfall instead of interception should be mentioned as interception was calculated from throughfall and rainfall. The header is repeated in the middle of the table on page 5, but should be repeated only on the next page.
Reply: Thank you for your comment. The header issue was caused by later typesetting. It should have been located at the page break. Since any study must measure open precipitation to estimate interception, we do not believe this parameter should be included in the table instead of “interception.” Moreover, many studies did not report open rainfall, but only reported interception estimates.
Line 171: Which studies? This information should be included in table 1.
Reply: Thank you for your comment. However, we believe that readers are given enough information in Table 1, the proceeding text discussing the studies, and the new supplemental materials (S1 & S2) to link specific ancillary details to the findings presented. As such, we prefer not to disrupt the sentence with the suggested (already often-referenced) citations.
Line 195: Which ones are the current estimates? Please add references.
Reply: We revised the sentence as requested.
Line 218: Change “patterns” to “variability”.
Reply: We revised the sentence as requested.
Line 234f: To calculate this “portion” of rainfall diverted to stems, one needs an arial basis for calculation. Was this done on basis of the canopy area of single canopies, on a field basis, or on a stem basal area basis?
Reply: Our text (and Ney’s study) mentions that Wollny, unfortunately, did not collect stemflow. Recent studies that collect stemflow of crops do provide such numbers (review by Sadeghi et al., 2019). Wollny observed throughfall under crop canopies within 4 m2planting beds using a rain gauge that was levelled with the soil surface. Information on plan density was not provided, but is also not required to estimate interception.
Stemflow reported in the early studies was either reported in [l] or in [mm], converted using the crown projected area.
Line 236: “limited number” Please indicate how many observation points were monitored.
Reply: We revised the sentence as requested.
Line 261ff: Please add how he calculated it. For better understanding, it would also be good to name at least the most important current methods.
Reply: Thank you for your comment and we have revised accordingly.
Line 275: Change forest to tree age? It is the same age, but the throughfall was estimated for single trees and not for the forest.
Reply: We changed the sentence.
Line 275: Why do you think that the annual estimates are more robust than the monthly ones? Please give details in the text.
Reply: Thank you for your comment. In general, averaged values for longer time periods are more robust as more measurements entered the calculation of the average. This, of course, only holds true in case the amount of observations are higher at annual than at monthly scale. This may be flipped for short intensive campaigns limited to a single month but, in general, more observations equal more statistical robustness.
Figure 3a: It is impossible for the reader to trace the values included in this figures. They are taken from the review of Ebermayer (1897), but not all values from this old review are plotted in figure 3a. E.g. why did you not include 98% throughfall from 20year old Fagus? The data shown in this figure need to be traceable.
Reply: Thank you for your comment. This 98% throughfall value was provided in Figure 3c. It is now also provided in Figure 3a.
Line 275ff and Figure 3a: The authors included forests with different closure in this table. As the collectors were located at the edges of canopies, it is obvious that this figure will not show a clear difference between needleleaved and broadleaved forests. And 19th Century were well aware of this fact.
Reply: Thank you for your comment. We now acknowledge this with a direct quote.
Line 279f: This only indicates that forest structure was different but it does not say anything about canopy structure of single trees. This is also what Ebermayer said in 1897.
Figure 3:
3a – what do dots and lines indicate? 3c – Why did you include Fagus values from Page 288 (Ebermayer, 1897), but not Picea values from the same table? They do not fit so well, to make your point. But what is the valid reason to omit them?
Reply: Thank you for your comment. We agree with the reviewer and note that we did not state anything about the structure of individual trees. As for the Picea values, these were omitted because they did not represent a chronosequence (only two different ages); whereas the data used represent a broader age range. However, these two Picea age points do support our point: Relative throughfall decreased with stand age.
Line 307f: Why should be the closure higher for older stands? These were managed forests of which part of the trees are harvested in more or less regular intervals. Is this really what Ebermayer says? You should check the text. And Figure 3c only shows the relation of throughfall to age and not what you mentioned in this line. Please correct.
Reply: Thank you for your comment. The exact relationships between stand age and canopy closure for individual species and individual sites with specific management practices certainly can confound interpretation. We have edited the sentence to simply describe the trend in Figure 3c.
Line 313: The crowns do not increase winter moisture, but they decrease less the incoming rainfall.
Reply: Thank you for your comment. This was a poor translation (by the English-speaking co-author) that has been revised.
Table 4 and line 344: Relative through can’t be collected. Only absolute rainfall and throughfall can be collected.
Reply: Thank you for your comment. Revised as requested.
Line 348f: “rarely reported” – in the 19th Century? Please rewrite. Now it sounds as today (since mid 20th Century) these events were rarely reported and this is not true.
Reply: Thank you for your comment. Sentence deleted.
Figure 4: Probably the data was not collected by Ebermayer himself. Therefore please indicate who published the data first and at which station they were collected.
Reply: Thank you for your comment. We do not see where Ebermayer cites the data source. If the reviewer has this information, we will gladly include it.
Line 362f: The number of gauges is reported twice. Please delete this in one of the two sentences.
Reply: Done.
Figure 5:
Add the sampling size per box. Indicate the meaning of the boxes whiskers as there is not one standard for it. Indicate why some boxes don’t have wiskers.
Reply: We have adapted the figure caption.
Line 380: “permitting the estimation”: Please specify which criteria need to be fulfilled to permit it.
Reply: Thank you for your comment. Revised for clarity.
Line 381: typo in (iii)
Reply: Done.
Line 383f: Please indicate if these ranges of recent studies have been calculated for the same tree species.
Reply: We have added more information.
Line 390: This is not true for the German language literature.
Reply: Thank you for your comment. If you know specific literature to support this statement, please do share. In the revised manuscript draft, based on our knowledge of the literature, we have modified the sentence.
Table 5:
Change “Class” to an expression with more information (e.g. distance to…). Also name the classes listed in the second line. Mention in the table header that these numbers are about throughfall.
Reply: This “class” indicates storm size classes (as described in the table caption).
Line 401: There was enough publications out to compile a review about SF in 2003 (Levia and Frost, 2003). Your statement is not correct.
Reply: We have changed the sentence to make this point clearer, as stemflow is, compared to e.g. throughfall, much less studied. However, we would like to note that simply having“enough publications out to compile a review" does not invalidate the statement that “stemflow was little-researched.” In fact, Levia and Frost (2003) explicitly state several times that “few studies” at that time examined fundamental aspects of stemflow generation and chemistry.
Line 405: This is not surprisingly as this reference reports a range of 0 to 34% of stemflow. Every measurement can fall in this range. I’m sure you can find better references to make your point.
Reply: Thank you for your comment. We have added clearer references.
Figure 6:
6a: indicate which data points were obtained from Ebermayer and which ones from Riegler. And as Ebermayer was probably citing other studies, please provide the original data source. Always indicate to what the stemflow depth (mm) is related to. Stem basal area or projected canopy area? The stemflow data should also be compared to common measures used in the recent literature as done for throughfall in this manuscript. Please use the funnelling ratio on basis of stem basal area and canopy projection area. Figure caption line 4: this is not from an example stand, but from an example tree, as only one (?) tree was sampled.
Reply: Thank you for your comment. The caption was corrected. All stemflow estimates are normalized by projected crown area. We did not calculate funneling ratios.
Since 427f: This is not correct. It has been observed much more often as summarized in recent reviews.
Reply: Thank you for your comment. We do not disagree per se, however, we cannot reply to this comment unless the respective references are provided that negate the references we chose.
We very much appreciate the level of detail and the many constructive comments given by R1, however, statements such as ‘summarized in recent reviews’ without providing detail is far from constructive.
Line 429: Rewrite this sentence as it is too wordy.
Reply: Thank you for your comment. We do not agree that a sentence with 7 words and one number is too wordy. In case you disagree with the content we are happy to revise.
The second sentence, covering about two lines, is, to our understanding also not excessively wordy. Given the journal’s reference style it made sense to add the publication years and mention the timespan between the publications.
“Indeed, 19thcentury stemflow observers were quite keen. Very early (in 1881) the stemflow process was already conceptually described as a “funnel” [29](similar to Herwitz [74]over a century later, in 1986).”
Line 439ff: It is not appropriate to repeat your own (long) text here. A lot of other researchers also described this before.
Reply: The reviewer is incorrect as the quote mentioned is notour own text – it is a translation of the first known text to make these conceptual descriptions of the stemflow process. As such, we disagree and did not delete this. Yet, we have added more information to clearly show the origin of the referenced text.
Line 452: “as mentioned earlier” Where? It was only shown in the figure, but I think it was not mentioned in the text.
Reply: Thank you for your comment. As mentioned in the quotation from Riegler.
Line 477ff: This translation is wrong. This is about the needed correction due to stemflow that was not measured. As it is written in the manuscript, it says that the researchers in the end of the 19thCentury didn’t know if interception was 20 or 80% and this is of cause not true.
Reply: True, we corrected the translation.
References
Ebermayer, G., 1897. Untersuchungs-Ergebnisse über die Menge und Vertheilung der Niederschläge in den Wäldern. Rieger, München.
Hoppe, E., 1896. Regenmessung unter Baumkronen. Mitteilungen aus dem förstlichen Versuchswesen Österreichs.
Levia, D.F., Frost, E.E., 2003. A review and evaluation of stemflow literature in the hydrologic and biogeochemical cycles of forested and agricultural ecosystems. Journal of Hydrology 274, 1–29.
Ney, K.E., 1894. Ueber die Messung des an den Schäften der Bäume herabfliessenden Regenwassers, in: Bericht Über Die ... Versammlung Des Internationalen Verbandes Forstlicher Versuchsanstalten : Zu Mariabrunn. Frick, Wien.
Ney, K.E., 1893. Der Wald und die Quellen. Pietzker, Tübingen.
Zon, R., 1927. Forests and water in the light of scientific investigation By Raphael Zon. Forest service, United States Department of agriculture. Govt. print.off., Washington.
Reviewer 2 Report
Dear Autors,
The article is very clear and is great to read.
In my opinion, such a synthesis of knowledge about penetration of precipitation through the crowns of trees is very much needed.
Charts and statements very legible.Well done!
Scientific positions to which the authors relate very well selected.
Author Response
R2
The article is very clear and is great to read.
In my opinion, such a synthesis of knowledge about penetration of precipitation through the crowns of trees is very much needed.
Charts and statements very legible.Well done!
Scientific positions to which the authors relate very well selected.
Reply: We thank the reviewer for the great compliment.
Reviewer 3 Report
This study reviews 'old' lierature related to rainfall partioning across Europe.
Personally, I find the article interesting and a good contribution to the literature and well written. I can't comment as to its relevence to a special issue, as the topic of that is unclear to me.
I have a few minor comments which I hope will help to clarify a few things.
I am not clear on the 'why' of this manuscript. Why is this information important and how can it help with water resource 'management outcomes' ie a short section discussing how this information is related to land use change (for example) and our understanding of water resource partitioning, especially when clearing natural forest to plant commercial forest or converting grassland to crops etc I would like to see a bit more international relevence. Pinus for example has published information on through-fall and stem flow from many regions with these two paper very relevent and from Australia (Benyon RG, Doody TM. 2014. Comparison of interception, forest floor evaporation and transpiration in Pinus radiata and Eucalyptus globulus. Hydrological Processes, 29(6), 1173-1187; Benyon RG, Thieveanathan T, Doody TM. 2006. Impacts of tree plantations on groundwater in south-eastern Australia. Australian Journal of Botany. 54: 181-192.) This data could be compered and discussed with Figure 2 from different continents to add more interest to global readers. 'how' can this data be used by others - why might it be of interest to your readers? Just something to think about.Please remove 'already' from ln 15
Can you please add year of publication to Table 1 - I had to keep looking at the reference list to identfy this which made it annoying (given the reference style of the journal).
Author Response
R3
This study reviews 'old' lierature related to rainfall partioning across Europe.
Personally, I find the article interesting and a good contribution to the literature and well written. I can't comment as to its relevence to a special issue, as the topic of that is unclear to me.
I have a few minor comments which I hope will help to clarify a few things.
I am not clear on the 'why' of this manuscript. Why is this information important and how can it help with water resource 'management outcomes' ie a short section discussing how this information is related to land use change (for example) and our understanding of water resource partitioning, especially when clearing natural forest to plant commercial forest or converting grassland to crops etc I would like to see a bit more international relevence. Pinus for example has published information on through-fall and stem flow from many regions with these two paper very relevent and from Australia (Benyon RG, Doody TM. 2014. Comparison of interception, forest floor evaporation and transpiration in Pinus radiata and Eucalyptus globulus. Hydrological Processes, 29(6), 1173-1187; Benyon RG, Thieveanathan T, Doody TM. 2006. Impacts of tree plantations on groundwater in south-eastern Australia. Australian Journal of Botany. 54: 181-192.) This data could be compered and discussed with Figure 2 from different continents to add more interest to global readers. 'how' can this data be used by others - why might it be of interest to your readers? Just something to think about.
Reply: Thank you for your comment. We have added a section on past and present land use and references (Fuchs et al.) that specifically address land cover change in Europe from 1900 to 2010.
With respect to international relevance we have read the two papers with great interest and have provided more references in the sections on throughfall and stemflow that put the early results in context to current findings. We did, however, not use the provided references but instead focused on references that stem from similar climate regions as the early European studies.
Please remove 'already' from ln 15
Reply: We changed the sentence.
Can you please add year of publication to Table 1 - I had to keep looking at the reference list to identfy this which made it annoying (given the reference style of the journal).
Reply: We added this information to a new table (S2) in the supplemental materials.
Round 2
Reviewer 1 Report
The manuscript has been improved substantially and most of my comments were addressed satisfactorily. There are only three issues that I still don’t agree with and which should be addressed before publication:
Line 18: Add the aim of data analysis and the comparison to recent studies.
Reply: The aim is provided at the end of the paragraph: “to reveal the roots of interest in precipitation partitioning processes and represent a generally forgotten piece of history shared by the hydrology, meteorology, forestry and agricultural scientific communities. … we hope modern scientists interested in plant-precipitation interactions will find new inspiration in our synthesis and evaluation of this literature.”
Reviewer: I have read this text. Line 18 is part of the abstract and I think that the aim of data analysis should be stated here (and in the end of the introduction). Data analysis is used to answer a question. Or perhaps the data analysis only consists of calculating location and dispersion measures of the data in order to compare them to recent studies. That is ok, it but should be stated what the aim of the data analysis was.
Line 275: Why do you think that the annual estimates are more robust than the monthly ones? Please give details in the text.
Reply: Thank you for your comment. In general, averaged values for longer time periods are more robust as more measurements entered the calculation of the average. This, of course, only holds true in case the amount of observations are higher at annual than at monthly scale. This may be flipped for short intensive campaigns limited to a single month but, in general, more observations equal more statistical robustness.
Reviewer: The robustness of an estimate depends on the existence of outliers. And as you said, the annual estimates are only more robust if their sample size is bigger. Therefore it is not the annual versus the monthly that influences the robustness, but the sample size. This should be corrected in the text.
Since 427f: This is not correct. It has been observed much more often as summarized in recent reviews.
Reply: Thank you for your comment. We do not disagree per se, however, we cannot reply to this comment unless the respective references are provided that negate the references we chose.
We very much appreciate the level of detail and the many constructive comments given by R1, however, statements such as ‘summarized in recent reviews’ without providing detail is far from constructive.
Reviewer: This is how you make a reviewer work for you… Thanks for the compliment! Examples are: Iida, Kakubari, & Tanaka (2005), Abrahams et al. (2003), Chinen (2007), Germer (2013), Crabtree & Trudgill (1985), Martinez-Meza & Whitford (1996). If this is not enough, than please continue yourself with the search!
Abrahams, A. D., Parsons, A. J., & Wainwright, J. (2003). Disposition of rainwater under creosotebush. Hydrological Processes, 17(13), 2555–2566. https://doi.org/10.1002/hyp.1272
Chinen, T. (2007). An observation of surface runoff and erosion caused by Acacia albida stemflow in dry savanna in the south-western Republic of Niger. Geographical Reports of Tokyos Metropolian University, 42, 21–30.
Crabtree, R. W., & Trudgill, S. T. (1985). Hillslope hydrochemistry and stream response on a wooded, permeable bedrock: The role of stemflow. Journal of Hydrology, 80, 161–178.
Germer, S. (2013). Development of near-surface perched water tables during natural and artificial stemflow generation by babassu palms. Journal of Hydrology, 507(0), 262–272. https://doi.org/10.1016/j.jhydrol.2013.10.026
Iida, S., Kakubari, J., & Tanaka, T. (2005). “Litter marks” indicating infiltration area of stemflow-induced water. Tsukuba Geoenvironmental Sciences, 1, 27–31.
Martinez-Meza, E., & Whitford, W. G. (1996). Stemflow, throughfall and channelization of stemflow by roots in three Chihuahuan desert shrubs. Journal of Arid Environments, 32(3), 271–287. (ISI:A1996UK36900005).
Author Response
Initial reviewer comment:
Line 18: Add the aim of data analysis and the comparison to recent studies.
Initial author response:
The aim is provided at the end of the paragraph: “to reveal the roots of interest in precipitation partitioning processes and represent a generally forgotten piece of history shared by the hydrology, meteorology, forestry and agricultural scientific communities. … we hope modern scientists interested in plant-precipitation interactions will find new inspiration in our synthesis and evaluation of this literature.”
R2 reviewer comment:
I have read this text. Line 18 is part of the abstract and I think that the aim of data analysis should be stated here (and in the end of the introduction). Data analysis is used to answer a question. Or perhaps the data analysis only consists of calculating location and dispersion measures of the data in order to compare them to recent studies. That is ok, it but should be stated what the aim of the data analysis was.
R2 author response:
We have revised the wording to show that the data analysis had the aim to make the early results comparable to recent findings.
Initial reviewer comment:
Line 275: Why do you think that the annual estimates are more robust than the monthly ones? Please give details in the text.
Initial author response:
Thank you for your comment. In general, averaged values for longer time periods are more robust as more measurements entered the calculation of the average. This, of course, only holds true in case the amount of observations are higher at annual than at monthly scale. This may be flipped for short intensive campaigns limited to a single month but, in general, more observations equal more statistical robustness.
R2 reviewer comment:
The robustness of an estimate depends on the existence of outliers. And as you said, the annual estimates are only more robust if their sample size is bigger. Therefore it is not the annual versus the monthly that influences the robustness, but the sample size. This should be corrected in the text.
R2 author response:
We agree with the reviewer regarding the statistical definition of robustness and have revised the text accordingly.
Initial reviewer comment:
Since 427f: This is not correct. It has been observed much more often as summarized in recent reviews.
Initial author response:
Thank you for your comment. We do not disagree per se, however, we cannot reply to this comment unless the respective references are provided that negate the references we chose. We very much appreciate the level of detail and the many constructive comments given by R1, however, statements such as ‘summarized in recent reviews’ without providing detail is far from constructive.
R2 reviewer comment:
This is how you make a reviewer work for you… Thanks for the compliment! Examples are: Iida, Kakubari, & Tanaka (2005), Abrahams et al. (2003), Chinen (2007), Germer (2013), Crabtree & Trudgill (1985), Martinez-Meza & Whitford (1996). If this is not enough, than please continue yourself with the search!
R2 author response:
As we and the reviewer simply disagree on what qualifies as “rarely observed,” we have removed the sentence to appease them. Specifically, the reviewers’ referenced studies represent a handful of local observations and (despite the reviewer’s insinuation that there are many more) they are in fact the majority of studies with these types of observations over the past 150 years …are even 12 studies reporting these kinds of observations over more-than-a-century not “rare” or “few” compared to other hydrologic processes?
With respect, we are not “making a reviewer work for us”; rather, we are asking the reviewer to support their claim with specific evidence.
