THE SHELL LEAKS FILES: 11 SEPTEMBER 2026
SLF-2007-054
The Sakhalin Papers XLIV: The Whales Moved — What the 2015 Data Eventually Revealed
The airguns stopped in 2015. The scientific argument did not. Seven years later, a series of peer-reviewed studies finally put numbers on what happened to western gray whales during the unprecedented seismic season off Sakhalin. The results resist both convenient extremes. The whales did respond: their local density fell as accumulated seismic and vessel noise increased, and individual animals changed movement and breathing behaviour. The researchers concluded explicitly that mitigation had not eliminated short-term behavioural responses. Yet models examining reproduction predicted little or no effect from the 2015 disturbance scenario, 14 calves were identified the following year, and the wider population continued its long-term recovery. The evidence therefore supports neither “no impact” nor “population damage proved.” It supports something scientifically more difficult: measurable disturbance, mitigation that reduced but did not eliminate response, and unresolved population-level consequences.
Archive reference: SLF-2007-054
Collection: The Sakhalin Papers
Principal scientific records: Glenn Gailey et al., Gray whale density during seismic surveys near their Sakhalin feeding ground; Glenn Gailey et al., Western gray whale behavioral response to seismic surveys during their foraging season, Environmental Monitoring and Assessment, 2022
Supporting scientific record: Lisa Schwarz et al., Gray whale habitat use and reproductive success during seismic surveys near their feeding grounds, 2022; Lisanne Aerts et al., Seismic surveys near gray whale feeding areas off Sakhalin Island, Russia: assessing impact and mitigation effectiveness, 2022
Contemporaneous scientific record: Western Gray Whale Advisory Panel Statement of Concern, 8 May 2015
Authenticated Shell record: Royal Dutch Shell plc Sustainability Report 2015
Subsequent conservation record: IUCN, 3 September 2016 and 24 April 2019
Court record: Export Credits Guarantee Department v Friends of the Earth [2008] EWHC 638 (Admin), used solely for the earlier Sakhalin II environmental and public-finance context
Evidence standard: Statistical association is distinguished from proof of injury or population-level causation. Observed behaviour is distinguished from modelled demographic consequences. Population recovery is not treated as proof that industrial disturbance was harmless. Industry financial and logistical support for parts of the research programme is disclosed rather than ignored.
Introduction
The previous instalment ended on 29 July 2015.
Sakhalin Energy’s seismic operation was over.
Its mitigation system had stopped the source when required.
The Western Gray Whale Advisory Panel would later regard the mitigation component as satisfactorily implemented, while recording shortcomings in monitoring, equipment preparation and training.
But the most important question remained unanswered.
What had the whales actually done?
That was never going to be answered from the vessel log.
A shutdown record could tell scientists when an airgun array stopped.
It could not tell them whether whales elsewhere had changed course.
An acoustic buoy could measure sound.
It could not by itself establish whether a feeding whale shortened its dive, accelerated, moved towards shore or abandoned an area temporarily.
And successful completion of a seismic programme could certainly not establish whether disturbance had energetic or reproductive consequences.
The answer required years of analysis.
In October 2022, much of that analysis finally appeared in a special collection of peer-reviewed papers in Environmental Monitoring and Assessment. Two papers in particular examined the questions at the heart of the 2015 dispute: whether whale density changed with accumulated sound exposure, and whether individual whale behaviour changed as seismic vessels and other vessels approached. (Springer)
The findings were not “nothing happened.”
Nor were they “the whales were driven away.”
The truth was more interesting.
1. The 2015 monitoring effort was enormous
Whatever criticism can be made of the 2015 programme, lack of data collection is not one of them.
For the density study, shore teams operated from 13 observation stations between 1 June and 30 September.
Across those stations they accumulated 10,042 distribution scans and 16,817 whale sightings. The researchers noted that the 2015 scan total was more than twice the number collected over the entire previous decade from 2004 to 2013. (Springer)
The behavioural programme was similarly substantial.
Five shore-based teams accumulated approximately 3,843 observer-hours.
They produced 1,270 whale tracks, 401 focal-follow sessions, 44,634 geographic whale positions and 30,735 recorded respiration events.
Around 7,403 movement-analysis intervals and 2,328 respiration-analysis intervals were ultimately derived from the observations. (Springer)
The acoustic environment was monitored with 40 automatic underwater acoustic recorders, nine of which could transmit real-time acoustic information. The system recorded not merely seismic impulses but the wider industrial soundscape: seismic vessels, ordinary vessel traffic, pile driving and other activity. (Springer)
This matters.
The 2015 controversy had originally been framed largely around seismic airguns.
The eventual scientific record showed that a whale inhabiting this coastline did not experience industrial activities one corporate project at a time.
It experienced sound.
2. The first result: whale density fell as vessel noise accumulated
The density analysis attempted to separate industrial sound from the natural factors already known to influence where whales feed.
That was essential.
Whales do not distribute themselves randomly across the ocean.
Water depth mattered.
Distance from shore mattered.
The time of the season mattered.
Geographic position mattered.
And the energy available from important prey groups — including amphipods, isopods and cumaceans — also significantly influenced whale density. (Springer)
After accounting for those factors, however, another relationship remained.
As cumulative vessel sound exposure increased, gray-whale density in the affected area decreased.
The researchers detected a statistically significant relationship over both an eight-hour exposure period and a seven-dayexposure period.
Their interpretation was cautious but clear: the pattern was consistent with both relatively short-term responses and longer responses when vessel activity continued in the same area. (Springer)
That did not mean every whale left.
It did not mean the feeding ground was abandoned.
It meant that, after the other major explanatory variables were considered, fewer whales were present in places experiencing greater accumulated vessel noise.
That is a much narrower proposition.
It is also scientifically significant.
3. The second result: seismic exposure produced a similar pattern
The researchers then examined seismic sound separately.
Again, the result was not an all-or-nothing displacement.
At the shorter timescale, gray-whale density declined slightly as cumulative seismic exposure over the preceding two hours increased.
The decline was greater when exposure over the preceding three days was considered.
Extending the analysis to seven days did not explain significantly more variation than the three-day measure.
The authors therefore interpreted the result as evidence that whales could respond on both shorter and longer timescales — while also potentially returning after periods of higher exposure had ended. (Springer)
This is the evidence behind the headline of this instalment.
The whales moved.
But that sentence needs its qualification attached.
They were not shown to have abandoned Sakhalin.
Some remained within the nearshore feeding area even during periods of relatively high sound exposure.
Others appear to have reduced their use of particular areas and later returned. (Springer)
“Local displacement” is therefore closer to what the data support than “flight from the feeding grounds.”
4. One whale demonstrated the process in real time
The behavioural study contains a particularly useful case example.
Observers were tracking a gray whale that was feeding while a seismic source vessel approached.
As the vessel turned towards the whale’s direction, the animal stopped its original feeding pattern and began travelling towards shore.
When the vessel subsequently moved farther away, the whale resumed feeding.
But it did so in a different area from the one it had originally occupied. (Springer)
One animal does not establish a population effect.
The researchers did not pretend that it did.
What the observation supplies is something more modest but valuable: a visible example of the type of response that the statistical analysis was detecting across the much larger dataset.
The whale did not strand.
It did not disappear permanently.
It altered what it was doing while an industrial source approached.
Then it resumed feeding elsewhere.
5. The whales were also breathing differently
Distribution was only half the story.
The five behavioural teams were recording how individual whales moved and breathed.
Natural behaviour remained the dominant explanation for much of the variation. Feeding whales behave differently from travelling whales, and water depth affects diving and respiration.
But industrial variables still accounted for statistically significant changes.
As vessels approached, whale respiration intervals shortened, surface blow rates increased and dive times decreased.
The approach of seismic vessels was also associated with changes in surface behaviour.
In the movement data, whales increased speed, range and distance from shore as seismic vessels approached, and tended to move more perpendicular to the approaching vessels.
With increasing seismic sound exposure, whales appeared to move closer to shore.
Increasing continuous vessel sound was associated with more reorientation and reduced directionality. (Springer)
The paper’s broader interpretation was straightforward: vessel distance and sound exposure significantly influenced movement and respiration. Whales tended to breathe faster and move faster when vessels were closer or sound exposure was greater. (Springer)
Again, none of that establishes physical injury.
It establishes behavioural response.
That was precisely one of the phenomena the mitigation programme was intended to reduce.
6. The researchers said mitigation did not eliminate the response
This is perhaps the most important sentence in the 2022 behavioural paper.
The researchers considered that the mitigation measures may have reduced larger or longer-term responses.
But they concluded that mitigation “did not eliminate behavioral responses” in the short term. (Springer)
That conclusion deserves to be preserved without embellishment in the historical record.
It does not say mitigation failed.
A mitigation regime can reduce an effect without reducing it to zero.
It does not say the whales suffered population-level harm.
Behavioural disturbance and demographic damage are different questions.
And it does not contradict the conclusion that operational mitigation measures were implemented.
It tells us something the operational record alone could never reveal:
compliance with the mitigation system did not mean the whales behaved as though the industrial activity was absent.
7. The 2015 WGWAP warning now looks prescient — but only up to a point
Seven years earlier, before the seismic season began, the Western Gray Whale Advisory Panel had warned about precisely this possibility.
Its formal Statement of Concern of 8 May 2015 arose after WGWAP concluded that the combined scale of planned seismic activity was exceptional and that the Panel lacked some information it considered necessary for final assessment.
The Panel’s concern included the possibility that cumulative acoustic exposure could cause whales to move away from preferred feeding areas. It urged serious consideration of postponing at least some seismic work. (IUCN Cetacean Specialist Group)
The 2022 density results are directionally consistent with that warning.
Higher cumulative industrial sound was associated with lower local whale density.
But it would be wrong to turn that consistency into retrospective certainty.
WGWAP’s pre-survey concern involved possible consequences across the combined industrial programme.
The later studies found measurable relationships within the actual 2015 dataset.
They did not establish that every predicted consequence occurred.
Nor did they establish that the whales permanently vacated preferred habitat.
That distinction is important if this archive is to remain documentary rather than polemical.
8. Sakhalin Energy was not responsible for the entire acoustic environment
The scientific papers reinforce another distinction already made in earlier instalments.
The 2015 season involved two oil and gas operators and up to four seismic source vessels operating across several licence blocks.
Sakhalin Energy conducted its Piltun-Astokh programme.
Exxon Neftegas conducted separate seismic operations.
The monitoring programme deliberately captured sounds and whale responses associated with activity beyond a single company boundary. (Springer)
Ordinary vessel traffic, fishing vessels, research vessels and support operations also contributed to cumulative sound exposure. (Springer)
The later results therefore cannot responsibly be expressed as:
“Shell’s survey caused all the whale movement observed in 2015.”
The data do not establish that.
They establish associations between whale responses and the combined acoustic and vessel environment during a season of unusually intensive offshore activity.
Sakhalin Energy was part of that environment.
It was not all of it.
9. A paradox: the better the dataset became, the more effects scientists could detect
There is an important methodological lesson buried in these papers.
Earlier analysis of the 2010 seismic operation had found little or no measurable behavioural response.
That result had helped support a highly positive public account of the Sakhalin mitigation model.
But the 2022 behavioural paper pointed out a limitation in those earlier studies: the datasets had been much smaller and therefore had limited statistical power to detect subtle or moderate effects. (Springer)
The 2015 programme changed that.
There were more vessels.
There was more seismic activity.
There were many more observations.
And the study had substantially greater ability to detect changes that a smaller dataset might miss.
The result was scientifically awkward but important.
Better monitoring did not merely validate mitigation.
It revealed responses that less powerful monitoring had struggled to measure.
That is not a failure of science.
It is exactly what more informative science is supposed to do.
10. But whale density was also being shaped by food
Any attempt to attribute every movement in 2015 to industrial sound would be contradicted by the same research.
Whales congregated strongly around the mouth of Piltun Bay and in other areas associated with favourable feeding conditions.
Their density increased through much of the early season before declining after early August.
And density was significantly related to the energetic value of several prey groups. (Springer)
Other research in the same 2022 collection found unusual use of areas farther offshore and considered whether concentrations of high-energy sand lance could help explain those distributions.
The modelling authors explicitly acknowledged that incomplete spatial and temporal information about prey limited the precision with which some habitat patterns could be reproduced. (Springer)
So there were at least two simultaneous forces shaping whale distribution.
Whales moved in response to what they wanted:
food.
And the evidence indicates that they also changed behaviour and local distribution in response to what they were exposed to:
industrial activity and noise.
The scientific challenge was separating one from the other.
11. Then comes the apparent contradiction: reproduction did not collapse
If behavioural disturbance reduced feeding opportunities, the ultimate conservation question was whether whales lost enough energy to affect survival or reproduction.
A separate 2022 study attempted to bridge that gap using a stochastic dynamic programming bioenergetics model.
The model allowed simulated pregnant whales to move among feeding areas, respond to disturbance and compensate by feeding elsewhere.
Its result was considerably more reassuring than the behavioural findings.
Under the actual 2015 acoustic-disturbance scenario, predicted reproductive success and broad habitat use were broadly similar with and without disturbance.
The researchers described the effect of disturbance on predicted reproductive rate as no-to-little. (Springer)
That finding is not inconsistent with whales changing behaviour.
It answers a different question.
A whale can interrupt feeding, travel to another location and resume feeding — exactly as the observed individual did — without necessarily losing enough seasonal energy to impair reproduction.
The bioenergetic model was designed to examine whether such compensatory behaviour could prevent short-term disturbance becoming a demographic consequence.
For 2015, it generally predicted that it could.
12. Fourteen calves appeared the next year
Field observations gave the modelling exercise a particularly important reality check.
Researchers identified 14 new calves in the Sakhalin feeding areas in 2016.
Eleven calves had been identified in 2015.
Depending on how the pool of females potentially capable of being pregnant in 2015 was defined, the study calculated an observed successful reproductive rate ranging from approximately 0.64 to 0.88. (Springer)
That is a powerful counterweight to any claim that the 2015 seismic season demonstrably caused widespread reproductive failure.
It did not.
At least 14 calves were observed the following season.
But that fact cannot be turned into proof that disturbance was irrelevant either.
The modelling study itself stresses uncertainty about the number and identity of reproductive females and limitations in assumptions about habitat use and prey distribution. (Springer)
Science again refuses the convenient binary answer.
13. “No population effect proved” is not the same as “no biological cost”
The behavioural researchers went further than simply documenting movement.
They considered what repeated interruption could mean if it accumulated.
Earlier bioenergetics work had suggested that sufficient loss of feeding opportunity could, in theory, affect pregnancy success.
But the 2022 behavioural paper explicitly said it remained unknown how often particular individuals — including pregnant females — were repeatedly disturbed, and whether the observed responses resulted in biologically significant consequences for growth, survival or reproduction. (Springer)
That is the boundary of the evidence.
Inside the boundary:
behaviour changed.
local density changed.
industrial sound was statistically associated with those changes.
Outside the boundary:
proof that the 2015 programme caused population decline.
proof that whales suffered reproductive failure.
proof of lasting physiological injury.
The distinction is not semantic.
It is the difference between an observed response and a demonstrated conservation consequence.
14. The population was recovering at the same time
In September 2016, IUCN published what, on its face, was very good news.
It reported that the population feeding off the Russian Far East had grown at approximately 3–4% per year, from an estimated 115 animals in 2004 to 174 in 2015.
IUCN also credited Sakhalin Energy with making important efforts during the preceding 12 years to limit the effects of its operations, while warning that industry activity remained a threat. (IUCN)
In 2018, the population’s IUCN Red List classification was changed from Critically Endangered to Endangered.
IUCN reiterated the 3–4% annual growth figure when it renewed the Western Gray Whale Advisory Panel in 2019, while stressing that continued cooperation across oil and gas, fisheries, shipping and other industries remained necessary. (IUCN)
That long-term recovery is real and important.
But it does not erase the 2015 behavioural findings.
A population can increase while individuals still respond to industrial disturbance.
Equally, evidence of behavioural response does not prove that the disturbance prevented population recovery.
Both statements can be true.
That is why population trajectory is a poor substitute for direct impact monitoring.
15. The Red List change itself needs a qualification
There is another reason for caution.
The 2022 behavioural study noted that the later IUCN status assessment incorporated information from both Sakhalin and Kamchatka.
It also recorded that the population would still have met the more severe threshold if assessment were restricted to animals observed only off Sakhalin because the number of reproductive females remained low. (Springer)
So the 2018 change from Critically Endangered to Endangered was significant.
It was not a declaration that the population was secure.
Nor did it mean that the feeding grounds had ceased to require protection.
IUCN continued the WGWAP process and, in 2018, separately warned that western gray whales faced very high entanglement risk from coastal salmon nets as well as continuing pressures from oil and gas development. (IUCN)
The conservation problem had expanded beyond Shell long before the Shell-related problem disappeared.
16. Shell’s own 2015 report was more careful than a “no impact” claim
Royal Dutch Shell plc’s authenticated Sustainability Report 2015 deserves to be read alongside the later papers.
Shell said it and IUCN had worked together since 2004 to minimise the effects of operations on western gray whales.
It said that, under WGWAP guidance, the company had worked to reduce the effects of operations on whales and their habitat.
Shell highlighted the fact that Sakhalin Energy was the only energy company operating at Sakhalin in 2015 with an IUCN independent observer on the team implementing seismic monitoring and mitigation.
Crucially, the report said the Panel would continue to examine the impact of oil and gas development following the 2015 season. (Shell)
Those verbs matter.
Minimise.
Reduce.
Examine.
Shell’s own authenticated report did not say that the operation created no disturbance.
The later science shows why such wording would have been difficult to sustain.
17. The research itself was peer reviewed — but it was not financially detached from industry
The 2022 papers should not be misdescribed as wholly independent research funded without operator involvement.
The behavioural study acknowledges financial, logistical and safety-management support from Exxon Neftegas Limited.
One of its authors, the late Rodger H. Melton, was affiliated with ExxonMobil Exploration Company. (Springer)
The density paper makes the same acknowledgement of Exxon Neftegas financial, logistical and safety support and likewise lists Melton’s ExxonMobil affiliation. (Springer)
The broader paper describing the 2015 mitigation and impact-assessment programme states that Exxon Neftegas funded preparation of that publication, and several authors were affiliated with ExxonMobil. It also records anonymous peer review and editorial review. (Springer)
Those facts do not invalidate the findings.
They belong in the provenance record.
Indeed, they make one feature of the published results especially noteworthy.
Research supported by an operator did not conclude that mitigation had eliminated behavioural responses.
The papers reported statistically detectable effects.
That is precisely why research provenance should be disclosed rather than used automatically either to dismiss or to sanctify a result.
18. WGWAP’s role also needs to be kept distinct
The Western Gray Whale Advisory Panel was not a court.
It was not the Russian environmental regulator.
And it did not possess legal power to prohibit the 2015 surveys.
It was an IUCN-convened scientific advisory mechanism.
The recovered WGWAP archive shows that the process eventually ran for 17 years and produced more than 600 formal recommendations, principally concerning the oil and gas industry and associated regulators. Its final meeting took place in November 2021. (IUCN Cetacean Specialist Group)
An independent evaluation later recorded that WGWAP’s May 2015 Statement of Concern represented one of those occasions when ordinary individual recommendations were considered insufficient to express the Panel’s level of concern. (IUCN Cetacean Specialist Group)
That history makes the 2022 science particularly valuable.
It provides evidence against which the Panel’s precautionary warnings can actually be tested.
Not vindicated in every detail.
Not disproved.
Tested.
19. The old English court record still supplies the historical backdrop
The British High Court had encountered the gray-whale issue years before the 2015 seismic season.
In Export Credits Guarantee Department v Friends of the Earth [2008] EWHC 638 (Admin), Mr Justice Mitting recorded that proposed British financial support concerned a Sakhalin offshore development in which Shell had originally held a majority interest.
The judgment noted the potentially serious consequences of the development for western gray-whale habitat and feeding grounds.
The actual case concerned access to environmental information surrounding proposed UK export-credit support, not scientific adjudication of seismic effects. (vLex)
The court did not determine whether the 2015 surveys were safe.
It did not establish acoustic thresholds.
It did not find that Shell caused whale injury.
Its relevance here is narrower and historical.
By the time scientists began collecting the extraordinary 2015 dataset, the importance of western gray-whale habitat had already been part of British governmental and judicial records for years.
Documentary Findings
Established
The 2015 Sakhalin season produced an unusually large dataset covering whale distribution, individual behaviour, acoustic exposure and prey conditions. The density study used more than 10,000 shore-based scans; the behavioural study analysed 1,270 tracks and 401 focal follows; acoustic monitoring included 40 underwater recorders. (Springer)
After accounting for important natural and spatial variables, higher cumulative vessel sound exposure was statistically associated with lower local whale density over both shorter and longer exposure windows. Higher cumulative seismic exposure was likewise associated with lower density, with relationships detected over two-hour and three-day periods. (Springer)
Individual whales also altered movement and respiration in relation to vessel proximity and sound exposure. The researchers documented changes including speed, direction, reorientation, respiration interval, surface blow rate and dive behaviour. (Springer)
The behavioural paper concluded that the mitigation regime did not eliminate short-term behavioural responses. (Springer)
Separate bioenergetic modelling nevertheless found no-to-little predicted effect on reproductive rate under the 2015 disturbance scenario, and 14 calves were identified during the following feeding season in 2016. (Springer)
The wider population was increasing during this period: IUCN reported an estimated rise from 115 animals in 2004 to 174 in 2015, approximately 3–4% per year, and the population was subsequently reclassified from Critically Endangered to Endangered in 2018. (IUCN)
Scientifically Observed — Causation Limits
The density relationships are statistical associations derived from observational data in a complex real-world environment, not a controlled laboratory experiment.
The researchers attempted to account for major natural influences including season, geography, water depth and prey availability. Those variables themselves explained substantial portions of whale distribution. (Springer)
Multiple seismic programmes, support vessels, fishing vessels and other maritime activities contributed to the 2015 soundscape.
Consequently, it would be incorrect to attribute every documented behavioural or distributional response specifically to Sakhalin Energy.
Likewise, reduced local density should not be converted into a claim that whales permanently abandoned their Sakhalin feeding grounds.
The data explicitly show that some whales remained during relatively high exposure and that animals could return after exposure diminished. (Springer)
Shell and Industry Position in the Documentary Record
Shell’s 2015 Sustainability Report presented the IUCN relationship as an environmental partnership intended to minimise impacts, emphasised WGWAP guidance and highlighted Sakhalin Energy’s use of an IUCN Independent Observer. It also acknowledged that scientific examination of oil and gas impacts would continue. (Shell)
IUCN itself subsequently gave Sakhalin Energy significant credit for efforts to reduce operational effects while simultaneously warning that industrial activity continued to pose risks to the recovering population. (IUCN)
The 2022 science does not justify rewriting either position into an absolute.
It does not show that mitigation was pointless.
It does show that mitigation did not make behavioural response disappear.
Not Established
It is not established that the 2015 seismic season caused a decline in the western gray-whale population.
It is not established that the 2015 activity caused widespread reproductive failure.
It is not established by these studies that seismic exposure caused permanent hearing damage or other physical injury to the whales.
It is not established that Sakhalin Energy or Shell was responsible for all of the industrial sound to which animals were exposed.
It is not established that the later increase in population occurred because of WGWAP, Shell or Sakhalin Energy’s mitigation measures.
And the population’s improvement in conservation status does not establish that industrial disturbance was biologically irrelevant.
Commentary
This may be one of the most instructive files in the entire Sakhalin archive.
For years the public argument tended towards two incompatible narratives.
One said that sophisticated mitigation demonstrated environmentally responsible offshore development.
The other said that industrial seismic activity beside the feeding grounds of an endangered whale population was inherently unacceptable.
The 2015 evidence does not fit comfortably into either narrative.
The mitigation system mattered.
Shutdown criteria mattered.
Scheduling mattered.
Acoustic modelling mattered.
Observers mattered.
The extensive monitoring programme mattered enormously.
And yet the whales still responded.
That is not surprising.
Mitigation is not a force field.
Its purpose is to reduce risk and disturbance, not to make a seismic vessel acoustically invisible.
The more interesting question is what happened after the response.
If a whale interrupted feeding, moved a short distance and successfully compensated later, the demographic cost might be negligible.
If the same whale were repeatedly disturbed during a constrained feeding season and could not recover the lost energy, the consequence could be much greater.
The 2015 data recorded the first part of that chain with considerable precision.
The modelling attempted to calculate the second.
For that particular year, the population-level picture was reassuring.
Reproduction did not collapse.
The population continued its wider recovery.
But a reassuring demographic outcome does not retrospectively convert measurable disturbance into no disturbance.
That is the central documentary point.
There is also an institutional lesson.
A company could have commissioned only enough monitoring to demonstrate compliance.
Instead, the combined 2015 research programme generated enough information to demonstrate that compliance and biological response can coexist.
That is scientifically valuable.
It is also uncomfortable for corporate communications departments, because “the mitigation plan was implemented” fits neatly into a sustainability report while “the animals still changed their behaviour, but we cannot demonstrate a population-level consequence” does not fit quite so neatly.
Yet the second formulation is closer to what the evidence ultimately showed.
There is one final irony.
WGWAP had been concerned before the survey that whales might move away from preferred feeding areas.
Years later, peer-reviewed analysis did indeed detect reductions in local density associated with cumulative industrial sound.
But those same years also produced a growing whale population and enough calves to make a simple catastrophe narrative untenable.
The scientists had not discovered that one side was right and the other wrong.
They had discovered something harder:
impact is not a binary condition.
A whale can be disturbed without being killed.
A feeding area can experience displacement without being abandoned.
A mitigation programme can work without eliminating response.
And a recovering population can still require protection from cumulative industrial pressure.
That is what the 2015 experiment eventually revealed.
Source Record
The principal behavioural paper is Glenn Gailey et al., “Western gray whale behavioral response to seismic surveys during their foraging season,” Environmental Monitoring and Assessment, Volume 194, Article 740, published 18 October 2022. It records the 1,270 tracks, 401 focal follows, acoustic exposure data, statistically significant behavioural responses and the conclusion that mitigation did not eliminate short-term behavioural response. (Springer)
The principal distribution paper is Glenn Gailey et al., “Gray whale density during seismic surveys near their Sakhalin feeding ground,” Environmental Monitoring and Assessment, Volume 194, Article 739, published 18 October 2022. It analyses more than 10,000 shore-based scans and reports declining local whale density associated with increased cumulative seismic and vessel sound after controlling for major natural and habitat variables. (Springer)
Gailey et al. — Gray whale density during seismic surveys near their Sakhalin feeding ground
The supporting bioenergetics paper is Lisa Schwarz et al., “Gray whale habitat use and reproductive success during seismic surveys near their feeding grounds: comparing state-dependent life history models and field data,” Environmental Monitoring and Assessment, Volume 194, Article 733, published 18 October 2022. It found broadly similar reproductive-success predictions with and without the 2015 acoustic-disturbance scenario and compared those predictions with field observations, including the 14 calves identified in 2016. (Springer)
Schwarz et al. — Gray whale habitat use and reproductive success during seismic surveys
The programme-level scientific paper is Lisanne Aerts et al., “Seismic surveys near gray whale feeding areas off Sakhalin Island, Russia: assessing impact and mitigation effectiveness,” Environmental Monitoring and Assessment, 2022. It documents the 2015 mitigation architecture, monitoring strategy and the relationship between behavioural, distributional, acoustic and bioenergetic studies. (Springer)
Aerts et al. — Assessing impact and mitigation effectiveness in the 2015 Sakhalin surveys
The contemporaneous warning is the Western Gray Whale Advisory Panel Statement of Concern with respect to proposed seismic activity on the Sakhalin shelf in 2015, dated 8 May 2015. It records WGWAP’s information concerns and its scientific apprehension about the cumulative scale of the planned seismic season. (IUCN Cetacean Specialist Group)
WGWAP — Statement of Concern, 8 May 2015
Sakhalin Energy’s contemporaneous response records the company’s operational reasons for proceeding, its engagement with WGWAP and the purpose of the Piltun-Astokh 4-D survey. (IUCN Cetacean Specialist Group)
Sakhalin Energy — Response to the May 2015 WGWAP Statement
Royal Dutch Shell plc’s authenticated Sustainability Report 2015 records Shell’s relationship with IUCN, its stated aim of minimising impacts on western gray whales and Sakhalin Energy’s use of an IUCN Independent Observer during the 2015 seismic programme. (Shell)
Royal Dutch Shell plc — Sustainability Report 2015
IUCN’s 3 September 2016 statement records the estimated population increase from 115 animals in 2004 to 174 in 2015 while warning that industrial activity continued to pose a threat. (IUCN)
IUCN — Western gray whales recovering, but industry still poses a threat, 3 September 2016
IUCN’s 24 April 2019 announcement records the 2018 change in conservation status from Critically Endangered to Endangered and the continuation of WGWAP work on cumulative impacts and underwater noise. (IUCN)
IUCN — Scientific panel to continue advising on endangered gray whales, 24 April 2019
The historical judicial record remains Export Credits Guarantee Department v Friends of the Earth [2008] EWHC 638 (Admin). It is used here only to establish that the sensitivity of western gray-whale habitat formed part of the British governmental and legal record surrounding Sakhalin II years before the 2015 survey. (vLex)
Export Credits Guarantee Department v Friends of the Earth — [2008] EWHC 638 (Admin)
Research-provenance note: Several papers in the 2022 special collection acknowledge financial, logistical or safety support from Exxon Neftegas Limited, and some authors held ExxonMobil affiliations. This is disclosed here as part of the provenance of the evidence. The papers were published in a peer-reviewed scientific journal and, importantly, reported measurable whale responses rather than concluding that industrial effects were absent. (Springer)
Archive disclaimer: The evidence distinguishes detectable behavioural and distributional responses from demonstrated population-level injury. No court judgment cited here determines the biological consequences of the 2015 seismic programme, and no population trend is attributed solely to Shell, Sakhalin Energy, Exxon Neftegas, WGWAP or any single conservation measure.
Site wide disclaimer also applies.
Next Archive File
SLF-2007-055 — The Sakhalin Papers XLV: From Critically Endangered to Endangered — Who Gets Credit for the Recovery?
In September 2016, IUCN announced a striking figure.
115 whales in 2004.
174 in 2015.
An estimated annual increase of 3–4%. (IUCN)
Two years later, the conservation classification changed from Critically Endangered to Endangered.
For Shell and IUCN, the Sakhalin collaboration could now be presented as evidence that independent scientific oversight and industrial development were capable of coexisting.
But the story was less simple than the headline.
The Red List reassessment drew upon a changing understanding of whales seen away from Sakhalin.
The number of reproductive females remained small.
Fishing-net entanglement emerged as a major threat.
Other oil companies were not bound into the WGWAP relationship in the same way as Sakhalin Energy.
And population recovery itself could not establish how much credit belonged to the Panel, to Shell’s mitigation measures, to international whaling protection, to changing survey knowledge — or simply to the biology of a remnant population recovering from historic exploitation.
SLF-2007-055 will examine the celebrated recovery claim, the evidence behind the Red List change, what WGWAP could legitimately claim to have achieved, and the limits of using a growing whale population as a corporate environmental success story. (IUCN)
*This website and sisters royaldutchshellgroup.com, shellnazihistory.com, royaldutchshell.website, johndonovan.website, shellnews.net, and shellwikipedia.com, are owned by John Donovan - more information here. There is also a Wikipedia segment, the Shell DPA Files, "Shell and the Spies", the Shell Leaks files, as well as books written and published by John Donovan - Kindle eBooks. Timeline of the Donovan Shell Feud. Toxic History of Royal Dutch Shell Group. Shell and the Donovans: The Full Media Record — 550+ Articles, 110 Books, 40 Years. *All created and supported by internet wizz, Nick Gill.
