Long-Term Stochastic Modeling of Sheet Pile Corrosion in Coastal Environment from On-Site Measurements
Round 1
Reviewer 1 Report
The article deals with the evaluation of long-term corrosion tests. The main attention is paid to possibilities of stochastic modeling of corrosion weakening of steel structures located in coastal environment. From the reviewer's point of view, the article is very interesting. The results are applicable in the reliability assessment and service/life prediction of steel structures exposed to coastal environment.
After minor corrections, reviewer recommends the article for publication in the journal.
Lines 106 and 117 till 120: Has the original thickness of the components been measured? This is not clearly discernible from the information in the article.
Figure 4, vertical axis: Number ow what? Number of changes per day?
Lines 166 till 168: Can the authors give at least consideration of the influence of the steel composition on the development of corrosion? Can the different steel composition have a significant effect on the measured corrosion losses?
Figure 6: Why are the results shown for reinforced concrete beam? The article deals with steel structures.
Table 3 and line 323: You correctly state that the “corrosion rate is varying with lime due to the reduction of corrosion speed with time”. What time period does the values of “average corrosion rate” in Table 3 correspond to?
Figures 7 and 8: The figures are interesting. On the horizontal axis are the “annual” values?
Line 368: Note: Gamma distribution is appropriate. The reviewer for similar cases (e.g. reliability assessment of structures; probabilistic corrosion prediction models) usually prefers 3-paremeter lognormal distribution or bounded histograms.
Figure 14 and Table 14: Interesting results.
Lines 30, 406, 407, 507: Correct typing errors.
Author Response
All changes in the text are in blue (except minor changes).
Comments from Reviewer #1
1. The article deals with the evaluation of long-term corrosion tests. The main attention is paid to possibilities of stochastic modeling of corrosion weakening of steel structures located in coastal environment. From the reviewer's point of view, the article is very interesting. The results are applicable in the reliability assessment and service/life prediction of steel structures exposed to coastal environment.
After minor corrections, reviewer recommends the article for publication in the journal.
The authors acknowledge the reviewer for the positive review and scientific interest of the paper.
2. Lines 106 and 117 till 120: Has the original thickness of the components been measured? This is not clearly discernible from the information in the article.
That is a key and interesting question. That was not measured because it was reported in the reports “as built”. Unlike piles, the thickness of sheet piles is not changed from the design basis; for piles, the thickness could depend on the availability of non-compliant pile-lines for oil or gaz transport. The paper focusses on sheet pile and doesn’t require this measurement. A sentence was added in the paper in view to clarify this point.
“Knowledge of the initial thickness can be an issue. For piles, the thickness of the ‘as built’ pile could depend on the availability of non-compliant pile-lines for oil or gaz transport. The change, if conservative, is not always reported in the construction documents. Unlike piles, sheet piles thickness is not changed from the design documents. Our study focusses on corrosion of sheet piles and the thickness used is the value reported in the design documents”.
3. Figure 4, vertical axis: Number ow what? Number of changes per day?
The reviewer is right. That is not the number but the Maximum variation of the sea water level (that has been changed in Figure 4).
4. Lines 166 till 168: Can the authors give at least consideration of the influence of the steel composition on the development of corrosion? Can the different steel composition have a significant effect on the measured corrosion losses?.
The reviewer is right and that is underlined by the second reviewer. Unfortunately the composition of the steel is not available when the pile is supplied. Another issue is that it can changed during the works if the supplier change or is the wharf is built in two periods separated for over 10 years. Get information requires exhaustive measurements from samples. They are expensive especially if not carried out during inspection. It was not possible to get this information in this study because of the cost of inspection of 23 structures in about 10 places. But that is an information that could be added to actual measurement campaign. This explanation has been added in the text:
“is clearly an issue. Unfortunately the composition of the steel is not available when the pile is supplied. Another issue is that it can changed during the works if the supplier change or is the wharf is built in two periods separated for over 10 years; that is common for extension of a wharf for its adaptation to the market. Get information requires exhaustive measurements from samples. That is an information that should be added to actual measurement campaigns in view to better assess the influence of this parameter. The age of the structures being various, there was a lot of suppliers, standards and quality of steel. Moreover, the number of influencing factors being huge, steel composition should be considered as an additional factor that acts in combination with others, with unknown relationship. We assume that the lack of knowledge increases the scatter of our data and that a knowledge of the steel composition could help to carry out a detailed statistical analysis”.
5. Figure 6: Why are the results shown for reinforced concrete beam? The article deals with steel structures.
Figure 6 is not for reinforced concrete but for steel piles and that was explained in the text. Reinforced concrete is only introduced in line 322. We add this sentence in view to better explain why the comparison is interesting:
“This comparison is interesting because reinforced concrete is steel with a protective chemical (concrete) layer.”
6. Table 3 and line 323: You correctly state that the“corrosion rate is varying with lime due to the reduction of corrosion speed with time”. What time period does the values of “average corrosion rate” in Table 3 correspond to?
The duration of exposure that was used for the computation is given in the second column of the table (from 10 to 63 years depending of the harbor and the exposure zone. It was probably not clear and we added this sentence:
“The time period at which is was computed is given in the second column of the table. “
7. Figures 7 and 8: The figures are interesting. On the horizontal axis are the “annual” values?
The values are the average measurements (and confidence interval – Fig. 7). That has been clarified in the text:
“at times of inspection 25 and 63 years respectively for BO1 and HA7d “ and line 356 “(mean values after 63 years)”.
8. Line 368: Note: Gamma distribution is appropriate. The reviewer for similar cases (e.g. reliability assessment of structures; probabilistic corrosion prediction models) usually prefers 3-paremeter lognormal distribution or bounded histograms.
We agree that several options can be chosen. We have underlined that statistically that was a good choice and that this model is suitable for regions with similar water characteristics than France.
9. Figure 14 and Table 14: Interesting results.
We thank the reviewer for this comment.
10. Lines 30, 406, 407, 507: Correct typing errors.
That was a corrected
Reviewer 2 Report
This is a very interesting paper about the modelling of a big amount of data (35 338 measurements both on structural and environmental points of view) regarding with sheet-pile corrosion in coastal environment from on site measurements for more than 40 years.
This paper can be considered included in one of the subject areas of the Journal of Marine Science and Engineering.
Under my opinion, the paper is well structured. I’d like to make some comments about different aspects of the paper.
The environmental parameters are not considered in this study. Authors justified the reason, but I think table 1 can be deleted and the section 2.3 could be reduced. It’s difficult to access to the thesis from reference 23 (it’s in French). An url must be useful.
The model is based on the loss of thickness measurements, but the protocol associated with the NDT tool is not referenced. A reference must be included.
The composition of the steel is not considered and it’s assumed that the steel is almost homogeneous in French harbours. I don’t agree with this sentence. As the main amount of data come from BO Harbour (86.2%), the supposition of the uniformity of the composition for modelling purposes can be accepted. The authors say that more than the 99% of the data correspond with HO harbour (lines 151-152). That’s not true. It represents the 86.2% of the total measurement and the 87.27% of the measurements corresponding to the tidal/spray zone. A total amount of 35 388 measurements was gathered. According to figure 5, the sum is about 34868. It’s seem to be a lack of information regarding to BO harbour. In the next section, information about corrosion in the spay and tidal zone are provided, but according to the figure 5, all the data correspond to the immersion zone.
One important aspect is that there is not corrosion data of the tidal/spray zone in BO, PL and SE harbours. What’s the reason? The lack of information related to atmospheric corrosion data is clear, but not with the tidal/spry zone.
The authors say that knowing that the measurements of corrosion and environmental parameters were not carried out at the same depth, the relationship between parameters and corrosion is highly difficult to capture. That’s true. For corrosion purposes, the nature of the corrosion products is also very important. Also, the different types of corrosion processes involved in this case make difficult to consider all the variables involved (i.e. MIC).
Table 3 is about the mean values and ranges (in brackets) of some physico-chemical parameters of seawater near the 308 studied structures in the immersed zone, but the information into the table is about corrosion rates as it’s said in line 304. That’s a mistake. As I said before, the physico-chemical parameters are not relevant in this paper.
The conclusions are in good agreement with the data.
References should be described according to the journal’s style.
The next observations are highlighted in the text:
Line 6: the initial comma must be deleted.
Line 29: change “uncertain-ties” by “uncertainties”.
Line 30: change “thick-ness” by “thickness”
Line 41: change “Paik” by “Paik et al.”
Line 53: change “Jonbloed” by “Jonbloed et al.”
Line 73: change “PH” by “pH”; “conductivity, nutrients” by “conductivity and nutrients”.
Line 91: Engineering Center of the French Ministry of Public Works.
Line 129: change “Figure 1” by “Figure 3”.
Lines 142 to 147; 151-152: size of the text is not correct.
Line 197: change “environ-mental” by “environmental”.
Lines 304-306: the line-spacing is different to the rest of the text.
Line 324: change “corrosion speed” by “corrosion rate”.
Line 373: I don’t understand the sentence “Number of measures whore distribution follows a given pdf versus the age”.
Line 376: the sentence of figure 12 is wrong. It can not be understood.
Line 407: delete the dot after “given”
Line 408: change “wit” by “with”.
Line 447: delete “M” before “Main”. The font of the text is different to the rest of the paper
Comments for author File: 
Comments.pdf
Author Response
All changes in the text are in blue (except minor changes).
Comments from Reviewer #2
0. This is a very interesting paper about the modelling of a big amount of data (35 338 measurements both on structural and environmental points of view) regarding with sheet-pile corrosion in coastal environment from on site measurements for more than 40 years. This paper can be considered included in one of the subject areas of the Journal of Marine Science and Engineering. Under my opinion, the paper is well structured. I’d like to make some comments about different aspects of the paper.
The authors acknowledge the reviewer for the positive review and scientific interest of the paper. The paper was carefully checked, especially the missing references and the too long sentences..
1. The environmental parameters are not considered in this study. Author justified the reason, but I think table 1 can be deleted and the section 2.3 could be reduced.
Offshore steel frame structures are called Jacket structures. Fatigue is checked where the local stresses are the highest, at the place of welded connection. This topic has been clarified in the paper and offshore steel wind turbines has been added in the title.
In the introduction: Offshore steel trussed-shaped structures, called Jacket structures are the most popular bottom fixed one for supporting oil and gaz platforms and offshore wind substation. This paper focusses on them. They are built by welding cylindrical beams that form a lattice-like structure.
The title was chanted to: A two scale probabilistic time-dependent fatigue model for offshore steel wind turbines.
2. It’s difficult to access to the thesis from reference 23 (it’s in French). An url must be useful..
The french title was given as well as the url and the DOI.
3. The model is based on the loss of thickness measurements, but the protocol associated with the NDT tool is not referenced. A reference must be included.
The protocol was described in Schoefs et al. 2009. That is now cited in the paper.
4. The composition of the steel is not considered and it’s assumed that the steel is almost homogeneous in French harbours. I don’t agree with this sentence. As the main amount of data come from BO Harbour (86.2%), the supposition of the uniformity of the composition for modelling purposes can be accepted. The authors say that more than the 99% of the data correspond with HO harbour (lines 151- 152). That’s not true. It represents the 86.2% of the total measurement and the 87.27% of the measurements corresponding to the tidal/spray zone.
This sentence about homogeneity of the steel did not reflect the meaning of the authors and was deleted. The reviewer is right and that is underlined by the first reviewer. Unfortunately the composition of the steel is not available when the pile is supplied. Another issue is that it can changed during the works if the supplier change or is the wharf is built in two periods separated for over 10 years. Get information requires exhaustive measurements from samples. They are expensive especially if not carried out during inspection. It was not possible to get this information in this study because of the cost of inspection of 23 structures in about 10 places. But that is an information that could be added to actual measurement campaign. This explanation has been added in the text:
“is clearly an issue. Unfortunately the composition of the steel is not available when the pile is supplied. Another issue is that it can changed during the works if the supplier change or is the wharf is built in two periods separated for over 10 years; that is common for extension of a wharf for its adaptation to the market. Get information requires exhaustive measurements from samples. That is an information that should be added to actual measurement campaigns in view to better assess the influence of this parameter. The age of the structures being various, there was a lot of suppliers, standards and quality of steel. Moreover, the number of influencing factors being huge, steel composition should be considered as an additional factor that acts in combination with others, with unknown relationship. We assume that the lack of knowledge increases the scatter of our data and that a knowledge of the steel composition could help to carry out a detailed statistical analysis”.
5. A total amount of 35 388 measurements was gathered. According to figure 5, the sum is about 34868. It’s seem to be a lack of information regarding to BO harbour. In the next section, information about corrosion in the spay and tidal zone are provided, but according to the figure 5, all the data correspond to the immersion zone.?
In fact, there was a typos and the total number was 35 460 in line 74, in the abstract and in line 80. That has been changed in the text. These data include the data from BO. Figure 5 was corrupted after the copy from the original data base and it was changed. It includes the number of data in the 3 zones for BO.
6. One important aspect is that there is not corrosion data of the tidal/spray zone in BO, PL and SE harbours. What’s the reason? The lack of information related to atmospheric corrosion data is clear, but not with the tidal/spry zone.
Reviewer is right and the reason is the same as above (Figure 5 was corrupted during the copy) and has been changed.
7. The authors say that knowing that the measurements of corrosion and environmental parameters were not carried out at the same depth, the relationship between parameters and corrosion is highly difficult to capture. That’s true. For corrosion purposes, the nature of the corrosion products is also very important. Also, the different types of corrosion processes involved in this case make difficult to consider all the variables involved (i.e. MIC).
We agree and we decided to place table 1 in Appendix 1.
8. Table 3 is about the mean values and ranges (in brackets) of some physico-chemical parameters of the seawater near the 308 studied structures in the immersed zone, but the information into the table is about corrosion rates as it’s said in line 304. That’s
a mistake. As I said before, the physico-chemical parameters are not relevant in this paper..
Title of table 3 has been corrected.
9. The conclusions are in good agreement with the data.
We thank the reviewer for this comment.
”.
10. References should be described according to the journal’s style.
It was checked and corrected.
11. The next observations are highlighted in the text:
Line 6: the initial comma must be deleted.
Line 29: change “uncertain-ties” by “uncertainties”.
Line 30: change “thick-ness” by “thickness”
Line 41: change “Paik” by “Paik et al.”
Line 53: change “Jonbloed” by “Jonbloed et al.”
Line 73: change “PH” by “pH”; “conductivity, nutrients” by “conductivity and nutrients”.
Line 91: Engineering Center of the French Ministry of Public Works.
Line 129: change “Figure 1” by “Figure 3”.
Lines 142 to 147; 151-152: size of the text is not correct.
Line 197: change “environ-mental” by “environmental”.
Lines 304-306: the line-spacing is different to the rest of the text.
Line 324: change “corrosion speed” by “corrosion rate”.
Line 373: I don’t understand the sentence “Number of measures whore distribution follows a given pdf versus the age”.
Line 376: the sentence of figure 12 is wrong. It cannot be understood.
Line 407: delete the dot after “given”
Line 408: change “wit” by “with”.
Line 447: delete “M” before “Main”. The font of the text is different to the rest of the paper
We thank the reviewer for these detailed comments. It was corrected and others typos were found and corrected also.
Round 2
Reviewer 2 Report
Perfect.
