Saturday, January 8, 2022
Friday, January 7, 2022
The danger of leaving weatherprediction to AI
From Wired by Meghan Herbst
When it comes to forecasting the elements, many seem ready to welcome the machine.
But humans still outperform the algorithms—especially in bad conditions.
HUMANS HAVE TRIED to anticipate the climate’s turns for millennia, using early lore—“red skies at night” is an optimistic sigil for weather-weary sailors that’s actually associated with dry air and high pressure over an area—as well as observations taken from roofs, hand-drawn maps, and local rules of thumb.
These guides to future weather predictions were based off years of observation and experience.
Then, in the 1950s, a group of mathematicians, meteorologists, and computer scientists—led by John von Neumann, a renowned mathematician who had assisted the Manhattan Project years earlier, and Jule Charney, an atmospheric physicist often considered the father of dynamic meteorology—tested the first computerized automatic forecast.
Charney, with a team of five meteorologists, divided the United States into (by today’s standards) fairly large parcels, each more than 700 kilometers in area.
By running a basic algorithm that took the real-time pressure field in each discrete unit and prognosticated it forward over the course of a day, the team created four 24-hour atmospheric forecasts covering the entire country.
It took 33 full days and nights to complete the forecasts.
Though far from perfect, the results were encouraging enough to set off a revolution in weather forecasting, moving the field toward computer-based modeling.
Over the ensuing decades, billions of dollars in investments and the evolution of faster, smaller computers led to a surge in predictive capability.
Models are now capable of interpreting the dynamics of parcels of atmosphere as small as 3 kilometers in area, and since 1960 these models have been able to include ever-more-accurate data sent from weather satellites.
In 2016 and 2018, the GOES-16 and -17 satellites launched into orbit, providing a host of improvements, including higher-resolution images and pinpoint lightning detection.
The most popular numerical models, the US-based Global Forecasting System (GFS) and European Center for Medium-Range Weather Forecasts (ECMWF), released significant upgrades this year, and new products and models are being developed at a faster clip than ever.
At a finger’s touch, we can access an astonishingly precise weather forecast for our exact location on the Earth’s surface.
Today’s lightning-speed predictions, the product of advanced algorithms and global data collection, appear one step away from complete automation.
But they’re not perfect yet.
Despite the expensive models, array of advanced satellites, and mega-computers, human forecasters have a unique set of tools all their own.
Experience—their ability to observe and draw connections where algorithms cannot—gives these forecasters an edge that continues to outperform the glitzy weather machines in the highest-stake situations.
THOUGH TREMENDOUSLY USEFUL with big-picture forecasting, models aren’t sensitive to, say, the little updraft in one small land quadrant that suggests a waterspout is forming, according to Andrew Devanas, an operational forecaster at the National Weather Service office in Key West, Florida.
Devanas lives near one of the world’s most active regions for waterspouts, marine-based tornadoes that can damage ships that pass through the Florida Straits# and even come onshore.
The same limitation impedes predictions of thunderstorms, extreme precipitation, and land-based tornadoes, like those that tore through the Midwest in early December, killing more than 60 people.
But when tornadoes occur on land, forecasters can often spot them by looking for their signature on radar; waterspouts are much smaller and often lack this signal.
In a tropical environment like the Florida Keys, the weather doesn’t change much from day to day, so Devanas and his colleagues had to manually look at variations in the atmosphere, like wind speed and available moisture—variations that the algorithms don’t always take into account—to see if there was any correlation between certain factors and a higher risk of waterspouts.
They compared these observations to a modeled probability index that indicates whether waterspouts are likely and found that with the right combination of atmospheric measurements, the human forecast “outperformed” the model in every metric of predicting watersprouts.
Similarly, research published by NOAA Weather Prediction Service director David Novak and his colleagues show that while human forecasters may not be able to “beat” the models on your typical sunny, fair-weather day, they still produce more accurate predictions than the algorithm-crunchers in bad weather.
Over the two decades of information Novak’s team studied, humans were 20 to 40 percent more accurate at forecasting near-future precipitation than the Global Forecast System (GFS) and the North American Mesoscale Forecast System (NAM), the most commonly used national models.
Humans also made statistically significant improvements to temperature forecasting over both model’s guidance.
“Oftentimes, we find that in the bigger events is when the forecasters can make some value-added improvements to the automated guidance,” says Novak.
Particularly in adverse conditions, great improvements to the model’s forecast were usually due to human augmentation, he adds.
This is even more true for local, severe events like thunderstorms and tornadoes, which rely on split-second decision-making in order to save lives.
As forecasters become more familiar with a particular model, they begin to notice its biases and failings, Novak adds.
Just like the model learns from us, we learn from the model.
Milrad entered the field in the early 2000s, an era when the dominant forecasting methods were shifting from older techniques to numerical weather models and automated observations.
These technologies were critical to recent advances in atmospheric science, but Milrad cautions his students against complacency and dependence on the automated data models.
“If they’re going to be forecasting precipitation, they should be able to defend it by analyzing the physical processes and mechanisms that they see on the maps,” says Milrad.
He sees utility in the continued use of rules of thumb and pattern recognition techniques, not only as teaching tools, but also to defend against losing the vital experience forecasters bring to bear in severe weather situations or when models are off-base.
“There’s an old adage that ‘all models are wrong, some are useful,’” says Milrad.
“Even if it’s a great forecast it’s going to be slightly wrong.
It's how you can add value to that model.”
Plus, even though computer-generated forecasts are likely to continue improving over time, a number of challenges stand in the way of anything resembling full automation, which requires a significant expansion of computing power, with a multibillion dollar price tag.
The Department of Energy bankrolled the development of three exascale computers—capable of performing 1018 calculations per second—in 2018.
The first of these, the Aurora supercomputer under development at Argonne National Laboratory in Illinois, is slated to go online in 2022 and will be able to perform 1 quintillion calculations per second, but several different scientific fields are vying for access to its immense processing power.
And current infrastructure could also be at risk, since full rollout of 5G threatens to interfere with several key weather satellites.
Radio interference could degrade the quality of satellite observations of water vapor and potentially set forecasting capability back by decades.
In truth, the future of accurate weather forecasting may not necessarily rely on automation, but on a more mundane solution: financial support.
Thanks to these technological advances in weather forecasting and meteorology, human forecasters who once juggled the more tedious aspects of the job now have the bandwidth to focus on severe weather, research, and communicating important information about risks and preparation to agencies and people living in their area.
If such important work is to continue, the National Weather Service, on which so much of our weather infrastructure relies, must remain adequately funded.
Though it’s the private weather companies—like Accuweather and Weather Underground—that can provide more frequent, pinpoint forecasts, their business models rely on advertisement, subscription revenue, and enhanced services offered at a premium, and most employ few meteorologists (Accuweather employs around 100, while the NWS has more than 2,000).
Prior attempts by legislators—with financial backing from Accuweather executives—to limit the NWS from sharing weather information with the public have been met with outrage by the meteorological community.
If we want to continue to receive in-depth weather forecasts and crucial warnings, touched by human hands, we need to preserve agencies and services that value human-augmented forecasts and the public’s right to know.
(The service’s budget fell considerably during the Trump administration but thankfully is now reaching new highs, with a $6.2 billion NOAA funding package proposed for 2022—the largest in the agency’s history.)
Devanas, the NWS forecaster in Key West, agrees that the private sector has a lot to contribute to forecasting but is wary of the amount of unreliable weather information that is circulated as a result.
Even as algorithms and models continue to improve, Devanas believes we can’t lose sight of the science behind everything.
“I'm not here to say ‘Today is going to be 92 degrees, and it’s going to be 80 degrees at night with a 20 percent chance of rain.’ I could in essence get a monkey to do that,” he says.
“Those are things where we need some local expertise.
Those are things where the rules of thumb come in, and that local knowledge becomes invaluable.”
- VentureBeat : DeepMind claims its AI weather forecasting model beats conventional models
- UN News : A.I. model shows promise to generate faster, more accurate weather forecasts
- The Guardian : Do I need a brolly? Google uses AI to try to improve two-hour rain forecasts
- BBC : AI can predict if it will rain in two hours' time
- ECMWF : AI & Machine Learning for weather predictions
- AI Oodles : AI and ML will replace the traditional weather forecast system
- Towards AI : AI is Predicting Faster and More Accurate Weather Forecasts
- GeoGarage blog : AI breakthroughs could improve weather forecasts : finding patterns in the data / Google AI model outperforms traditional methods of weather ... / Tracking hurricanes with artificial intelligence / Can Artificial Intelligence help build better, smarter climate ... / Why are all my weather apps different?
Thursday, January 6, 2022
Australian icebreaker maps deepsea mountain
From Phys by Australian Antarctic Program
The summit of an underwater mountain, higher than Mount Kosciuszko, has been mapped for the first time by the Australian icebreaker Research and Supply Vessel (RSV) Nuyina.
At 2500 meters high, 2900 meters wide and 4500 meters long, the seamount had been identified from satellites at about 50 degrees South, on the edge of the "Furious Fifties."
While an earlier voyage had skimmed one side of the seamount, no detailed mapping at its summit had ever been done.
Seamounts are usually formed from extinct volcanoes and can be biological hotspots, attracting plankton, corals, fish and marine mammals.
As Nuyina's passage to Davis research station took the ship directly over the seamount, the onboard acoustics team took the opportunity to switch on the ship's hull-mounted multibeam echosounder, to find out what lay beneath.
Senior Acoustics Officer Floyd Howard said the echosounder "sees" seafloor features by emitting pings of sound in a fan-like pattern beneath the ship.
When the sound hits an object or the seafloor, it bounces back towards the ship, allowing scientists to build a picture of the seafloor—similar to echolocation used by dolphins.
As the ship cruised at 8 knots over the seamount, its surface structure and height were gradually revealed, ping by colorful ping, with different colors representing different depths.
"The highest point we reached was about 500 meters below the ocean surface, so it's a significant feature," Mr Howard said.
The team has informally named the feature "Ridgy-Didge Seamount" until an official name can be bestowed upon it.
The information collected during the seamount pass, and future mapping efforts by Nuyina and other ships, will contribute to global efforts to map the world's oceans by 2030.
Links :
Wednesday, January 5, 2022
How dead whales seem to connect deep-sea life
From HakaIMag by Fanni Szakal
On a regular day, Wirth runs Pico Sport, a whale watching company in the Azores.
“You have to lift these 500-kilogram blocks one by one, and you never know when you reach the point that the whale goes down,” he says.
Wirth was pulled away from his day job when he was approached by the British Broadcasting Corporation (BBC).
The sunken whale offered one of the most impressive shots in the Blue Planetseries.
Within just 25 minutes of the whale hitting the bottom, an attached camera recorded the arrival of the first scavenger, an enormous sixgill shark.
Over the course of a year, the research and film team descended to the carcass six more times.
At the bottom, the team filmed and sampled the now-exposed skeleton of the whale, picked clean by the creatures of the deep.
Years on, Cuvelier and her team’s analysis of the denizens of this deliberately sunken whale have revealed a potential overlap between the polychaete worm species that colonized the cow bones off Portugal and the whale bones near Brazil, suggesting that a dead whale in the mid-Atlantic could indeed be a stepping stone connecting the two sides of the ocean.
MaurĂcio Shimabukuro, a researcher at the University of Southern Denmark who was not involved in the study, says that to confirm this finding Cuvelier and her team will need to collect some of the worms and study their DNA.
Cuvelier wants to fill this gap, too.
Tuesday, January 4, 2022
Great Barrier Reef explodes into life in 'magical' spawning event
Scientists working beneath the waves say they witnessed the event, in which coral simultaneously release sperm and eggs en masse, overnight Tuesday off the coast of Cairns, Queensland, hailing it as a positive sign the reef was able to regenerate despite ecological threats.
"Nothing makes people happier than new life -- and coral spawning is the world's biggest proof of that," Australian marine biologist Gareth Phillips, who had a front row seat to this year's coral spawn, said in statement via Queensland Tourism and Events.
Philips, from research center Reef Teach, and his team of marine biologists, divers, students and photographers dived to the bottom of the ocean to capture footage of coral off the coast of Cairns, Queensland. Over the next couple days, they'll be traveling to other reef sites to film and observe.
This footage will allow scientists to monitor this year's coral crop -- and keep tabs on the general health of the Great Barrier Reef, which is protected by UNESCO and this year avoided an "in danger" rating from the World Heritage Committee.
"Magical" conditions
Gabriel Guzman/Calypso Productions
Philips called monitoring this year's coral spawn off Cairns "the ultimate treasure hunt."
"I've seen the corals all go off at once, but this time there seemed to be different species spawning in waves, one after the other. The conditions were magical with the water like glass and beautiful light coming from the moon," he said.
Philips said his team swam around looking for coral on the verge of spawning.
"Once we found a ripe coral, we watched as it took about 30 seconds for each colony to complete its spawning. It was the ultimate treasure hunt ... it was so exciting that we even grabbed the skipper and got him in the water."
Gareth Phillips/Reef Teach
The Great Barrier Reef's coral spawn is a coordinated annual effort -- for much of the year, coral multiplies by splitting and dividing, but once a year the coral simultaneously releases bundles of sperm and egg into the ocean.
"Each coral larvae drifts until it lands and settles on the sea floor," said Philips. "Spawning takes place over several days with different species spawning on different nights."
The Great Barrier Reef Foundation explains that coral bundles need to find another bundle from the same coral species in order to reproduce, so by releasing bundles at the same time, coral increase that likelihood.
The annual coral spawn usually takes place from October to November, but timings can vary due to factors like water temperature and currents. The date of the spawn can also fluctuate across the length and breadth of the 2,300 kilometer (1,429 mile)-long Great Barrier Reef.
Marine biologist Gareth Philips said this year's conditions were "magical."
Gareth Phillips/Reef Teach
Sign of hope
For Philips, the coral spawning is a sign of hope in the face of ecological troubles that recently prompted UNESCO to ask Australia for a report on the state of the Great Barrier Reef's conservation by February 2022.
Earlier this month, a study from James Cook University in Australia found that only 2% of the reef has escaped bleaching -- a consequence of heat waves -- since 1998.
Philips said it was "gratifying" to see the reef give birth, an event loosely coinciding with Australia's decision to start relaxing some of the world's toughest Covid travel restrictions.
"It's a strong demonstration that its ecological functions are intact and working after being in a recovery phase for more than 18 months," he said.
"The reef has gone through its own troubles like we all have, but it can still respond -- and that gives us hope. I think we must all focus on the victories as we emerge from the pandemic."
Monday, January 3, 2022
The Bermuda triangle: what science can tell us about the mysterious ocean region
From DiscoverMag by Nathaniel Scharping
A region of the ocean purported to swallow ships whole has fascinated us for decades.
But is there any truth to the tales?
Just off the southeast coast of the United States, there lies a span of ocean that’s long held a fearsome reputation.
Ships traversing its choppy breadth disappear without a trace.
Flights routed above the waters blink from radar screens, never to be seen again.
The mysterious happenings have conjured stories of supernatural interference, alien kidnappings and an area somehow outside the normal bounds of physical reality.
The Bermuda Triangle, it’s said, is a haunted place.
That’s just one version of the story, of course.
The Bermuda Triangle has been the site of a number of high-profile and still-mysterious naval and aviation disappearances.
But that those disasters are the result of anything sinister, as opposed to the logical conjunction of environment and statistics, is extremely doubtful.
Still, a number of people have proposed scientifically valid explanations for the disappearances of ships and planes in the Bermuda Triangle over the years.
The ocean is a dangerous place, after all, and it’s not uncommon even today for things to go wrong.
In the storm-tossed waters of the North Atlantic, safety is never a guarantee.
Where is the Bermuda Triangle?
The Bermuda Triangle, as it’s most commonly defined, stretches between Miami, San Juan, Puerto Rico and the island of Bermuda.
In all, it encompasses hundreds of thousands of square miles in the North Atlantic Ocean, a huge area.
The region also sees heavy traffic from ships coming and going from the East Coast and Gulf of Mexico.
The Bermuda Triangle got its name from a 1964 article in the pulp magazine Argosy, which linked together a few disappearances in the region.
“The Deadly Bermuda Triangle” didn’t offer up any explanations for the occurrences, though it did heavily emphasize the mysterious nature of the area.
The article features the disappearance of the U.S.S Cyclops, a Navy supply ship, in 1918, and the loss of a flight of bombers during a practice run in 1945, as well as one of the search and rescue planes sent out after them.
These incidents, and others, have today become part of the lore of the Bermuda Triangle.
These stories are often stitched together to hint at something untoward lurking beneath the surface of the Atlantic Ocean.
In addition to the supernatural explanations, a number of more realistic explanations for the phenomenon have been put forward throughout the years, ranging from wayward magnetism to dangerous bubbles.
The fact that the area within the Bermuda Triangle is heavily trafficked could account for some of the mystery.
Any region with lots of ships going through it is bound to see more accidents than a place with less activity.
Pair that with the fact that the Bermuda Triangle is often swept by hurricanes, and it’s not hard to see why ships might occasionally sink there.
Another common explanation for the Bermuda Triangle rests on magnetism.
The Earth’s magnetic North Pole isn’t the same as its geographic North Pole, which means that compasses usually don’t point exactly north.
Only along what’s known as agonic lines, which line up magnetic and geographic north, are compasses truly accurate.
One agonic line runs from Lake Superior down through the Gulf of Mexico near the Bermuda Triangle.
One theory holds that mariners, usually accustomed to accounting for a discrepancy in their compass readings, may make mistakes when very near to the agonic line that lead them astray.
Paired with the often shallow waters of the island-strewn Caribbean Sea, navigational errors could lead to boats running aground on hidden shoals.
Another theory posits that the Bermuda Triangle might be home to a large-scale magnetic anomaly, a region where the Earth’s magnetic field lines are warped and twisted.
This, too, could cause navigational mistakes.
But, as others have noted, there’s no evidence the Bermuda Triangle contains any unusual magnetic disturbances, something that’s clear when looking at a magnetic map of the region.
More recently, some scientists have suggested that ship sinkings in the Bermuda Triangle could be due to massive bubbles released from undersea methane deposits.
The seafloor in the region is known to contain large pockets of gas that could be released suddenly, turning the ocean into a frothy soup that swallows ships.
A similar process likely created huge seafloor craters near Norway.
But though the mechanism itself makes sense, there’s no evidence of any recent methane release from the area around the Bermuda Triangle.
The last time anything similar happened in the region was around 15,000 years ago, according to U.S.
Geological Survey geologist Bill Dillon.
Another explanation for the Bermuda Triangle that checks out on paper is the presence of rogue waves.
These huge waves can form unexpectedly and rise two or even three times above surrounding waves.
As Vice reports, British researchers used lab and computer models to simulate the effects of rogue waves more than 100 feet tall on ships as part of an investigation into the Bermuda Triangle.
Ships that were sufficiently long could get caught suspended between two wave peaks with nothing supporting them from below and snap in half, one researcher theorizes.
But, while rogue waves are certainly capable of capsizing or breaking a ship, we have no definitive evidence tying them to any of the naval disasters in the Bermuda Triangle.
The U.S. government doesn’t recognize the Bermuda Triangle, and the area doesn’t appear on any official maps.
And the Coast Guard and Department of Defense have repeatedly refrained from giving the area, or its legends, any outsized significance.
Furthermore, there’s no evidence suggesting the region sees higher rates of maritime or aviation disasters than anywhere else in the world, after accounting for the amount of traffic that passes through.
The Human Factor
The true explanation for the Bermuda Triangle may ultimately reside not in the ocean, but in our minds.
Our minds are often biased toward bizarre or otherwise memorable events, and have trouble accurately accounting for statistical discrepancies.
For example, we’re more likely to remember things that seem exceptional — such as a ship that disappears with no explanation— than something more ordinary, like a ship sinking in a hurricane.
And once something stands out to us, it can form the basis for further attention.
It’s a form of what’s called a frequency illusion, sometimes referred to as the Baader-Meinhof effect.
Essentially, once we’re introduced to something once, we tend to notice it more often all around us.
That can lead us to think whatever we’ve noticed is becoming rapidly more common, when, in reality, we’re just noticing it more.
Whatever is ultimately responsible for the legend of the Bermuda Triangle, be it psychological or otherwise, it’s worth remembering that there’s never been any evidence that the region is any more dangerous than anywhere else.
So go ahead and take that vacation to Bermuda — but, as always, make sure to wear a lifejacket when you’re out on the water.
It’s just common sense.
Links :