A Review of Detection Distance for Search Dogs Based on Environment and Target Odor
Summary: Working dogs are expected to perform well in a huge variety of environments. Understanding what the research tells us about the effects of different environmental factors, as well as target odor, could help handlers modify their search strategy and expectations to better suit the conditions their dog is in. This article summarizes the relevant research.
When designing a search strategy to use with a detection dog, it is imperative to know the distance at which the dog can reasonably encounter odor and find the target. Without this understanding, searchers may needlessly increase search time, miss targets, or incorrectly estimate populations. The question of detection distance and search strategy is important for all detection dogs, but this article will primarily focus on conservation detection dogs rather than dogs searching for drugs, bombs, or other small and/or concealed targets. In conservation and some SAR applications, detection distance is measured in meters or tens of meters rather than centimeters.
In the field of search and rescue or explosives detection, a missed target could lead to the loss of human life. In the field of conservation detection dogs, a missed target could spawn a new generation of invasive species. An overly granular search strategy can also be detrimental by reducing the area covered and using precious time, energy, and money to walk over areas that the dog has already cleared thanks to airflow.
Furthermore, it is desirable to quantify the area effectively searched by a dog team. This can be used to identify areas that need to be revisited after a hasty search or be paired with probability of detection calculations to determine population of a species in an area. Determining the area searched is impossible without understanding the distance at which the dog can reasonably detect a target, because it is far more complex than simply looking at where the dog’s GPS says the dog walked.
Detection distance for dogs varies based on the target odor and its characteristics, environmental factors such as vegetation cover and terrain, weather, location of source (for example, buried versus suspended in a tree) and more. As such, there is no one-size-fits-all answer for detection distance.
Weather conditions may also affect a dog’s ability to work; both dry conditions and hot conditions can decrease a dog’s physiological ability to scent thanks to excessive panting, rapid energy depletion, and/or a dried nose.1-4
Although limited, there have been scientific studies conducted that attempt to identify optimal transect spacing for various targets and local areas. We’ll outline a few studies and their findings below.
- A study from Graham et al (1994)5 looking at dogs finding live human subjects classified the detection distance based on the air stability class. We can’t dig too deeply into air stability class here, but essentially classifications range from very unstable (Class A) to very stable (Class F). The study found that at 100 meters, dogs had a 95% detection rate for very stable air and 13% for very unstable air. At 25 meters from the subject, dogs had an 82% detection rate for very unstable wind, ranging up to 99% for very stable air. This paper underlines the importance of air stability, not just topography, for determining transect spacing. We would expect detection distances to vary for different targets.
- Buried land mines have a “halo” of ground odor that depends on the size of the mine but is usually about 60 centimeters in diameter. Anecdotal reports from handlers say detection distance can be up to 5 meters, suggesting airborne plumes as well as the ground plumes (Settles and Kester, 2001).6
- Reed (2011) found that scat detection dogs working in northern California oak woodlands found more than 75% of scats that were within 11 yards of the transect.7 The dogs’ performance dropped to 30-40% when scats were 27 yards off-transect.
- Detection distances for knapweed dogs were as high as 68 yards, although the paper does not specify percentage detected at various distances – 68 yards is just a maximum (Goodwin 2010).8
- Desert tortoise detection dogs could locate tortoises from up to 70 yards away but mean detection distance was 15 yards. The paper also found that detection rates were 70% when humidity was 16-18% and that higher wind allowed for greater detection distance (Cablk et al., 2008 and Nussear et al., 2008).3,9
- Dogs trained to find tree snakes in Guam had an average detection rate of 35%, and 30% of alerts were within 1 yard of the snake (Savidge et al., 2011).10 This suggests that in hot, humid, calm conditions with an elevated target, extremely small detection distances can be expected.
- Whale scat detection dogs succeeded at distances of 164 to 616 yards (Rolland et al., 2006).11 With a large sample, good scenting conditions, and very little turbulence or thermals over open ocean, the scent cone traveled very far.
- Chambers et al (2015)12 found that bat roost detection dogs were much more successful with guano bags hung lower above ground. The dogs had a 69% success rate with samples at 2 meters versus a 40% success rate when the samples were 6 meters above ground.
These papers and their findings are summarized in the table below. Because Graham,5 Settles and Kester6 and Chambers et al12 had such different study design than the other papers, they are not included in the table.
|Paper Authors||Target||Max Detection Distance||Average Detection Distance||Detection Rate||Search Conditions|
|Reed, 20117||scat||not given||not given||70% at 10m||northern California oak woodlands|
|Goodwin, 20108||knapweed (plant)||62m||not given||77%||short vegetation, flat area|
|Cablk et al., 2008;9Nussear et al., 20083||desert tortoises (live)||64m||13.7m||70% when humidity was 16-18%||open shrub desert|
|Savidge et al., 201110||tree snakes (live)||11.9m||not given, 30% were within 1 yard||35% (26-44% average between teams)||hot. humid, calm air; elevated source|
|Rolland et al., 200811||whale scat||1930m||not given||not given||open ocean|
Aside from detection distance, there are other factors to consider when determining transect spacing and search strategy.
- Study Goals. Do you want to find every single individual sample? In that case, you want narrower transects to ensure that samples are not missed. This is common when working with invasive plants, where a miss of a single plant could result in tens of thousands of seeds being released into the ecosystem. If your goal is to rely on the dogs to find “hotspots” for further investigation or monitoring, wider transects may allow the dog team to cover more ground and find more of these hotspots. Similarly, if your goal is simply to collect scat samples for other analysis and it’s okay if you miss some, wider transects may be appropriate. When the consequence of missing a sample is high, err on the side of narrower transect width.
- Time and Financial Budget. Unfortunately, money and time often limit what we can accomplish during a study. If you have limited time or money to survey a given area, a “hasty search” with wider transects may be a simple reality of your situation. This may only give presence/absence of a species or yield a few samples, but this information could be used to target further funding and research next season. That said, in some cases it may make more sense to reduce the study area and use narrower transects rather than increase transect width to survey a larger area. This comes down to study goals and the other factors below.
- Target Odor Availability. Samples that are buried, covered by duff, hidden inside of tree knots, tucked under bark, or otherwise protected from air flow will have less available scent for a detection dog. It’s not that a buried scat doesn’t stink; it’s that not as much of the odor can leach out of the soil as quickly. If you expect many of your samples to be tucked away, buried, or otherwise hidden from wind, expect to need narrower transects.
- Target Odor Volatility. When you walk into a room containing a dead mouse or fresh flowers, you know it right away. But if you walk into a room containing a bowl of uncooked greens or even dried flowers, you likely wouldn’t notice the odor. Part of this is based on the scent receptors in your nose and how they’re interpreted by your brain, but it is also because of the volatility of the molecules emanating from a sample. More volatile molecules mean that there is more odor available in the air for a dog to detect. This is partially why scat detection dogs and carcass detection dogs often have a greater detection distance (and therefore can tolerate wider transects) than dogs finding plants, cryptic animals, and nests. This (plus humidity) relates to why fresher scat samples are often detectable from farther away than desiccated samples.
- Target Odor Size. It’s intuitively clear that a dead whale smells more than a dead bat and that adult male grizzly bear scat is easier to sniff out than that of a young cub. With increased surface size of a target, more odor is given off for the dogs to detect. Generally, larger targets stink more and therefore allow for wider transects.
- Local Weather Conditions. We’ll dig more into weather conditions below, because this gets complicated. At an extremely basic level, higher humidity and consistent winds make for easier scenting and therefore facilitate wider transects. Blustery conditions, shifting wind, low humidity, and inversions can all complicate a search and may require narrower transects. Since the weather shifts constantly, it is often wise to devise study methods that allow the handler to update search strategy and transect width based on the conditions on the ground if your goal is to locate as many samples as possible.
- Local Topography. Odor flowing with the wind may get channeled down a riverbed, caught in an eddy behind a hill, or pool in a depression in the ground. Air flow (and odor with it) can change throughout the day based on slope direction and aspect. Varied topography makes for challenging search conditions and may require narrower transect spacing. If your goal is to find as many samples as possible, it’s best to give handlers permission to deviate from their transects to allow their dogs to access air flow that may not fall on-transect. For example, when searching wind farms in Nebraska in 2021, I often deviated off-transect to allow my dog to search steep gullies that may have trapped odor.
- Local Vegetation. Thick underbrush can limit air flow. The edges of forest can suck in or expel air depending on other local conditions and time of day. Trees can create eddies of scent in their wake, suck odor up in a “chimney” effect, or capture odor in their canopy. Forests can also protect the understory from overhead weather conditions. Extremely short grass or pavement on a flat surface can allow odor to move very far without collecting in many spots, making it difficult for the dog to “read” the scent plume. This stuff is complicated! Generally speaking, denser vegetation needs narrower transect spacing. Handlers should allow their dogs to investigate vegetation that may have collected odor in its bark or in its eddy, and use their knowledge of air flow to aid the dog in locating the final source.
- Individual Dog Team Characteristics. Not all dog-handler teams are created equal. Some dogs excel at detailed, intensive searches. Other dogs naturally work fast and far. Some handlers are extremely in-tune and read their dog’s minor changes of behavior with ease; other handlers expect the dog to spell everything out for them. Some dogs tire easily or lose motivation; others are absolute workhorses. Knowing the strengths and weaknesses of a detection dog team will help you design methodology that meets your needs.
- Smith, D. A., Ralls, K., Hurt, A., Adams, B., Parker, M., Davenport, B., Smith, M. C., & Maldonado, J. E. (2006). Detection and accuracy rates of dogs trained to find scats of San Joaquin kit foxes (Vulpes macrotis mutica).Animal Conservation, 6(4), 339–346.
- Homan, H.J., Linz, G. & Peer, B.D. (2001) Dogs increase recovery of passerine carcasses in dense vegetation. Wildlife Society Bulletin 29(1), 292-296.
- Nussear, K.E., Esque, T.C., Heaton, J.S., Cablk, M.E., Drake, K.K., Valentin, C., Yee, J.L., & Medica, P.A. (2008). Are wildlife detector dogs or people better at finding desert tortoises (Gopherus agassizii)? Herpetological Conservation and Biology 3(1),
- Gazit, I., & Terkel, J. (2003). Explosives detection by sniffer dogs following strenuous physical activity. Applied Animal Behaviour Science, 81(2), 149–161.
- Graham H. (1994). Probability of detection for search dogs, or How long is your shadow? Response Magazine, Winter 1994.
- Settles, G. S., & Kester, D. A. (1998). The external aerodynamics of canine olfaction. 323–335 in Sensors and Sensing in Biology and Engineering, F.G. Barth, J.A.C. Humphrey, and T.W. Secomb, Springer, Vienna & NY.
- Reed, S.E., Bidlack, A.L., Hurt, A., & Getz, W.M. (2011). Detection distance and environmental factors in Conservation detection dog surveys. The Journal of Wildlife Management, 75(1), 243–251.
- Goodwin, K., Engel, R., & Weaver, D. (2010). Trained dogs outperform human surveyors in the detection of rare spotted knapweed (Centaurea stoebe). Invasive Plant Science and Management, 3(2), 113-121.
- Cablk, M.E., Sagebiel, J.C., Heaton, J.S., & Valentin, C. (2008). Olfaction-based detection distance: a quantitative analysis of how far away dogs recognize tortoise odor and follow it to source. Sensors, 8(4), 2208–2222.
- Savidge, J.A., Stanford, J.W., Reed, R.N., Haddock, G.R., & Adams, A.A.Y. (2011). Canine detection of free-ranging brown treesnakes on Guam. New Zealand Journal of Ecology, 35(2), 174–181.
- Rolland, R., Wasser, S., Gillette, R., Davenport, B., Kraus, S., &; Hamilton, P. (2006). Faecal sampling using detection dogs to study reproduction and health in North Atlantic right whales (Eubalaena glacialis). Journal of Cetacean Research and Management 8(2):121-125.
- Chambers, C., Vojta C.D., Mering, E.D., & Davenport, B. (2015). Efficacy of scent-detection dogs for locating bat roosts in trees and snags. Wildlife Society Bulletin, 39(4), 780–787.