This blog post is an overview of my thesis as presented at recent conferences. I call it an “elevator-talk” but it’s more like a 10 minute elevator ride, hope you enjoy it!
Extreme weather events have become a common occurrence and coastal communities are adversely affected by it. Studies have shown that the changing climate has increased the frequency and severity of storms, surging sea levels, and floods, as was seen with Hurricane Sandy (2012) and Typhoon Haiyan (2013). The need to be proactive in preparing for these events, as a means of climate change adaptation and disaster risk reduction, is evident. This study focuses on the formal definition, measurement and simulation of coastal community preparedness and response to severe storm events. The community of Isle Madame in Richmond County Nova Scotia was used as a case study to demonstrate the framework.
Canadians adhere to gradated (bottom–up) emergency management system (Figure 1), where responsibility initially lies with the individual and gradually increases as their capacity is exhausted. The issue with this system is that that money, knowledge and resources flow top-down.
We often associate storms with larger communities (Hurricane Sandy – New York, Hurricane Katrina – New Orleans), but what about the smaller communities along the coast that may not have the resources to bring in a large consulting or planning firm to access emergency management? They require a framework in order to evaluate their state of preparedness for these severe environmental events. In such a bottom-up system, if you don’t ask you don’t get.
We proposed a framework, containing four components (Figure 2), to determine what it means for smaller coastal communities to be prepared for severe storm events. In order to be prepared, communities would need to evaluate their:
- Resources (e.g., blankets, generators, fire trucks, sandbags),
- Emergency plans (a blue print in order to assign responsibilities and determine procedures),
- Decision making in a time of stress, and
- Ability to manage uncontrollable events (Sometimes, things just don’t go your way. A fallen tree or washed out road may hinder your response despite having the right resources, plans and well informed decision making)
In order to evaluate resources and emergency plans, we have created a suite of indicators of preparedness. Thirty-one (31) indicators were created from literature for preparedness. The literature ranges from local emergency plans that included the phone number of individuals up to international frameworks such as the United Nations Hyogo Framework for Action. The indicators were classified into a hierarchical structure that includes three levels: Dimensions, Attributes, and Indicators (Figure 3). A scoring metric was created for each indicator in order to capture its essence. The hierarchical structure also enabled us to derive weights for each indicator through a pair-wise comparison. With the scores and weights of the hierarchy, communities would be able to conduct sensitivity and gap analyses to identify areas of improvement.
A table-top exercise was created to assess decision making. Table-top exercises are generally seen as open ended discussions where a narrative is presented and an informal discussion follows. During the discussion, participants are often so concerned with what they would do individually that it is difficult to gauge how the organization would navigate through the emergency as a whole. Therefore a more structured table-top exercise was proposed. Five exercise phases were created starting from the tail end of preparedness, moving into response and then the early stages of recovery (Figure 4). For each phase, two to three detailed events were created. The detailed events cover 12 of the 16 hazards defined as “very likely” and “nearly to occur” hazards in Richmond County’s Emergency Plan. Hazards left out were not within the scope of this study.
The phased table-top exercise was presented to Richmond County in collaboration with the Canadian Red Cross in early May 2014. The exercise gained positive feedback from the participants and was also mention in the local newspaper (link). Different from normal table-top exercises, members of the community were invited to participate as well. It was interesting to see that community members were the most proactive and at times took initiative to address the gaps identified during the exercise.
With well-informed decision making, things are not guaranteed to succeed. For that reason, a simulation model was used to evaluate how different decision strategies would fair in storm situations. The simulation model comprises of a series of decision trees, one tree for each detailed event from the table-top. Figure 7 shows the decision tree for the event of alerting the community for the arrival of a storm. Three decision alternatives were propose: direct (e.g., door-to-door notification), passive with articulation (e.g., message through the local radio station), and passive without articulation (e.g., sirens). Of course, there is a trade-off for each decision. A direct approach will take longer to implement, however, it will notify more people. Following the decision, a future uncontrollable state will occur. In this case, the storm will either take a long, medium, or short time to arrive. The outcome for this event is the percentage of community members alerted.
The decision trees were constructed in the Arena Simulation program and different strategies can be evaluated if the communities provide empirical data for response time. Possible strategies for evaluation include: top-down, bottom-up, minimize cost, minimize response time, and random decision making.