Improving Community Resilience and Emergency Plans by Mapping Risk and Preparedness at the Neighborhood Scale
Round 1
Reviewer 1 Report
This is an interesting study and the authors have clearly identified the challenges in community response for this hazard and this location. Robust data collection methodology. The seismic risk side of the study is robust. The survey perception side of this I have additional question. My concern is sample size and representation in the five polygons identified in the study. This is not addressed in the paper. Additionally what were the questions asked in the PRE survey (how a question is phrased influences the response)? How were the interviews done and by what standards (e.g. trained interviewers or criteria). How were the responses recorded and were any comments provided? Did the PRE survey address both seismic concerns and extreme weather?
I see this paper as a limited case study and suggest it be identified that way in the paper. Overall this is a good paper and is a good addition to the literature.
Author Response
The authors wish to thank the reviewers for constructive and important comments. We have addressed all comments and made the requested clarifications in a way that benefits the manuscript and its future readers.
Following is a detailed list or revisions based on reviewer 1 comments:
a. What is the sample size of results presented in the 5 polygons in Figure 3?
Answer: The sample size for DI (building characteristics used to calculate seismic risk) is 90-100%. This is simply because each neighborhood was constructed using a single building design, and in the following decades very few buildings were retrofitted or altogether replaced by newer buildings. While we did distribute DI questionnaires, and mostly received uniform DI estimates (as expected), some people did not know the date of construction of their house, so we had to complete/correct this based on the local council’s data. As we had full access to this data we presented DI scores for each neighborhood.
As for PRE (readiness) – we collected a sample size of 115 responses from the ‘new’ and ‘old’ neighborhoods (which represents 13-17% of households in these neighborhoods; this is stated in lines 274-275). In the PRE calculation we batched the two polygons of new neighborhoods, and the two of old ones, so we only have two average values of PRE (Av. PRE=5.2, SD=1.5 for the new neighborhoods and Av. PRE=3.9, SD=1.5 for the old neighborhoods). We did not calculate the PRE of all other neighborhoods because we did not have enough data. This is reiterated in the text in lines 293-295:
“While the typical risk score (in terms of MMI) was calculated for each neighborhood separately, readiness (PRE) averages were determined for the new and old neighborhoods (where we retrieved responses from approximately 15% of households)”.
b. What were the questions asked in the PRE survey (how a question is phrased influences the response)?
Answer: We simply adopted the questions from previous surveys (CEG, ARC Public Readiness Index, 2006; references 5, 53), and we translated them to Hebrew for the local public. In the manuscript we added the following examples of questions and reference (lines 261-265):
“All questions were closed-ended (e.g. “Do you (or your neighbors) have a fire extinguisher?”, “Do you (or your neighbors) have a first aid kit?”) with three optional answers representing three degrees of community preparedness (Yes (1 point), A Neighbor has (0.5 pt), No (0 pt)). The questions and their exact wording were based on previous surveys [5,53].”
c. How were the interviews done and by what standards?… How were the responses recorded and were any comments provided?
Answer: The online and in-person surveys were based on closed-ended questions with minimal (or zero) interference by the interviewer (this was stated in line 261). No comments were collected.
d. Did the PRE survey address both seismic concerns and extreme weather?
Answer: The PRE survey was only about preparedness, not about seasonal effects such as weather conditions. The weather analysis was based on data from the national Meteorological Service (line 417, reference 54).
d. I see this paper as a limited case study and suggest it be identified that way in the paper.
Answer: This was added to lines 146, 609. All case studies are local, specific and limited in the sense that gained insights might be case-dependent and should be carefully considered before any general application. Our paper reflects this modest interpretation of any case study.
Reviewer 2 Report
This paper presents a case study of Mitzpe Ramon and the use of a tool for self-assessment of risk and preparedness. The authors outline the necessity of such tools and the inclusion of spatial and temporal patterns for emergency planning. This paper provides a thoughtful and well-organized case study on a very important and timely topic. The practical implications for this work are clear and the conclusions support the complexity of planning and support in multi-hazards contexts, particularly in locations that are geographically (and possibly socially) isolated or where tourism is a vital part of the local economy.
This piece requires minor revisions for grammatical errors. Concerning content, I found the Introduction to be quite thin in its handling of the complex nature of community resilience. There is one paragraph on page 2 about the challenges associated with building community resilience; This is a key area where this paper can be improved. Can you tell us more about why engaging communities in emergency preparedness and community resilience is such a difficult task? This is very important as it relates to the central point of this work as the tool is designed for the public. How do we know the public will use and understand it? And as an extension of this line of questioning, I would be very curious to know if there will be other work published or that is in preparation for understanding how the tool actually informed community behavior. For example, you say that studying change in preparedness behavior is not something you did and goes beyond the scope of this specific paper, but is this something we can look for in future work? The tool you have developed and the argument for a greater understanding of spatial and temporal variation are incredibly important, yet if no one uses the tool, how do we address the variation that you argue is so important? This connection needs to be made more clearly in the first few pages of the manuscript.
Sections 2 and 3 are well written and provide sufficient detail.
In section 4, why did you only calculate tourism rates based on weekend stays rather than at any point in the week? Is there reasoning that purely weekend visitation is more appropriate than weekly or monthly averages?
Section 5 provides a great level of detail to think about how to calculate spatial and temporal variation into emergency planning at a local or regional scale. Key questions that kept returning based on these results are 1) Has this study had been shared with the community? 2) What was the community's initial reaction? and 3) What can the community do -- particularly older neighborhoods that are more at risk -- for planning? This connection back to community resilience and actually using this information for improving resilience and preparedness behavior is important and should be addressed in the discussion or summary of the paper.
Overall, this paper is a thoughtful contribution to the literature. There is significant value in this localized approach and the authors make a well-reasoned argument for including spatial and temporal data into emergency planning decisions.
Author Response
The authors wish to thank the reviewers for constructive and important comments. We have addressed all comments and made the requested clarifications in a way that benefits the manuscript and its future readers.
Following is a detailed list or revisions based on review #2:
1. The introduction can be improved:
a. Can you tell us more about why engaging communities in emergency preparedness and community resilience is such a difficult task?
Answer: The major challenges referred to in this paper in relation to engaging communities to improve resilience are:
- The challenge of evaluating community resilience at a resolution relevant for spatial analysis on the scale of single neighbourhood (lines 40-47; references 14-22).
- Effectively engaging the community, particularly in remote periphery towns and villages. This is presented as a global challenge which is being addressed in many recent papers (lines 40-44, references 14, 15 and lines 57-61, references 25, 26), and specifically for Israel (lines 64-68, reference 29-31 and lines 83-88, reference 37). On the topic of periphery vulnerability, we also mention the important inverse response law outlined by Phibbs et al., 2018 (reference 31, lines 89-100).
- Evaluation of long-term effectiveness of various intervention methods that aim to improve resilience (lines 62-64, references 27, 28).
b. How do we know the public will use and understand it? Will there be other work published to evaluate how the tool actually informed community behaviour. …Is this something we can look for in future work?
Answer: The survey is tailored for self-assessment by the public – it is short, simple and is worded based on laymen terms. While we did not encounter any difficulties in understanding the questions, we realize that in any wide-spread implementation, the questions regarding building characteristics may need to be adjusted to describe the house types that exist in the study area. In the case study of Mitzpe Ramon this was easily achieved as 90-100% of houses in each neighbourhood were designed and built by the same characteristics, at the same time and by the same building company. Similarly, while we had near-complete cooperation from the public, we do not know for sure this would be the case in every community (it probably helped that we are local scientists and we had help from local students and therefore the people we addressed were exceptionally cooperative).
As for long-term evaluation of the effectiveness of our intervention method – this was originally planned to be assessed in a second larger case study in which we would only use online surveys (and where the scientists are not part of the small community). This part of the research has been delayed due to budget constraints. The publication of the method and first case study will hopefully strengthen the case to actually complete the second phase of the study. A statement was added to the discussion section 5.2 regarding future work: “A larger scale, long term study is being planned to evaluate the effectiveness of our intervention method” (lines 558-559).
c. If no one uses the tool, how do we address the spatial and temporal variation that you argue is so important? This connection needs to be made more clearly in the first few pages of the manuscript.
Answer: The tool is mostly useful if it is used by a large sample of residents of each neighbourhood. In our case study we were able to confirm the DI assessments made by community members and were able to collect sufficient PRE surveys in the major neighbourhoods of town. Where we did not have enough data, we did not calculate or present an average PRE score. This was restated in the caption of Figure 3 (lines 397-398).
2. In section 4, why did you only calculate tourism rates based on weekend stays rather than at any point in the week?
Answer: The weekend number of tourists (per day) are 2-4 times higher than during weekdays (not including peak events which attract many thousands of tourists). It is taken for granted that during weekdays, the amount of transient people is lower and the burden on emergency and recovery efforts/resources is therefore lower too. This was added in lines 457-458.
3. Section 5 provides a great level of detail to think about how to calculate spatial and temporal variation into emergency planning at a local or regional scale. Key questions that kept returning based on these results are:
a) Has this study had been shared with the community? b) What was the community's initial reaction? and c) What can the community do? This connection back to community resilience and actually using this information for improving resilience and preparedness behavior is important and should be addressed in the discussion or summary of the paper.
Answer: We thank the reviewer for this important comment.
While the implications for Mitzpe Ramon were stated in Section 6 (lines 592-597), the process of communicating the implication and the response of the local council are indeed an important addition. We added this in Section 5, lines 533-544:
“Communicating our results and insights to Mitzpe Ramon Council and its emergency response officer, a few improvements were outlined for local emergency plans. Water and supply distribution plans should account for potential extensive damage severity and transient population in the old neighborhoods. The capacity for local evacuee management should also account for increased transient population and weather conditions (with SOB peaking in August). In addition, as instrumental PRE scores were low in the old neighborhoods (lacking fire extinguishers and first aid kits), a program was proposed to encourage residents and tenant associations/committees to get such emergency equipment and improve their risk awareness and resilience. As these improvements do not require much funding, and as they were derived in collaboration with council representatives and with a direct focus on local plans, the feasibility of implementation is high (but ultimately depends on local authorities)”.
