Technical Rescue Interface: Search and Rescue in the Non-Snow/Glacier/Mountaineering Environment

Technical Rescue Interface: Search and Rescue in the Non-Snow/Glacier/Mountaineering Environment

Keith Conover

Ryan Circh

Robert J. Koester


In this chapter, we provide an overview of wilderness search and rescue (SAR) for those who:

  • practice or intend to practice wilderness EMS (WEMS),

  • supervise WEMS providers, or

  • are interested in WEMS, and are not trained members of a wilderness SAR team.

The chapter focuses primarily on two aspects of wilderness search and rescue that are not covered fully in other chapters. One is search management. The other is the medical aspect of force protection: providing incidental medical care to SAR team members to keep them functioning, and perhaps doing health screening on SAR team members.

It is difficult to know how many SAR operations occur. Information-gathering is spotty at best. The National Park Service keeps reliable statistics, and they show about 3,000 SAR operations per year.1,2 However, most SAR incidents probably occur outside of national parks: in national forests, in state parks, on other public lands, and on private lands. Some of these are straightforward rescues without much searching, but a significant proportion involves at least some searching.

The standard model of emergency services training these days, at least in the United States, tends to follow a four-level training ladder, likely originating in the regulations and four training levels established for handling hazardous materials by the U.S. Occupational Health and Safety Administration. The following is an informal interpretation of these levels as applied to other emergency services specialties.

Awareness: you know enough to recognize the hazards, know enough not get yourself killed, and know when to call for expert assistance.

Operations: you know enough to complete simple operations in the specialty, if you are supervised by someone with more experience and training.

Technician: you know enough to complete simple operations without supervision, and to participate in complex operations supervised by someone with more experience and training.

Specialist: you know enough to run even complex operations.

This chapter reviews the awareness level of wilderness search management, participation with search operations in the field, tools to interface with the leaders of SAR teams, and concepts of medical force protection.


The English language, as with the Internet, grows without top-level supervision. It’s messy. New terms emerge, old terms acquire new meanings, and sometimes terms have multiple meanings. And like any specialty, SAR has its own special terms. Interfacing with SAR teams is easier when you can “talk the talk.” It also provides you with some credibility with SAR team members,
though perhaps not so much as when you can “walk the walk” and tell stories about all the difficult rescues you have done. (Some embellishment is expected but it must be done artfully and with at least a modicum of modesty and self-deprecation.)

Search and rescue itself is one of those terms that has come to mean many different things.

Looking for people trapped in a burning building? That’s search and rescue.

Looking for live people or dead bodies in collapsed buildings? That’s search and rescue.

Looking for and rescuing a downed pilot behind enemy lines? That’s search and rescue.

Using SCUBA gear to retrieve bodies from a bus that went off a bridge into the bay? That’s search and rescue.

Looking for the wreckage of Malaysia Airlines flight MH370 on the floor of the Indian Ocean using oceanographic sonar? That’s search and rescue.

Looking to see if anyone was affected by widespread flooding or a tornado? That’s search and rescue.

Looking for a hunter who has activated a Personal Location Beacon (PLB) or a commercial Satellite Emergency Notification Device (SEND)? That’s search and rescue.

The difference between “rescue” and “recovery” is critical to understand. In a rescue, the subject is believed to be a patient who will need assistance and potentially medical care. In a recovery, the subject is believed to be a body without chance of survival. Significant risks might be taken to save a life of a patient, but the risk profile of a body recovery operation should be very low.

The Land Search and Rescue Addendum to the National Search and Rescue Supplement to the International Aeronautical and Maritime Search and Rescue Manual Version 1.0 (which, despite the lengthy and impenetrable title, is well worth reading) provides the following definition of SAR.

Search: An operation using available personnel and facilities to locate persons in distress

Rescue: An operation to retrieve persons in distress, provide for their initial medical or other needs, and deliver them to a place of safety.3,*

SAR that best fits the context of WEMS is sometimes called land search and rescue. We distinguish land search and rescue from air search and rescue, which is (mostly) looking for downed aircraft from the air. The problem with this definition is that ground teams (“land search and rescue teams” in some definitions) form a significant portion of the effort to find downed aircraft, and aircraft are sometimes used to look for lost persons (rather than downed aircraft) on the ground.

We also use the term land search and rescue to distinguish it from maritime search and rescue: looking for missing vessels (or aircraft) lost at sea.

While the term land search and rescue has some currency as the main context where WEMS is done, you can make a good argument that the Coast Guard, which some would argue is a maritime SAR agency, also deals with significant amounts of wilderness SAR. In terms of remoteness, difficult coastal terrain, and length of transport to an emergency department (ED), many Coast Guard rescues fit the bill for WEMS, and some Coast Guard personnel have been trained as wilderness EMT (WEMTs) specifically to deal with such issues.

The term land search and rescue is occasionally used, but we most often talk about wilderness search and rescue as the preferred term for the type of SAR that connects most directly to WEMS.

You can argue that U.S. wilderness SAR teams only do a fraction of their work in Congressionally designated wilderness areas, or even state-designated wilderness areas. On the other hand, a lot of wilderness SAR work is in areas that are at least relatively wild, and the term seems to get across the idea better than any other. In addition, as discussed in the Introduction chapter and Chapter 1, the term “wilderness” in the context of medical care is far more expansive than simple governmental designations.

There are three more SAR disciplines that should be distinguished from wilderness SAR.

The term urban search and rescue (USAR) has famously come to be synonymous with searching collapsed buildings and trying to rescue people trapped in them. For the most part, this is not really the type of SAR where WEMS should apply; this is usually in urban areas with somewhat-intact EMS and medical systems. In severe or widespread disasters, though, the existing EMS and medical systems may be entirely disrupted, and you can reasonably call it a WEMS context. If wilderness SAR teams respond to support such operations, which is a reasonable and likely highly effective response, wilderness SAR team members should have extra training for the environment and hazards after such a disaster: the hazards are different, at least in some respects, than the environment for which they train and in which they usually respond.

Sometimes wilderness SAR teams help manage lost-person searches in urban and suburban areas. Some such areas contain big parks that are relatively wild, especially at night or in deep winter or after a bad storm that has toppled many trees. Even if it is in a suburban area without such relatively wild area, we tend to call this urban search (not USAR which is different). Urban
search has its own specific strategies, tactics and hazards, as does a police missing person investigation. It is common for wilderness SAR teams with expertise in search management to assist urban or suburban law enforcement with such an urban search. In some areas, such as the San Francisco Bay Area, some SAR teams do more urban search than wilderness SAR. A text and reference on urban search techniques is available.4

A more comprehensive glossary of SAR terms and acronyms can be found in the Land Search and Rescue Addendum published by the National Search and Rescue Committee. This list is a subset of the more complete glossary found in the National Search and Rescue Supplement (NSS). In addition, a more complete discussion of SAR terms can be found from Selected Inland Search Definitions which is an appendix within Sweep Width Estimation for Ground Search and Rescue.5


Wilderness SAR in the open desert southwest and in the densely forested wet-cold Appalachians might seem very different, but there are many similarities. Wilderness SAR teams can and do take advantage of mechanical devices such as helicopters, boats, four-wheel drive vehicles and all-terrain vehicles (ATVs). See Chapter 28 for further discussion about mechanical vehicle use in WEMS and wilderness SAR. But the primary transport mechanism for most SAR team members on most operations is the human foot, usually encased in a wool sock (SAR team members mostly, and appropriately, despising cotton socks*) and an appropriately sturdy hiking or climbing boot. SAR team members are expected to be able to travel efficiently long distances on foot. The expectations for WEMS personnel may be lower, but SAR team members are generally expected to be able to navigate from point A to point B with flair and élan, regardless of terrain or weather, often using a map, occasionally a compass, but mostly disdaining their GPSs (at least when others can see them). They are expected to be keen-eyed searchers, capable team managers, expert communicators, and survival experts. Misquoting the inscription on the New York Post Office: Neither snow nor rain nor heat nor gloom of night stays these SAR team members from the swift completion of their appointed tasks.

Wilderness SAR teams vary widely in their size and capabilities. A few teams are just search teams… they find lost people, but do not provide any first aid, medical care or rescue. Mostly these are teams that field air-scenting or trailing dogs, though most teams that have such dogs also provide at least first-aid-level care and do some rescue. Some teams provide a full range of SAR services, including technical cave and mountain rescue. Some states offer certification of team by certain minimum requirements, which provides some assurance of quality. Sometimes this certification is by a state agency, sometimes it is by a statewide association of SAR teams. In North America, the generally accepted highest level of wilderness SAR team competence is that provided by the Mountain Rescue Association (MRA).


SAR resources (things, people, or animals that can search: planes, trains, and automobiles, as well as horses, dogs, humans, helicopters, drones, and the like) can use different strategies. A strategy can be carried out using different tactics, depending upon the task’s specific requirements. The strategy of confinement ensures that the subject does not leave the search area unbeknownst (it has happened). Attraction is a strategy for mobile responsive subjects who will move toward a noise or light source. Investigation collects additional information or sightings about the missing subject. Hasty searches follow well-defined linear features, a known route, or go to specific spots where the subject might be located. Area searches cover larger areas with either multiple resources or a single resource following a well-defined search pattern.

Most SAR tasks are designated to use one of these strategies, using resources such as human ground searchers, dogs, mounted teams, ATVs, snowmobiles, or mountain bikes. Man-trackers and tracking/trailing dogs try to follow the subject by visible tracks or scent. Man-trackers sometimes learn the subject’s direction of travel, or document clues, or do cutting for sign (described later). Aeronautical resources (low-flying light aircraft, helicopters, or drones) can cover an area using different search patterns. They can do a hasty search by following a known route, search specific linear features or likely crash sites (such as where the flight path crosses a mountain range), or use electronic equipment to search for radio or radio-beacon signals.


There are various human being search tactics: techniques for looking (or sniffing) for a lost person, or looking for clues to the lost person’s whereabouts. Some textbooks classify human search tactics as Type I (emphasizes speed more than thoroughness),
Type II (a balance of speed and thoroughness), and Type III (emphasizes thoroughness more than speed). Most SAR people, though, rely on the roughly equivalent terms hasty, sweep, and line (or saturation) for tasks.

Early in a search, especially when searching for a likely responsive subject, it makes sense to use available resources for less-thorough but more widespread searching, using hasty and sweep tasks. With a wide-spaced sweep task, searchers may be spaced far beyond their visual sweep width for detecting an unresponsive subject and certainly for small clues. However, they likely have much larger and overlapping sweep widths for hearing a responsive subject.

In the past, it was taught that repeated non-thorough (eg, sweep) tasks were more effective than a single line/saturation task with the same amount of searchers and searcher effort. This was based on a mathematical model that has since been shown to be incorrect.6

For an aircraft searching a segment, it therefore is best to do a single pass over the segment with close track spacing instead of multiple passes over the segment with wide track spacing. Applying this finding to ground search is a bit trickier, however. Close-spaced human-searcher saturation or line search tasks are usually done by large teams that have high operational friction. Operational friction consists of those things that suck up time and effort, or otherwise impede operations, but do not contribute directly to the search effort.

Convoys move at the same speed as the slowest vehicle, and the more vehicles in a convoy, the more likely you will have a slow vehicle. If a vehicle needs to stop for gas or some other reason, the entire convoy needs to stop, and the more vehicles in a convoy, the more likely a vehicle will need to stop. Even in this day of ubiquitous GPS apps on smartphones, dividing up a convoy still seems to cause major complications and is best avoided.

Hiking groups move at the same speed as the slowest hiker, and the more hikers in a group, the more likely you will have a slow hiker. If a hiker needs to stop to retie a boot, the entire group must stop, as breaking up a hiking group is even worse than breaking up a convoy. And since saturation/line search teams are basically large synchronized-hiking groups, this applies to them as well. Large saturation/line search tasks have other sources of operational friction, such as parts of the line drifting downhill, so that the leader must call a halt and move searchers back and forth to re-dress the line.

The higher operational friction of line searches might mean that, unlike aircraft searches, repeated sweep searches actually might be a more effective use of searchers compared with a line search. Until someone does a comparative study, carefully not controlling for operational friction, we won’t know for sure. Even if most search managers don’t believe that repeated sweep searches are better than a line search, sometimes a repeated sweep is appropriate. If a segment has already been searched by a sweep search, but a new clue makes it much more likely that the subject is in that segment, it may be appropriate to get another sweep task into that area quickly, as a sweep task is usually quicker to dispatch into the field than a line search task.

Trailing and air-scenting are tactics which work best if you have a long nose, pointy ears and four feet, and we will discuss search dogs in the next section. If you’re a dog, you may consider the human sense of smell laughable. But there are searches where the smell of fire or aviation fuel led human searchers to a small-aircraft crash site, which leads to advice to human searchers to use as many senses as you can to search for clues: vision (including checking out suspicious clumps of brush, and from time to time turning around and looking with a different view, and even looking up in trees), hearing (“JAKE, CAN YOU HEAR ME?! JAKE?!” [then stop and listen intently]), and smell.

A hasty search task is often sent to search along a linear feature such as a trail or a stream. Another type of search, used either after hasty search tasks or sometimes at the same time, is searching an assigned area rather than a linear feature. This can be with an air-scenting dog, zig-zagging through the area. It can also be with a team of humans in a line traversing the area, sometimes called area search. When the humans are very widely spaced, we call this a sweep search; when close-spaced, we call this line search or saturation search. Sometimes hasty search and sweep search are combined; a linear feature can be searched with flankers out to either side of the linear feature looking for clues as well as a responsive subject, though this slows down the team and may delay them in finding a responsive subject along the trail.

Search resources (field teams) vary in their ability to find clues. An air-scenting dog and handler can rapidly search an area and find, or exclude the possibility of finding, a human being in that area. A sweep task with human searchers, though slower, is much more likely to find clues, such as tracks that can be identified as the subject’s, or something left by the subject.

One of the authors once found what are arguably the two best clues of which we have heard, both on the same task, off the Appalachian Trail in a ravine in Virginia’s Blue Ridge Mountains. First, a plastic bag of clothes with the subject’s name on tapes sewn into each item. Second, after man-tracking from that point (see below) and calling out the subject’s name, a response of “I’m over here, dammit!”

Man-tracking (usually just shortened to tracking) is a technique that has long been used in law enforcement. It probably started by using guides skilled at tracking wild game applying their skills to track humans. Man-tracking was introduced to SAR teams in the 1970s by those such as the late Ab Taylor of the U.S. Border Patrol. The Border Patrol uses man-tracking to locate illegal immigrants, but Ab also put his skills to work to find lost children, and brought these skills to the attention of SAR teams, developing a cadre of SAR tracking instructors.

Teaching searchers how to search for, identify, protect, and follow human tracks is now part of the training of most wilderness
SAR teams. Trained searchers are expected to be clue-conscious: to know how to identify human tracks and appreciate their value as clues, especially in untracked wild areas, and to protect them. One of the authors, searching such an untracked area, found a track crossing perpendicular to his assigned hasty task, going from north to south. This directed the search strategy to the area south of his assigned search task, where another team quickly found the lost subject, a 92-year-old woman who had been mushroom-hunting and had fallen and gotten her leg trapped between two rocks. She had been stranded there for days, but luckily was right next to a small stream with water. This points out how a single track can serve as a clue and result in a save.

Searchers are sometimes tasked to cut for sign (also known as sign-cutting). This means to search, either in circles around a clue, or perhaps perpendicular to the subject’s projected line of travel, looking for tracks (“sign”).

Some SAR team members go on to advanced training in man-tracking, and may be dispatched to a potential track to start tracking at that point, using the step-by-step method taught by Ab Taylor and others. Man-trackers may start at the Point Last Seen (PLS), or if a good clue establishes it, the Last Known Position (LKP), but often investigators have trampled the tracks there. Scent-specific trailing-dog tasks are sometimes used from the PLS instead, though with frustratingly low rates of success.


Many animals have highly refined senses of smell, and could theoretically be used for searching—in particular, pigs and buzzards seem to feature frequently in SAR humor—and horses used by mounted teams have a keen sense of smell compared to humans, which adds to their baseline usefulness as mounts for humans. But the animal most used for lost-person search is “man’s best friend,” the dog. There are arguments about which breed of dog is best for SAR, but this is best left for informal discussion (probably both lengthy and heated) with a group of knowledgeable dog handlers, as there is no consensus even as to whether one breed is best, much less which breed. Search dogs can be highly effective at finding those lost in a wild area.

As with the term search and rescue, the term dog team can be more than a bit confusing when used in conversation, and we know of four separate meanings of the term. On the one paw, a dog team can be a wilderness search organization, all of whom are dog handlers. On the other paw, a dog team can also be a search organization, only some of who are dog handlers, although this more commonly is called a wilderness SAR team with dogs. On the third paw, a dog team can be a team sent out on a search task, consisting of a handler and dog working together, along with one or a few other humans who are called walkers or flankers: SAR team members who accompany the handler stay well back and often handle communications and some navigation. The usual term for such a group heading out for a search or rescue is field team. But the teams that are specifically tasked to use a dog, usually for an air-scenting or trailing task, are sometimes informally called a dog team to distinguish them from the other, non-dog teams, which are called just field teams.

On the fourth and final paw, we have a dog team of one human handler and one dog who have trained together, tested together, and have been credentialed as competent in a specialty such as air-scenting, trailing, or human remains detection (HRD). It takes lots of time and dedication, on the part of both the dog and the handler, to become a competent and credentialed dog team. There are quite a variety of credentialing agencies for such search dog teams, with different testing standards; if you ask a dog handler about them, you will probably get a strong response about which are of high quality and which are not. But, as a dog handler in the Appalachian Search and Rescue Conference (ASRC) once famously observed: “If you get three dog handlers together, about the only thing you’ll get two of them to agree on is that the third one is wrong.”

Dogs have different training and capabilities. There are a variety of dog specialties, including water search (searching from a boat), cadaver/human remains detection (HRD), collapsed-structure search, avalanche search, and evidence search. The two most commonly used in wilderness SAR are air-scenting and trailing.

First, let us describe how to do an air-scenting task. To make it easier to appreciate, we will describe this from the dog’s view.

Your human handler and the other humans will usually follow a trail, a stream, or perhaps steer a fairly straight course through the middle of an assigned search area (SAR teams often call this a segment). You should stay ahead of the humans; stay close enough that you can hear them, but being out of sight is OK, at least for brief periods. While they are struggling along behind you (humans can be quite slow in the woods), you should run back and forth ahead of them, sniffing carefully for the distinctive scent of a human, any human. As any competent dog knows, individual animals (including humans) all have a slightly different scent, but animals have a distinctive species-specific smell. It is said that foxes are particularly sharp and acidic, whereas humans are warm and complex with overtones of oak and cedar, especially if the human has been eating beef, and often a yeasty finish if the human has been eating bread or drinking beer, but perhaps this is just one dog’s interpretation. This scent is created by small bits of skin, hair, and evaporating skin oils that animals give off. This material floats downwind, spreading as it goes, in a cone of scent. When you are air-scenting, keep your nose up and sniff periodically. Ignore the scent of the humans with you, but keep sniffing for a different human.

When you are air-scenting, you are just sniffing for an unexpected human scent, any unexpected human scent. As soon as you scent any human other than your team, check the wind direction and remember it. When training your human, you
should have worked out a standard way to communicate this “alert”; whatever it is, run back to your human (for some reason, an alert never happens when you are right next to your handler) and give your alert signal, whatever it is. Once you are sure your human has paid attention and acknowledged your alert, it’s time to head out and try to catch that scent again. Winds shift, so you will usually have to range back and forth until you can smell it again. And sometimes, you won’t find it again; c’est la vie. Still, even a single alert can be useful to those back at Base who are plotting these things on a map. If you are lucky, you will get another noseful of that same scent, at which time your job is to follow that scent upwind until you find the search subject. If the wind shifts, you may need to range back and forth a bit more to pick it up. It is important to remember that old search-dog mantra: humans are slow. While there is a certain competitive urge to get to the subject as fast as possible, you may need to slow down a bit so your humans don’t get out of barking range.

When you find a search subject, you need to communicate this with your handler who, as usual, is probably lagging far behind. Run back to your handler, give the signal that you have taught your handler, get a response that you have been understood (“Show me!” seems to be pretty standard) and then lead your handler back to the subject. This is called a refind.

With air-scenting, there is lots of scent in the air, at least when you get close. But for trailing, you have got your nose down near the ground, trying to find some of that scent that has drifted down onto the ground. That makes it harder, as there is less scent, and the older the trail the less scent is left; sometimes they have you try to follow trails that are a couple of days old, which is well-nigh impossible. What is worse is that you have to pick out the right person’s trail from other people’s scent trails; unlike air-scenting, trailing is scent-specific. If you are lucky, your handler will have a good scent article in a paper or plastic bag for you to check from time to time. Ideally this is from someone who has been trained how to collect a good scent article without contaminating it, but you will have to work with whatever you have got.

As discussed above, man-tracking is a well-trained human visually following someone’s footprints or other signs of passage; do not confuse it with a dog’s trailing.*


When searching for a lost person in a wilderness area, searching may be complicated by the fact that the subject may still be moving, resulting in an ever-expanding search area. Thus, we arrive at the key concept of containment: knowing if the subject leaves the established search area.

There are many tactics that can help contain the search area. You could:

  • Post notes at trail intersections “this way out.”

  • Run string lines with flagging tape on them, and small signs saying “this way out.”

  • Put camp-ins (a couple of searchers camping out) at locations where a lost person might reasonably end up, such as a major trail junction, or the main approach to a mountain climb. A camp-in team may carry in a tent, sleeping bag, sleeping pads, a stove and food, and may serve as a rest and resupply area for more mobile field teams.

  • Create a “track trap” in an area where a mobile subject might travel: sweep an area of mud, dirt or dust flat, so it is ready to accept good tracks, and then send teams to check the track trap on a regular basis.

  • Have searchers do slow patrols along roads around the area from a vehicle. Have searchers similarly walk to patrol trails that bound the area. These tasks may be particularly useful to get less-trained and less-fit searchers into the search effort.

However, patrolling or searching by vehicle is not nearly as sensitive for either clues or subjects as foot-based searchers. Once upon a time, in a large wilderness area traversed by the Appalachian Trail in southwestern Virginia, both foot searchers and trail-bike motorcycles looked for the subject for many days. When finally found by the foot searchers after almost a week, the subject commented, “the only time I was afraid for my life was once when I almost got run over by a motorcycle.”


For many years, researchers have worked to get search management into a more scientific framework. This has resulted in a fair amount of literature, and several computer programs designed to assist search managers. Here we will review only the most prominent aspects. If you are interested in more details you can consult the literature on this topic.5,7,8,9 In particular, a rigorous but very readable introduction to search theory, prepared for the U.S. Coast Guard by J.R. Frost, is available free online.10,

The central equation of search theory is:


POA is Probability of Area: the probability* that the subject is in that circumscribed area. POD is Probability of Detection: how likely the search technique will find the subject if the subject indeed is in that area. Multiply the two, and you get the POS or Probability of Success: the probability that you will find the missing subject.

With this equation, we try to quantify, and then combine, two uncertainties. The first uncertainty is probability that the subject is in a particular search area segment (Probability of Area = POA). The second uncertainty is the probability that your search tactic will find the subject if the subject is in the area searched (Probability of Detection = POD).

If you are 100% certain the subject is in the area (POA = 100%), and you search it with a tactic that never misses a subject (100% POD), then you have 100% × 100% = 100% chance that you will find the subject (100% POS). Note that the math is easier if you do it using probability rather than percentage. A probability of 100% is the same as a probability of 1.0; in this case, 1.0 × 1.0 = 1.0.

If you are 50% certain (probability 0.5) that the subject is in the area (50% Probability of Area), and search it with a tactic with a 50% Probability of Detection (probability 0.5), then multiplying them together (0.5 × 0.5 = 0.25) gets you a 25% Probability of Success.

The goal of search theory is to find the subject in the shortest amount of time. The most powerful SAR tactical decision aids calculate the probability of success rate (PSR), which is a measure of how effectively you are using your available resources to find the subject. The aid then tells you how to allocate your resources appropriately to maximize it.3

Defining the Search Area

The first step in planning the search is to plot an Initial Planning Point (IPP), using either the PLS (point last seen, where the subject was last seen by a human observer) or the LKP (last known position, a position established by a reliable clue). The next step is to define the overall search area: What will I search, and where will I send some (or no) resources? Too small an area, and you may miss the subject. Too large, and you may never be able to finish searching the area with your available resources. Textbooks traditionally describe establishing the search area through a four-step process of Theoretical, Statistical, Subjective, and Deductive.11 Actual practice tends to involve your looking up the 95% distance the subject is likely to travel from the IPP, based upon statistical models.12 Then, you reduce the search area where there are obvious travel barriers (eg, an impassible river). Finally, you match the boundary of the search area to features a field team could find on the ground.

We mentioned containment above, but there are many times when containment is not a major factor for search planning. When a person has been missing for several days, the possible travel distance makes the potential search area huge, usually orders of magnitude larger than could be covered by available resources. In such cases, the actual search area is much smaller than this theoretical area.

To aid us in deciding how large an area to search, and which areas within that we should search first, we can use data on lost person behavior. Robert Koester, a long-time search manager of the ASRC and one of this chapter’s authors, also authored a book called Lost Person Behavior that provides this information, which is also available in a smartphone app.12 For example, if you are searching for a lost hiker, you can consult Lost Person Behavior, which recommends concentrating on trails, cutting for sign around decision points: points where the trail route is unclear and a hiker might go astray. Cutting for sign means traveling carefully around the decision point, searching intently for human tracks. If we look at the statistics in Lost Person Behavior, we find that 50% of hikers are found within 1.9 miles (3 km) of the IPP, and 52% were found downhill from the IPP. Thus, even if the subject cannot be contained, search efforts can be focused on the most probable areas.

Probability of Area

Search theory rests on the premise that, while the location of the subject is unknown, some areas are more likely to contain the subject than others. Much like looking for your lost keys, they are more likely to be in some specific locations than others (coat pockets?) The actual probability for each area ranges from near zero to approaching one. You can use three main methods to determine the initial POA.

For decades, the traditional land SAR method has been the Mattson consensus method.13 This is based on information from your investigations, and is a consensus of subject matter experts you gather together, calculated mathematically. The Mattson consensus may also include information from the other two methods.

The second method is a statistical method, also known as the stochastic approach. It takes various models (or a single statistical model) of where people (or aircraft, or ships) tend to be found, typically calculated from an IPP. Figures 30.10, 30.11, 30.12, 30.13, 30.14 provide examples of this approach. Wherever the subject was last seen by a human (seeing them on live video or on a time-stamped video recording counts) is the PLS. Wherever the subject can last be located (for example, by a good clue) is the LKP. The point first chosen as the starting point of the search, whether it is the PLS or an LKP, is the IPP. Segments of the search area are then assigned POA. This is the model most commonly used to look for missing aircraft based upon track information.
As mentioned above, one of the authors once found, deep in a ravine, a plastic bag of clothes with the subject’s name on name tapes sewed into each. The PLS was back, up on a ridge along the Appalachian Trail. This clue reliably established a new LKP, and it refocused search efforts.

The final model is a particle motion or Markov model. This is the model used by the U.S. Coast Guard; it considers how the subject may move due to wind and currents in the ocean.14 A particle motion model creates a mathematical set of rules defining how a particle moves, and essentially rolls the dice (introduces probability) for each discrete move. You then run a Monte Carlo simulation on hundreds or thousands of particles. Then, using specified time parameters, where the particles end up define the probable locations. This particle motion technique is seldom used to predict the location of missing people on land.

Some computer programs allow you to combine these techniques to calculate one composite POA for each of your search area planning regions or segments.

Segmenting the Search Area

When planning area searches (air-scenting, sweep or line/saturation tasks), search managers generally segment (using the word segment as a verb) the area into small, searchable segments (using the word segment now as a noun). A rule of thumb—actually the rule of two thumbs—says that on a standard-scale USGS topographic map, the area covered by your two thumbprints is about the right size for an air-scenting dog or human search task. A field team can usually search such a segment in 4 to 6 hours, and can usually complete two tasks in a 12-hour shift. Linear features may also be assigned a segment number for purposes of planning hasty search tasks. Some method is then used to assign a POA to each segment, and you generally send teams into the segments those with the highest POA first. Segmentation is an art taught to SAR managers; the segments must not only be of a reasonably searchable size, but must have boundaries that can be well seen on both the map and in the field (Figure 30.1).

More terminology: search area generally refers to the entire area currently being searched, and searchable segment (using the word again as a noun) usually refers to a small portion of the search area, assigned to a particular field team to search during a search task. Sometimes people to refer to a small segment as a search area; refer to the context to figure out the usage.

There is one “segment,” or better, region that is not on the map, and is referred to by the acronym ROW (Rest of World). Searching particularly high-probability locations in the ROW, particularly nearby bars, is a part of some searches and has given rise to the common term bastard search, though a much more politically correct term, especially if you are dealing with a subject with dementia, is investigative search.

The Mattson Consensus

The original method for assigning land search POA is the Mattson Consensus Method, named after the U.S. Air Force’s Robert J. Mattson, who taught this method at the joint U.S. Air Force—U.S. Coast Guard National Search and Rescue School in the 1970s.13

The original, basic, pencil-and-paper method works as follows. First, choose a small number of people who have some sort of qualifications to provide an educated guess as to where the subject might be. (The guesses of psychics are usually classed as “uneducated” and not included.) Things that might lead to an educated guess include:

  • specific knowledge about the search subject;

  • experience at running searches;

  • knowledge of the local terrain, including such things as popular hiking trails, good fishing streams, or favorite hunting areas;

  • thorough knowledge of lost person behavior; or perhaps

  • training in using Geographic Information System (GIS) to, for example, predict travel times on and off trails.

Next, divide the search area, not into searchable segments, but into planning regions, and assign each region a letter. These regions may be larger (or smaller) than searchable segments, as they are for assigning POA, not for creating specific tasks.

Then, using a pencil and paper (see Figure 30.2), make a list of the search region letters down the left side of the paper, then a draw a grid next to this. The grid needs a horizontal row for each planning region, plus an extra one for ROW.* It also needs a vertical column for each of the people who will be contributing their thoughts; you can call them your Mattsoners. Add another column at the far right for the averages.

Have each Mattsoner assign a POA to each region, including the ROW “region.” For a traditional Mattson, each Mattsoner must be capable of some mental math: the total POA for all regions, including the ROW, must add up to 100%. (This is why computer programs are so popular for doing this.) Mattson recommended that all the Mattsoners use a separate sheet of paper, and list their percentages privately. This avoids peer-pressure effects that might dilute the wisdom of this particular crowd. Then, have someone enter the values in the grid illustrated in Figure 30.2 and do the calculations.

Finally, it is a simple matter (at least if you are a computer) to average all the readings. You use the averaged POA for each of the regions to direct your search strategy: search the regions with the highest POA first. Given the results of the Mattson Consensus in Figure 30.2, and the reality that at the time of this consensus there were just three field teams currently available, those teams should be assigned to Regions A, B, and C. If the
first set of planning regions corresponds with the segments on the map in Figure 30.1, A=1, B=2, C=3 and so forth, then send teams to segments 1, 2 and 3.

FIGURE 30.1. Segmenting a Search Area. Initial segments for a man who “…went up the holler to do a bit a huntin’ on Calf Mountain.” Illustration by Keith Conover, MD, FACEP. Used with permission.

There are issues with the classical Mattson method. Mattsoners will sometimes give you a set of probabilities that add up, not to 100%, but to 90% or 120%. Scaling these entries so they do total to 100% is sometimes called “coherentizing” the entries. Or, Mattsoners assign probabilities for a few of the more likely segments, then simply provide a similar low probability for all the rest (cheating).

Only gold members can continue reading. Log In or Register to continue

Oct 16, 2018 | Posted by in EMERGENCY MEDICINE | Comments Off on Technical Rescue Interface: Search and Rescue in the Non-Snow/Glacier/Mountaineering Environment

Full access? Get Clinical Tree

Get Clinical Tree app for offline access