Increase and Spatial Variation in Soil Infiltration Rates Associated with Fibrous and Tap Tree Roots
Round 1
Reviewer 1 Report
- line 106-108: unit of bulk density: Mg/m3, not kg/m3, trunk diameter: is it the root collar dimeter?. if it is trunk diameter, you should present the measuring height. please use cm as the unit of tree height and diameter.
- 187-189: is It any reason of P<0.005 for the significance test. Most sicentific papers use P<0.05 for signicicant test.
- line 192, 193: difficult to understand this expression.
-line 393: Q. robur : italic
Author Response
Response to Reviewer 1 Comments
Point 1: Line 106-108: unit of bulk density: Mg/m3, not kg/m3, trunk diameter: is it the root collar dimeter? If it is trunk diameter, you should present the measuring height. Please use cm as the unit of tree height and diameter.
Response: Corrections and revisions have been made to the text as shown below:
Line 106 (Line 106 in WATER-531691): The soil is a loam, with a bulk density of 1.412 g/cm3 and total soil porosity of 0.46. The trunk diameters of M. baccata and S. japonica at 30 cm from soil surface were 5.15 and 6.46 cm, respectively. The root lengths were 23 cm for both M. baccata and S. japonica.
Point 2: 187-189: is It any reason of P<0.005 for the significance test. Most scientific papers use P<0.05 for significant test.
Response: It should be “P<0.05”. Corrections and revisions have been made to the text as shown below:
Line 188-190 (Line 187-189 in WATER-531691): The F test for the mean value was conducted (Table 1) and the p-value of mean value was determined to be <0.05, which indicates a significant mean value.
Point 3: Line 192, 193: difficult to understand this expression.
Response: Corrections and revisions have been made to the text as shown below:
Table 1. Soil steady-state infiltration rates (m/d) at the bottom outlets
of three test tanks.
|
Tank |
Blank test |
M. baccata |
S. japonica |
|
Run 1 |
0.28 |
0.34 |
0.61 |
|
Run 2 |
0.28 |
0.33 |
0.62 |
|
Mean value of soil steady-state infiltration rates |
0.28±0.003 |
0.33±00.009 |
0.61±0.007 |
|
F test of mean value of soil steady-state infiltration rates |
11.55a (3.02×10-3)b |
78.51 (3.56×10-8) |
21.69 (1.72×10-4) |
Note: a The F statistics of mean value of soil steady-state infiltration rates with df 1 and 19.
b p-value of mean value of soil steady-state infiltration rates from F test.
Point 4: Line 393: Q. robur: italic.
Response: Corrections and revisions have been made to the text as shown below:
Line 395-396 (Line 393-394 in WATER-531691): Chandler, K.R.; Chappell, N.A. Influence of individual oak (Quercus robur) trees on saturated hydraulic conductivity. Forest Ecol Manag. 2008, 256, 1222–1229.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors response to my comments on the earlier version of the manuscript clarified some, but not all of the issues.1. The new version of Fig.6 clarified the details of constant head experiments. It would be good if the authors also mention in the text that in each experiment outflows from only one level were used.
2. In the falling head tests the authors used the horizontal cross-section of the whole tank (A) to calculate hydraulic permeability. It means that they assumed uniform distribution of vertical flow velocity over the whole cross-section. If this was the case, they should have obtained the same results for each measurement point. Clearly, the infiltration capacity increases with the distance from tree trunk, but I have doubts if formula (7), which refers to one-dimensional flow, can be applied in this case.
3. The authors should provide values of hydraulic conductivities together with the infiltration rates obtained from their measurements. For example, they can can calculate hydraulic conductivity of different soil layers from constant head tests and compare them with values obtained from falling head tests at depth 0.35m. The hydraulic conductivity from falling head tests is mentioned in Lines 159-162, but no values are provided. These values were shown in Fig.7 in the previous version of the manuscript, but now they are replaced by infiltration rates.
4. Continuing the previous comment, the comparison of hydraulic conductivity values from the earlier version of Fig. 7 and the infiltration rates from the current version indicates that the hydraulic gradient is about 4, which seems too much. Please clarify this.
Author Response
Response to Reviewer 2 Comments
Point 1: The new version of Fig.6 clarified the details of constant head experiments. It would be good if the authors also mention in the text that in each experiment outflows from only one level were used.
Response: Corrections and revisions have been made to the text as shown below:
2.2. Infiltration rate measurements
…
where V is the volume of the measuring vessel/container (V = 5 × 10-4 m3) and t is the time (in days, converted from recorded minutes) taken for the vessel to be filled with water running from each horizontal monitoring tube. Mean infiltration rates (I, m/d) of the soil at depths of 0.35 m and 0.70 m were calculated by the four infiltration rates at the 01, 02, 03, and 04 monitoring tubes and by the four infiltration rates at the 05, 06, 07, and 08 monitoring tubes (Figure 3).For each test, monitoring tubes at only one depth were used, and other monitoring tubes at the other depths were closed.
Figure 3. Schematic view of (a) the constant-head monitoring method experimental setup, showing the position of (b) the horizontal monitoring tubes.
3.3.1. Infiltration rates at different soil depths
Figure 6. Mean soil infiltration rates at different depths in the three test tanks: (a) blank test, (b) M. baccata, and (c) S. japonica. (In each test, outflows from only one depth were used.)
Point 2: In the falling head tests the authors used the horizontal cross-section of the whole tank (A) to calculate hydraulic permeability. It means that they assumed uniform distribution of vertical flow velocity over the whole cross-section. If this was the case, they should have obtained the same results for each measurement point. Clearly, the infiltration capacity increases with the distance from tree trunk, but I have doubts if formula (7), which refers to one-dimensional flow, can be applied in this case.
Response: In this study, the hydraulic conductivities (m/d) at different distances (10 cm, 65 cm, and 130 cm) from the tree trunks or center of blank tank were measured under saturation condition, and water with a depth at 1 cm on the soil surface was kept during the falling head tests. Therefore, we assume that water flow in this study follow one-dimensional flow.
Additionally, same value of hydraulic conductivities at different distances (10 cm, 65 cm, and 130 cm) from the tree trunks or center of blank tank are still difficult to obtained due to the heterogeneity of soil caused by the tree roots and the influence of sidewall effect of the tank. The tree roots maybe mainly affect the hydraulic conductivities at 10 cm, 65 cm, and the sidewall effect of tank maybe mainly affect the hydraulic conductivities at 130 cm.
Point 3: The authors should provide values of hydraulic conductivities together with the infiltration rates obtained from their measurements. For example, they can calculate hydraulic conductivity of different soil layers from constant head tests and compare them with values obtained from falling head tests at depth 0.35m. The hydraulic conductivity from falling head tests is mentioned in Lines 159-162, but no values are provided. These values were shown in Fig.7 in the previous version of the manuscript, but now they are replaced by infiltration rates.
Response: line 159-162: “hydraulic conductivity” should be “infiltration rate”. Corrections had been made and shown below:
Line 161-163 (Line 159-162 in WATER-531691): Mean infiltration rates (I, m/d) of the soil at distances of 0.65 m and 1.30 m were calculated by the four infiltration rates at the 02, 03, 04, and 05 monitoring tubes and by the four infiltration rates at the 06, 07, 08, and 09 monitoring tubes (Figure 4).
In the revised manuscript, all “hydraulic conductivity” were calculated to “infiltration rate” by equation below:
I =K × J
Where, K is the hydraulic conductivities (m/d), J is the hydraulic gradient.
In addition, the value of soil infiltration rates in vertical (0.35 m, Fig. 6) and horizontal directions (0.35 m, Figure 7) represent different soil layer. The value of infiltration rate in vertical (0.35 m, Figure 6) represents the soil permeability at depth from 0-0.35 m, but the value of the infiltration rates in horizontal directions (0.35 m, Fig. 7) represent the soil permeability at depth from 0.35-1.05 m.
Point 4: Continuing the previous comment, the comparison of hydraulic conductivity values from the earlier version of Fig. 7 and the infiltration rates from the current version indicates that the hydraulic gradient is about 4, which seems too much. Please clarify this.
Response: Higher hydraulic gradient (3.86) was obtained due to the selected 2-m plexiglass tube. In this study, hydraulic gradient (J) was calculated by the equation below:
J = h1/L, (6)
Where, h1 is the initial water level in the plexiglass tube (h1 = 2.70 m), L is the distance between the bottom of plexiglass tube and the bottom of the tank (L = 0.70 m).
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Unfortunately, the pdf file seems to be the old version of the manuscript, since the corrections mentioned in the author's response are not included.
Anyway, I'm still not convinced by the validity of falling head measurements. Eq. (7) is valid for one dimensional flow DRIVEN BY THE FALLING HEAD. It is based on the assumption of continuity - the volume of water leaving the supply tube in a unit of time is the same as the volume of water seeping through the soil sample. As you stated in the response, during the falling head tests in tubes, the surface of the tank was submerged in water with constant head, so actually you had a mixture of constant head and variable head boundary conditions.
Moreover, Eq. (7) is based on the assumption of variable (decreasing) infiltration rate. So the values of infiltration rates that you report for falling head tests are only instantaneous, corresponding to the initial stage of infiltration.
Also, please add to the manuscript all the explanations from your response to my comments, this will be helpful for the readers.
I suggest that you calculate the average infiltration rate in falling head tests as a*(h1-h2)/delta_t, where a is the cross section of the tube. In this way you can remove Eq. (7) and all references to hydraulic conductivity.
Author Response
Response to Reviewer 2 Comments (Round 2)
Please also see the attachment.
Point 1: Unfortunately, the pdf file seems to be the old version of the manuscript, since the corrections mentioned in the author's response are not included.
Anyway, I'm still not convinced by the validity of falling head measurements. Eq. (7) is valid for one dimensional flow DRIVEN BY THE FALLING HEAD. It is based on the assumption of continuity - the volume of water leaving the supply tube in a unit of time is the same as the volume of water seeping through the soil sample. As you stated in the response, during the falling head tests in tubes, the surface of the tank was submerged in water with constant head, so actually you had a mixture of constant head and variable head boundary conditions.
Moreover, Eq. (7) is based on the assumption of variable (decreasing) infiltration rate. So the values of infiltration rates that you report for falling head tests are only instantaneous, corresponding to the initial stage of infiltration.
Also, please add to the manuscript all the explanations from your response to my comments, this will be helpful for the readers.
I suggest that you calculate the average infiltration rate in falling head tests as a*(h1-h2)/Δt, where a is the cross section of the tube. In this way you can remove Eq. (7) and all references to hydraulic conductivity.
Response: Reponses to two reviewers and revised manuscript with changes marked had been supplied in this version.
We fully agree with the suggestions to Eq. (7), average infiltration rate (m/d) in the falling head measurements was re-calculated by the equation below:
I = a × (h1 - h2)/ (A × Δt)
where a is the cross-sectional area of the plexiglass tube (a = 7.85 × 10-5 m3), h1 and h2 are the initial and final water levels in the plexiglass tube (h1 = 2.70 m and h2 = 1.15 m), A is the horizontal planar area of soil inside of the tank (A = 4.0 m2), and Δt is the time interval between the starting water level.
During the falling head tests in tubes, the surface of the tank was submerged in water with a depth at 1 cm, and only want to make the soil under saturation condition. All responses had been added in this revised version as shown below:
Line 149-163: After the tank soil had been saturated with water for 3 hours, the horizontal variation in the infiltration rate of the soil was measured by the variable-head method, and water with a depth at 1 cm on the soil surface was kept to make the soil under saturation condition. The bottoms of the vertical monitoring tubes were set at distances of 0.10 m (1 tube), 0.65 m (4 tubes), and 1.30 m (4 tubes) from the tree trunk or tank center (Figure 4). The vertical monitoring tubes were made from plexiglass (internal diameter = 0.01 m, external diameter = 0.02 m, length = 2.0 m), and were inserted to a depth of 0.35 m from the soil surface. The mean infiltration rates (I, m/d) at distances of 0.10, 0.65, and 1.30 m were calculated using the equation below:
I = a × (h1 - h2)/ (A × Δt), (5)
where a is the cross-sectional area of the plexiglass tube (a = 7.85 × 10-5 m3), h1 and h2 are the initial and final water levels in the plexiglass tube (h1 = 2.70 m and h2 = 1.15 m), and Δt is the time interval between the starting water level, h1, and the final water level, h2. Mean infiltration rates (I, m/d) of the soil at distances of 0.65 m and 1.30 m were calculated by the four infiltration rates at the 02, 03, 04, and 05 monitoring tubes and by the four infiltration rates at the 06, 07, 08, and 09 monitoring tubes (Figure 4). For each test, only the outlet at bottom was open.
Line 149-163: t is noteworthy that the value of soil infiltration rates in vertical (0.35 m, Figure 6) and horizontal directions (0.35 m, Figure 7) represent different soil layer. The value of infiltration rate in vertical (0.35 m, Figure 6) represents the soil permeability at depth from 0-0.35 m, but the value of the infiltration rates in horizontal directions (0.35 m, Figure 7) represent the soil permeability at depth from 0.35-1.05 m. Additionally, same value of infiltration rates at different distances (10 cm, 65 cm, and 130 cm) from tree trunks or center of blank tank are still difficult to obtained due to the heterogeneity of soil caused by the tree roots and the influence of sidewall effect of the tank. The tree roots maybe mainly affect the hydraulic conductivities at 10 cm, 65 cm, and the sidewall effect of tank maybe mainly affect the hydraulic conductivities at 130 cm.
Figure 7. Mean soil infiltration rates at different distances from the tree trunks in the three test tanks: (a) blank test, (b) M. baccata, and (c) S. japonica.
Author Response File: Author Response.docx
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The experiments are interesting and the general conclusion about the increase of infiltration caused by tree roots seems plausible. But I have serious doubts regarding the spatial variability of soil permeability and infiltration rates reported by the authors.
From Fig. 5 results that in each test after about 3 hours steady state conditions were reached. The hydraulic gradient was close to unity, so the equivalent vertical permeability is approximately equal to the infiltration rate in each case, i.e. 0.28, 0.33 and 0.61 m/d for the the blank case, M. baccata an S.japonica respectively. How should we then interpret the vertical variation in the infiltration rate, shown in Table 2 and Fig. 6 ? In steady state the infiltration rate must be constant through the column. If these results refer to transient stage, they are not meaningful.
Also, the spatial variability of permeability reported in Table 3 in Fig. 7 is not consistent with the result of steady state test. The average permeability calculated from the spatial distribution shown in Fig. 7 is smaller than the one obtained from steady flow measurements in all three cases. Eq. 5 is suitable for one dimensional vertical flow condtions, note that A refers to the corss section area of the whole tank. However, during the falling head tests the flow conditions were probably far from one dimensional vertical, closer to radial and distorted by impermeable walls and the presence of drainage layer at the top. It is not known what was the position of the initial water table in the tank during the tests. So these results do not seem reliable.
Reviewer 2 Report
This paper addresses data describing “Increase and spatial variation in soil infiltration rates associated with fibrous and tap tree roots” by using an experimental tank. Overall, I believe that the paper has scientific merit, and the questions about how root types affects soil infiltration rates in artificially manipulated tank systems. Although the paper is well written, the description of some very important point was inadequate or completely missing in the manuscript. Main concern about the analysis is with respect to the tree and soil information. In addition, more detailed soil information is required such as soil bulk density, total porosity, and soil texture with planted tree age, tree height, and tree diameter in each treatment. The other thing is how to control evapotranspiration during the study period. I suppose the study results may be affected by these factors with root types. Finally, some statistical analyses to compare the infiltration effects of two root types should be conducted in the paper.
Comments
1. Line 37, 41, 50, 52, 66, 69, 217: need to follow Journal’s citation rule.
2. Line 82, 168: crabapple?
3.Table 1, 2, 3: How to calculate % standard deviation?
4.Table 2 and Figure 6. Some data were duplicated between Table 2 and Figure 6.
5.Table 2 and Figure 7. Some data were duplicated between Table 2 and Figure 7.
6. Line 390-391: check the citation rule
