IBM Deep Thunder

Last updated

Deep Thunder is a research project by IBM that aims to improve short-term local weather forecasting through the use of high-performance computing. It is part of IBM's Deep Computing initiative that also produced the Deep Blue chess computer.

Contents

Deep Thunder is intended to provide local, high-resolution weather predictions customized to weather-sensitive specific business operations. [1] For example, it could be used to predict the wind velocity at an Olympic diving platform, destructive thunderstorms, and combined with other physical models to predict where there will be flooding, damaged power lines and algal blooms. The project is now headquartered at IBM's Thomas J. Watson Research Center in Yorktown Heights, New York.

History

The Deep Thunder project is headed by Lloyd Treinish, who joined IBM in 1990, after working for 12 years at NASA's Goddard Space Flight Center. [2]

The project began in 1995 as an outgrowth of a project designed to help provide accurate weather forecasts for the 1996 Atlanta Olympic Games. In collaboration with the National Oceanic and Atmospheric Administration, IBM scientists built one of the first parallel processing supercomputers to be used for weather modeling, based on the IBM RS/6000 SP. It was installed at the National Weather Service office in Peachtree City, Georgia, in 1996, where it ran for several months and produced multiple forecasts daily. After a few years of development, the team set up an implementation in New York City in 2001 to test the project. The group is currently working on establishing the Rio de Janeiro operations center. [3]

The name Deep Thunder arose after the IBM Deep Blue system played and defeated the world chess champion Garry Kasparov in May, 1997. In the following November, a journalist used the name Deep Thunder in an article, which stuck with the developers. [3] Current members of Deep Thunder are Lloyd Treinish, Anthony Praino, Campbell Watson and Mukul Tewari.

Technology

Deep Thunder uses a 3D telescoping grid where data from one model feeds into another and is verified with historical data. For example, they start with a global model from NOAA, and as they zoom in the resolution decreases exponentially, down to models with resolutions of 1 kilometer, and sometimes as small as 1 meter. Using this method, IBM can cut down on the amount of processing required. IBM uses many sources of data to feed Deep Thunder, including public satellite sources, and many other private sources, as well as whatever local sensors and data a location, may have. [4]

The Watson computer system will be used to generate the Deep Thunder weather forecasts. [5] Input data will be collected from over 200,000 Weather Underground personal weather stations, weather satellite data, smartphone barometer and data from other sources.

Applications

Utility Companies

IBM worked with a North American utility company that has over 90,000 poles, wires, and transformers to develop a prediction service that can pinpoint where incoming storms will bring down trees and power lines. The service can be used to call in the needed number of repair crews and station them near where the damage will occur, drastically decreasing downtime. [6] [7]

Agriculture

Deep Thunder could be used to determine optimal times to plant, irrigate, and harvest crops, based on the dynamic environmental conditions of individual farm locations. Precision agriculture using Deep Thunder could lead to better price points for crops by saving water, allocating labor more effectively, and improving supply chain efficiency. Using these methods of predictive weather farming, crop losses from the weather can be cut by over 25%. The island nation of Brunei is working with Deep Thunder to develop precision weather forecast models to increase rice production. [8]

In The Olympics

Deep Thunder was used during the 1996 Atlanta Olympic Games to accurately predict that there would be no rain during the closing ceremony. [2]

The Jefferson Project

The Jefferson Project at Lake George (New York) is a global example of sustained protection of freshwater ecosystems. The project uses Deep Thunder to predict the weather at an unprecedented 333 m resolution, simulating complex airflow patterns over the lake that are crucial to lake currents and nutrient cycling. [9]

New York

New York City was the first city to test a full-scale implementation of Deep Thunder. IBM is experimenting with using a mobile app to distribute location-specific predictions and issue alerts. Data from the app can be used by maintenance crews to determine if wind levels are too high to work, or it could be used to get a weather forecast at a certain address. [4]

Rio de Janeiro

IBM is currently expanding the project to Rio de Janeiro to predict floods, and anticipate where storms might trigger mudslides. The city is collaborating with IBM on a multimillion-dollar plan to improve emergency responsiveness, by providing comprehensive information about rainfall estimates, wind speed and direction, probabilities of landslides and floods to responders. It is expected to be utilized during the 2014 FIFA World Cup and 2016 Summer Olympics. [6]

Dublin, Ireland

The Dublin City Council is working with IBM to help make Dublin, Ireland, the third city in the world to implement the Deep Thunder forecasting model in an attempt to predict and issue warnings about incoming flash floods that are impacting businesses and homeowners. [10]

Related Research Articles

<span class="mw-page-title-main">European Centre for Medium-Range Weather Forecasts</span> European intergovernmental weather computation organisation based in the UK

The European Centre for Medium-Range Weather Forecasts (ECMWF) is an independent intergovernmental organisation supported by most of the nations of Europe. It is based at three sites: Shinfield Park, Reading, United Kingdom; Bologna, Italy; and Bonn, Germany. It operates one of the largest supercomputer complexes in Europe and the world's largest archive of numerical weather prediction data.

<span class="mw-page-title-main">Meteorology</span> Interdisciplinary scientific study of the atmosphere focusing on weather forecasting

Meteorology is a branch of the atmospheric sciences with a major focus on weather forecasting. The study of meteorology dates back millennia, though significant progress in meteorology did not begin until the 18th century. The 19th century saw modest progress in the field after weather observation networks were formed across broad regions. Prior attempts at prediction of weather depended on historical data. It was not until after the elucidation of the laws of physics, and more particularly in the latter half of the 20th century the development of the computer that significant breakthroughs in weather forecasting were achieved. An important branch of weather forecasting is marine weather forecasting as it relates to maritime and coastal safety, in which weather effects also include atmospheric interactions with large bodies of water.

<span class="mw-page-title-main">Weather forecasting</span> Science and technology application

Weather forecasting is the application of science and technology to predict the conditions of the atmosphere for a given location and time. People have attempted to predict the weather informally for millennia and formally since the 19th century. Weather forecasts are made by collecting quantitative data about the current state of the atmosphere, land, and ocean and using meteorology to project how the atmosphere will change at a given place.

<span class="mw-page-title-main">Meteorologist</span> Scientist specialising in meteorology

A meteorologist is a scientist who studies and works in the field of meteorology aiming to understand or predict Earth's atmospheric phenomena including the weather. Those who study meteorological phenomena are meteorologists in research, while those using mathematical models and knowledge to prepare daily weather forecasts are called weather forecasters or operational meteorologists.

Forecasting is the process of making predictions based on past and present data. Later these can be compared (resolved) against what happens. For example, a company might estimate their revenue in the next year, then compare it against the actual results. Prediction is a similar but more general term. Forecasting might refer to specific formal statistical methods employing time series, cross-sectional or longitudinal data, or alternatively to less formal judgmental methods or the process of prediction and resolution itself. Usage can vary between areas of application: for example, in hydrology the terms "forecast" and "forecasting" are sometimes reserved for estimates of values at certain specific future times, while the term "prediction" is used for more general estimates, such as the number of times floods will occur over a long period.

<span class="mw-page-title-main">National Weather Service</span> U.S. forecasting agency of the National Oceanic and Atmospheric Administration

The National Weather Service (NWS) is an agency of the United States federal government that is tasked with providing weather forecasts, warnings of hazardous weather, and other weather-related products to organizations and the public for the purposes of protection, safety, and general information. It is a part of the National Oceanic and Atmospheric Administration (NOAA) branch of the Department of Commerce, and is headquartered in Silver Spring, Maryland, within the Washington metropolitan area. The agency was known as the United States Weather Bureau from 1890 until it adopted its current name in 1970.

<span class="mw-page-title-main">Met Office</span> United Kingdoms national weather service

The Meteorological Office, abbreviated as the Met Office, is the United Kingdom's national weather service. It is an executive agency and trading fund of the Department for Business, Energy and Industrial Strategy and is led by CEO Penelope Endersby, who took on the role as Chief Executive in December 2018 and is the first woman to do so. The Met Office makes meteorological predictions across all timescales from weather forecasts to climate change.

<span class="mw-page-title-main">Numerical weather prediction</span> Weather prediction using mathematical models of the atmosphere and oceans

Numerical weather prediction (NWP) uses mathematical models of the atmosphere and oceans to predict the weather based on current weather conditions. Though first attempted in the 1920s, it was not until the advent of computer simulation in the 1950s that numerical weather predictions produced realistic results. A number of global and regional forecast models are run in different countries worldwide, using current weather observations relayed from radiosondes, weather satellites and other observing systems as inputs.

The National Severe Storms Laboratory (NSSL) is a National Oceanic and Atmospheric Administration (NOAA) weather research laboratory under the Office of Oceanic and Atmospheric Research. It is one of seven NOAA Research Laboratories (RLs).

<span class="mw-page-title-main">Ensemble forecasting</span>

Ensemble forecasting is a method used in or within numerical weather prediction. Instead of making a single forecast of the most likely weather, a set of forecasts is produced. This set of forecasts aims to give an indication of the range of possible future states of the atmosphere. Ensemble forecasting is a form of Monte Carlo analysis. The multiple simulations are conducted to account for the two usual sources of uncertainty in forecast models: (1) the errors introduced by the use of imperfect initial conditions, amplified by the chaotic nature of the evolution equations of the atmosphere, which is often referred to as sensitive dependence on initial conditions; and (2) errors introduced because of imperfections in the model formulation, such as the approximate mathematical methods to solve the equations. Ideally, the verified future atmospheric state should fall within the predicted ensemble spread, and the amount of spread should be related to the uncertainty (error) of the forecast. In general, this approach can be used to make probabilistic forecasts of any dynamical system, and not just for weather prediction.

Flood forecasting is the use of forecasted precipitation and streamflow data in rainfall-runoff and streamflow routing models to forecast flow rates and water levels for periods ranging from a few hours to days ahead, depending on the size of the watershed or river basin. Flood forecasting can also make use of forecasts of precipitation in an attempt to extend the lead-time available.

<span class="mw-page-title-main">Tropical cyclone forecast model</span> Computer program that uses meteorological data to forecast tropical cyclones

A tropical cyclone forecast model is a computer program that uses meteorological data to forecast aspects of the future state of tropical cyclones. There are three types of models: statistical, dynamical, or combined statistical-dynamic. Dynamical models utilize powerful supercomputers with sophisticated mathematical modeling software and meteorological data to calculate future weather conditions. Statistical models forecast the evolution of a tropical cyclone in a simpler manner, by extrapolating from historical datasets, and thus can be run quickly on platforms such as personal computers. Statistical-dynamical models use aspects of both types of forecasting. Four primary types of forecasts exist for tropical cyclones: track, intensity, storm surge, and rainfall. Dynamical models were not developed until the 1970s and the 1980s, with earlier efforts focused on the storm surge problem.

<span class="mw-page-title-main">Atmospheric model</span>

An atmospheric model is a mathematical model constructed around the full set of primitive dynamical equations which govern atmospheric motions. It can supplement these equations with parameterizations for turbulent diffusion, radiation, moist processes, heat exchange, soil, vegetation, surface water, the kinematic effects of terrain, and convection. Most atmospheric models are numerical, i.e. they discretize equations of motion. They can predict microscale phenomena such as tornadoes and boundary layer eddies, sub-microscale turbulent flow over buildings, as well as synoptic and global flows. The horizontal domain of a model is either global, covering the entire Earth, or regional (limited-area), covering only part of the Earth. The different types of models run are thermotropic, barotropic, hydrostatic, and nonhydrostatic. Some of the model types make assumptions about the atmosphere which lengthens the time steps used and increases computational speed.

A wind power forecast corresponds to an estimate of the expected production of one or more wind turbines in the near future, up to a year. Forecast are usually expressed in terms of the available power of the wind farm, occasionally in units of energy, indicating the power production potential over a time interval.

<span class="mw-page-title-main">Tropical cyclone forecasting</span> Science of forecasting how a tropical cyclone moves and its effects

Tropical cyclone forecasting is the science of forecasting where a tropical cyclone's center, and its effects, are expected to be at some point in the future. There are several elements to tropical cyclone forecasting: track forecasting, intensity forecasting, rainfall forecasting, storm surge, tornado, and seasonal forecasting. While skill is increasing in regard to track forecasting, intensity forecasting skill remains unchanged over the past several years. Seasonal forecasting began in the 1980s in the Atlantic basin and has spread into other basins in the years since.

Specialized wind energy software applications aid in the development and operation of wind farms.

<span class="mw-page-title-main">Wind wave model</span> Numerical modelling of the sea state

In fluid dynamics, wind wave modeling describes the effort to depict the sea state and predict the evolution of the energy of wind waves using numerical techniques. These simulations consider atmospheric wind forcing, nonlinear wave interactions, and frictional dissipation, and they output statistics describing wave heights, periods, and propagation directions for regional seas or global oceans. Such wave hindcasts and wave forecasts are extremely important for commercial interests on the high seas. For example, the shipping industry requires guidance for operational planning and tactical seakeeping purposes.

<span class="mw-page-title-main">Nowcasting (meteorology)</span>

Nowcasting is weather forecasting on a very short term mesoscale period of up to 2 hours, according to the World Meteorological Organization, and up to six hours, according to other authors in the field. This forecast is an extrapolation in time of known weather parameters, including those obtained by means of remote sensing, using techniques that take into account a possible evolution of the air mass. This type of forecast therefore includes details that cannot be solved by numerical weather prediction (NWP) models running over longer forecast periods.

<span class="mw-page-title-main">UBD IBM Centre</span>

The UBD|IBM Centre is research collaboration between the Universiti Brunei Darussalam and International Business Machines (IBM) Corporation. The centre was launched by His Majesty Sultan Haji Hassanal Bolkiah Mu'izzadin Waddaulah, the Sultan and Yang Di-Pertuan of Brunei Darussalam, on October 15, 2011. IBM Research Labs in India, New York, USA (Watson), Australia, and Brazil are participating in this joint research.

The National Centers for Environmental Information (NCEI), an agency of the United States government, manages one of the world's largest archives of atmospheric, coastal, geophysical, and oceanic data, containing information that ranges from the surface of the sun to Earth's core, and from ancient tree ring and ice core records to near-real-time satellite images.

References

  1. "IBM100 - Deep Thunder". Research.ibm.com. 2010-04-06. Retrieved 2014-01-12.
  2. 1 2 "Lloyd Treinish's home page". Research.ibm.com. 2005-12-20. Retrieved 2014-01-12.
  3. 1 2 "IBM100 - Deep Thunder". 03.ibm.com. 2010-04-06. Retrieved 2014-01-12.
  4. 1 2 Gallagher, Sean (14 March 2012). "How IBM's Deep Thunder delivers "hyper-local" forecasts 3-1/2 days out". Ars Technica. Retrieved 24 October 2013.
  5. Jancer, Matt (26 August 2016). "IBM's Watson Takes On Yet Another Job, as a Weather Forecaster". Smithsonian. Retrieved 29 August 2016.
  6. 1 2 "IBM100 - Deep Thunder". 03.ibm.com. 2010-04-05. Retrieved 2014-01-12.
  7. http://earthnetworks.com/Portals/0/pdf/IBM_Deep_Thunder_Case_Study.pdf [ bare URL PDF ]
  8. Treinish, Lloyd (23 July 2013). "How To Prevent Hunger In Upcoming Decades? Try Precision Agriculture". Huffington Post. Retrieved 6 November 2013.
  9. "Jefferson Project at Lake George". Rensselaer. Retrieved 3 April 2019.
  10. Brophy, Daragh. "IBM's 'Deep Thunder' system could be used in Dublin to predict flash floods". thejournal.ie. Retrieved 22 October 2013.