Route du Rhum 2018 takes off with flying foiling monsters

Jeremie Beyou with 'Charal'

Alex Thomson with 'Hugo Boss'

François Gabard with 'Macif'

Armel Le Cléac'h with 'Banque Populaire IX'

Sébastien Josse with 'Gitana 17' 

Francis Joyon with 'IDEC'

Links :

• Route du Rhum 2018 website

Startling new research finds large buildup of heat in the oceans, suggesting a faster rate of global warming

A post-sunset swimmer at Moonlight Beach in Encinitas, Calif., this month.

(Mike Blake/Reuters)

From WashingtonPost by Chris Mooney and Brady Dennis

The findings mean the world might have less time to curb carbon emissions.

The world’s oceans have been soaking up far more excess heat in recent decades than scientists realized, suggesting that Earth could be set to warm even faster than predicted in the years ahead, according to new research published Wednesday.

Over the past quarter-century, Earth’s oceans have retained 60 percent more heat each year than scientists previously had thought, said Laure Resplandy, a geoscientist at Princeton University who led the startling study published Wednesday in the journal Nature.
The difference represents an enormous amount of additional energy, originating from the sun and trapped by Earth’s atmosphere — the yearly amount representing more than eight times the world’s annual energy consumption.

 www.argo.ucsd.edu

In the scientific realm, the new findings help resolve long-running doubts about the rate of the warming of the oceans before 2007, when reliable measurements from devices called “Argo floats” were put to use worldwide.
Before that, differing types of temperature records — and an overall lack of them — contributed to murkiness about how quickly the oceans were heating up.

The higher-than-expected amount of heat in the oceans means more heat is being retained within Earth’s climate system each year, rather than escaping into space.
In essence, more heat in the oceans signals that global warming is more advanced than scientists thought.

“We thought that we got away with not a lot of warming in both the ocean and the atmosphere for the amount of CO2 that we emitted,” said Resplandy, who published the work with experts from the Scripps Institution of Oceanography and several other institutions in the United States, China, France and Germany.
“But we were wrong. The planet warmed more than we thought. It was hidden from us just because we didn’t sample it right. But it was there. It was in the ocean already.”

The United Nations panel on climate issued a report warning of unprecedented temperature rise between 2030 and 2052 if global warming continues.

(Reuters)

Wednesday’s study also could have important policy implications.
If ocean temperatures are rising more rapidly than previously calculated, that could leave nations even less time to dramatically cut the world’s emissions of carbon dioxide, in the hope of limiting global warming to the ambitious goal of 1.5 degrees Celsius (2.7 degrees Fahrenheit) above preindustrial levels by the end of this century.

The world already has warmed one degree Celsius (1.8 degrees Fahrenheit) since the late 19th century.
Scientists backed by the United Nations reported this month that with warming projected to steadily increase, the world faces a daunting challenge in trying to limit that warming to only another half-degree Celsius.
The group found that it would take “unprecedented” action by leaders across the globe over the coming decade to even have a shot at that goal.

Meanwhile, the Trump administration has continued to roll back regulations aimed at reducing carbon emissions from vehicles, coal plants and other sources and has said it intends to withdraw from the Paris climate accord.
In one instance, the administration relied on an assumption that the planet will warm a disastrous seven degrees Fahrenheit, or about four degrees Celsius, by the end of the century in arguing that a proposal to ease vehicle fuel-efficiency standards would have only minor climate impacts.

 Ocean heat content data.

Illustration: USGCRP report, originally from Cheng et al., 2017

The new research underscores the potential consequences of global inaction.
Rapidly warming oceans mean that seas will rise faster and that more heat will be delivered to critical locations that already are facing the effects of a warming climate, such as coral reefs in the tropics and the ice sheets of Greenland and Antarctica.

“In case the larger estimate of ocean heat uptake turns out to be true, adaptation to — and mitigation of — our changing climate would become more urgent,” said Pieter Tans, who is the leader of the Carbon Cycle Greenhouse Gases Group at the National Oceanic and Atmospheric Administration and was not involved in the study.

The oceans absorb more than 90 percent of the excess energy trapped within the world’s atmosphere.

The new research does not measure the ocean’s temperature directly.
Rather, it measures the volume of gases, specifically oxygen and carbon dioxide, that have escaped the ocean in recent decades and headed into the atmosphere as it heats up.
The method offered scientists a reliable indicator of ocean temperature change because it reflects a fundamental behavior of a liquid when heated.

“When the ocean warms, it loses some gas to the atmosphere,” Resplandy said.
“That’s an analogy that I make all the time: If you leave your Coke in the sun, it will lose the gas.”

This approach allowed researchers to recheck the contested history of ocean temperatures in a different and novel way.
In doing so, they came up with a higher number for how much warming the oceans have experienced over time.

Quantification of ocean heat uptake from changes in atmospheric O

“I feel like this is a triumph of Earth-system science.
That we could get confirmation from atmospheric gases of ocean heat content is extraordinary,” said Joellen Russell, a professor and oceanographer at the University of Arizona.
“You’ve got the A team here on this paper.”

But Russell said the findings are hardly as uplifting.

The report “does have implications for climate sensitivity, meaning, how warm does a certain amount of CO2 make us?” Russell said, adding that the world could have a smaller “carbon budget” than once thought.
That budget refers to the amount of carbon dioxide humans can emit while still being able to keep warming below dangerous levels.

The scientists calculated that because of the increased heat already stored in the ocean, the maximum emissions that the world can produce while still avoiding a warming of two degrees Celsius (3.6 Fahrenheit) would have to be reduced by 25 percent.
That represents a very significant shrinkage of an already very narrow carbon “budget.”

The U.N. panel of climate scientists said recently that global carbon emissions must be cut in half by 2030 if the world hopes to remain beneath 1.5 Celsius of warming.
But Resplandy said that the evidence of faster-warming oceans “shifts the probability, making it harder to stay below the 1.5-degree temperature target."

Understanding what is happening with Earth’s oceans is critical, because they, far more than the atmosphere, are the mirror of ongoing climate change.

According to a major climate report released last year by the U.S. government, the world’s oceans have absorbed about 93 percent of the excess heat caused by greenhouse gases since the mid-20th century.
Scientists have found that ocean heat has increased at all depths since the 1960s, while surface waters also have warmed.
The federal climate report projected a global increase in average sea surface temperatures of as much as nearly five degrees Fahrenheit by 2100 if emissions continue unabated, with even higher levels of warming in some U.S. coastal regions.

Image: NASA Goddard Space Flight Center

The world’s oceans also absorb more than a quarter of the carbon dioxide emitted annually from human activities — an effect making them more acidic and threatening fragile ecosystems, federal researchers say.
“The rate of acidification is unparalleled in at least the past 66 million years,” the government climate report stated.

Paul Durack, a research scientist at the Lawrence Livermore National Laboratory in California, said Wednesday’s study offers “a really interesting new insight” and is “quite alarming.”

The warming found in the study is “more than twice the rates of long-term warming estimates from the 1960s and ’70s to the present,” Durack said, adding that if these rates are validated by further studies, “it means the rate of warming and the sensitivity of the Earth’s system to greenhouse gases is at the upper end.”
He said that if scientists have underestimated the amount of heat taken up by the oceans, “it will mean we need to go back to the drawing board” on the aggressiveness of mitigation actions the world needs to take promptly to limit future warming.

Beyond the long-term implications of warmer oceans, Russell added that in the short term, even small changes in ocean temperatures can affect weather in specific places.
For instance, scientists have said warmer oceans off the coast of New England have contributed to more-intense winter storms.

“We’re only just now discovering how important ocean warming is to our daily lives, to our daily weather,” she said.

Links :

• EcoWatch : Oceans Have Absorbed 60% More Heat Than Scientists Thought
• Business Insider : Earth's oceans may be way hotter than scientists realized — here's how worried you should be 
• ScienceDaily : Earth's oceans have absorbed 60 percent more heat than previously thought
• BBC : Climate change: Oceans 'soaking up more heat than estimated'
• Motherboard : Earth’s Oceans Have Absorbed Far More Heat Than Experts Believed, Study Claims
• CNN : World's oceans have absorbed 60% more heat than previously thought, study finds
• Wired : The Sea May Be Absorbing Way More Heat Than We Thought 

Maps of Clipperton island or Isle de la Passion

First map of the island Clipperton, then "Isle de la Passion",

after the French landing on the atoll in April 1711

by vessels of Mr Piecourt coming from China to Peru

source Gallica BNF

  Localization of Clipperton island with the GeoGarage platform (SHOM nautical chart)

Clipperton island with the GeoGarage platform (SHOM chart)

 Clipperton island with the GeoGarage platform (NGA chart based on SHOM 1937)

DEM 20 m / SHOM

 Aerial view (Ocean Polaire website)

 For their expedition to Clipperton, conducted in partnership with the University of French Polynesia, the Pristine Seas team shot hundreds of hours of underwater and topside footage for a documentary film.

Their scientific findings supported the creation of a marine reserve, announced at the September 2016 Our Ocean conference in Washington, D.C., that encompasses Clipperton’s territorial waters.

Links :

• Le Figaro : Clipperton, un atoll français paradisiaque noyé sous les déchets plastiques
• Ile de clipperton (Christian Jost blog) : cartes anciennes
• YouTube : Clipperton Island from above - Aerial Video of Ile de la Passion 

Acoustic detection of humpback whales using a convolutional neural network

HARP deployment locations. Green: sites with currently active recorders.

Red: previous recording sites.

From Google by Matt Harvey

Over the last several years, Google AI Perception teams have developed techniques for audio event analysis that have been applied on YouTube for non-speech captions, video categorizations, and indexing.
Furthermore, we have published the AudioSet evaluation set and open-sourced some model code in order to further spur research in the community.
Recently, we’ve become increasingly aware that many conservation organizations were collecting large quantities of acoustic data, and wondered whether it might be possible to apply these same technologies to that data in order to assist wildlife monitoring and conservation.

As part of our AI for Social Good program, and in partnership with the Pacific Islands Fisheries Science Center of the U.S.
National Oceanic and Atmospheric Administration (NOAA), we developed algorithms to identify humpback whale calls in 15 years of underwater recordings from a number of locations in the Pacific.
The results of this research provide new and important information about humpback whale presence, seasonality, daily calling behavior, and population structure.
This is especially important in remote, uninhabited islands, about which scientists have had no information until now.
Additionally, because the dataset spans a large period of time, knowing when and where humpback whales are calling will provide information on whether or not the animals have changed their distribution over the years, especially in relation to increasing human ocean activity.
That information will be a key ingredient for effective mitigation of anthropogenic impacts on humpback whales.

Passive Acoustic Monitoring and the NOAA HARP Dataset

Passive acoustic monitoring is the process of listening to marine mammals with underwater microphones called hydrophones, which can be used to record signals so that detection, classification, and localization tasks can be done offline.
This has some advantages over ship-based visual surveys, including the ability to detect submerged animals, longer detection ranges and longer monitoring periods.
Since 2005, NOAA has collected recordings from ocean-bottom hydrophones at 12 sites in the Pacific Island region, a winter breeding and calving destination for certain populations of humpback whales.

The data was recorded on devices called high-frequency acoustic recording packages, or HARPs (Wiggins and Hildebrand, 2007; full text PDF).
In total, NOAA provided about 15 years of audio, or 9.2 terabytes after decimation from 200 kHz to 10kHz.
(Since most of the sound energy in humpback vocalizations is in the 100Hz-2000Hz range, little is lost in using the lower sample rate.)

From a research perspective, identifying species of interest in such large volumes of data is an important first stage that provides input for higher-level population abundance, behavioral or oceanographic analyses.
However, manually marking humpback whale calls, even with the aid of currently available computer-assisted methods, is extremely time-consuming.

Supervised Learning: Optimizing an Image Model for Humpback Detection

We made the common choice of treating audio event detection as an image classification problem, where the image is a spectrogram — a histogram of sound power plotted on time-frequency axes.

Example spectrograms of audio events found in the dataset, with time on the x-axis and frequency on the y-axis.

Left: a humpback whale call (in particular, a tonal unit), Center: narrow-band noise from an unknown source, Right: hard disk noise from the HARP

This is a good representation for an image classifier, whose goal is to discriminate, because the different spectra (frequency decompositions) and time variations thereof (which are characteristic of distinct sound types) are represented in the spectrogram as visually dissimilar patterns.
For the image model itself, we used ResNet-50, a convolutional neural network architecture typically used for image classification that has shown success at classifying non-speech audio.
This is a supervised learning setup, where only manually labeled data could be used for training (0.2% of the entire dataset — in the next section, we describe an approach that makes use of the unlabeled data.)

The process of going from waveform to spectrogram involves choices of parameters and gain-scaling functions.
Common default choices (one of which was logarithmic compression) were a good starting point, but some domain-specific tuning was needed to optimize the detection of whale calls.
Humpback vocalizations are varied, but sustained, frequency-modulated, tonal units occur frequently in time.
You can listen to an example

If the frequency didn't vary at all, a tonal unit would appear in the spectrogram as a horizontal bar.
Since the calls are frequency-modulated, we actually see arcs instead of bars, but parts of the arcs are close to horizontal.

A challenge particular to this dataset was narrow-band noise, most often caused by nearby boats and the equipment itself.
In a spectrogram it appears as horizontal lines, and early versions of the model would confuse it with humpback calls.
This motivated us to try per-channel energy normalization (PCEN), which allows the suppression of stationary, narrow-band noise.
This proved to be critical, providing a 24% reduction in error rate of whale call detection.

Spectrograms of the same 5-unit excerpt from humpback whale song
beginning at 0:06 in the above recording.

Top: PCEN.

Bottom: log of squared magnitude.

The dark blue horizontal bar along the bottom under log compression has become much lighter relative to the whale call when using PCEN
Aside from PCEN, averaging predictions over a longer period of time led to much better precision.
This same effect happens for general audio event detection, but for humpback calls the increase in precision was surprisingly large.
A likely explanation is that the vocalizations in our dataset are mainly in the context of whale song, a structured sequence of units than can last over 20 minutes.
At the end of one unit in a song, there is a good chance another unit begins within two seconds.
The input to the image model covers a short time window, but because the song is so long, model outputs from more distant time windows give extra information useful for making the correct prediction for the current time window.

Overall, evaluating on our test set of 75-second audio clips, the model identifies whether a clip contains humpback calls at over 90% precision and 90% recall.
However, one should interpret these results with care; training and test data come from similar equipment and environmental conditions.
That said, preliminary checks against some non-NOAA sources look promising.

 A female humpback whale swims with a male (below) in Maui, Hawaii.

Long thought silent, females may have their own, low-frequency songs.

Unsupervised Learning: Representation for Finding Similar Song Units

A different way to approach the question, "Where are all the humpback sounds in this data?", is to start with several examples of humpback sound and, for each of these, find more in the dataset that are similar to that example.
The definition of similar here can be learned by the same ResNet we used when this was framed as a supervised problem.
There, we used the labels to learn a classifier on top of the ResNet output.
Here, we encourage a pair of ResNet output vectors to be close in Euclidean distance when the corresponding audio examples are close in time.
With that distance function, we can retrieve many more examples of audio similar to a given one.
In the future, this may be useful input for a classifier that distinguishes different humpback unit types from each other.

To learn the distance function, we used a method described in "Unsupervised Learning of Semantic Audio Representations", based on the idea that closeness in time is related to closeness in meaning.
It randomly samples triplets, where each triplet is defined to consist of an anchor, a positive, and a negative.
The positive and the anchor are sampled so that they start around the same time.
An example of a triplet in our application would be a humpback unit (anchor), a probable repeat of the same unit by the same whale (positive) and background noise from some other month (negative).
Passing the 3 samples through the ResNet (with tied weights) represents them as 3 vectors.
Minimizing a loss that forces the anchor-negative distance to exceed the anchor-positive distance by a margin learns a distance function faithful to semantic similarity.

Principal component analysis (PCA) on a sample of labeled points lets us visualize the results.
Separation between humpback and non-humpback is apparent.
Explore for yourself using the TensorFlow Embedding Projector.
Try changing Color by to each of class_label and site.
Also, try changing PCA to t-SNE in the projector for a visualization that prioritizes preserving relative distances rather than sample variance.

A sample of 5000 data points in the unsupervised representation.

(Orange: humpback / Blue: not humpback.)

Given individual "query" units, we retrieved the nearest neighbors in the entire corpus using Euclidean distance between embedding vectors.
In some cases we found hundreds more instances of the same unit with good precision.

Manually chosen query units (boxed) and nearest neighbors using the unsupervised representation.

We intend to use these in the future to build a training set for a classifier that discriminates between song units.
We could also use them to expand the training set used for learning a humpback detector.

Predictions from the Supervised Classifier on the Entire Dataset

We plotted summaries of the model output grouped by time and location.
Not all sites had deployments in all years.
Duty cycling (example: 5 minutes on, 15 minutes off) allows longer deployments on limited battery power, but the schedule can vary.
To deal with these sources of variability, we consider the proportion of sampled time in which humpback calling was detected to the total time recorded in a month:

Time density of presence on year / month axes for the Kona and Saipan sites.

The apparent seasonal variation is consistent with a known pattern in which humpback populations spend summers feeding near Alaska and then migrate to the vicinity of the Hawaiian Islands to breed and give birth.
This is a nice sanity check for the model.

We hope the predictions for the full dataset will equip experts at NOAA to reach deeper insights into the status of these populations and into the degree of any anthropogenic impacts on them.
We also hope this is just one of the first few in a series of successes as Google works to accelerate the application of machine learning to the world's biggest humanitarian and environmental challenges.
To find out how this project was started, read the NOAA Fisheries blog post by Research Oceanographer Ann Allen.

Links :

• ScienceDaily : Shifting presence of North Atlantic right whales tracked with passive acoustics / Whales use song as sonar, psychologist proposes
• National Geographic : Mysterious New Humpback Whale "Song" Detected?

Sulphur-emissions rules for shipping will worsen global warming

Carriage ban adopted to implement Sulphur 2020 limit

Owners planning to install ‘emissions cheat’ systems to avoid having to buy cleaner,

more expensive fuel 

 From The Economist

The IMO’s rules could also wipe 3% off America’s GDP

Dp world london gateway, a container terminal on the Thames estuary, is Britain’s fastest-growing port.
The borough of Thurrock, where the port is situated, has the country’s third-worst levels of air pollution, in part because of fumes spewed out by ships in the port.
Upriver, in London, the International Maritime Organisation (imo), the United Nations agency for shipping, began a meeting on October 22nd aimed at taking action against air pollution.
But new rules to make ships cleaner will impose crippling costs on the industry while worsening global warming.

The imo will cut emissions of sulphur either by reducing its content in marine fuel from 3.5% to 0.5% from 2020 or by requiring ships to remove it from exhaust fumes.
Sulphur from ships causes acid rain and air pollution, which contributes to between 212,000 and 595,000 premature deaths a year and 14m cases of childhood asthma, according to research published in Nature Communications in February.

Installing a scrubber is accepted by flag States

as an alternative means to meet the sulphur limit requirement.

Shipowners can meet the rules by installing “scrubbers”, but these are expensive.
Only around 2,000 of 90,000 commercial vessels on the world’s seas will have them by the deadline.
And dealing with sulphur has other ill-effects, points out Paddy Rodgers of Euronav, an oil-tanker firm.
The most popular system washes sulphur out of engine fumes with seawater, which is then chucked overboard.
The wastewater kills marine wildlife and causes cancer in humans, according to some research.

Most shipowners will switch to pricier low-sulphur fuels.
But if all ships did so in 2020, demand for them would double (see chart) and the industry’s fuel bill would rise by $60bn, roughly the entire sum spent in 2016, say analysts at Wood Mackenzie, a research firm.
It would also have a dramatic impact on aviation and road transport.
Ships run on a heavy residue that remains after petrol, diesel and other lighter hydrocarbons are extracted from crude oil in refining.
Competition for lighter fuel that clean ships require could raise the price of diesel for lorries by 50% and for jet fuel by 30-40% in 2020, reckons Philip Verleger, an energy economist.
The resulting spike in global transport costs, he says, would hit world trade and wipe a staggering 3% off America’s gdp and 1.5% off the whole world’s in 2020.

Big shipping lines such as Denmark’s Maersk, which can afford scrubbers or pricier fuels, support the changes.
Smaller firms could be forced to scrap older ships, says Basil Karatzas, a consultant.
Less capacity will mean higher rates for the rest.

Worse still is the effect of the new rules on global warming.
The Intergovernmental Panel on Climate Change, a un-backed body, says sulphur emissions have a net cooling effect because they scatter sunlight in the atmosphere.
Sulphur also helps to form and thicken clouds that reflect sunlight away from the Earth.

Some studies find that by burning heavy marine fuel the industry is slowing global warming, as the cooling effects of sulphur emissions outweigh the warming caused by those of carbon dioxide.
Scientists at the Centre for International Climate and Environmental Research in Oslo calculate that shipping in net terms reduced man-made warming by 7% in 2000.
The imo’s new rules will undo much of this effect.
The paper in Nature Communications found that the use of lower-sulphur fuels after 2020 will reduce the cooling effect from shipping by around 80%.

The imo does not accept that this might kill more people in the longer term than the number who succumb each year to air pollution.
“This is the imo's biggest impact in its 60-year history,” beams Kitack Lim, the organisation’s secretary-general.
Alas, for efforts to combat climate change, it is an impact of the wrong sort.

Links :

• The Guardian : Thousands of ships could dump pollutants at sea to avoid dirty fuel ban
• Lloyds : IMO adopts 2020 sulphur cap fuel carriage ban /  Cutting CO2 output is the biggest challenge facing shipping
• FT : Maritime body puts off proposed change to fuel sulphur rule / US seeks late change to sulphur-cap fuel rules
• Hellenic Shipping news : IMO: Implementation of sulphur 2020 limit – carriage ban adopted
• Bunkerspot : Concawe: full 0.50% sulphur marine fuel compliance not straightforward for Europe’s refiners
• GeoGarage blog : Guilt-Free on the Sea? How Norway is using oil and gas riches to engineer a future in emission / The shipping industry sets sail toward a carbon-free future / Lower emissions on the high seas /