Conquering "snarge": an update from the Feather Lab

Flying Safety, Sept, 2004 by Carla J. Dove, Marcy Heacker

In 2003, the Feather Identification Laboratory at the Smithsonian Institution's National Museum of Natural History received a record number of bird strike cases for accurate species identification (2042 Air Force cases and 286 cases for civil aviation). Although most of these identifications were completed using whole feathers or feather parts, more than 260 of the Air Force identification cases consisted exclusively of a single paper towel swipe taken from the impact point on the aircraft.

[ILLUSTRATION OMITTED]

Collecting bird strike remains by spraying the impact point on the aircraft with a water bottle and swiping it down with a paper towel is a collecting technique that has been in use throughout the military BASH programs since 2001 (see http://afsafety.af.mil/AFSC/Bash/wild.html for collecting methods). Paper towel swipes containing only small amounts of blood, bird tissue and small microscopic bits of downy fluff from feathers is what we in the Feather Lab have termed "snarge." Smelly by nature, and often packed with mold spores, gooey bird fat or amorphous bits of flesh, "snarge" can provide valuable information about the species of birds your aircraft just smacked.

For many years now, the number of "snarge" samples sent for identification has increased. This is due to better awareness of bird strike issues, more outreach efforts to inform field personnel of collecting methods, and the need to have accurate data on problem species. Also, aircrew and maintenance personnel are much better at spotting the diagnostic signs of bird strikes and are becoming experts at finding the evidence. Eugene LeBoeuf (Chief, USAF BASH Team) once described "snarge" as looking like something you find in your handkerchief after a bad cold, but we see it as a goldmine of information and the ultimate identification challenge!

[ILLUSTRATION OMITTED]

Feather Identification

The majority of bird strike cases we receive have enough whole feather material to identify species by using the morphological characters such as size, shape, pattern, color and texture of the feather. These whole feather samples are then compared to bird specimens housed at the Smithsonian Institution until a perfect match is found.

[ILLUSTRATION OMITTED]

For cases with small or fragmented feathers, we use the unique microscopic characteristics in the fluffy plumulaceous (downy) barbs of the feather to give us clues as to the "group" or family of birds from which the unknown sample has come. This helps focus our efforts in the right direction for comparing any whole bits of feather and getting a final identification. These identification cases involving whole feather material are identified to the species level in nearly 93 percent of the cases.

The Problems With "Snarge"

As the amount of feather material decreases, the difficulty in identifying a bird increases. While our goals here at the Feather Lab strive for specific identification in each case, many times the microscopic evidence can only be identified to a general level.

Identifications based solely on microscopic characters (many of which were paper towel swipes) reached the highest number in 2003 at 559 cases. A third of these cases were identified only to the major group (Order) of birds because the microscopic structures are very similar in closely related birds. For example, many of the microscopic samples are identified as Passeriformes (the group of birds including the songbirds, or perching birds). Although we think of songbirds as being small birds, in 2003 Passeriformes caused a combined estimated $450,000 in damage to US military aircraft. Many times we cannot determine which songbird or even the family the particular songbird belongs to, because feather micro-characters are similar within this order of birds. This is also true in other groups (ducks, hawks, etc.) making microscopic identifications difficult if no other feather material is present.

While only approximately 12 percent of the bird strike cases received for identification are "snargy" paper towel swipes, they can be some of the most important in terms of identifications. Many times there are downy feather barbs embedded in the goo. With a lot of practice, patience and squinting, we have gotten pretty good at finding these minute pieces of feather. Even so, processing and identifying these minute samples is much more labor intensive and at least 50% more time-consuming than a strike with ample material. For example, an average bird strike case with many feathers and adequate data (location, date, etc.) usually takes about one hour to prepare (sometimes we have to actually wash the feathers to remove dirt, grease and blood), identify and report the results. A swiped sample currently requires a minimum of two hours.

Ultimately, we would like to identify every bit of remains we receive to the exact species level to increase our overall knowledge of bird strike damage and prevention. With the increased numbers of microscopic cases, decreased level of identification for these cases, and increased amount of time to work them, you can imagine that "snarge" has presented us with some challenges.