Groups Similar Search Look up By Text Browse About

ID: 98222

URL: https://venturebeat.com/2018/09/13/how-the-weather-company-tracks-storms-like-hurricane-florence-using-predictive-modeling/

Date: 2018-09-13

How The Weather Company tracks storms like Hurricane Florence using predictive modeling

As Hurricane Florence bears down on the East Coast, its intensity is coming into sharp relief. The Category 2 hurricane had sustained winds of 105 miles per hour at the time of writing, with tropical storm-force winds stretching more than 335 square miles in all directions. Rainfall was predicted to reach 40 inches in regions of coastal Carolina, and Wilmington — which just had its rainiest year to date — might get eight months worth of rain in three days. One firm closely tracking Florences progress is The Weather Company, the weather forecasting division of IBM whose consumer-facing brands include the weather.com, intellicast.com, and Weather Underground. Its systems analyze more than 100 terabytes of third-party data and generate 25 billion customized regional models daily. Our models provide a forecast evolution of what the atmosphere is going to look like in the coming days, Dr. James Belanger, a senior meteorological scientist for The Weather Company, told VentureBeat in an interview. The system keeps in memory what the forecasts are all across the globe. One of the predictive technologies its data scientists tap is Deep Thunder, an IBM research project spun out of the companys Deep Computing initiative. Leveraging public satellite imagery, proprietary datasets, and sensors — including more than 250,000 personal weather stations and smartphone barometer readings — its able to produce short-term, hyperlocal weather forecasts for local governments and corporate clients alike. In a pilot in Rio de Janeiro, Deep Thunder predicted floods and anticipated where storms might trigger mudslides. And in the U.S., it enabled a utility company to pinpoint where storms were likely to bring down power lines. Its capable of even greater precision, Belanger explained. Using historical weather data, it can create probability distributions by modeling synthetic storms (think a computer-generated tropical cyclone.) And later this year, itll be used to issue probabilistic snowfall reports seven days in advance. Another tool in The Weather Companys arsenal are forecast runs from government agencies like the National Weather Service, which are typically released 10-15 days in advance of shifting weather patterns.  Theyre ingested and run through adaptive regression models — statistical models that automatically plot interactions between variables  — that apply weight and bias corrections for any given target location. Its a really scalable system, Belanger said. The calibration information is continually updated … Were able to distribute temperature and precipitation forecasts for anywhere in the globe. So hows The Weather Data use that backend to track high-profile storms like Florence? While it defers to the National Weather Service on forecasted storm impacts — going so far as to impose what Belanger calls guardrails and adjustments that ensure its messaging remains in alignment — its consumer team employs analytics to identify the best times to issue alerts. They look at the times of day when consumers are most actively engaging our content, Belanger said, and make sure to keep people informed of rainfall and flood threats. Its an important way we deliver messages and communications to people who are the most vulnerable.



ID: 98237

URL: https://www.theverge.com/2018/9/13/17857478/hurricane-florence-storm-surge-flooding-the-weather-channel-video-graphics

Date: 2018-09-13

This terrifying graphic from The Weather Channel shows the power and danger of Hurricane Florence

A new video from The Weather Channel shows in real time the danger of flood waters already rising in parts of the Carolinas as Hurricane Florence starts battering the coast. The storm is moving slowly and is anticipated to bring deadly storm surges to the region as well as torrential rains. Thats a recipe for a flooding disaster, meteorologist Marshall Shepherd told The Verge in an interview on Monday. The National Hurricane Center is predicting storm surges anywhere from two to more than 11 feet high. But its hard to visualize what those numbers actually mean for someone near the water. The National Hurricane Center tried to make it clear with a cartoon graphic that shows rainbow colored water levels rising over the heads of a family in a house. You'll be seeing the NHC potential storm surge flooding graphic a lot over the next few days as #Florence nears the coast. Here's how to interpret it; The graphic is more effective than numbers, or even maps. But The Weather Channel takes the visuals a step further using mixed reality that show the waters surrounding the on-screen meteorologists, including Greg Postel and Ericka Navarro. The flood rises above their heads as the wind howls and floating cars slosh at the surface. This @weatherchannel visualization of storm surge is an amazing and sobering use of technology to show what hurricanes like Florence can do The mixed reality graphics, created in partnership with augmented reality company The Future Group, harness the Unreal Engine — a popular video game development platform. Rather than creating effects and rendering them in post-production, the process used to create visuals for most films, the Unreal Engine builds effects in real time, Ren LaForme reported for Poynter when The Weather Channel unveiled the tech in a tornado demo. The Weather Channel has since used the immersive reality for an in-studio lightning explainer, and now, to visualize Hurricane Florences floods. The business that were in is safety, Michael Potts, The Weather Channels vice president of design, told The Verge in an interview. The weather is a visceral, physical thing, and were trying to recreate that in the most realistic way possible. On screen, the water towers over the meteorologist as fish swim by — adding emphasis to her words that Hurricane Florences floods will threaten lives. If you find yourself here, please get out, she says. If youre told to go, you need to go.



ID: 98253

URL: https://arstechnica.com/science/2018/09/florence-is-now-only-a-category-2-hurricane-that-wont-matter-much/

Date: 2018-09-13

Florence is now “only” a Category 2 hurricane. That won’t matter much

When it comes to water, wind speed matters less than the size of the wind field. Ten years ago today, Hurricane Ike made landfall as "only" a Category 2 hurricane along Galveston Island near Houston. Though Ike had 110mph winds at landfall, it had grown very large over the Gulf of Mexico, and this large size allowed it to develop an enormous amount of "integrated energy" that manifested itself as a devastating storm surge. With about $30 billion in damages, Hurricane Ike was, at the time, the second-costliest US hurricane on record. As of Thursday morning, Hurricane Florence has weakened to 110 mph as it contends with slightly increased wind shear and drier air. Technically this means Florence is no longer a "major" hurricane, and it may not be when it reaches the North Carolina coast early on Friday morning. Practically, however, that won't matter when it comes to storm surge and inland rainfall. The simplest, most common metric for the measurement of a storm's intensity is maximum wind speed, and certainly this matters in terms of pure destructive potential when it comes to, say, losing a roof or propelling debris through the air. But when it comes to water, the maximum wind speed matters less than the size of the wind field for both storm surge and the destructive power of waves moving onshore. And Florence is a large storm, with hurricane-force winds extending outward up to 80 miles from the center and tropical-storm-force winds outward up to 195 miles. The primary example of this is 2012's Hurricane Sandy, which had modest 80mph winds by the time it reached the New York and New Jersey areas. But because the storm was so large it produced a destructive storm surge in New York City and caused an estimated $65 billion in damage in the United States. (At the time, it became the second costliest US hurricane but has since been supplanted by Harvey and Maria, both in 2017). Thursday, September 13th is the 10 year anniversary of Hurricane Ike. The storm's 15'+ surge devastated the Bolivar peninsula. The outer banks of North Carolina may face a similar threat from Florence. Our thoughts are with you Carolina. Prep for the worst, and hope for the best The area to the north of Florence's landfall along the North Carolina coast—storm surge is always worst to the "right" of a storm's center, where its rotation pushes water onshore—is less populated and developed than the New Jersey and New York shores. But the storm surge will be no less devastating for homes and businesses located there.  Areas from Cape Fear to Cape Lookout face a potential 9 to 13 feet of storm surge, the National Hurricane Center warns, on top of the normal tides. Florence's slow movement over the next two or three days will add another element of uncertainty. Meteorologist Greg Diamond has noted that areas along the coast of North Carolina on the right side of the storm could experience a significant storm surge through multiple high tide cycles, which would have a cumulative battering effect. The National Hurricane Center has, accordingly, warned that "Life-threatening, catastrophic flash flooding and prolonged significant river flooding are likely over portions of the Carolinas and the southern and central Appalachians late this week into early next week, as Florence is expected to slow down as it approaches the coast and moves inland. " It is important to recognize that, while attention will be focused on coastal impacts from Florence, inland flooding will remain a problem for some areas well into next week.



ID: 98294

URL: https://arstechnica.com/science/2018/09/first-ever-pre-storm-study-looks-at-climate-change-and-hurricane-florence/

Date: 2018-09-13

A quick simulation of Hurricane Florence done without climate change

Simple analysis suggests global warming boosts Florences rain by 50 percent. In the last few years, teams of scientists have developed a consistent protocol for rapidly analyzing the influence of climate change on extreme weather events. Within a week of the disaster, reports have been available to inform the conversation about whether we can expect more events like it in the future. But on Wednesday, we saw the first example of something new—an analysis published before the event even happened. A group led by Stony Brook Universitys Kevin Reed ran a very simple computer model experiment on Hurricane Florence—which isnt due to make landfall until Friday—and quickly released the top-line results. The rapid studies weve been seeing are done by examining the historical weather record to estimate how rare and extreme a given storm or heat event would be in that area of the globe. From there, climate model simulations are used to see if climate change is expected to change the frequency of that type of event. In this case, theres obviously no data available for a thing that hasnt happened yet. Instead, the researchers focused on a much more limited question that is faster to answer: how does a warmer world change this storm? In the counterfactual world where global warming never happened, its impossible to say if Hurricane Florence would even have been born. Even small changes can have complex consequences on the atmosphere, such that events would play out completely differently. But thats not the point. Since Hurricane Florence is occurring in this warmer world, we can simply examine the effect of warmer temperatures. To do this, the researchers took the current state of the world on Tuesday, dropped that into their model as a starting point, and pressed play to simulate ahead to Sunday. For a comparison simulation, they took those starting conditions and essentially subtracted out global warming. In this counterfactual world, the storm looks significantly different. Hurricanes are fueled by energy from the evaporation of warm seawater, so its no surprise that warmer sea surface temperatures should give the storm a boost. The size of the boost in this case is pretty remarkable, though. The model analysis showed the real-world Florence dumping 50 percent more rain near the coast than it would in a world without human-caused warming. The modeled hurricane clearly stays stronger when simulated under current-day conditions, but its also larger. The diameter of the storm is about 80 kilometers (50 miles) greater than in the cooler simulation, which would translate into higher storm surge flooding on the coast. Team member Michael Wehner told Ars that the team is working to repeat this analysis with updated observations as the storm barrels down on the Carolinas, so we'll get to see how similar the results are for each iteration. The researchers also plan to repeat their work after the storm and carefully compare with the forecast analyses. That will help show how useful this trial run of pre-storm analysis was. For their part, the group behind the within-one-week studies explained Thursday that they wont be providing an analysis of Hurricane Florence in the near future (for reasons ranging from complex historical data to swamped workloads). But they did comment on these pre-storm results, writing, More analyses are needed to assess the robustness of this quick analysis, although the basic result that global warming increases the precipitation is a very robust one supported by observations and modelling studies.