Climate Change Preparedness: Comparing Future Urban Growth and Flood Risk in Amsterdam and Houston
Round 1
Reviewer 1 Report
I have only one question. LTM uses maps of 2006 and 2012 for Asmsterdam and the 2001-11 period for Houston. Can not use more updated data? for example, for European cities URBAN ATLAS should be more up-to-date. more recent data would ensure greater accuracy to the forecast?
Author Response
Point 1: I have only one question. LTM uses maps of 2006 and 2012 for Amsterdam and the 2001-11 period for Houston. Cannot use more updated data? For example, for European cities URBAN ATLAS should be more up-to-date. More recent data would ensure greater accuracy to the forecast?
Response 1 (P.6): Yes, recent data would increase prediction accuracy, and, in this research, we have used all the recent data which was available in May, 2018 (as stated in Line 252 in p.6). It is a limitation in data availability. In addition, multi-time analysis will raise prediction validity such as 2001 & 2006, 2006 & 2012, and 2012 & 2017.
In Amsterdam case, we tried to find the most recent land cover as you mentioned, but the land cover 2017 was not available last year. However, the prediction result of Amsterdam with 2006 and 2012 land cover shows very good prediction capability with 82% PCM value and satisfies with all the calibration.
For Houston, the U.S. National Land Cover Data provides land cover raster data every decade (e.g. 1991, 2001, and 2010), but the raster image format in 1991 does not match with others 2001 and 2010. Thus, we use the most recent (2001 and 2011) and available data for Houston analysis.
Reviewer 2 Report
The quality of the text needs to be improved, the connectors between the different sections of the paper are missing, and there is a need for greater justification about why some methods or decisions selected are taken. As an example, section 4.4 presents 4 types of accuracy measures but it is not clear which of them is more relevant or what each of them provides to the research.
Other commentaries follow:
Lines 42-46. Please include the references of the studies that you are referring to.
Section 2.1.2. The title of this section is a bit misleading, maybe it is more accurate “Existing research or state of the art in LCM Studies”.
Line 129. What do you mean with a global? Worldwide?
Lines 147_154. This paragraph isn’t easy to understand. Please, rewrite it and explain the categorizations and approaches in more detail.
Lines 256-260. Please explain the impact that these different methods for each case study could have in the analysis of the results. As it will be clear in section 5.3., this will make difficult to compare both cities and. Therefore, one of the aspects that are most highlighted as innovative in this paper is that two cities are compared so the comparability of the methodology used is essential for the results. Please justify how this different approach will affect the results.
Line 260. Areas is plural.
Line 266-267 & Table 1. Can you categorise the input factors according to the driving factors? It would be more appropriate to locate the table closer to the text where it is mentioned for first time.
Lines 278-288. Include the references (website address or other) when available.
Section 5.1. 4th line. Can you explain why the calibration results are worse for Houston?
Section 5.2. Comment on these results.
Section 5.3. 3rd-4rd paragraph. Please add these nomenclature in the Figure. In the case of Houston, add the % of each case in order to be able to compare in relative terms with Amsterdam.
Author Response
Point 1: The quality of the text needs to be improved, the connectors between the different sections of the paper are missing, and there is a need for greater justification about why some methods or decisions selected are taken. As an example, section 4.4 presents 4 types of accuracy measures but it is not clear which of them is more relevant or what each of them provides to the research
Response 1 (P.1):
• Text quality and Connectors between the different sections – We have thoroughly revised, edited, and reorganized the manuscript to increase its readability, make the writing flow better, and create a more substantive piece of literature.
• Justification of methods or decisions selected (ex. Calibration) – added additional information about calibration (see p.8). The four calibration methods presented are the typically used methods for prediction modelling. Most researchers only use one. We used all 4 to ensure validity of the model. All variables utilized were chosen from existing literature on urban prediction modelling and our models proved to be good fit according to the outputs and calibration methods.
Point 2: Lines 42-46. Please include the references of the studies that you are referring to.
Response 2 (P.1&2): added Reference; [7] Ecology, [8] hydrology, [9] soil erosion, [10, 11] flood in single locations.
Point 3: Section 2.1.2. The title of this section is a bit misleading, maybe it is more accurate “Existing research or state of the art in LCM Studies”.
Response 3 (P.3&4): divided the previous Section 2.1.2 and created Section 2.1.3 (p.3) and 2.1.5 (p.4).
Point 4: Line 129. What do you mean with a global? Worldwide?
Response 4 (p.2): Yes, global means the study area is worldwide scale. changed “a global” to “the world.”
Point 5: Lines 147_154. This paragraph isn’t easy to understand. Please, rewrite it and explain the categorizations and approaches in more detail.
Response 5 (P.3) : Rewrote the paragraph.
Point 6: Lines 256-260. Please explain the impact that these different methods for each case study could have in the analysis of the results. As it will be clear in section 5.3., this will make difficult to compare both cities and. Therefore, one of the aspects that are most highlighted as innovative in this paper is that two cities are compared so the comparability of the methodology used is essential for the results. Please justify how this different approach will affect the results.
Response 6 (P.6): added explanation about two flood risks and the expected result.
Point 7: Line 260. Areas is plural.
Response 7 (P.6): changed “areas is” to “areas are.”
Point 8: Line 266-267 & Table 1. Can you categorise the input factors according to the driving factors? It would be more appropriate to locate the table closer to the text where it is mentioned for first time.
Response 8 (P.7): Added driving factor categories (e.g. natural environment, built environmental, and socio-economy) in Table 1. Move the table close to the Line 266-267.
Point 9: Lines 278-288. Include the references (website address or other) when available.
Response 9 (P.8): Added all citation and data source in reference.
Point 10: Section 5.1. 4th line. Can you explain why the calibration results are worse for Houston?
Response 10 (P.10&11): Houston’s low calibration value may be due to driving factor’s fitness and prediction scale (predicting urban pixel numbers vs population pixel numbers). Added explanation in Section 5.1.
Point 11: Section 5.2. Comment on these results.
Response 11 (P.11): Added comment for flood risk result, related to Point 6.
Point 12: Section 5.3. 3rd-4rd paragraph. Please add these nomenclature in the Figure. In the case of Houston, add the % of each case in order to be able to compare in relative terms with Amsterdam.
Response 12 (P.12): Added % of urban flood exposure in Houston in the text. We have utilized percentages and measurements through the paper now to help alleviate such issues.
Author Response File: 
Author Response.pdf
Round 2
Reviewer 2 Report
The paper was greatly improved with the changes performed, a couple of details to review before publication.
Line 123 - help comparing instead of help compare
Page 9, 1st paragraph - replace “there exists no universal... “ by it doesn’t exist universal...
Section 6 - I don’t agree with the following sentence: “The projected 3ft SLR does not impact a large amount of urban area.” I think the results already show an impact on a large amount of urban area.
Author Response
Response to Reviewer 2 Comments (Round 2)
R2_Point 1: Line123 - help comparing instead of help compare.
R2_Response 1 (P.3): changed from “help compare” to “help comparing”
R2_Point 2: Page 9, 1st paragraph - replace “there exists no universal... “ by it doesn’t exist universal...
R2_Response 2 (P.9): changed from “there exists no universal” to “it does not exist universal a”
R2_Point 3: Section 6 - I don’t agree with the following sentence: “The projected 3ft SLR does not impact a large amount of urban area.” I think the results already show an impact on a large amount of urban area.
R2_Response 3 (P.14): changed from “The projected 3ft SLR does not impact a large amount of urban area. However, sea-level could increase to 6ft or more in the future, this will greatly exacerbate the land under flood risk in Houston.” to
“The projected 3ft SLR does impact a large amount of urban area, and, moreover, 6ft or more SLR increase in the future will greatly exacerbate the land under flood risk in Houston.”
Additional Spelling or Numbering Change
R2_AD1: Line 44 (P.2): changed from “have” to “have,”
R2_AD2: Line 59 (P.2): changed from “2.2.1” to “2.1.1”
R2_AD3: Line 131 (P.3): changed from “A new hybrid model” to “New hybrid models”
R2_AD4: (P.12): changed from “46km2” to “46 km2”
R2_AD5: Eighth line in the third paragraph in section 6 (P.15): changed from “For more accurate risk analysis” to “For a more accurate risk analysis”
Author Response File: 
Author Response.docx
