Since there has been some discussion of oiled bird rescue/"rehabilitation"
on Tweeters recently, the enclosed paper, published in the April 1996
issue of the Ibis (138:222-228), may shed some light on the question of
the effectiveness of oiled bird rescue, cleanup, and treatment.


Brian Sharp
2234 NE 9th Av
Portland, Or 97212
503-287-6501
ecopers at teleport.COM



POST-RELEASE SURVIVAL OF OILED, CLEANED SEABIRDS IN NORTH AMERICA

Brian E. Sharp
Ecological Perspectives
2234 NE 9th Ave
Portland, OR 97212
USA


The number of days between ringing and recovery of oiled, cleaned and
released seabirds was extremely low, usually a matter of a few days or
weeks, and for three species was 5-100 times lower than for non-oiled
birds. For oiled, cleaned Common Guillemots Uria aalge, post-release
life expectancy was 9.6 days, and long-term recovery rates were 10-20%
those of non-oiled birds. Measures of survival were not greater for
oiled birds treated in recent years with modern methods. The cost and
effectiveness of rehabilitation efforts for oiled seabirds need to be
re-examined in the light of results showing low post-release survival.

Oil spills at sea very often result in the oiling and deaths of large
numbers of seabirds, but a proportion (0.3-30% after some recent spills)
of birds that are oiled are retrieved alive and taken to 'rehabilitation'
centres, where they are cleaned and treated (Hope Jones et al. 1970,
Brown et al. 1973, Bibby & Lloyd 1977, Bibby 1981, Monnat & Geurmeur
1979, Ford & Casey 1989, Piatt et al. 1990, Page et al. 1990, Burger &
Fry 1993, Ford et al. in press). Of the birds rescued, the percentage
successfully cleaned, treated and released has increased from about 2%
after the 1967 Torrey Canyon spill (Bourne 1970), to as high as 60% after
the American Trader spill in California in 1990 (White & Sharp 1994),
though the release rate is often lower (e.g., 18% and 9% after the Arco
Anchorage and Orphan spills in Washington, 1986 and California, 1991,
respectively) (Jan White, pers comm). The percentage released has been
used as a measure of the success of bird rescue efforts, and release
itself is sometimes claimed to be tantamount to rehabilitation (Holcomb
1991, Monahan & Maki 1991). The post-release survival of cleaned
seabirds has not been used to assess whether treatment and cleaning
efforts are effective and in fact surprisingly few data are available on
the question.

This paper uses North American ringing data for marine birds (including
grebes and scoters, which winter in marine environments) to 1) measure
post-release recovery rates, number of days survived, and survival rates
of oiled birds; 2) compare measures of survival and movements of oiled
birds and non-oiled controls; 3) determine whether measures of survival
of oiled birds have increased since 1989; and 4) examine whether measures
of physiological condition which are used as criteria of suitability for
release are related to post-release survival.


METHODS

Ringing and recovery records, obtained from files maintained by the Bird
Banding Laboratory (BBL), U.S. Fish and Wildlife Service, Laurel,
Maryland, were examined for North American seabirds that were oiled,
cleaned and released between 1969 and 1994. Recovery records included
reports from the general public to the BBL of ringed birds that were
shot, found dead, released alive, or for which present condition was
listed as unknown, but excluded recoveries of skeletons, ring only, and
'returns' (resightings by the ringer at breeding sites). Copies of
ringing schedules were checked to ascertain whether probably oiled birds
coded 'miscellaneous' (because they were also transported, treated, or
held in captivity) were in fact oiled. Copies of letters reporting
unlikely or unusual recoveries were examined for possible errors.
Records of 'released and returned' birds ('R&R's'), i.e., birds that were
returned, dead or alive, to treatment centres within a few days of
release, and for which recovery records are not submitted to the BBL by
the ringer) were examined separately and not included in survival
analyses.

'Days elapsed' or 'Days survived' is the interval in days between release
and recovery. For recoveries with imprecise dates (e.g. unknown day in
June), the midpoint of the imprecise interval was used as the recovery
date. Recoveries of non-oiled Velvet Scoters Melanitta fusca used as
controls were largely derived from ringings in Saskatchewan and British
Columbia, non-oiled Western Grebe Aechmophorus occidentalis and A.
clarkii recoveries were derived from birds ringed in Manitoba, and
non-oiled Common Guillemot Uria aalge recoveries were obtained from
California, Oregon, and Newfoundland. There were insufficient normal
wild Surf Scoters Melanitta perspicillata with which to compare
recoveries of oiled Surf Scoters.

T-tests were used to compare mean days elapsed for oiled and non-oiled
birds. Oiled : non-oiled differences in overall and long-term recovery
rates (recovery rate being the number of recoveries divided by the number
ringed) were tested by chi-square and binomial tests. T-tests and
regressions were used to compare mean days elapsed of oiled birds cleaned
after recent spills (since 1990) with mean days elapsed for spills in
previous years. Statistical tests, except the binomial test used for
long-term recovery rates, were performed by the statistical program
MINITAB.

Whenever possible, the age of non-oiled controls was the same as that of
oiled birds. Oiled Common Guillemots were not aged or sexed when ringed,
but since most spills occurred in winter, oiled guillemots were
predominantly birds in at least their second calendar year of life,
whereas most non-oiled Common Guillemots were ringed at breeding colonies
as flightless young.

Oiled Common Guillemots provided the largest sample of usable recoveries
(n = 78) for any one species, and their survival was estimated by the
maximum likelihood method (Brownie et al. 1978), using the program
ESTIMATE. Ringings and recoveries were combined into 20-day periods, and
survival was estimated over the length of the period. The maximum number
of recovery periods allowed by the program is 20, and two recoveries
after 400 days were assigned to recovery period 20. The survival rate
obtained for oiled guillemots was compared with published rates for
non-oiled guillemots.

Intake and treatment records from rehabilitation centres were examined to
obtain measurements of oiling, days in captivity, and physiological
condition. Regressions were computed for 'days elapsed' in relation to
degree of oiling (rated on a scale of 1-3), days between capture and
cleaning, number of days in captivity, weights at intake and release
expressed as a percentage of normal weight for the species (Dunning
1993), packed blood cell volume at intake and release, concentration of
leucocytes at release, and total blood proteins at release. T-tests were
also used to compare differences in the means of the above variables for
'long'- and 'short'-term recoveries (< or >175 days). Records included
in these analyses included released and returned birds for Nestucca
(Washington 1989), American Trader, and Exxon Valdez (Alaska 1989) oil
spills.

For oiled and non-oiled Common Guillemots, post-release movements were
analyzed using 10-minute blocks of latitude and longitude, approximately
6 miles by 10 miles (c. 10 x 16 km) on a side.


RESULTS


Database

The available data consist of 127 recoveries from approximately 3,200
oiled marine birds ringed of 13 species. There were also 101 released
and returned records available since 1989. Almost all available recovery
data for North American seabirds derived from Pacific coast oil spills.
Three species--Western Grebe, Velvet Scoter, and Common
Guillemot--provided 10 or more recoveries. The recovery rate for all
species combined for all west coast spills since 1969 excluding Alaska
was 4.18%. The Alaska recovery rate for seabirds is negligible, e.g.
only one recovery has accrued from 1102 guillemots ringed since 1955.


Mean and median days elapsed

Table 1 shows mean and median days elapsed for individual species and for
all species combined. The distribution of recoveries of oiled birds is
skewed: most recoveries occurred within a few days or weeks of release.
For Common Guillemot, for example, 73 of 78 recoveries (94%) occurred
within 60 days after release. Longer term recoveries were few, e.g. only
2 of 78 Common Guillemots survived more than a year (only 4 more than 5
months). Median days survived were low, 4-11 days for oiled birds of any
species.

Mean days survived of 101 'released and returned' birds from various
spills since 1989 was 4.0 days. The inclusion of such records would
reduce mean and median days elapsed for all species to 39 and 6,
respectively, and for Common Guillemots to 21 and 5 days, respectively.
Comparisons of mean days survived and overall recovery rates of oiled and
non-oiled birds

The pattern of recoveries of oiled and non-oiled birds of three species
differed. Mean days survived for oiled Western Grebes, oiled Common
Guillemots, and oiled Velvet Scoters were all significantly lower than
means of non-oiled controls (P < 0.001). Mean days survived for
non-oiled birds were 5-100 times higher than means for oiled birds of
those three species (Table 1).

Recovery rates were compared for oiled and non-oiled birds of 3 species.
Recovery rates of oiled Western Grebes (0.088, n = 10) and Common
Guillemots (0.052, n = 80) exceeded those of non-oiled controls (0.037, n
= 33, 0.014, n = 23, X21 5.70, 32.8, respectively) whereas the recovery
rate of immature oiled Velvet Scoters (0.025, n = 10) was less than that
of non-oiled scoters (.079, n = 18, X21 8.75). There was no difference
in the recovery rates of oiled and non-oiled adult Velvet Scoters (Table 2).

Survival of birds oiled and treated in recent spills

Survival of birds oiled, cleaned and treated since 1990, using state of
the art methods, was compared with birds oiled and cleaned 1969-1989.
Mean days survived were not different for birds oiled and treated prior
to and after 1990, for all species combined (means 38 and 111, P = 0.17),
for Western Grebes (means 178 and 11, P = 0.26), or for Common Guillemots
(means 29 and 107, P = 0.36). Mean days survived for oiled Common
Guillemots ringed after 1989 and for non-oiled guillemots were
significantly different (means 107 and 485 days, P < 0.001) (Table 1).

For all species combined, Western Grebe, Common Guillemot, and Velvet
Scoter, regressions of days survived against year oiled were also not
significant. For Western Grebe and Velvet Scoter, slopes of regressions
were almost significantly negative, suggesting that days survived had, if
anything, declined in recent years. There were no relationships between
recovery rates and year.

For oiled Common Guillemots, long-term recovery rates (of birds surviving
at least a year after release) were compared with non-oiled controls from
California, Oregon, and Newfoundland. Long-term recovery rates for oiled
guillemots were between 10% and 20% of, and were significantly lower
than, those of all three non-oiled controls (Table 3). Long-term
recovery rates of birds ringed since 1989 also did not increase when
examined separately.


Estimate of survival for Common Guillemots

>From recoveries of Common Guillemots (n = 78), survival was estimated by
the maximum likelihood method (Brownie et al. 1978). Data were too few
to estimate survival separately for spills prior to and after 1989 (n =
62 and 16, respectively), and data for all years were combined.
Estimates of survival and mean life span are presented in Table 4. The
survival rate was 12.6% per 20-day period, and mean life expectancy of
oiled, cleaned, and treated guillemots after release was 9.6 days.

Goodness-of-fit tests indicated that the data best fit Model 3, which
assumes constant recovery and survival rates, when recoveries were
combined into 20-day periods. The assignment of two recoveries greater
than 400 days to recovery period 20 made a negligible difference in
estimates of survival and life expectancy. Likewise, omitting those two
recoveries altogether made negligible differences in the estimates.


Movements

Disproportionate numbers (68%, n = 79) of oiled guillemots were recovered
in the same or adjacent 10-minute blocks, and only 2.5% were recovered
outside the borders of the state in which they were ringed. In contrast,
45% (n = 65) of non-oiled birds were recovered out of state, and only 21%
in the same or adjacent blocks. The four oiled guillemots that survived
more than five months also moved the longer distances. The pattern of
movements for oiled and non-oiled Common Guillemots differed
significantly (X26 48.895, P < 0.005) (Table 5).


Days elapsed in relation to condition

Current treatment protocols for oiled birds include blood and serum
chemistry tests, some of which are used as criteria of suitability for
release. Release criteria such as blood parameters and weights were poor
predictors of survival after release, for all species and for Common
Guillemots examined separately (Table 6). Number of days survived for
all species combined was positively associated with degree of oiling
(r263 = 6.6%, P < 0.05) and number of days in captivity (r296 = 13.8%, P
< 0.0001), and negatively associated with lower final PCV (r252 = -8.9%,
P < 0.03). Weights at intake or release were not related to number of
days survived. Leucocytes of birds recovered shortly after release were
not significantly higher in t-tests than those of long-term recoveries
for all species combined (t17 2.00, ns), or for Common Guillemots.
Number of days in captivity was negatively associated with weight at
intake (r28 = 53.0%, P < 0.03), but not with degree of oiling (r245 =
0.0%, ns) or PCVfinal (r235 = 4.9%, ns).


DISCUSSION

The overwhelming preponderance of the North American ringing data
indicate that post-release survival of oiled seabirds is low. 1) The
data show a markedly skewed pattern of recoveries for oiled birds in the
first few days or weeks after release, with low mean and lower median
days elapsed for all species. 2) Mean days elapsed of oiled Common
Guillemots, Velvet Scoters, and Western Grebes were 5-100 times lower
than those of non-oiled controls. 3) Oiled birds cleaned and treated in
recent years did not show any increase in measures of survival compared
to oiled birds cleaned before 1990. Mean days elapsed for recently oiled
Common Guillemots was still significantly lower than that of non-oiled
guillemots. 4) First year and overall recovery rates for oiled Common
Guillemots and Western Grebes were significantly higher than expected,
and long-term recovery rates for oiled Common Guillemots were
significantly lower than those of non-oiled controls. The possible
problem of truncated distributions of recoveries is minor; any late
occurring recoveries would not change long-term recovery rates
significantly. 5) There was a scarcity of oiled seaduck recoveries in
subsequent hunting seasons. Of 395 oiled Velvet Scoters ringed, which
comprised the largest sample for any hunted species, one would have
expected 16 such recoveries, but there were none. And 6) recovery
locations for oiled Common Guillemots were closer to release locations
than for non-oiled guillemots.

On the other hand, there were only few recovery data suggestive of
long-term survival of oiled birds after release, i.e., a small number of
oiled bird recoveries of 6 months to three years, suggesting limited
dispersion of treated birds into the wild population. These exceptions
were too few to measurably influence mean and median days survived, and
certainly the small number of birds involved would not contribute
significantly to the wild population of non-oiled birds.

The survival of oiled Common Guillemots calculated by the maximum
likelihood method, 0.13 per 20 day period, translates into a negligible
annual survival rate. In contrast, typical annual survival for non-oiled
Uria is 0.90-0.95 for adults and 0.20-0.40 for young over 3-4 years to
first breeding (Birkhead & Hudson 1977, Hatchwell & Birkhead 1991, Gaston
1991, Sydeman 1993, Harris & Wanless 1995).

It might be argued that recoveries during the first year few days after
release are biased, but such possible biases cannot account for the low
measures of survival obtained here. The low long-term recovery rates of
oiled guillemots, which are not subject to such possible
recovery-inflating factors as onshore winds after release, or publicity,
can only be a consequence of abnormally high mortality in the year after
release. The low long-term recovery rates reinforce the conclusion that
the low number of days survived are real and not merely a result of
either publicity or onshore winds after release. Telemetry data (from
radio-tagged pelicans - see below), which are not dependent on reporting
by the general public, with its attendant biases, corroborate the ringing
recovery results presented here. In addition, since most oiled
Guillemots were ringed in at least their second calendar year of life,
the latter birds had survived much of the post-fledging mortality which
occurs during the first autumn and winter (Mead 1971). Thus, the
differences in mean days survived found between non-oiled and oiled
Guillemots probably underestimate actual differences. Excluding
released and returned birds from calculations of mean days sruvived and
survival rates also results in inflated measures of survival for oiled
bird samples.

In recent years, the mean proportion of oiled birds that survive
treatment and are released has been 0.35 (range 0.09-0.60). If the
proportion of birds that survive a year after release is 0.15 (range
0.10-0.20), survival from initial rescue to 1 year after release is about
0.05 (5%). The long-term recovery rates of oiled Guillemots may
represent an approximation of the proportion of birds that survive and
behave like non-oiled birds and might be considered the proportion that
is truly rehabilitated. A mortality rate greater than 90% in the first
year would result in the reduced long-term recovery rates observed.

Chris Mead, then Senior Ringing Officer with the British Trust for
Ornithology, stated (in litt., August 1991), "All our experience with the
rehabilitation of oiled points towards a very low success rate....the
result from many hundreds (possibly several thousand) cleaned
Auks--mostly Guillemots are really quite depressing. In most instances,
the birds are found close to the release point either dead or moribund
within a few days of release or never again....we do have a few ringing
recoveries which show survival for several years and birds returning to
distant potential breeding areas....I would say that the overall average
must be less than 10% of the birds apparently fit on release survive
longer than one month".

There are few other studies, but all, with one exception, indicate
that oiled, cleaned birds do not usually survive long after release.
J.P. Croxall (in Swennen 1977) found that the recovery rate of oiled auks
released into the North Sea was 11% in the first 6 months, as opposed to
a recovery rate for normal birds of 3% over their entire lifespan of
several years. Swennnen (1977) found that oiled, cleaned birds
"released" into large enclosures had an annual mortality rate of 35-37%,
even though fed and protected from mortality factors that would have been
operative in the natural ecosystem, compared with a mortality rate of 7%
for non-oiled controls. Daniel Anderson (pers. comm.) found that the
survival to 6 months of oiled radio-tagged Brown Pelicans Pelecanus
occidentalis was one-third (P<0.05) and survival to 2 years was one-sixth
(P<0.01) that on non-oiled controls. In an extensive radio-telemetry
study after the Exxon Valdez oil spill, C. W. Monnett (pers. comm.) found
that 2/3 of 45 oiled, prime-aged sea otters Enhydra lutris were dead
within 2 years, whereas normal mortality for that age class was 6%.
Estes (1992) further stated that rescue efforts themselves caused a
stress-induced mortality rate of 5-10%. Oiled and cleaned Magellanic
Penguins Spheniscus magellanicus were not seen again after release, and
rehabilitation was stated to be "of little use ... even in the best case"
(D. Boersma, paper presented at 4th International Conference on Effects
of Oil on Wildlife, Seattle, April 1995).

Exposure of seabirds to oil at sea evidently affects them in ways that
are not reversed by cleaning and treatment. After the Exxon Valdez oil
spill, a significant proportion of oiled birds autopsied by the Madison
Laboratory of the U.S. Fish and Wildlife Service showed gastrointestinal
haemorrhaging and pneumonia (U.S. Department of Interior, unpubl. data).
D. M. Fry (unpubl. data) reported the causes of death of 50 seabirds
taken to rehabilitation centers after oil exposuire in the Exxon Valdez
spill included stress (due to oiling, cleaning, handling and captivity),
haemolytic anemia, emaciation, hypothermia, and secondary infections in
captivity., Khan and Ryan (1991) and Wood and Heaphy (1991) found from
autopsies that, after cleaning and treatment, oiled Guillemots were
affected by various phsyiological disorders, including liver, kidney and
intestinal malfunction, and that oiling and oil ingestion cause internal,
physiological damage which, they suggested, must affect post-releasse
survival.

The only work that supports the view that cleaning and treatment of
oiled seabirds results in a significant level of rehabilitation was with
oiled, cleaned Jackass Penguins Spheniscus demersus, a high percentage of
which (37%-84%) were later seen alive at colonies (Morant et al. 1981).
It appears that the "Jackass Penguin is ... the only seabird to have been
successfully rehabilitated on a substantial scale" (ibid. p.279). A
rigorous analysis of possible differences in the survival of oiled and
non-oiled penguins, however, has not been performed (A. J. Williams,
pers. comm.).

Claims that cleaning and treatment are the equivalent of
rehabilitation are thus unsubstantiated. Almost all the
evidence--ringing data, other studies of survival, autopsies and
radio-telemeetry--points to the conclusion that cleaned and treated birds
are unfit, in the survival sense, at release. The available data
therefore do not support a decision that oiled seabirds should be
rescued, treated and cleaned. Whether there is a valid humanitarian
basis for the human impulse to attempt to alleviate the distress of oiled
birds is a separate, ethical consideration.

Bird rescue, cleaning and treatment are costly. After the Exxon
Valdez event, approximately U.S. $41 were spent in the rescue, treatment
and release of approximately 800 birds.

Because their mortality rates are so high, oiled seabirds that have
been cleaned and released should be added to the total of dead birds for
the purpose of assessing damages of oil spills to seabirds and for the
purpose of seeking compensation. Cleaning and treatment do not provide
effective mitigation for and cannot be considered as even partial
restoration of damage. The high cost of treatment, coupled with poor
post-release survival, strongly suggests that programs for oiled birds
rescue should be critically reexamined. Oil spill response planning and
resources should be redirected to the prevention of damage, rather than
focussing on ineffective attempts at rehabilitation after the damage has
occurred.


Acknowledgements:
The California Department of Fish and Game, Office of Oil Spill Prevention
and Response, and the U.S. Department of Justice provided financial
support. International Bird Rescue Center, Berkeley; Peninsula Humane
Society, San Mateo; George Whittle Wildlife Rescue Center, Monterey; San
Francisco Bay Bird Observatory; Fresno Wildlife Rescue and
Rehabilitation; Nisqually National Wildlife Refuge and U.S. Fish and
Wildlife Service, Anchorage, Alaska, inter alia, ringed oiled birds after
various spills. D. Bystrak and N. Mullis at the Bird Banding Laboratory,
Laurel, Maryland, extracted ringing and recovery data from computer and
manual files. J. White, University of California, Davis, and
rehabilitation centers provided treatment records for oiled birds. J.
Hines, Patuxent Wildlife Research Center, Laurel, Maryland, furnished the
program ESTIMATE. U. Wilson, J. Piatt, D. Heinemann, R.G. Ford, J.
White, C. Mead, V. Mendenhall and anonymous reviewers commented on the
manuscript.

REFERENCES

Bibby. C.J. 1981. An experiment on the recovery of dead birds from the
North Sea. Ornis Scan. 12: 261-265.

Bibby, C.J. and Lloyd, C.S. 1977. Experiments to determine thefate of
dead birds at sea. Biol. Conserv. 12: 295-309.

Birkhead, T.R. and Hudson, P.J. 1977. Population parameters for the
Common Guillemot Uria aalge. Ornis Scand. 8: 145-154.

Bourne, W.R.P. 1970. Special review--After the 'Torry Canyon"
disaster. Ibis 112: 120-125.

Brown, R.G.B., Gillespie, D.I., Lock, A.R., Pearce, P.A., and Watson,
G.H. 1973. Bird mortality from oil slicks of eastern Canada,
February-April 1970. Can. Field-Nat. 87: 227-234.

Brownie, C., Anderson, D.R., Burnham, K.P. and Robson, D.S.
1978. Statistical Inference from Band Recovery Data--A Handbook.
Research Publication No. 131. Washington, D.C.: U.S. Fish and Wildlife
Service.

Burger, A.E. and Fry, D.M. 1993. Effects of oil pollution on seabirds in
the northeast Pacific. In Vermeer, K., Briggs, K.T., Morgan, K.H. and
Siegel-Causey, D. (eds.) The Status, Ecology and Conservation of Marine
Birds of the North Pacific. Special Publication. Ottawa: Canadian
Wildlife Service.

Dunning, J.B. 1993. Body Weights of 686 Species of North American
Birds. West. Bird Banding Assoc. Monogr. No. 1. Reprinted by Internat.
Wildl. Rehab. Council.

Estes, J.A. 1992. Catastrophes and Conservation: Lessons from Sea
Otters and the Exxon Valdez. Science 254: 1596.

Ford, R.G., Bonnell. M.L., Varoujean, D.H., Page, G.W., Sharp, B.E.,
Heinemann, D., and Casey, J.L. In press. Assessment of direct seabird
mortality in Prince William Sound and the Western Gulf of Alaska
resulting from the Exxon Valdez oil spill. In Proc. Exxon Valdez Oil
Spill Symposium, Anchorage, February 1993.

Ford R.G. and Casey, J.L. 1989. Seabird mortality resulting from the
Nestucca oil spill incident winter, 1988-89. Ecological Consulting Inc.
Portland, Oregon.

Gaston, A.J. 1991. Project on the Population Dynamics of the
Thick-billed Guillemot, Uria lomvia: Interim Report on Studies at Coats
Island. Tech. Rept. Series No. 134. Canadian Wildlife Service,
Headquarters. Ottawa.

Gibson, M.J. and J. White. 199anless, S. 1995. Survival and
non-breeding of adult Common Guillemots Uria aalge. Ibis 137:192-197.

Hatchwell, B.J. and T.R. Birkhead. 1991. Population dynamics of Common
Guillemots Uria aalge on Skomer Island, Wales. Ornis Scand. 22:55-59.

Holcomb, J. 1991. Overview of bird search and rescue and response
efforts during the Exxon Valdez oil spill. Pp. 225-228 In Proc. 1991
Internat. Oil Spill Conf. Am. Petroleum Inst. Washington D.C.

Hope Jones, P., Howells, G., Rees, E.I.S., and Wilson, J. 1970. Effect
of 'Hamilton Trader' oil on birds in the Irish Sea in May 1969. Brit.
Birds 63:97-110.

Khan, R.A. and Ryan, P. 1991. Long Term Effects of Crude Oil on Common
Guillemots (Uria aalge) Following Rehabilitation. Bull. Env. Contam.
Toxic. 46:216-222).

Mead, C.J. 1971. Seabird mortality as seen through ringing. Ibis 113:418.

Monahan, T.P. and Maki, A.W. 1991. The Exxon Valdez 1989 wildlife
rescue and rehabilitation program. Pp. 131-136 In Proc. 1991 Internat.
Oil Spill Conf. Washington ca.

Page, G.W., Carter, H.F., and Ford, R.G. 1990. Numbers of seabirds
killed or debilitated in the 1986 Apex Houston oil spill in central
California. Studies in Avian Biol. 14: 164-174.

Piatt, J.F., Lensink, C.J., Butler, W., Kendziorek, M., and Nysewander,
D.R. 1990. Immediate impact of the 'Exxon Valdez' oil spill on marine
birds. Auk 107:387-397.

Swennen, C. 1977. Laboratory Research on Seabirds. Neth. Inst. for Sea
Research. Texel, The Netherlands.

Sydeman, W.J. 1993. Survivorship of Common Guillemots on Southeast
Farallon Island, California. Ornis Scand. 24:135-141.

White, J. and Williams, T. 1991. Saving endangered species in major oil
spill cleanup efforts. Pp 221-223 In Proc. 1991 Internat. Oil Spill
Conf. Am. Petroleum Inst. Washington D.C.

Wood, M.A. and Heaphy, N. 1991. Rehabilitation of oiled seabirds and
bald eagles following the Exxon Valdez oil spill. Pp. 235-239 In Proc.
1991 Internat. Oil Spill Conf. Am. Petroleum Inst. Washington D.C.



Table 1. Mean and median days survived of oiled and non-oiled seabirds
after release in North America, based on ringing recoveries, excluding
R&R's (unless stated).
-----------------------------------------------------------------
Species N Mean Median SEmean Range 95% CL
--------------- --- ---- ------ ------ ------ ------- -------
OILED:
All species 118 51 6 13.9 0-919 23.4-78.6
R&R's 101 4 - 0.44 0-27 3.14-4.86
Guillemot
all years 78 39 6 16.0 0-919 6.0-70.5
<1990 68 29 6 14.1 0-919 1.06-56.3
>1990 10 107 8 79.9 0-810 0-263.2
With R&R's 160 21 5 7.89 0-919 5.51-6.67
R&R's 82 4 - 0.44 1-27 3.14-4.86
Western Grebe 10 111 11 68.6 1-763 0-248.5
Velvet Scoter 10 8 7 0.32 1-16 4.5-10.9
Surf Scoter 8 39 5 30.6 2-255 0-101

NON-OILED:
Western Grebe 37 763 624 114 19-658 532-993
Guillemot 641 485 216 31 1-9259 424-546
Velvet Scoter 22 1019 466 269 8-4939 459-1579
------------------------------------------------------------------
"R&R" = Released and returned to rehabilitation centre: dead on
collection, dead on arrival, euthanized, or rewashed and re-released.


Table 2. Recovery rates for oiled and non-oiled Western Grebes, Velvet
Scoters, and Guillemots.
------------------------------------------------------------
Number Number Recovery Chi-sq
Species Banded Recovered Rate (df=1)
----------------- ------ --------- -------- ------
Western Grebe
Oiled 113 10 0.088
Non-oiled 889 33 0.037 5.697*
Velvet Scoter
Immature
Oiled 397 10 0.025
Non-oiled 229 18 0.079 8.752**
Adult
Oiled 397 10 0.025
Non-oiled 449 17 0.038 1.028
All ages
Oiled 397 10 0.025
Non-oiled 678 35 0.052 4.039*
Guillemot
Excluding AK
Oiled 1526 80 0.052
Non-oiled 1590 23 0.014 32.843**
Including AK
Oiled 1823 80 0.044
Non-oiled 2192 23 0.010 42.060**
------------------------------------------------------------
* = P < 0.05, ** = P < 0.01.


Table 3. Long-term recovery rates (in years 2 and beyond) of oiled and
non-oiled Guillemots.
-------------------------------------------------------------------
No. Oiled recovery
long- Long rate as percent
term term of non-oiled rate*
No. recov- recovery ------------------
Dataset Ringed eries rate All years > 1989
----------------- ------ ----- -------- --------- ------
Oiled:
All years 1272 2 .0016
Recent > 1989 751 1 .0013
Non-oiled:
Newfoundland** 23475 274 .0117 13.4 11.4
(.0011) (.0015)
California** 574 5 .0087 18 15.3
(.0005) (.0115)
Oregon** 2716 21 .0077 20.3 17.2
(.027) (.0202)
Oregon adults 148 2 .014 11.6 9.9
(.0016) (.0007)
------------------------------------------------------------------------
*Figures in parentheses are probabilities that oiled rates do not differ
from non-oiled rates.
**Most ringed as flightless young


Table 4. Maximum likelihood estimate of survival for oiled Guillemots,
recoveries from all years, assuming constant survival, by 20-day recovery
period.
-------------------------------------------------------------------
20-day Number 20-day Recovery Period
Period Ringed 1 2 3 4 5 6 7 8 9 10
------ ------ -- -- -- -- -- -- -- -- -- --
1 967 53 8 1 0 0 1 0 0 1 1
2 269 9 0 0 0 0 0 1 0 0
3 36 1 1 0 0 0 0 0 1

Survival rate 12.6% per 20-day period
SE 3.8%

Mean life span 0.48 of 20-day period
SE 0.07
95% CI 0.34-0.62

Test of Model
X23 4.68, theoretical chi-square.05 7.81, P = 0.20. A
probability > 0.05 indicates acceptance of the model (Brownie et al,
1978).
--------------------------------------------------------------------



Table 5. Post-release movements of oiled and non-oiled Guillemots in
relation to release location.
------------------------------------------------------------
Recoveries by 10-minute block
of latitude and longitude
------------------------------------------------
Out of
Same +1 +2 +3 +4 +5 >=+6 State n
---- --- --- --- --- --- ---- ---- --
Oiled 26 25 8 6 1 4 7 2 79
Non-oiled 7 7 3 7 3 1 8 29 65
-------------------------------------------------------------------


Table 6. Probabilities (P) associated with regressions (R2) of days elapsed
against parameters for condition.
-------------------------------------------------------------------
All species Guillemot
--------------------- --------------------
Parameter R2(%) n P R2(%) n P
--------- ---- -- ----- ----- -- ----
PCV initial* -0.1 31 0.889 -0.4 18 0.813
PCVfinal -8.9 53 0.030 -8.0 40 0.077
Degree of oil 6.6 64 0.040 4.8 47 0.139
Leucocytes -2.6 18 0.522 -2.6 8 0.705
Total protein 0.4 66 0.627 0.1 50 0.801
Daystoclean -1.6 76 0.272 0.0 59 0.884
Dayscaptive 13.8 97 0.000 31.9 74 0.000
Weightinitial -5.1 22 0.312 -25.8 9 0.162
Weightfinal -0.1 17 0.892 0.0 7 0.974
--------------------------------------------------------------------
* PCV = packed cell volume.