Wednesday, June 25, 2014

How many meals do we need?

The first, critical task of an emergency manager at the beginning of an event is to determine the scale of the disaster. The scale of the disaster determines the resource requirements, which leads to an inventory of resources on hand and identification of what is short. A big first question in mass care is: How many meals do we need?

This is a basic, even fundamental, question for mass care practitioners at the beginning of a disaster. Yet, at the 2014 National Mass Care Exercise we spent an inordinate amount of time and energy on the first day of the exercise trying to answer that question. A lot of the practitioners at the Exercise had gained their knowledge and experience operating at the local level in the impact area. Most of my experience was gained at the State Emergency Operations Center (EOC). This variance in our backgrounds influenced how we viewed both the question and the answer.

The Mass Care Chief at a small disaster addresses the question of how many meals are needed by looking first at how many people are in the shelters. Job #1 in mass care is to resource the shelters. If 50 people are in the shelters then they need 2 meals/day and that's 100 meals. Next step is to address disaster feeding in the community through mobile feeding (Yes, I understand it's more complicated than this but bear with me). A Food Service Delivery Vehicle (check the link for the definition) normally carries 300 meals/trip. Based on our initial estimate of the scale of the disaster we need 300 + 100 meals or 400 meals/day. If the mobile feeding vehicle comes back empty then we send them out the next day on 2 trips and up our requirement to 700 meals/day.

At the State EOC I'm looking at this question in an entirely different way. Ten years ago this summer I worked 4 hurricanes in a 6 week period. I would've been pretty damn stupid if I didn't start figuring out what I was supposed to do after the 3rd or 4th hurricane. I learned pretty quick that my reaction time to a request that I hadn't planned for was at least 48 hours, and was more often 4 or 5 days. If you asked me for something and I had it on hand I could get it to you today or tomorrow.

If I didn't have it in hand I had to go get it and then physics got in the way. First, I had to convince someone else in the EOC that what you needed was both necessary and a priority. Next, if it had to come from FEMA, we had to wait breathless and unknowing while our request circumnavigated the federal approval process. Then, someone not under my control had to buy whatever it was you needed, arrange to load it on a truck somewhere in the United States and then drive it to wherever I had designated as the delivery location.

As you can see, 4 or 5 days sounds about right.

But if you're in the middle of a BIG disaster (and believe me, if you're involved in the disaster and the one making the request, it's BIG) and you ask the State for something you want it delivered yesterday, not in 4 or 5 days. That means that if I want to do a good job then I have to figure out what you need, order it, and have it on hand ready to deliver to you by the time you get around to figuring out that you need it. And you'll still complain that I didn't get it to you yesterday.

As a result of this experience I learned to look at the question of how many meals needed in a different manner. I don't know and don't care how many people we need to feed. I need to know how many meals/day of production and distribution resources are required to meet the estimated scale and scope of the disaster.

Furthermore, we need to estimate this meals/day requirement 24 hours before landfall on a Noticed Event, or within 12 hours on a No-Notice Event. Why so early? Because if we don't have enough we need to give FEMA the 4 or 5 days to get and deliver whatever it us we're short of.

So - how do we estimate the meal requirement before the hurricane has even made landfall? All mass care resource requirements are a function of the size of the population affected, the intensity of the event and the geography of the affected area. When estimating the requirements for a hurricane, for example, we use the Hurricane Center forecast to determine the geography of the affected area and the intensity of the event. The population affected can then be determined using census data.

But everyone affected by the disaster has not been impacted by the same intensity. Some people were hit by Category 4 winds while others by Category 1. Their disaster meal requirements would be different so we need to estimate how many people are affected by different intensity levels. So how do we do that? We looked at different factors associated with Low, Medium and High Intensity and put them in Table 1 below.

Table. 1. Estimating Disaster Intensity for use in
forecasting mass care resources
Intensity Level
  • Structural damage to buildings characteristic of a Category 4/5 hurricane or Mercali Intensity levels of X/XI/XII.
  • Up to 80% or more of customers without power
  • Up to 50% or more of  Potable Community Public Water Systems inoperable
  • Wastewater collection system is NOT providing wastewater treatment in accordance with permit conditions and regulations.
  • Structural damage to buildings characteristic of a Category 3 hurricane or Mercali Intensity levels of VIII/IX.
  • Up to 50% or more of customers without power
  • Up to 30% or more of  Potable Community Public Water Systems inoperable
  • Wastewater collection system is properly conveying and providing wastewater treatment, but at a compromised capacity.
  • Structural damage to buildings characteristic of a Category 1/2 hurricane or Mercali Intensity levels of VI/VII.
  • Up to 20% or more of customers without power
  • More than 10% or more of  Potable Community Public Water Systems inoperable
  • Wastewater collection system is properly conveying and providing wastewater treatment with limited disruptions.

Once we have determined the population affected by each intensity level using Table 1, we use Table 2 to calculate an estimate of the meals/day in production and distribution capability required.

Table 2. Estimating Disaster Meals/Day required
Estimate procedure
Sum of population affected by Low Intensity event X 5%
Sum of population affected by Medium Intensity event X 15%
Sum of population affected by High Intensity event X 25%
Estimate of Meals/Day production & distribution capability required
X + Y + Z

These percentages are a crude approximation but they are based on meal counts we collected during the 2004-2005 hurricane season. Experienced mass care practitioners have some version of this algorithm in their head. The result of this estimate process is not the final answer but the starting point of the discussion on the daily state mass care conference call.

The state and the voluntary agencies reach a consensus on the meals/day disaster feeding requirement during the daily state mass care conference not later than 24 hours before a Notice Event or not later than 12 hours after a No-Notice Event. This consensus meal estimate becomes a very powerful tool. Once we agree upon what is required, then we can look at what is available to meet that requirement.

If the requirement is 100,000 meals/day and we have 5 Type II Field Kitchen Units available then we know that we have sufficient production capacity to meet the requirement. Furthermore, if we assume that a Food Service Delivery Vehicle can distribute, on average, 1,000 meals a day then we know that we will need at least 100 vehicles for this event.

As you can see the number of people in shelters and the number of meals that we are feeding each affected person in the community plays no part in my calculations at the state, or macro level. At the micro level, where the Mass Care Chief is preparing a Feeding Plan, those numbers come into play.

Job #1 for the State Mass Care Coordinator is to determine if the voluntary agencies have sufficient resources on hand to meet the consensus meals/day estimate. If the answer is no, then Job #2 is to figure out what is short and go get it. And you better hurry.