Marian Hardy
Mid-Atlantic D.O.G.S.
May 1992
May '92 - NASAR Conference


The results of the first analysis of the National Water Search Report is presented along with a discussion of how the SAR dogs can be deployed, the types of water that can be searched, the SAR dog team's limitations, where to find SAR dog teams and some examples of SAR dog teams and divers working together to detect and locate drowned victims.

In searches on land we have found that not just one search resource holds all the answers, to solve the mystery of where the victim is located. The skills and experience of many types of resources are needed. Most times it is the coordinated effort of the search that finds the victim, not a single person's or one group's "superior skill". For instance, in the past 10 years in the Mid-Atlantic States, the various land search resources have been not only working together on searches more and more, but the various groups have been training together periodically in order to better understand and coordinate their respective skills.

In water searches, no one resource holds all the answers either. There are, however, some added complications. A water search is a three dimensional problem being conducted under adverse conditions for the searchers. There are depth, current, snags, temperature, and various states of "blindness" for either the diver or the dragging crews.

The tools and techniques available to searchers to find someone under water are increasing in type and in numbers. For instance, there are in addition to the dragging tongs, new techniques being developed and employed by the public service divers and the swift water rescue people. There are dog teams trained to detect human scent coming from under water. There are under water TV cameras and sonar hooked up to recorders to make visual records of an under water search area. And there are the LORAN-C and the Geodetic Positioning System (GPS) to determine position in open or relatively open water so that you can relocate a search area.

My objectives today are to present the capabilities of the search dog teams, to give you some incite as to what to look for in a water search dog team, to suggest strategies for using the teams and to invite you to get acquainted with your nearest SAR Dog Unit who does water search.

For those of you who are new to water search with dogs - a little background. In the early '80s, the interest in using Search and Rescue (SAR) dogs to detect victims under water began to spread across the United States. A few SAR dog units and individuals had already become aware that their dogs were alerting on victims in watery graves, but the concept was neither widely accepted, understood nor practiced.

One problem was that people did not believe that dogs could find someone under water. After all, "everyone knows" that if you are a bad guy, the best way to give the police dog the slip is to walk through water ! The other part of the problem was that, when a dog detects a submerged victim, it was an "indirect find" -- someone else still had to physically locate and recover the body, and that recovery might not happen until a later date - either by diving or dragging, or the body might surface by itself. In any case, the relationship of the dog's alert to the ultimate recovery was overlooked -- and the myth perpetuated..

A few of us concerned about the use of dogs in water search developed the NATIONAL WATER SEARCH REPORT in 1984 (Appendix A), as a mechanism to systematically collect data about water searches throughout the SAR dog community. You will note that the report is interested in three time frames - the incident, the search (with dogs) and the recovery.

The objective was, and still is, to document the fact that dogs have been responsible for detecting drowned victims and to identify some basic rules-of-thumb that might help dog handlers and water searchers to identify the most likely areas to be searched in certain circumstances - but more about that later.

In 1988, a compilation and analysis of the National Water Search Reports was published and a second iteration is currently in process.

The 1988 Study was based on 122 water search reports sent to NASAR from SAR dog units around the country. Twenty-six different SAR dog units were involved, sometimes cooperating in multi-dog unit searches. Of the 130 victims involved, 84 were found by dogs, 24 were recovered out of the area searched and 22 have not yet been recovered, to my knowledge. Of the 22 victims not recovered or found, the dog alerts in 9 instances could not be followed up by divers or draggers because the location was too hazardous to the divers, too deep (150 feet) or, in the case of flooded valleys, the remaining trees, buildings and bridges underwater were not suitable for either dragging or diving. The bottom line is that dogs can detect a person under water, and I'd like to discuss how it works, how you can deploy dogs, the types of water that can be searched, the SAR dog team's limitations, where to find SAR dogs and give you some documented examples of SAR dog teams and divers working together to detect and locate drowned victims.


Although a lot of scientific work has been done on scent, little is still actually known about the true mechanism as it relates to the olfactory ability of the dog. This is especially true in the detection of a drowned victim.

One thing we do know is -- that it works.!
For those who would like an explanation, we have a theory which is based on some reasonable assumptions about what scent is and on two laws of chemistry - the Law of Diffusion and the Phenomenon of Evaporation.

A person gives off scent all the time - whether they are on the surface or under water, under ground or under rubble. We are all like "Pig Pen" in the comics with a little cloud of dust (or scent) around us all the time.

If we assume that scent is composed of either the soluble or insoluble solids, liquids and gases associated with the human body, then the Law of Diffusion tells us that these soluble solids, liquids and gases will dissolve in the surrounding water much like table salt.

As these "scent" molecules from a body diffuse into the water, the lighter ones will diffuse to the surface, where evaporation takes place. Once in the gaseous state, dogs can easily detect the scent on air currents.

Those insoluble molecules lighter than the surrounding water will also rise to the surface and float. These insolubles provide scent to our dogs, too.

This is only a partial explanation, but from observation in the field, we know that human-associated scent comes to the surface or out of the water and is detected by the dogs.

You might wonder how long a body can produce scent sufficient for a dog to detect. Nobody knows exactly, but we have one documented case of a dog detecting a body under water in the Kern River, in California, after 192 days. According to the report, "The body was so tightly wedged up into a cavern in the rocks, that it had to be pried free. The diver noted that, since the body wasn't on the bottom, he might have missed it, had it not been for the dog's alert." During the same search, the dog alerted on human bones which, according to the pathologist, had been in the river for not more than a year. In my opinion, I don't think 192 days is the limit.


Usually the dogs work

In a boat, the dog and handler work from a position usually in the bow. When the dog detects the victim's scent, she alerts or signals the handler by dropping her nose to the water, or she may become frustrated and anxious, bite at the water, dig at the bottom of the boat, whine, bark or may try to jump into the water to reach the source of the scent. As the boat takes the dog out of the scent cone, the dog follows the scent to the back of the boat, if possible, or will just relax as if nothing had happened.

My dog, PC, after detecting human scent from the water will usually look around the boat for a paddle, oar or something to reward herself.

Rechecking the area with the same dog or another and observing their behavior will confirm the alert. It is generally thought that from search to search, the depth, the temperature (air and water), the length of time of submersion, the experience of the dog and handler, the presence of current (surface and otherwise), the weather, the presence of thermoclines, and more all have an effect on the intensity of the dog's alert. The dog will alert in some fashion to the presence of human scent. Its up to the handler to read and interpret that alert - which may be subtle and which may take more than one pass of the area.

If working a shore line, when the dog detects human scent, she will probably enter the water and swim to the area (or point) where the scent is coming from the water. The dog may even swim back to shore to get the handler in a "refind" - meaning "come with me and see what I've found". Usually the handler does not comply.

The scent that rises from the victim to the surface is first acted upon by the currents in the water and then, if conditions are right, the scent evaporates and the air currents take over and a dog can detect the scent from shore or in a boat. You can see why it is important for the handler, the divers and the local fishermen to communicate and together work out the problem of where to look for the victim in terms of the various water currents and wind vectors. The answers are not necessarily obvious.

As an example, the handler should be aware that what you see at the surface, current-wise is not necessarily the same under the surface. A recent search in West Virginia is a case in point. A young man had slipped off one of the cross-river ledges about 15 feet from the shore opposite the National Park at Harpers Ferry. The location of the point last seen (PLS) was well identified by witnesses. In the late afternoon, two dogs alerted mildly at the PLS, but one dog, while returning to the Park side, alerted strongly into the water just off the Park shore and about 40 feet down stream from the incident.

The next morning dive teams started a systematic search from the PLS down stream. Later in the day, while the dive teams were still working, dragging teams found the young man's body just down stream of the dog's strong alerts on the Park side of the river. Divers said that the bottom current was very strong. In addition there must have been a very strong cross current to carry the body across the river and a short down-stream distance.


There are different considerations with respect to each type of water, so they were categorized and grouped during the analysis for the final report in 1988 (Reference 1). Lakes and ponds can often be covered from the shore line. If the body of water is large enough, small boats can be used to grid the area. Any alerts can be marked with a buoy of some kind or observed from shore points so that divers will have a reference point from which to start their underwater search operations.

You will note in the Water Search Fact Sheet (Appendix B), that the rate of success using dogs is at least 84 % in the lake, pond or reservoir situations. You will also note that 68 % of the victims found were recovered by divers (or in some cases by dragging operations) from the bottom. The "live" find noted on figure A was a person up to his shoulders in a lake when the dog air scented him from shore. The victim was stuck in the mud. I might add that, had he been stuck in the mud in a tidal pond or in fast current, the find would not have been a happy one.

Flooded quarries should be searched from a boat - the sides of the quarry are most often be too high, steep or hazardous for a search from shore. Quarries are also usually deep and very cold, which may put divers in jeopardy, especially if they are sports divers and not trained or equipped for deep underwater search and recovery. An example of a flooded quarry search took place near Front Royal, VA. The sheriff thought that a young man might have drowned in an 80 foot deep flooded quarry with submerged caves. Then again, he might have been playing a joke on his friends.

A dog alerted near the entrance to the cave and at the waterline inside the cave, down wind of the entrance. Although the divers found nothing, we suggested that the sheriff run a large motor boat around -- like stirring a bath tub -- and the body might rise. The sheriff did so the next day and the young man's body came up near the entrance where the dog had alerted. Unfortunately he had not been playing a joke.

Rivers, streams and creeks can also be searched from the shores or from boats. One of the "rules of thumb" developed from the review of the NATIONAL WATER SEARCH REPORTS, to date, is related to rivers, streams and creeks in flood. The water conditions at the time of the drowning or incident may well suggest the most likely place to look.

Generally, the local river and rescue people say that the victims are "always found within a mile downstream of the PLS" - it could be 1/2 mile or 3/4 mile, or what ever is appropriate for the particular stream. It has been our experience and that of other dog units across the country that this is true even in a stream or river with normally fast current. Rivers and streams seem to have their own patterns.

However, it is a different ball game when dealing with a river or stream in "torrential flood," particularly when the water is rising. Under these conditions, even a mild-mannered little stream can carry a pickup truck 150 yards downstream, as has happened in Virginia three times in the past few years. In a flood situation when substantial non-buoyant weights such as boulders and trucks are carried by the water flow, a victim doesn't have the weight to overcome the horizontal force of the water flow and sink to the bottom. We have found that the victim may be carried at or just under the surface of the water until something stops it -- which may be 23 miles. In such cases the best places to search are in the area of those features which will change the direction of travel of something being carried by the torrential current. For instance --

In catastrophic flooding, such as in Puerto Rico and in West Virginia in 1985, the forest and fences on either side of the river formed strainers where we found vehicles and houses as well as people. In West Virginia many of the thousands of farm animals and turkeys found there were from other counties up stream.

When looking at the statistics for flooded rivers and streams generated by the water search analysis, at least 42 % of the victims are found outside the area searched by dogs or others. Forty-two percent compared to 16 % (normal flow rivers) and 14 % (lakes) may be significant. The "finds" in the flooded river situations were at 23, 40, 10, 13, 12, 17, 2, 3/4, and 1/2 miles, respectively. A lot of territory to search.

One thing to remember in water searches as well as land searches is that, if you can determine where the victim isn't, you have positive information -- as a result, diving or dragging operations will not have to be conducted in those areas cleared by dog teams. The areas cleared by dogs can be significant when you are dealing with many miles of river or hundreds of acres of a lake or reservoir.

In the past couple of years, we have had some finds on land that were water related and water search experience (and common sense) was helpful.

In this case, Mary R. had wandered off near the Liberty Reservoir, near Baltimore. A search was conducted by the State Police. We were contacted late on day seven of the search. Five and a half search hours later a dog in a canoe on the reservoir, air scented Mary and took her handler to shore and 200 yards into the woods where Mary was found - ALIVE.

In CHARLES COUNTY, MD, The authorities suspected a robbery/murder had been committed -- even had a suspect -- but no body. A few weeks later after getting a tip from an informant, the Sheriff's Dept. asked for our assistance.

The dog's initial alert was toward the water passing through a semi-clogged conduit under the dirt road in the area to be searched. This focused our attention on the flooded area from whence the water was flowing. The find was made by our dogs between the edge of the flooded area and an active dumping area for crab shells.

The dogs were curious and sniffing around the base of the thick growth of weeds along the road. Because of their subsequent actions over the next couple of minutes, I beat down the weeds for them -- not far off the road was the body of the victim.

In CHESTERFIELD CO., VA, a little 10 year old girl went missing in October. An extensive search was conducted at the time - with no results. in April, the following year, we were asked to search an area previously searched by others for a possible buried victim. The report was that a dog had alerted into a pond associated with the area of interest, but dragging found nothing.

We checked the pond and my dog alerted strongly in a limited but general area where the land sloped to the water . The water contained lots of tree roots but was too shallow to hide a body. Meanwhile, my colleague was near by doing a systematic, disciplined grid search of the hillside leading to the pond, when she made the find. The body was buried a foot underground. Apparently the scent, like water, was flowing down hill and collecting in the root filled water.

By the way, the trial of the suspect was conducted. He was convicted and sentenced to the electric chair.


When people ask about the dog's limitations, it is important to reiterate that the dogs are only one aspect of the search, rescue and/or recovery of victims when non-direct finds take place, such as in collapsed buildings and mud slides. Water searches are no different.

In water search, the dogs are "human scent detectors" and can only detect where the scent is emerging from the water. The subsequent location and recovery of the victim is accomplished using other skills and equipment.

There is evidence that, if the air temperature is colder than the water, such as 35 degrees F. or below, the rate of evaporation as well as wind force and barometric pressure effect the scent emerging from the water and, although the scent pools at the surface of the water, it does not get airborne. In such a circumstance, the dog may pick up the scent only at the water surface while swimming.

A case in point was the search for a 7 year old boy, in Illinois, who had fallen through the ice in mid-December. The search was joined by three dog teams in mid-March. While waiting for the boats to be launched, the one handler walked the bank to get an idea of what had happened and to look for possible "from-shore alerts". While walking back to the point last seen (PLS), the handler threw a stick into the water for her dog to retrieve (and to search while in the water). The first two times the dog came back with the stick. The third time (each at a different location) the dog let the stick go and alerted up stream by casting back forth in a small area about 10 feet off the bank. A few days later in the rain an attempt was made to investigate the area with a long pole. The only result was the remark that the pole had hit something that didn't feel like the sand, stumps and other debris in the area. Three days later when the weather cleared the boy's body was recovered at the bend in the river a quarter of a mile or so down stream, where the prevailing wind had pushed it. That day the air temperature was in the 40s and the dog air scented the body just after leaving the PLS.

Extreme temperatures whether hot or cold are certainly limiting factors. Another potential limitation we have to recognize is the presence of methane gas - which is frequently found in swampy areas. Methane gas is associated with cadaver material and can mislead or confuse the dogs.


The SAR dogs are a relatively scarce resource - not only in the United States, but in the world. There are about 100 established SAR dog units in this country. The NATIONAL SAR DOG DIRECTORY, currently available from the NASAR Bookstore, gives the 24-hour alerting phone number and an information source for each known unit in the country as well as the environments in which they work and their communications systems.

If you have the responsibility for body recovery or search in general and want to use SAR dogs, the NATIONAL SAR DOG DIRECTORY will help identify the nearest dog units.

Please Note that NASAR is not a "certifying agency" for Search and Rescue Dogs, however most established dog units have written standards. Find out through preplanning what specific skills those units or unaffiliated handlers near you have to offer. Ask them for a demonstration and a briefing about their specific training, experience, and equipment. Ask for a copy of their standards, call out procedures and their availability.

When you request a trained water search team, you can expect to get a team that has trained to search in the water environment. Some handlers however may elect not to participate in a swift water search because of the special skills required, so if that environment is a factor, it is best to discuss the matter beforehand.

The training in Mid-Atlantic D.O.G.S. includes both flat and swift water safety and self-rescue skills, equipment, knowledge of river hazards and search strategies and lots of on-the-water practice.

As a potential user of Water Search Dog teams, you might be interested in some of their logistics and what to expect when calling them to a mission or to work on a case. I will speak in terms of my unit, Mid-Atlantic D.O.G.S., because I know the answers there, but I would expect other volunteer units to operate in a similar manner.

For perspective, our members are all volunteers who have regular employment in many fields and range from senior executives in the federal government to self-employed individuals. All of them have made arrangements with their employers (or selves) to take off work to join searches, when the need arises. The time taken is their vacation leave or, if self-employed, their business may be temporarily closed. Upon entering this service, each handler assumes the responsibility of paying for their own training and mission expenses. We feel that because of the urgency of the victim's need -- a life may be at stake -- our response can not be limited by lack of funding - so we just do it!

There are some considerations, however. Our handlers will leave work anytime for a potential live victim or when the search is time-critical. In order to maintain this high level of response and commitment, we suggest that searches for bodies be done on weekends when possible,. There are, however, a few of us -- retired types -- who can and do respond anytime and anywhere.

So what does this mean to you?
In terms of cost, the is no fee for the services of the dog teams. In my opinion however, responding to a law enforcement investigation is different from a lost person-type mission, in that for the latter we are working for the victim or their family. In the former, we are assisting in an on-going investigation. Both are important, but the burden of total personal expense is harder to justify to our handlers. It is very helpful, if the requesting agency can house and feed the handlers and when the distance travelled is significant, help with the mileage expense -- our time is still volunteered. In some instances we have had to fly, which is an expense handlers can not personally handle and must be picked up by the requesting agency. Search dogs are accepted by most airlines for travel with the handler in the cabin -- with the same status as a seeing eye dog -- but this in-cabin transportation has to be arranged. Travel in cabin minimizes the dog's stress level and is important, especially when you expect the team to get off the plane and go to work.


River Rescue
	by Les Bechdel and Slim Ray
		NASAR Bookstore  P.O. Box 3709  Fairfax, VA  22038
White Water Rescue Techniques   
	by	Judy and Ted Waddell, National Park Service
		NASAR Bookstore  P.O. Box 3709  Fairfax, VA  22038
River Rescue - A Training Manual for Rescue Personnel
	by Ohio Department of Natural Resources, Division of Watercraft
		Vocational Insturctional Materials Laboratory  Ohio State 		University 
SAR Dogs and Water Searching
	by	East Coast SAR Dog Confederation   1984
		4  Orchard Way North   Rockville, MD  20854
Training for Water Search
	by Hatch and Judy Graham
		Dog Sports Magazine   1984
Water Search: Reading your Dog to the Depths
	by Hatch and Judy Graham
		Dog Sports Magazine  1985
Search Dog Training
	by Sandy Bryson
		NASAR Bookstore  P.O. Box 3709  Fairfax, VA  22038
Underwater Searches Using Dogs
	by	Judy Graham     RESPONSE! Magazine   1985
		NASAR Bookstore  P.O. Box 3709  Fairfax, VA  22038
Utilizing Dogs for Body Searches
	by	Marcia Koenig  c.1984
		NASAR Bookstore  P.O. Box 3709  Fairfax, VA  22038
Utilizing Air-Scenting Search Dogs to Locate Drowning Victims
	by Alice J. Stanley, Virginia Search and Rescue Dog Association
		NASAR Bookstore  P.O. Box 3709  Fairfax, VA  22038
Water Search Strategy and Handler Safety
	by Marian Hardy    1987      NASAR Bookstore  P.O. Box 3709  Fairfax, VA  22038
Dive Rescue and SAR Canines,  The Odd Couple That Works
	by Marian Hardy   1988    
	  	NASAR Bookstore   P.O. Box 3709   Fairfax, VA  22038
Water Search with Dogs
	by  Marian Hardy  1989
		NASAR Bookstore   P.O. Box 3709   Fairfax, VA  22038