Satellite Weather Information and Intelligent ... Weather Information and Intelligent... ·...

White Paper for PhD Lab Rotation Mohammed Ali, 10/05/2004

Satellite Weather Information and Intelligent Transportation System Introduction

Photo courtesy of the California Department of Transportation

On her way to pick up her daughter from King Magnet Elementary at Downtown Little Rock, Amy notices the dark clouds of a thunderstorm rapidly approaching. Meanwhile computers at a private meteorological service is using data from the National Weather Services and the Arkansas Department of Transportation to produce a highly specific “pathcast” for the storm, indicating which areas are expected to be affected. After picking up the girl, she is supposed to drive for home at Bryant about 12 miles west of Little Rock. Amy’s in vehicle communication systems beefs few times and relays a message that flash flooding will occur on highway 5. Due to the on going construction work, interstate 30 operates only a single lane and probability of its functioning is 20%. Concerned about the flooding on her usual return path interstate 5, Amy tell the hands free communication systems her destination and requests an alternate route. The system integrates both weather and traffic information and provides an alternate route that allows Amy to reach her home.

Weather significantly affects the safety and capacity of the nation’s roadways. When it turns wintry with snow and ice, it can be deadly too. Each year, adverse weather is associated with more than 1.5 million vehicular accidents, which result in about 800,000 injuries and 7,000 fatalities [3]. Over 17% of all fatal crashes occur during winter weather conditions [4]. Of these, 60% happen in rural areas (most on non-interstate roadways). Too often, people believe that little can be done about the adverse effects of weather on roadway transportation. On the contrary, Intelligent Transportation Systems is tantalizingly close to providing drivers and traffic managers with the kind of real-time weather and routing information described above in Amy’s scenario, thanks to advances made by computational science, transportation and meteorological communities. Over the course of the next 15 years, a focused road weather research program that brings these communities together could deliver much better road weather services to the nation, saving lives and reducing injuries while improving efficiency of highway systems. Weather Satellite Technology Weather data collection has been facilitated greatly by remote sensing devices such as satellites, and computer based data organization tools such as geographic information systems [6]. Spontaneous observation of weather satellites, subsequent weather maps and radar have expanded

1


our predictability of catastrophe and preparedness for emergency response. The National Oceanic and Atmospheric Administration (NOAA) is dedicated to enhancing economic security and national safety through the analysis and research of weather and climate related events and providing environmental stewardship for USA. NOAA supports Department of Homeland Security (DHS) and Federal Emergency Management Agency (FEMA) by incorporating GIS data that include storm surge forecasts, precipitation forecasts, flash flood guidance, real-time and forecast wind data and wave heights [1]. The NOAA satellite information services operates nation’s environmental satellites are used for ocean and weather observation & forecasting, climate monitoring. The system has geostationary operational environmental satellites (GOES) for short-range warning and polar-orbiting environmental satellites (POES) for long-term forecasting. The weather data gathered by GOES satellites combined with data from Doppler radars and automated surface observing systems, aids weather forecasters greatly in providing warnings of thunder storms, winter storms, flash floods, hurricanes and other sever weather [2]. GOES significantly advanced the ability to observe weather systems by providing frequent interval visible and infrared imagery of the earthy surface, atmospheric moisture and cloud cover.

Parabolic Mesh Dish Antenna, Collects Weather Data from GOES Courtesy: GPS, Remote Sensing and GIS Lab, University of Arkansas at Little Rock

2


Remote Sensing Mechanism of Weather Satellites Weather satellites have sensors aboard that detect both visible light and infrared or heat radiation. The sensors providing views of reflected sunlight are engineered to be more detailed than infrared, so that smaller objects can be seen. However, visible images are only available during the day, limiting their continuous monitoring of weather conditions. Although less detailed, infrared views are temperature maps of surfaces viewed from the satellite's vantage point, whether land, water or clouds. The temperature variations of the surfaces may be enhanced to highlight certain features of interest to meteorologists. The sensors themselves depended upon television research for their images. Later sensors were outgrowths of this and went on to solid-state extensions where heat radiation, as well as light, from the Earth could be measured. Finally, the signals that are measured electrically, are converted to digital values for storage and are later transmitted down to Earth. There, the visual images we are familiar with, are produced. This last step is highly dependent on computer technology for the assembly, organization, and interpretation of the data. The drawing of figure 1 shows an earth surface and atmospheric cross-section. A temperature scale at the left shows the decrease in temperature with an increase in height in the atmosphere. Shade in the temperature-scale blocks with your pencil. Start by making the bottom block darkest, the next one up lighter, the next lighter, and so on. Leave the top block white. The numbers in the drawing below indicate temperatures of various surfaces. For example, the lake surface is at +23oC, the upper surface of the fog bank is +18oC and the thunderstorm top is at a very cold -64oC. The rates of infrared (heat) radiation from objects are related to their surface temperatures. The higher the surface temperature, the greater the radiation. The lower the temperature, the less the radiation. Because of this, the cold tops of high clouds appear white while the tops of warmer low clouds appear grey in infrared pictures (unless the images have been enhanced).

Figure 1: Earth surface and Atmospheric Cross Section, Courtesy: Environment Canada

3


Figure 2a: GOES captured image on the weather of entire earth

Figure 2b: GOES Climate Image (20 Oct. 04, 3.05 pm) Processed by TRANSCAD

4


Weather Satellite Systems There are two types of weather satellites orbiting the earth [5]. One is Polar Orbiting Weather Satellites. It passes near the Earth's poles making north and south journeys at an altitude of about 800 kilometers. Polar orbiting satellites scan a strip of Earth, taking less than 2 hours to complete an orbit. With each pass, they survey a strip approximately 1900 km wide that is further west because of the Earth's eastward rotation. These satellites provide us with information on the condition of the ozone "hole" and composite pictures of snow cover and ocean surface temperatures. Another type is Geostationary Weather Satellites. Its orbit is located 35,800 kilometers directly over the equator. These satellites make one revolution, moving in the same direction as the Earth's rotation, in the time it takes Earth to make one rotation. This keeps them above the same spot on the equator, making them appear stationary, hence their name, Geostationary Operational Environmental Satellites (GOES).

Figure 3: Hurricane Charley just west of Fort Myers, Florida at 3.15 PM EDT 08/13/2004, Courtesy: NOAA

There are two geostationary satellites covering Canada and the United States. Each one has a field of view covering about one-third of the Earth's surface and their view remains the same; so sequential images may be viewed in rapid succession to show development and movement of

5


weather systems. Successive views from the same geostationary satellite can be provided to observe development of storm systems. They do not picture details as well as the closer polar-orbiting type of satellite, but they do provide more frequent views, every half hour, of the same Earth surfaces. Weather Satellite Characteristics Weather satellites are orbiting platforms from which onboard instruments can sense light and heat energy from the atmosphere and underlying surfaces. Because weather satellites can view a large area at one time, anywhere on Earth, they provide meteorological information over the oceans and sparsely populated land regions. Weather satellite pictures are received as composites of tiny blocks (called pixels) of varying energy intensities, often shown in shades of grey or in color. The area each block covers determines how detailed the image can be. In addition to sending back pictures of Earth, weather satellites can determine the temperature and water vapor content at different heights in the atmosphere. They can also monitor the ozone layer and detect energetic particles in the space environment. Weather Features in Satellite Imagery Hurricanes look like pinwheels of clouds. More often than not, the beginnings of hurricanes are detected from satellite views, because they occur over broad expanses of oceans [5]. Large comma-shaped cloud shields give shape and form to mid-latitude low-pressure systems. Clouds from which showers fall can look like grains of sand, especially on visible satellite pictures. Thunderstorms appear as "blobs" or "chains of blobs". Their high tops spread downwind from them as wispy cirrus clouds. They may have neighboring lower clouds appearing as tiny curved "tails" to the southwest. Such "tails" can also be indicators of the possibility of tornadoes. Movements of cloud patterns detected by viewing sequential satellite images, indicate the circulations of broad-scale weather systems. Wind speeds can be estimated at different levels and even upper-air jet streams can be identified. Meteorologists use satellite images to determine cloud shapes, heights, and type. Changes in these cloud properties, along with cloud movement, provide valuable information to weather forecasters to determine what is happening and what is likely to happen to weather in the hours and days ahead. Visible, infrared, and water vapor satellite imagery complement one another. There are weather features that can be clearly seen in one kind of image that are difficult to see in the others.

6


Visible Satellite Images Visible satellite images are views produced from reflected sunlight. Thus, these pictures look similar to pictures made with an ordinary camera. On visible satellite imagery, clouds appear white and the ground and water surfaces are dark grey or black. Since this imagery is produced by sunlight, it is only available during daylight hours. Low clouds and fog are usually distinguishable from nearby land surfaces. In addition, the hazy conditions associated with air pollution can be tracked. The shadows of thunderstorm clouds can be seen cast on lower clouds in the late afternoon. Snow cover can be monitored because it does not move as clouds do. Land features, such as streams, can be visible.

Figure 4: GOES visible satellite image for 00Z March 28, 1998, Courtesy: Environment Canada

7


Infrared Satellite Images Infrared satellite images are produced by the infrared (heat) energy Earth radiates to space. Since Earth is always radiating heat, infrared images are available day and night. On infrared images, warm land and water surfaces appear dark grey or black. The cold tops of high clouds are white and lower-level clouds, being warmer, are grey. Low clouds and fog are difficult to detect in the infrared when their temperatures are nearly the same as the nearby Earth surfaces. The enhanced images make it possible to keep track of land and oceanic surface temperatures. These surface temperatures play major roles in making and modifying weather. The high, cold clouds associated with severe weather are also easily monitored. Enhanced imagery can be interpreted to produce rainfall rate estimates. This information is used in flash flood forecasting.

Figure 5: GOES infrared satellite image for 13Z March 28, 1998, Courtesy: Environment Canada

Water Vapor Images Solid, liquid and vapor forms of water interact with specific ranges of infrared energy. Specially tuned geostationary weather satellite sensors can detect water vapor in the atmosphere, in addition to clouds. The water vapor sensors aboard weather satellites reveal regions of high atmospheric water vapor concentration in the troposphere between altitudes of 3 and 7 km. These regions, sometimes resembling gigantic swirls or plumes, can be seen to flow within and through broad scale weather patterns.

8


Recent studies suggest that, at any one time, atmospheric water vapor may be found concentrated in several large flowing streams forming the equivalent of "rivers in the sky".

Figure 6: GOES water vapor image for 2315Z on November 6, 2000, Courtesy: Environment Canada

Geographic Information Systems (GIS) GIS integrate layers of spatially oriented data through a variety of analytical approaches [6]. General advantages of GIS are: a) The easy of data retrieval; b) The discovery and display of information gained by observing; c) The capacity to process a large amount of data for spatial evaluation; d) Ability to make scale and projection changes, remove distortions, and perform coordinate rotation and translation; and e) The analysis of spatial relationships through the application of empirical and quantitative models [7]. Various GIS software packages such as: TransCAD, ARC/INFO, ARCView, SPANS, GRASS, MapInfo etc. available in the market. Irrespective of context, one should select a GIS tool which supports storing, displaying, managing, and analyzing weather and transportation data.

9


In addition it should provide:

• Powerful information retrieval engine with special extensions for transportation; • Mapping, visualization, and analysis tools designed for transportation applications; • Application modules for routing, travel demand forecasting, public transit, logistics, site

location, and territory management; • Fully integrated GIS with demand modeling and logistics functionality; • Creating and customizing maps, build and maintain geographic data sets, and perform

many different types of spatial analysis; • Include sophisticated GIS features such as polygon overlay, buffering, and geo-coding, and

has an open system architecture that supports data sharing on local- and wide-area networks.

• Able to predict network distances and travel times based on the actual shape of the road network and a correct representation of highway interchanges.

• Specify complex road attributes such as truck exclusions, delays at intersections, one-way streets, and construction zones;

• Provides a graphical solution that is easily understood; users can be able to convey highly technical information to the non-practitioner in a very straightforward and understandable manner.

Data Merging

This procedure is used to combine image data for a given geographic area with other geographically referenced data sets for the same area [6]. The other data sets are image data generated on other dates by the same sensor, by other remote sensing systems. All the weather forecasts are simple examples of data merging. Convenient images of cloud cover captured by weather satellites are of little uses unless it is references against jurisdictional boundaries such as nations, states, counties and the like. Figure 3, 4 & 5 show the shows overlay of national and state boundaries. Such pictures represent two data layers, one from satellite and another from archive image. The use of advection within the grid provides an accurate merge. Lakshmanan described a new way of merging data from multiple sensors, by constantly updating three-dimensional grid of data with data from the sensors. This allows a more current view than traditional merging methods. The merger processes was tested on a Linux Pentium III desktop, which easily keeps up with real-time feed from three weather service radars while using less than 100 MB of memory [10]. Researchers Group lead by the Georgia Institute of Technology developed a real-time, three-dimensional visualization system to help severe weather scientists improve the timeliness and accuracy of forecasting the formation, path and possible effects of storms. The system will allow weather researchers to use personal computers, as well as large-screen projections, to view, interrogate and analyze large observational data sets, including information from radar stations, severe weather detection software, high-resolution weather models, geographic information systems, satellites and aerial photography.

10


These sources will not only provide weather information, but also data on terrain, building locations and even human activities, such as rush-hour traffic. All of this data will be merged in a platform called the Virtual Geographic Information System (VGIS) previously developed by the project's lead researchers [11].

Figure 7: Merging of three weather service radars’ (KTLX in Oklahoma City, KSRX in Hot Springs and KINX in Tulsa) data on May 20, 2001 at 21.45 UTC, Courtesy: Lakshmanan 2003.

Intelligent Transportation Systems (ITS) ITS is a major focus of the last decade, seek to take existing detection, computer, and communication technologies and apply in them in an integrated way to increase the safety and efficacy of road transportation. For example, in Minneapolis-St. Paul, real-time data from freeway traffic counters are input into freeway capacity algorithms, which, in turn, automatically regulate traffic lights at entry ramps to improve traffic flow. ITS plays the major role in improving the efficiency in moving vehicles, goods and travelers over a transportation network, as well as increasing their safety by providing a real-time transportation systems (RTTS). RTTS include systems such as real-time traffic adaptive signal control systems, real-time ramp metering systems, real-time traveler information systems, real-time bus scheduling systems, real-time freight dispatch systems etc.[8] There are various complexities involved with implementing RTTS. Mirchandani (2004) has made an appreciable progress in studying these complexities, which includes: (i) uncertainties - the future traffic flow and future travel demand on a network is uncertain, therefore real-time systems must appropriately blend actual real-time

11


data with uncertain future data to make effective decisions; (ii) stochastic dynamics – traveler decision on routes, speeds, departure times, modes are randomly distributed, making traffic flow on a network is a stochastic dynamic phenomenon having components with different response time inertia; and (iii) online and real-time decision making – when data is streamed into the decision making platform using appropriate communication channels, are the associated online/real-time algorithm is the server’s problem of getting the best performance fore a given time budget allocated by the decision making client. Overview of Road Weather System Management Recent and anticipated advances in meteorology, roadway technology, and vehicle systems offer great opportunities to improve road weather decision making. Road weather systems could advance very quickly— probably within the next 15 years—if these can be implementation of existing technologies together. A robust new observation and communication “infostructure” that will make information available to roadway users where and when they need it [3]. This infostructure overlays the roadway infrastructure and would include: a data collection network, such as sensors embedded in the roadway and video monitoring devices at key points; a telecommunication system; and traffic management centers that receive and work with the data and then disseminate information to users. The infostructure would take advantage of advances in sensor technology, information technology, GPS, data management, computing, and geographical information systems, among others. Important decisions—such as when to slow down, when to salt roads for snow and ice, and when to pour or not to pour concrete—would be supported by “end-to-end” model-based tools to support decisions that integrate several elements: real-time observations of current weather, traffic, and road conditions; numerical weather predictions; models of traffic flow given observed conditions; and rules of practice. Cars and trucks of the future will be able to detect and respond to road weather conditions with ease and could stay in constant communication with weather information providers, traffic control centers, and other vehicles. These smart vehicles will inform drivers immediately of poor road conditions and of any obstacles ahead, with access to tools that determine optimum routes [3]. High costs of controlling snow and ice on roadways demand taking the guess work out of winter maintenance decisions. Enhanced roadway maintenance will be able to provide highly targeted weather information and decision support models that codify best practices and include cost-benefit analyses. Prototypes of this sort of decision support tool are already being developed by different private agencies. Enhanced traffic and emergency management will utilize traffic simulation models that dynamically forecast how traffic would most likely respond to weather, construction, accidents, and other road closures. Advances in Implementing Intelligent Transportation Systems Several advances have been made so far for the intelligent management of road weather system of the future. They are in place or under development, and roadway users are already benefiting from some recent advances. They are:

12


A. Dynamic Message Signs (DMS) DMSs are silent messengers strategically placed along highways to disseminate time-sensitive information to travelers on traffic flow rates, weather conditions, road closures, and alternate routes. Although this technology has been used for many years, today’s signs can now display real-time information received from a remote location. Drivers pay close attention to these signs, but it is not yet understood how drivers actually respond to the messages.


This center-to-roadside application area covers the interface between a traffic management subsystem and a specific type of roadway equipment that provides information to a vehicle operator -- the dynamic message signing (DMS). These signs may be deployed using various technologies, such as "flip panels," multiple lights, or light-emitting diodes and is capable of displaying a limited number of messages or a fully customized message using various fonts and colors. The primary purpose of these signs is to convey traffic conditions, weather conditions, and other traveler-advisory information to the vehicle operator. B. Vehicle Telematics VT include a variety of in-vehicle information and communication technologies and services. For example, since 1997, General Motors’ OnStar technology has offered drivers the ability to contact live, knowledgeable advisors for assistance if broken down, lost, or otherwise in need of service or information. Other companies are developing similar technology; there could be as many as 12 million telematics users by 2008. Advances in this field hold promise for improving how weather and traffic information is communicated to drivers.


13


C. Specific Road Weather Websites (SRWW) SRWW has been developed by several states and private firms that display real-time weather, road condition, and traffic information, including pavement temperature, air temperature, wind direction and speed, dew point, precipitation, pavement condition, and subsurface temperatures. Here are some examples of states, which are currently offering weather websites. Much of the focus is on winter weather driving hazards related to snow and ice. State Specific Road Weather Website

California http://www.dot.ca.gov/dist2/rwis/rwissites.php Colorado http://www.fhwa.dot.gov////////winter/roadsvr/CS117.htm Georgia http://gtalumni.org/StayInformed/magazine/win95/fog.html Illinois http://www.fhwa.dot.gov////////winter/roadsvr/CS039.htm Iowa http://www.fhwa.dot.gov////////winter/roadsvr/CS043.htm Minnesota http://www.fhwa.dot.gov////////winter/roadsvr/CS029.htm New Jersey http://www.fhwa.dot.gov////////winter/roadsvr/CS087.htm Tennessee http://www.tdot.state.tn.us/information-office/const.htm Washington http://www.fhwa.dot.gov////////winter/roadsvr/CS069.htm Wisconsin http://www.fhwa.dot.gov////////winter/roadsvr/CS038.htm D. Sensor Embedded in the Road Surface (SERS) The surface condition of roads is of great concern to Road Authorities and Road Maintenance people. During the winter months it is especially important to be informed of the changing surface conditions to enable road maintenance personnel to take the proper course of action. By knowing the road conditions, using anti-icing chemicals may be optimized and in many cases reducing maintenance cost. SERS relays weather, road condition, and traffic data to maintenance personnel and traffic managers. These data can be used to make forecasts, for example, about when road temperatures will reach the freezing point. In some locations, automatic de-icing devices are coupled with the temperature sensors so the devices will spray an anti-icing chemical on the road when it reaches freezing and water is present. E. Traffic Incident Management Systems (TIMS) Weather related traffic incidents cause bottlenecks on the roadways, slowing down or even stop the traffic. When traffic incident occurs, one or more lanes get blocked and a queue builds by the upstream vehicles due to the reduction of the capacity. These congestion and congestion related problems can be solved by applying efficient incident management systems. Traffic flow restoration unit (TFRU) is responsible for restoration of traffic. Each unit is equipped with multiple tow trucks, ambulances and so forth. In case of managing incidents they use two dispatching policies namely First Come First Serve (FCFS) – serve the incident that occurs first, or Nearest Neighbor (NN) – serve the nearest incident regardless when it happened. Both policies

14

http://www.fhwa.dot.gov////////winter/roadsvr/CS029.htm


have advantage and drawbacks. But the situations when there are not many TFRUs are active but the weather caused high traffic incidents, a very efficient TFRUs management is imminent. Xiao and Ozbay (2004) proposed a response surface methodology to manage TFRUs efficiently.

Response Surface Methodology (Xiao and Ozbay (2004)

The methodology had two assumptions. First, if the incident is not on the freeway service petrol (FSP) route, then the incidents have to be cleared by the TFRUs dispatched from the depot. Second, if the incident is not on the petrol route, which TFRU will respond to the incident will be determined by the shortest path rule. Based on the analysis, the result demonstrated that both depot and FSP are important in the incident management systems. For the depot service, besides the number of TFRUs, the location of the depot and the dispatching policies also has significant effect on the average incident duration [9]. If the total number of TFRUs assigned to depot and FSP remains constant, the decision makers will determine an optimal allocation of the total response vehicles between these two options.

15


16

Need Future Works Based on the study, research and analysis, the following five key research and development foci: 1. A robust, integrated observational network and data management system specifically designed to meet the needs of enhanced road weather research and operational capabilities; 2. A coordinated research effort to increase understanding of road weather phenomena and

develop options for increasing safety, mobility, and efficiency of the nation’s roadways during all types of weather;

3. Improved modeling capabilities and forecast tools designed to provide relevant, useful

information to those who build, maintain, operate, and use the nation’s roadways; 4. Multiple mechanisms for communicating road weather information to the range of users in

ways that support better informed decision making; and 5. An infrastructure that takes advantage of new technologies to monitor and predict road

weather conditions and then effectively convey road weather information to end users. REFERENCES 1. Leslie, J. (2004), NOAA News Online Magazine 2. Gerald J. Dittberner (2001), NOAA’s GOES satellite System – Status and Plans 3. NRC based Committee Report (2004), Where the Weather Meets The Road 4. www.foretell.com, US DOT Supported weather dissemination system 5. www.msc-smc.ec.gc.ca, Weather Satellites Teacher’s Guide Module 13 6. Chan, Y. (2001), Location Theory and Decision Analysis, ITP/Southwestern 7. Lee, Y. C.; Zhang, G. Y., (1989). “Development of Geographic Information Systems

Technology.” Journal of Surveying Engineering 115, No. 3:304-323. 8. Mirchandani, P. B., (2004), Intelligent Transportation Systems: A platform for Real-Time

Decisions, DCISSE and CSAM Colloquium of UALR. 9. Xiao, W.; Ozbay K.; (2004), “Response Surface Methodology Applied to Configuration of

Traffic Incident Management Systems.” Peer Reviewed Paper. 10. Lakshmanan, V., (2003), “Real-Time Merging of Multi-source Data.” NSF, FAA, and NSSL

funded Research Presentation at University of Oklahoma. 11. Sanders, J. (2001), “Seeing 3D in Real Time”. Visualization systems could improve severe

weather forecasting, http://www.gtresearchnews.gatech.edu/reshor/rh-f01/weather.html

http://www.foretell.com/

http://www.msc-smc.ec.gc.ca/

http://www.gtresearchnews.gatech.edu/reshor/rh-f01/weather.html

Satellite Weather Information and Intelligent ... Weather Information and Intelligent... ·...

Documents

Transcript of Satellite Weather Information and Intelligent ... Weather Information and Intelligent... ·...