Probability of Detection (POD) - PART 2
by John Mindock
This article originally appeared in Volume 2, Issue 10 of Lost... and Found.
This article was first published in October of 1997, and definitely contains information that does not reflect the current state of the art of search management. Please be sure to read the editorial notes that have been added since its original publication.
The previous mini-lesson (POD - PART 1) discussed the concept of POD and some of the factors involved in calculating it. This mini-lesson will continue with that discussion.
Probability of Area (POA)
This is an estimate of the probability that the subject is within a specific area. The total of the POA's of all areas being considered for searching must equal 100%. In order to cover all possibilities, a search segment known as `Rest of the World' (ROW) is also declared. This makes allowance for the situation where he might not be in the areas being considered (home, in the bar, at his girlfriend's, out of the historically-indicated search range, etc.) The POA of an area is estimated by the Incident Management, and an area's POA can change as a result of POD's reported by returning field teams.
As a simplified example, a fisherman is more likely to be near the stream than on top of the hill, so the POA's of areas near the stream would be higher than the ones near the top of the hill. However, after the areas adjoining the stream have been searched to high Unresponsive POD's, their POA's may be lowered and the top of the hill area's POA increased.
In theory, teams assignments correspond to the POA of the areas, with highest POA areas covered first. In reality, Incident Management does not have enough time to chitchat about POA's in the `initial attack' phase of a mission, so those assignments are made on intuition and history rather than a formal POA strategy. Later assignments might be more explicitly based on POA considerations.
Probability of Coverage (POC)
[Ed. Note added 27 Feb 2006: The "POC" referenced here was defined and used in a 1996 article in the NASAR journal "RESPONSE", and its usage was being encouraged in New Mexico SAR management at the time this article was written. Unfortunately this usage is in conflict with the usage of the acronym "POC" in all established literature on search theory since World War II, in which "POC" is used for "Probability of Containment", a synonym for "Probability of Area". The usage described below is not standard and is no longer in use by anyone, anywhere. Please refer to our more recent article on Search Theory and the references therein.]This is a predetermined set of values for Unresponsive POD, based on the factors that influence POD. It attempts to constrain the incoming POD's from field teams within the realm of reality. A simple illustration: a team of two searchers covering a six-square mile `difficult terrain' area in two hours in snowy nighttime weather should equate to a low POD.
If the team returns with a much higher POD, Incident Management can supersede their POD estimate with that prescribed by the POC tables. Another application of POC would be to tell a team of four people to search an area of four square miles, and do it for six hours. Using the POC tables, this would theoretically ensure that the team got a POD of `y' %, which might be more legally defensible than any subjective estimate the team might provide on their own.
When team members are in the area search mode, they spread out from each other. There are two formal designations for the type of separation - `visual' and `critical'. Visual separation means that the team members generally can see the person on either side of them. Critical separation means that they can generally see some midpoint between them, but do not attempt to stay in visual contact with each other.
Unless Incident Management specifies otherwise, visual separation is the tactic expected of teams in an area search mode. POD for visual separation would usually be larger than that for critical separation.
Terrain Types (easy, moderate, difficult)
One of the factors in POD is the type of terrain. In Hiking Guides, this usually is a function of the steepness and altitude. But in the context of POD, it refers to the difficulty of seeing every place where a person could be concealed. So `easy' terrain might be a grassy field, `moderate' a pinyon/juniper foothills area, and `difficult' a canyon filled with downed timber, boulders, etc. But a grassy, albeit steep, slope could also be `easy' in the context of POD.
Rate of Progress
Another factor in POD is rate of progress. Although there are no mandatory guidelines, one could generally expect a hasty team to progress at least two miles per hour, while an area search team should be no faster than one mph (even less on difficult terrain).
Estimating Responsive POD
There are many subjective factors in estimating responsive POD. The following are some general characteristics that may be used as parameters to devise that estimate.
Low (0 - 25%):
- not able to use standard attraction techniques (whistles, yells, etc.);
- high winds, blizzard, etc.;
- noisy running water;
- densely vegetated/treed areas;
- large areas;
- daylight search (sound doesn't carry as well).
- standard attraction techniques used somewhat often;
- moderate winds, rain, moderate snow;
- not much noisy running water;
- somewhat dense vegetation/trees;
- small canyons, small flat areas so searchers attraction noises carry somewhat well;
- evening search (sound carries better).
- standard attraction techniques used very often;
- little wind, rain, snow;
- no running water;
- non-dense vegetation/trees;
- large canyons, tops of ridges, places where sound carries well;
- night search (sound carries well at night).
There are many more factors in this estimate (See previous mini-lesson).
Low (0 - 25%)
- rate of progress greater than two mph;
- few searchers in a large area;
- difficult terrain;
- subject wearing camo/green, etc.;
- bad weather;
- efficient (as opposed to thorough) search tactics.
- rate of progress between one mph and two mph;
- adequate number of searchers to cover the area;
- moderate terrain;
- adequate light;
- subject wearing easily-seen clothing;
- decent weather;
- somewhat thorough search tactics.
High (70 - 90%)
- rate of progress less than one mph;
- more than enough searchers to cover the area;
- easy terrain;
- bright daylight;
- subject wearing bright clothing,;
- very nice weather;
- very thorough search tactics.
This exercise is intended to give one a `feeling' for the rate of progress that 20% POD and 80% POD searches entail. Find an area where there are many boulders, trees, and/or bushes. Search the area looking behind only every fifth obstacle. Record your time for this, a 20% POD search. Now return the other way, looking behind four-out-of-five obstacles. Record the time for this, an 80% POD search. In theory, it should take @4 times as long as the 20% approach. (Actually, since parts of your search area are visible without looking behind an obstacle, the POD's would be somewhat higher than 20% and 80%.)
Self-quiz on POD - PART 2
- What is POA?
- How can an area's POA change during a mission?
- How might POA be used in determining team assignments?
- What is POC?
- How might POC be used to predict POD?
- Compare visual vs. critical separation.
- Give examples of easy, moderate, and difficult `terrain type' in the context of POD?
- What rate of progress is generally expected from a hasty team?
- What are some situations that might lead to a low Responsive POD?
- What are some factors that might lead to a high Unresponsive POD?
Copyright and Disclaimer:: The contents of this website and the newsletter contained here are copyright © 1996-2006 by their respective authors or by Cibola Search and Rescue, Inc., and individual articles represent the opinions of the author. Cibola SAR makes no representation, express or implied, with regard to the accuracy of the information contained in these articles, and cannot accept any legal responsibility or liability for any errors or omissions that may be made. Articles made available at this website may be reproduced, with attribution given to Cibola SAR and the author, by any member of the Search and Rescue community for use in other team's publications.