Overall, it is an excellent contribution, the findings of which significantly advance the state of research which the authors have presented very well. The methodological approach is well considered and very well documented. The classification of the results is for the most part convincing. However, there are some issues that need to be adressed:

1. Please clarify whether retreat, relocation and resettlement are synonyms or define different approaches. If they are considered as synonysms please make use of only one of these terms. Otherwise, please explain the differences.
2. A fundamental question is whether the relatively small selection of 3 or 7 case studies  for in- and off-citiy resettlement is sufficient to draw general conclusions about the advantages and disadvantages of the two strategies such as " The second major insight of the assessment is that post-relocation conditions only slightly vary between the two resettlement types"
3. It is surprising that access to services of general interest (health care, education, culture etc.) was not considered as enabling factor. I can be assumed that in-city resttlement areas perform much better because of their proximity to infrastructure clusters 
4. The policy recommendations are rather generic. One would like to see more specific recommendations regarding the possible need for amendments to existing legislation and policies for resettlement in the Philippines

This research investigates crucial and timely issues around managed retreat options. It represents a thorough literature review on the subject along with the relevant background on the case study. The research methodology is robust, and its implementation is well justified. The research provides interesting results, particularly in comparing the in-city and off-city relocations that have already occurred. However, there is room for improvement in the presentation of both methodology and results, requiring further clarification.

Introduction: It is not clear what the national water code (1979), based on the danger areas identified, is about. It is mentioned that it does not include flood probability and vulnerability, but what does it include for defining flood risk zones? And how the new settlement locations are identified? Do people know about the flood probability, vulnerability, and exposure of new areas before their relocations?

Figure 1: This figure is confusing: a circular graph, particularly with the arrow connecting the outcome to the start, indicates an iterative process. However, the data collection and analysis process from phases 1 to 3 was a one-off process. The way it is presented in Figure 1 implies that the entire process has been repeated. A linear graph can better explain the process of this study.

Figure 1: If the whole process is connected, with each step building upon the findings of the preceding one, this should be shown by directed arrows indicating the direction of input and output. The type of data transferred between steps can also be shown in these areas, e.g., ‘enabling factors for resilient retreat’.

Figure 1 and methodology: I do not understand how the factors identified in the FGD, which is the second step, were already included in the household survey and were then evaluated in phase 3. Is it that the predefined factors identified and included in the HH survey are only ranked (but not identified) in the FGDs? In this case, how is Phase 3 built upon the finding of phase 2? Please elaborate a bit more on the relationships between the three phases and why the analysis needed to be done in separate phases.

The overarching survey questions need to be presented, particularly because the survey and FGD questions are not attached as SM.

Line 218: Mai should be May

While graphs are shown for all questions, explaining the content of the graphs in the text is redundant.

Table 2: why an aggregated response on the previous settlement is presented instead of separate in-city and off-city responses to allow for comparison with the current settlement?

Limitations: the on-site facilities and services of the new settlements are not included in the analysis. For example, availability or access to basic services might be related to the already available basic services in the new sites rather than their distance from the city. In addition, the period of time since the relocations is also important in what people think about social cohesion or even experiencing flood and there might be a difference between those who moved recently and a long time ago. These kinds of limitations can be acknowledged in the paper.

Future study: More granular analysis of relationships among factors and demographic characteristics of the population (age, gender, income, etc.)