Effects of Infiltration Amounts on Preferential Flow Characteristics and Solute Transport in the Protection Forest Soil of Southwestern China
Round 1
Reviewer 1 Report
General comments:
Overall, the study by Li et al sought to characterize preferential flow characteristics in forest soils of southwestern China (Simian Mountain, Chongqing). The manuscript is well-written and I thought the tables and figures were well-done too. I found the data interesting and useful for many interpretations but the current study has several drawbacks that prevent it from being useful by a broader audience. By addressing my comments, I believe the study would be worth publishing.
The study lacks a testable hypothesis and due to the absence of a testable hypothesis/inquiry the results and discussion section simply meander through the results instead of using them to make a point about how forest soils function or impacts of biological or physical features. This needs to be improved for this study to be of broader interest beyond southwestern China forest soil scientists.
While I recognize this is an important new contribution to describing macroporosity and preferential flow in southwestern China, the authors do a poor job using existing literature in the region (Sichuan, Guizhou, Hunan, Guangxi provinces have this data!) to inform readers of existing studies in the introduction and discussion. Further, descriptions from studies in Europe, North America, Japan, Russia, and other forested areas should be used to provide context. This is a severe setback and needs to be addressed.
Abstract:
Line 11: replace “:” with “as” since the authors not introducing a list and it can be one phrase instead of two by using the colon.
Line 15: change “namely” to “named”.
Line 19: change “the increase in” to “greater” for clarity since increasing can be thought of as upwards.
Introduction:
Line 45: Briefly state how soil texture and structure affect macropores. This is of importance here for the readers and since biological activities are described, the authors should describe soil texture and soil structure effects.
Lines 43 – 57: This paragraph would be bolstered by describing these variables in context with southwestern China.
Line 76: Change “south-western” to “southwestern” and be clear if fertilizers and pesticides are being used in natural forests or managed forests.
Results and Discussion section:
Line 147: “the depth of matrix flow increased with the increase in precipitation amounts.” This should be described in its own paragraph to explore why this occurred more clearly. Lines 163-165 on infiltration and Lines 174 – 176 potential root effects are not effectively discussed here. Do the authors have data (textural, measured infiltration rates, root density, tree species) to support these hypotheses or explore them further?
Lines 209 – 243: I found the discussion on enhanced solute transport to be unconvincing and meandering when using the observations. This paragraph would be greatly improved by using existing literature to help interpret observations and avoid poor logic in some of the interpretations. For example, “Consequently, high Br− concentration was found at the ends of preferential flow paths.” This data cannot be seen in Figure 3 and doesn’t support a hypothesis of enhanced solute transport.
Author Response
Response to Reviewer #1
General comments:
Comments to the Author
Overall, the study by Li et al sought to characterize preferential flow characteristics in forest soils of southwestern China (Simian Mountain, Chongqing). The manuscript is well-written and I thought the tables and figures were well-done too. I found the data interesting and useful for many interpretations but the current study has several drawbacks that prevent it from being useful by a broader audience. By addressing my comments, I believe the study would be worth publishing.
The study lacks a testable hypothesis and due to the absence of a testable hypothesis/inquiry the results and discussion section simply meander through the results instead of using them to make a point about how forest soils function or impacts of biological or physical features. This needs to be improved for this study to be of broader interest beyond southwestern China forest soil scientists.
While I recognize this is an important new contribution to describing macroporosity and preferential flow in southwestern China, the authors do a poor job using existing literature in the region (Sichuan, Guizhou, Hunan, Guangxi provinces have this data!) to inform readers of existing studies in the introduction and discussion. Further, descriptions from studies in Europe, North America, Japan, Russia, and other forested areas should be used to provide context. This is a severe setback and needs to be addressed.
Response to general comments:
We greatly appreciate the reviewer’s positive comments. Necessary revisions have been made with respect to the comments of the reviewer, and the responses regarding each of comments list below.
Specific comments:
Comment 1: Abstract:
Line 11: replace “:” with “as” since the authors not introducing a list and it can be one phrase instead of two by using the colon.
Line 15: change “namely” to “named”.
Line 19: change “the increase in” to “greater” for clarity since increasing can be thought of as upwards.
Response: Thanks for the reviewer’s comment. We revised the abstract in light of the above comments.(Line11; Line15; Line19)
Comment 2: Introduction:
Line 45: Briefly state how soil texture and structure affect macropores. This is of importance here for the readers and since biological activities are described, the authors should describe soil texture and soil structure effects.
Lines 43 – 57: This paragraph would be bolstered by describing these variables in context with southwestern China.
Line 76: Change “south-western” to “southwestern” and be clear if fertilizers and pesticides are being used in natural forests or managed forests.
Response: Thanks for the reviewer’s comment. We are sorry that we did not introduce the background of the forest clearly and we have added this section to the introduction.( Line47-54; Line79-88)
Soil types and structure have a complex effect on preferential flow because of their spatial heterogeneities, which can directly change the hydraulic properties, quantities and distribution of soil macropores. Soil clay content could change the soil pore structure and thus affect the type of preferential flow occurring in the soil [11]. The more obvious the macropore structure in the soil, the lower the exchange rate of water and solutes with the soil matrix, and the more favorable the formation of preferential flow [12].
The Three Gorges reservoir area is located in the combination of the middle and upper reaches of the Yangtze River, with complex topography, large spatial variation of natural resources and strong anthropogenic interference. It is a sensitive ecological area and an important functional area for soil and water conservation in China. The protection forest optimization project of nearly 4 million mu has been implemented to solve the problems of unreasonable spatial distribution and structure of forest species, poor quality of forest stands and low ecological protection effectiveness in the Three Gorges reservoir area of Yangtze River. However, fertilisers and pesticides are widely applied in the process of optimizing the construction of protective forests has an impact on the quality of water resources in the Three Gorges reservoir area. High mean annual precipitation reaching 1031 mm increases groundwater contamination risks in this region [12]. More than 80% of rainfall is received during April to October, with the region receiving the highest amount of rainfall during June to August. The utilisation rate of water, fertilisers and pesticides may seriously be influenced by preferential flow during these months. In addition, groundwater contamination is promoted by the increase in the amount of solutes that infiltrate deep soil together with preferential flow. Thus, determining preferential flow characteristics and their effect on solute transport under different precipitation amounts is highly desirable. The objective of the present study was to (1) characterise the distribution of preferential flow and (2) determine solute transport with preferential flow under different precipitation amounts via multiple-tracer experiments.
Comment 3: Results and Discussion section:
Line 147: “the depth of matrix flow increased with the increase in precipitation amounts.” This should be described in its own paragraph to explore why this occurred more clearly. Lines 163-165 on infiltration and Lines 174 – 176 potential root effects are not effectively discussed here. Do the authors have data (textural, measured infiltration rates, root density, tree species) to support these hypotheses or explore them further?
Response: Thanks for the reviewer’s comment. We revised the discussion in light of the above comment. We have also added root data to support our discussion (Table 3).
We compared and calculated the data extracted from vertical sections under different precipitation amounts. The dye coverage distributions and the images of dye flow patterns in vertical sections are shown in Figure 2. The nonuniform or dissimilar distributions of dye areas under different precipitation amounts suggested that different preferential flow patterns had developed. As shown in Figure 2, most of the vertical slices were stained with a dye coverage of more than 80% in the topsoil, indicating that the topsoil experienced matrix flow [12,17]. Meanwhile, the matrix flow depths for P20, P40 and P60 were 1.6, 7 and 9.3 cm, indicating the depth of matrix flow increased with the increase in precipitation amounts The increase in infiltration water was corresponding to a longer rainfall period, increasing the potential energy of water supply, and the increase in water potential gradient increased the uniform infiltration of soil water. Also, the increase in infiltration water increased the water content in the lower soil layer, slowed down the rate of water infiltration, and promoted the lateral movement of water in the top soil layer to form matrix flow [16]. This result is similar to the conclusion of a study in plantation forests in the southwest karst region. However, the depth of matrix flow at 55 mm infiltration was 2-3 cm, which was significantly smaller than our results. This could be due to the fact that the sand content and total soil porosity in the soils of our study area (66.33%; 60.30%) were higher than those of the above study area (27.59%; 37.91%).
The maximum infiltration depths for P20, P40 and P60 were 25, 30 and 40 cm, respectively. The consistency of these results with the findings of several other researchers [15,16] indicated the extensive variation range of movement of the water in the soil under high precipitation amounts. However, in the plantation forests in the southwest karst region, the maximum infiltration depth at 55 mm infiltration volume was about 24 cm and the spatial variability of preferential flow was small. In our study, precipitation amounts had a substantial effect on the spatial heterogeneities of preferential flow despite the similar properties of the three plots (Figure 2). Reductions in the patterns of dye coverage showed different trends amongst P20, P40 and P60. Compared with that in P40 and P60, which showed considerably different patterns amongst vertical slices with the reduction in depth, the pattern of decrement in P20 showed no evident change, indicating small spatial heterogeneity. This result may be attributed to the small depth of matrix flow that lacked sufficient head pressure to allow the formation of preferential flow path networks. Thus, the dye tracer infiltrated into deep soil layers only through a small part of preferential flow paths under low precipitation. In P40, unexpected increases in dye coverage were detected as the soil depth increased to 15 cm in P40, demonstrating that high amounts of the dye tracer had infiltrated into these layers. Therefore, lateral preferential flow became increasingly evident with the increase in precipitation amount because the precipitation amount exceeded the capacity for the vertical transport of the preferential flow to block preferential flow paths. This phenomenon could also be associated with preferential flow path characteristics, resulting in the high lateral movement of the dye tracer. Root channels are considered as primary preferential flow paths in forest soils [9,18]. An investigation into the root systems in the study area revealed that the total root length of the root with root diameter < 1 mm in the 10–20 cm soil layers was greater than that in the other soil layers except topsoil (0–5 cm). Consequently, in this region, the large number of intricate root distributions affected the connectivity of priority flow paths and thus increased lateral preferential flow along root channels. The dye coverage distribution in P60 was consistent with that in P20. Specifically, it lacked a peak and decreased gradually with soil depth. The dye coverage in the vertical slices from P60 had large standard deviations. Three of the five dyed coverage distribution curves showed sharp reductions in the 10–15 cm soil layers and an obvious trailing phenomenon at the soil depths of 15–40 cm, indicating the existence of penetrating macropore flow. This result was largely explained by the fact that the increasing amount of infiltration increased the head pressure of the preferential flow. This effect then prompted the formation of the preferential flow network and increased the connectivity of the preferential flow network. In this case, the transport patterns of preferential flow could be divided into two types under the influence of spatial heterogeneity caused by soil texture and roots and biological activities [4,18]. In one case, preferential flow paths were connected vertically to form macropores, and water was transported rapidly through macropores to deep soil layers, thus bypassing most of the soil matrix. By contrast, in the absence of macropores, the unstable wetting front continues to expand because the infiltration rate was lower than the saturated hydraulic conductivity of the soil. This result, in turn, led to the finger-like preferential flow pattern of water and solutes. At the same time, as the infiltration amount was increased, the flow pattern transformed from one dominated by preferential flow to one dominated by matrix flow. Furthermore, macropore flow and finger-like flow appeared as illustrated in Figure 2. The appearance of these flows provided support for the above explanation. Combining these results, the soil in the study area was dominated by sandy loam, and when the precipitation was less than 60 mm, the occurrence of preferential flow could increase the connectivity of soil pores and enhance the water retention capacity of the soil. However, when the precipitation was greater than 60mm, macropore flow might be formed, which would transport water directly to the deeper soil layers to reduce the water utilization rate, increase the risk of groundwater contamination and might reduce the stability of the soil body to induce geological disasters such as landslides.
Table 3. Root length (cm)
|
Soil Depth (cm) |
Root Diameter |
||||
|
<1mm |
1-3mm |
3-5mm |
5-10mm |
>10mm |
|
|
0-5 |
2549 ± 322 |
1284 ± 211 |
499 ± 142 |
125 ± 23 |
19 ± 12 |
|
5-10 |
1504 ± 97 |
528 ± 200 |
139 ± 53 |
54 ± 8 |
20 ± 11 |
|
10-15 |
2058 ± 88 |
426 ± 44 |
149 ± 40 |
138 ± 22 |
38 ± 10 |
|
15-20 |
2068 ± 282 |
394 ± 86 |
189 ± 40 |
104 ± 58 |
23 ± 10 |
|
20-25 |
720 ± 182 |
476 ± 112 |
111 ± 62 |
87 ± 54 |
38 ± 13 |
|
25-30 |
565 ± 300 |
212 ± 0 |
243 ± 13 |
44 ± 0 |
0 ± 0 |
|
30-35 |
373 ± 58 |
396 ± 58 |
158 ± 31 |
27 ± 1 |
0 ± 0 |
|
35-40 |
366 ± 21 |
122 ± 6 |
48 ± 6 |
0 ± 0 |
0 ± 0 |
Note：In each soil layer, the root length under different root diameter range is the sum of the root length of that root diameter, and the volume of each soil layer is 125 cm3.
Comment 4: Results and Discussion section:
Lines 209 – 243: I found the discussion on enhanced solute transport to be unconvincing and meandering when using the observations. This paragraph would be greatly improved by using existing literature to help interpret observations and avoid poor logic in some of the interpretations. For example, “Consequently, high Br− concentration was found at the ends of preferential flow paths.” This data cannot be seen in Figure 3 and doesn’t support a hypothesis of enhanced solute transport.
Response: Thanks for the reviewer’s comment. We are sorry for a vague express. We revised the discussion in light of the above comment.
As shown in Figure 3, the concentration of Br− at 0–40 cm soil depth under different infiltration intensities followed the order of plot 5 > plot 3 > plot 1, suggesting that increases in infiltration could increase in concentration. Additionally, in these plots, Br− concentration distributions displayed small, W-shaped serrated patterns with the increase in soil depth and was highest in the topsoil layer. In other words, the Br− concentration distributions in these plots showed similar trends and decreased with soil depth but peaked in different soil layers. In plot 1, Br− concentration peaked at the soil depth of 20 cm. In plots 3, Br− concentration peaked at the soil depth of 15 cm and then again at 25 cm. In plot 5, the first peak of the Br− concentration was observed at the soil depth of 20 cm and the second at the soil depth of 35 cm. These results indicated that with the increase in infiltration amount, the peak of Br− concentration appeared in deep soil layers and at near the end of the preferential flow. The last peak of Br− concentration observed in each plot was most likely attributed to the solution reaching the end of the preferential flow paths where accumulation occurred and the wetting fronts continued to extend to the surrounding area. [24]. Consequently, high Br− concentration was found at about 5 cm before the maximum infiltration soil depth. The other peaks might be attributed to the fact that when the upper infiltration water volume was increased to 40 mm and 60 mm, infiltration exceeded the capacity for the vertical transport of preferential flow, resulting in the accumulation of water and the development of lateral infiltration [25]. Dye coverage in the layer where the peak was higher than that in other layers also supported this interpretation (Figure 2), suggesting that the distribution pattern of preferential flow was the key to influencing solute transport.
In comparison to plot 3 and plot 5, plot 4 and plot 6 used the solutions of KI and KNO3 instead of KI and KBr in the final infiltration phase. The main purpose of this measure was to analysis the effect on the Br− concentration in the soil for infiltration volumes of 20-40 mm and 40-60 mm, respectively. As shown in Figure 3, the concentration of Br− in plot 4 was similar to that in plot 3, indicating that the infiltration volume of 20-40 mm did not change the distribution of Br− concentration, but only decreased with the increase of infiltration. The distribution of Br− concentration in plot 6 was significantly different from that in the preferential flow dyeing area. In plot 6, unexpected increases in Br− concentration were detected as the soil depth increased from 0 cm to 20, demonstrating that a large amount of Br− in the surface layer was transported to the deeper soil by the preferential flow when the infiltration volume was at 40-60 mm.
The NO3− concentrations in plot 2, plot 4 and plot 6 represented the distribution of NO3− at 0-20 mm, 20-40 mm and 40-60 mm of infiltration water respectively. The NO3− concentration shown in plot 1 was the original soil NO3− content when the NO3− -containing tracer was not added. The original NO3− concentration in the soil was low and therefore had a slight influence on the distribution of NO3− concentration in the solute transport experiments. The pattern of NO3− concentration in plot 2 was consistent with that of Br− concentration in plot 1. This result indicated the absence of a significant difference in the concentration distribution of different solutes transported with preferential flow under a 20 mm infiltration volume. The significantly lower concentration of NO3− than that of Br− at infiltration of 20 mm may attribut to the NO3− is more readily available to plants and microorganisms than Br− and converted into other forms, such as N2 and NH4+.In plot 4, the highest NO3− concentration was detected in the topsoil, and NO3− concentration in other soil layer was significantly lower than that in plot 2, suggesting that this component of the solution was mainly concentrated on the soil surface and did not transport to deeper soil layers with the preferential flow. In plot 6, NO3− concentration decreased with the increase in soil layers without significantly peaking mainly because increasing vertical infiltration from 40 mm to 60 mm reduced the capillary effect of the soil layer under preferential flow and promoted the vertical connectivity of preferential flow paths, resulting in rapid solute transport to deep layers. Additionally, the absence of peaks might be due to the presence of macropores that promoted downward solute transport to reduce lateral infiltration. The NO3− concentration in plot 6 was significantly higher than that in plot 4, which could also indicated that the preferential flow in plot 6 might be a macropore flow with strong accumulation.
These observations suggested that when the amount of water infiltration exceeded 40 mm, preferential flow infiltration depth and development increased, and solutes could infiltrate into deep soil layers along with macropore preferential flow during rainfall events. Therefore, in the future management of protective forest cultivation, fertilizer application and pesticide spraying before rain storm should be avoided as much as possible in order to reduce the pollution of shallow groundwater by pesticides and nutrients, etc.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear authors,
I have read your article with caution. I have read your article carefully. The manuscript contains valuable and interesting data that deserve to be published. Abstract is adequate but it would improve if the authors include some data. Keywords are adequate. References are current and easy to consult. However, But I have observed some serious defaults that the authors must correct before the article is published.
Authors analyze the movement of water in the soil (with special attention to preferential flow) in five plots. One plot for each treatment, without repetitions. This conditions the results obtained. This aspect is for me the main problem that this work has.
To study the movement of Br-, they make of 1 to 3 applications with equal doses of KI + KBr (plots 1, 3 and 5), which is adequate. But to analyze the movement of NO3- they only add nitrate in the last application, so plots 4 and 6 cannot be well interpreted. If they had added nitrates in all the additions, the result perhaps would have been different. Therefore, I recommend deleting plots 4 and 6. Why in P40 and P60 do they only apply KNO3 in the last dose?. I think that plots 4 and 6 should not be used to get any conclusions.
I include minor comments in an attached pdf file and that the authors should see and attend to the comments that are mentioned. This might improve final version of this manuscript for its publishing.
Comments for author File: 
Comments.pdf
Author Response
Response to Reviewer #2
General comments:
I have read your article with caution. I have read your article carefully. The manuscript contains valuable and interesting data that deserve to be published. Abstract is adequate but it would improve if the authors include some data. Keywords are adequate. References are current and easy to consult. However, But I have observed some serious defaults that the authors must correct before the article is published.
Response to general comments:
Response: Thanks for the reviewer’s comment. We thank the reviewer for the positive comments and greatly appreciate the detailed comments for improving our manuscript. Responses to the reviewer’s concerns in general comments are listed below.
Specific comments:
Comment 1: Authors analyze the movement of water in the soil (with special attention to preferential flow) in five plots. One plot for each treatment, without repetitions. This conditions the results obtained. This aspect is for me the main problem that this work has.
Response: Thanks for the reviewer’s comment. We are sorry that we did not present the complete layout of the experiment, and we have added this section to the materials and methods. ( Line123-126)
In fact, three replications were set up for our experiments, and the standard errors in Table 1, Figure 3, and Figure 4 in the manuscript were calculated from the replicate data. In Figure 2, only the image of one experiment was shown to avoid redundancy, and we can provide images of other replicate experiments if necessary.
In this study, three areas with similar site conditions were selected for replication of the field multiple-tracer experiments. In each area, three different precipitation levels of 20, 40 and 60 mm were established to simulate three different precipitation amounts under rain storm, which were designated as P20, P40 and P60, respectively. Each level had two plots, which were treated with different solutions. For each plot, two rectangular iron frames, namely, an inner frame with dimensions of 60 cm (length) × 60 cm (width) × 50 cm (height) and an outer frame with dimensions of 80 cm × 80 cm × 50 cm, were concentrically embedded into the soil to a depth of 30 cm after the experimental surface had been cleaned and smoothed. After embedding, the soil within 5 cm was compacted by using a wooden hammer to prevent dye tracer infiltration along the frames.
Comment 2: To study the movement of Br-, they make of 1 to 3 applications with equal doses of KI + KBr (plots 1, 3 and 5), which is adequate. But to analyze the movement of NO3- they only add nitrate in the last application, so plots 4 and 6 cannot be well interpreted. If they had added nitrates in all the additions, the result perhaps would have been different. Therefore, I recommend deleting plots 4 and 6. Why in P40 and P60 do they only apply KNO3 in the last dose?. I think that plots 4 and 6 should not be used to get any conclusions.
Response: Thanks for the reviewer’s comment. We are sorry for a vague express. We have revised the Results and Discussion to make the description clearer in the revised manuscript.
In comparison to plot 3 and plot 5, plot 4 and plot 6 used the solutions of KI and KNO3 instead of KI and KBr in the final infiltration phase. The main purpose of this measure was to analysis the effect on the Br− concentration in the soil for infiltration volumes of 20-40 mm and 40-60 mm, respectively. As shown in Figure 3, the concentration of Br− in plot 4 was similar to that in plot 3, indicating that the infiltration volume of 20-40 mm did not change the distribution of Br− concentration, but only decreased with the increase of infiltration. The distribution of Br− concentration in plot 6 was significantly different from that in the preferential flow dyeing area. In plot 6, unexpected increases in Br− concentration were detected as the soil depth increased from 0 cm to 20, demonstrating that a large amount of Br− in the surface layer was transported to the deeper soil by the preferential flow when the infiltration volume was at 40-60 mm.
The NO3− concentrations in plot 2, plot 4 and plot 6 represented the distribution of NO3− at 0-20 mm, 20-40 mm and 40-60 mm of infiltration water respectively. The NO3− concentration shown in plot 1 was the original soil NO3− content when the NO3− -containing tracer was not added. The original NO3− concentration in the soil was low and therefore had a slight influence on the distribution of NO3− concentration in the solute transport experiments. The pattern of NO3− concentration in plot 2 was consistent with that of Br− concentration in plot 1. This result indicated the absence of a significant difference in the concentration distribution of different solutes transported with preferential flow under a 20 mm infiltration volume. The significantly lower concentration of NO3− than that of Br− at infiltration of 20 mm may attribut to the NO3− is more readily available to plants and microorganisms than Br− and converted into other forms, such as N2 and NH4+.In plot 4, the highest NO3− concentration was detected in the topsoil, and NO3− concentration in other soil layer was significantly lower than that in plot 2, suggesting that this component of the solution was mainly concentrated on the soil surface and did not transport to deeper soil layers with the preferential flow. In plot 6, NO3− concentration decreased with the increase in soil layers without significantly peaking mainly because increasing vertical infiltration from 40 mm to 60 mm reduced the capillary effect of the soil layer under preferential flow and promoted the vertical connectivity of preferential flow paths, resulting in rapid solute transport to deep layers. Additionally, the absence of peaks might be due to the presence of macropores that promoted downward solute transport to reduce lateral infiltration. The NO3− concentration in plot 6 was significantly higher than that in plot 4, which could also indicated that the preferential flow in plot 6 might be a macropore flow with strong accumulation.
These observations suggested that when the amount of water infiltration exceeded 40 mm, preferential flow infiltration depth and development increased, and solutes could infiltrate into deep soil layers along with macropore preferential flow during rainfall events. Therefore, in the future management of protective forest cultivation, fertilizer application and pesticide spraying before rain storm should be avoided as much as possible in order to reduce the pollution of shallow groundwater by pesticides and nutrients, etc.
Comment 3: I include minor comments in an attached pdf file and that the authors should see and attend to the comments that are mentioned. This might improve final version of this manuscript for its publishing.
Response: Thanks for the reviewer’s comment. We are very sorry for our carelessness. We have revised the wrong words and sentences according to the suggestions in the PDF.
1：Mineralization (nitrification) of soil organic matter (SOM) releases nitrate ions. SOM contents are high, therefore nitrate levels may not be negligible.
Response: The initial concentrations of Br− in the soil were ignored because they were significantly lower than the applied concentrations.The concentration of NO3− in plot1 was considered as the initial concentration of NO3− in the soil.
2: Dark phrase. I don't understand what the authors meant. Rewrite again.
Response: By contrast, in the absence of macropores, the unstable wetting front continues to expand because the infiltration rate was lower than the saturated hydraulic conductivity of the soil.
Author Response File: Author Response.docx
Reviewer 3 Report
I think that ther is an nappropriate self citation: [25]. It is not necessary.
In general, what type of soils are they? (according to USDA or FAO classification, whichever you prefer) What artificial rainfall or permeameter system - if no artificial rain was applied - did you use? Was the volume per day, per hour, per minute? Missing unit of time used in the experiment and through the manuscript.
Line 141: Was the amount of precipitation real or simulated? You need to clarify it in methods chapter. Otherwise, results are incosistents.
To clarify in Lines 92-94: What kind of forest are there? Managed? Natural? Forest crops? What kind of agricultural activities are those forest suffering from?
Author Response
Response to Reviewer #3
Specific comments:
Comment 1:I think that ther is an nappropriate self citation: [25]. It is not necessary.
Response: Thanks for the reviewer’s comment. As suggested, we removed this reference.
Comment 2: In general, what type of soils are they? (according to USDA or FAO classification, whichever you prefer) What artificial rainfall or permeameter system - if no artificial rain was applied - did you use? Was the volume per day, per hour, per minute? Missing unit of time used in the experiment and through the manuscript. To clarify in Lines 92-94: What kind of forest are there? Managed? Natural? Forest crops? What kind of agricultural activities are those forest suffering from?
Response: Thanks for the reviewer’s comment. We are sorry that we did not introduce the background of the forest clearly and we have added this section to the introduction and Materials and Methods. ( Line79-88; Line101-115)
The Three Gorges reservoir area is located in the combination of the middle and upper reaches of the Yangtze River, with complex topography, large spatial variation of natural resources and strong anthropogenic interference. It is a sensitive ecological area and an important functional area for soil and water conservation in China. The protection forest optimization project of nearly 4 million mu has been implemented to solve the problems of unreasonable spatial distribution and structure of forest species, poor quality of forest stands and low ecological protection effectiveness in the Three Gorges reservoir area of Yangtze River. However, fertilisers and pesticides are widely applied in the process of optimizing the construction of protective forests has an impact on the quality of water resources in the Three Gorges reservoir area. High mean annual precipitation reaching 1031 mm increases groundwater contamination risks in this region [12]. More than 80% of rainfall is received during April to October, with the region receiving the highest amount of rainfall during June to August. The utilisation rate of water, fertilisers and pesticides may seriously be influenced by preferential flow during these months. In addition, groundwater contamination is promoted by the increase in the amount of solutes that infiltrate deep soil together with preferential flow. Thus, determining preferential flow characteristics and their effect on solute transport under different precipitation amounts is highly desirable. The objective of the present study was to (1) characterise the distribution of preferential flow and (2) determine solute transport with preferential flow under different precipitation amounts via multiple-tracer experiments.
Field multiple-tracer experiments were conducted in protection forest soil in Simian Mountain (28°36′ N, 106°23′ E), Chongqing Province, Three Gorges Area, China. This area is located at the end of the Three Gorges reservoir area where has a subtropical humid monsoon climate with 17 - 19°C annual mean temperature and 1000 - 1250 mm mean annual precipitation received mainly from April to October. The soil is classified as yellow earth. To improve the protective forest system, people have been optimizing the species and spatial configuration of the protection forests in this region from 2000 to the present. Our study slected the protection forests which were mainly dominated by the Cunninghamia lanceolate and Quercus acutissima, and the understory companion plant species include Itea oblonga, Eurga loquaiana, Plagiogyria distinctissima and Aster ageratoides. In the first 3 years, the protection forest were conducted for the pest and disease control regularly. The pesticides (Use 0.75%~1% Bordeaux, 75% chlorothalonil wettable powder, or 65% Dyson zinc) were applied during July to August once a week for 3 consecutive weeks, with 50 to 70 kg of the pesticides per mu. The type of fertilizer, the amount of fertilizer applied, and the time of fertilizer application depend on the actual forest growth.
Comment 3: Line 141: Was the amount of precipitation real or simulated? You need to clarify it in methods chapter. Otherwise, results are incosistents.
Response: Thanks for the reviewer’s comment. We are sorry for a vague express. We revised the Materials and Methods in light of the above comment.
In this study, three areas with similar site conditions were selected for replication of the field multiple-tracer experiments. In each area, three different precipitation levels of 20, 40 and 60 mm were established to simulate three different precipitation amounts under rain storm, which were designated as P20, P40 and P60, respectively. Each level had two plots, which were treated with different solutions. For each plot, two rectangular iron frames, namely, an inner frame with dimensions of 60 cm (length) × 60 cm (width) × 50 cm (height) and an outer frame with dimensions of 80 cm × 80 cm × 50 cm, were concentrically embedded into the soil to a depth of 30 cm after the experimental surface had been cleaned and smoothed. After embedding, the soil within 5 cm was compacted by using a wooden hammer to prevent dye tracer infiltration along the frames. Due to the relatively high initial water content of the soil in the study area, accumulation of water will rapidly form on the soil surface layer when the rain storm occurs. Therefore, the solution was quickly applied to the soil surface of the inner frame to simulate instantaneous ponding infiltration. In accordance with double-ring infiltration, the same depth of freshwater was simultaneously applied to the soil surface of the outer frame at each step to force the solution to infiltrate into the inner frame fully. The details of the field multiple-tracer experiments are shown in Figure 1 and Table 2. These details include the layout of the plots, the solute of the solution, the volume of the solution and the time consumed for solution infiltration.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Li et al., have made moderate changes to the manuscript and addressed most of the comments by myself and the other reviewers. They have added substantially to the introduction, methods, and results section. However, for this study to be fully developed, there are still areas to grow intellectually. The authors do not have a firm grasp of the existing literature and do not use it effectively in the discussion still. Therefore, I believe there are still minor to somewhat major revisions required.
Line 49-52: Please be more specific.
“Soil types and structure have a complex effect on preferential flow because of their spatial heterogeneities, which can directly change the hydraulic properties, quantities and distribution of soil macropores.” Please indicate how more clay or sand affects preferential flow. Does it increase it or decrease it?
“Soil clay content could change the soil pore structure” briefly describe how does clay content affect pore structure instead of this general statement that is not useful for readers.
Line 83: What is “mu” here? Please use SI units.
The use of the existing literature still is not adequate. On Lines 234 – 241, Lines 296 - 309, and Lines 319 – 324, these interpretations should be compared with existing literature. Have previous studies found similar findings or the opposite effects? providing context to your results is needed to turn this study from a report on findings to a study evaluating preferential flow.
Author Response
Response to Reviewer #1
General comments:
Comments to the Author
Li et al., have made moderate changes to the manuscript and addressed most of the comments by myself and the other reviewers. They have added substantially to the introduction, methods, and results section. However, for this study to be fully developed, there are still areas to grow intellectually. The authors do not have a firm grasp of the existing literature and do not use it effectively in the discussion still. Therefore, I believe there are still minor to somewhat major revisions required.
Response to general comments:
We greatly appreciate the reviewer’s positive comments. Necessary revisions have been made with respect to the comments of the reviewer, and the responses regarding each of comments list below.
Specific comments:
Comment 1:
Line 49-52: Please be more specific.
“Soil types and structure have a complex effect on preferential flow because of their spatial heterogeneities, which can directly change the hydraulic properties, quantities and distribution of soil macropores.” Please indicate how more clay or sand affects preferential flow. Does it increase it or decrease it?
“Soil clay content could change the soil pore structure” briefly describe how does clay content affect pore structure instead of this general statement that is not useful for readers.
Response: Thanks for the reviewer’s comment. We revised the sentence in Line49-52 to “ The greater the clay content in the soil, the more obvious the macropore structure and the more favorable the formation of preferential flow ”. ( Line 52)
Comment 2: Line 83: What is “mu” here? Please use SI units.
Response: Thanks for the reviewer’s comment. We are sorry for an unscientific express. We have revised “4 million mu” to “270,000 ha”.( Line83)
Comment 3: The use of the existing literature still is not adequate. On Lines 234 – 241, Lines 296 - 309, and Lines 319 – 324, these interpretations should be compared with existing literature. Have previous studies found similar findings or the opposite effects? providing context to your results is needed to turn this study from a report on findings to a study evaluating preferential flow.
Response: Thanks for the reviewer’s comment. We revised the discussion in light of the above comment.
Lines 234 – 241
This part of the sentences was actually a summary of the first part of the discussion, and therefore no comparison with previous studies was made. However, we have modified the first part of the discussion by adding references and a comparative analysis of some of the results. ( Line174-177; Line186-202)
We compared and calculated the data extracted from vertical sections under different precipitation amounts. The dye coverage distributions and the images of dye flow patterns in vertical sections are shown in Figure 2. The nonuniform or dissimilar distributions of dye areas under different precipitation amounts suggested that different preferential flow patterns had developed. As shown in Figure 2, most of the vertical slices were stained with a dye coverage of more than 80% in the topsoil, indicating that the topsoil experienced matrix flow [14,19]. Meanwhile, the matrix flow depths for P20, P40 and P60 were 1.6, 7 and 9.3 cm. These results were similar to the conclusion of a study in plantation forests in the southwest karst region [20] that has shown the depth of matrix flow increased with the increase in precipitation amounts. The increase in infiltration water was corresponding to a longer rainfall period, increasing the potential energy of water supply, and the increase in water potential gradient increased the uniform infiltration of soil water. Also, the increase in infiltration water increased the water content in the lower soil layer, slowed down the rate of water infiltration, and promoted the lateral movement of water in the top soil layer to form matrix flow [18].
Due to the increased the depth of matrix flow, it was also expected that dye would infiltrate deeper under the higher infiltration amounts. Indeed, our results showed that the maximum infiltration depths for P20, P40 and P60 were 25, 30 and 40 cm, respectively. The consistency of these results with the findings of several other researchers [17,18] indicated the extensive variation range of movement of the water in the soil under high precipitation amounts. However, our results showed the matrix flow depth and the maximum infiltration depth were occurred in deeper soil layers, compared to study in the plantation forests in the southwest karst region. This could be due to the fact that the sand content and total soil porosity in the soils of our study area (66.33%; 60.30%) were higher than those of the above study area (27.59%; 37.91%). Previous studies have been reported that water movement in clay soils was dominated by preferential flow with less matrix flow compared to loamy soils, as the clay soils were fine-textured and well-structured, which were more conducive to the presence of macropores. Additionally, Yan et al., (2016) also found that the matrix flow depth was 12 cm and the maximum infiltration depth was 28 cm in sandy soils with the sand content about 80%. Therefore, in future studies of preferential flow, the depth of occurrence of preferential flow and the maximum depth of infiltration could be predicted based on soil conditions and infiltration amounts.
Lines 296 - 309
This part of the discussion was modified in accordance with the comments.( Line302-306 ; Line 317-334)
In comparison to plot 3 and plot 5, plot 4 and plot 6 used the solutions of KI and KNO3 instead of KI and KBr in the final infiltration phase. The main purpose of this measure was to analysis the effect on the Br− concentration in the soil for infiltration volumes of 20-40 mm and 40-60 mm, respectively. As shown in Figure 3, the concentration of Br− in plot 4 was similar to that in plot 3, indicating that the infiltration volume of 20-40 mm did not change the distribution of Br− concentration. The distribution of Br− concentration in plot 6 was significantly different from that in the preferential flow dyeing area. In plot 6, unexpected increases in Br− concentration were detected as the soil depth increased from 0 cm to 20, demonstrating that a large amount of Br− in the surface layer was transported to the deeper soil by the preferential flow when the infiltration volume was at 40-60 mm. These results indicated that the solution from the second stage of infiltration mixed with the solution from the first stage of infiltration in the soil and transported downward. The third stage of infiltration pushed the pre-infiltrated solution downward with little interaction (as an approximately piston flow) at depths of 0-10 cm soil layers, probably because the first two stages of infiltration made the soil water saturated.
The NO3− concentrations in plot 2, plot 4 and plot 6 represented the distribution of NO3− at 0-20 mm, 20-40 mm and 40-60 mm of infiltration water respectively. The NO3− concentration shown in plot 1 was the original soil NO3− content when the NO3− -containing tracer was not added. The original NO3− concentration in the soil was low and therefore had a slight influence on the distribution of NO3− concentration in the solute transport experiments. The pattern of NO3− concentration in plot 2 was consistent with that of Br− concentration in plot 1. This result indicated the absence of a significant difference in the concentration distribution of different solutes transported with preferential flow under a 20 mm infiltration volume. This is consistent with other studies that have shown rapid downward transport of water through the native preferential flow paths at low infiltration volume. The significantly lower concentration of NO3− than that of Br− at infiltration of 20 mm may attribute to the NO3− is more readily available to plants and microorganisms than Br− and converted into other forms, such as N2 and NH4+[]. In plot 4, the highest NO3− concentration was detected in the topsoil, and NO3− concentration in other soil layer was significantly lower than that in plot 2, suggesting that the second stage solution was mainly concentrated on the soil surface and did not transport to deeper soil layers with the preferential flow. In plot 6, NO3− concentration decreased with the increase in soil layers without significantly peaks. This result was consistent with previous studies indicating that increasing infiltration reduced the capillary effect of the soil layer and promoted the vertical connectivity of preferential flow paths, resulting in rapid solute transport to deep layers []. However, differently from previous studies, there were no peaks observed in this study and the NO3− concentration in plot 6 was significantly higher than that in plot 4, suggesting that the solute infiltration process in the third stage was accumulated and rapidly transported. This result may be due to the fact that, when the amounts of infiltration volume reached a certain threshold, the formation of a penetrating macropore flow promoted the downward transport of solutes and reduced lateral infiltration.
These observations suggested that when the amount of water infiltration exceeded 40 mm, preferential flow infiltration depth and development increased, and solutes could infiltrate into deep soil layers along with macropore preferential flow during rainfall events. Therefore, in the future management of protective forest cultivation, fertilizer application and pesticide spraying before rain storm should be avoided as much as possible in order to reduce the pollution of shallow groundwater by pesticides and nutrients, etc.
Lines 319 – 324
This part of the sentences was a summary of the second part of the discussion, so this part was retained. We have added a discussion to enrich the article before this part of the sentence.
Author Response File: Author Response.docx
Reviewer 2 Report
Experimental design does not allow deducing the movement of nitrates in the soil since in plots 4 and 6 they add KNO3 only in the final phase.
The authors' explanations for the movement of nitrates in the soil are not convincing.
Author Response
Response to Reviewer #2
Specific comments:
Comment 1: Experimental design does not allow deducing the movement of nitrates in the soil since in plots 4 and 6 they add KNO3 only in the final phase.
The authors' explanations for the movement of nitrates in the soil are not convincing.
Response: Thanks for the reviewer’s comment. We are sorry that we were not able to explain the purpose of the experimental design clearly to you.
In fact the KNO3 was added only to better observe the transport of preferential flow at different stages of infiltration, not to compare the transport of two different ions.
For example, when the total infiltration volume was 40 mm, KBr was applied in the first stage (0-20 mm infiltration) and KNO3 was applied in the second stage (20-40 mm infiltration). KBr represented the movement trace of the staining solution in the first stage and KNO3 represented the movement trace of the staining solution in the second stage.
We modified the discussion based on the comments, analyzed and discussed the distribution of KNO3 and KBr respectively, and deleted the comparison between KNO3 and KBr to avoid disturbance.
Author Response File: Author Response.docx
