Spinning points, soggy polygons
The extensive coverage, localized detail, and frequent refresh rate of this
nationwide array of Doppler sensors, combined with the importance of weather
prediction, makes Meteorlogix’s boast of serving "more then 22,000 customers
with a focus on public safety, broadcast media, transportation, energy, and
aviation industries" seem reasonable.
Dr. Clive Reece, GIS products manager for Meterologix, explained the company’s
history as follows:
We’re the merger of three weather companies: DTN Weather Services
focused on real-time satellite delivery, Kavouras out of Minneapolis focused
on media weather graphics and high-end display (such as an SGI workstation)
of weather for television stations, and Weather Services Corporation out of
Boston focused on long-term climatology prediction. We were brought under one
umbrella [forgive the pun!] about two years ago. One of our business initiatives
has been weather-enabling GIS. We began by taking our primary workstation product
that received satellite information and created graphics, and added operational
conversion routines to produce point and polygon shapefiles of our real-time
weather feed. Since then, the business has expanded to GIS-based automated
weather decision-support systems and custom services.
Meteorlogix has discovered that the GIS community wants Internet access.
Most users, says Reece, "don’t want to punch a hole through their roof and put a
satellite feed in." Consequently, Meteorlogix is expanding its options to include
Internet delivery, providing weather shapefiles through FTP and ESRI (www.esri.com)
ArcIMS image and feature services. Satellite delivery, with its timely and reliable
data stream that remains free of Internet traffic, still remains the best choice for
more mission-critical applications in public safety and transportation, however.
Whether with weather -- decision support applications
Organizations that integrate weather services data into their spatial systems
want to know local weather conditions in either the present, the future, or both.
The rain in Spain (on the plain!). For weather conditions right now in a
specific location, the Meteorlogix feed is accurate to within 5 or 10 minutes,
depending on the kind of data required. Organizations such as the military or
utilities that require an accurate field response stand to benefit. Likewise,
awareness of real-time storm tracks improves routing, logistics coordination,
and operating efficiencies for local and national transportation systems. The
efficiencies can be realized in more-precise air traffic routing around rapidly
changing weather and the transport of hazardous materials around electrical
storms and high-wind events (See Figure 3). For instance, trucking companies
want to protect their property and personnel, but also want to keep whatever
they’re transporting moving as fast as possible. Adding meteorlogical data to
the balance sheet of a delivery more accurately accounts for the time lost when
avoiding weather that might damage cargo or hurt personnel.
Figure 3: Aviation traffic controllers use
meteorlogical data to better estimate flight delays and to reroute planes to
safer landing locations, as in this example from the Detroit region. Rather
than searching through a textual list of weather alerts, a traffic controller
can hover the mouse over airports on the ESRI ArcMap display to see their current
textual weather report.
Red sky at morning? Forecasting the weather is an age-old practice seeing
sharp rises in sophistication thanks to technological convergence of data and models.
Homeland security managers, for instance, can now integrate wind data and 3D building
data with plume dispersion models to create timely and highly accurate projections of
airborne concentrations of chemical gases from accidents or terrorist attack (see
Figure 4: This hypothetical arsine plume
dispersion model overlays Orlando’s population with areas of gas exposure. The wind
conditions from Meteorlogix feed into SAIC’s CATS (cats.saic.com) toolset before
being passed to ALOHA for plume generation. Map display is ESRI’s ArcMap.
Forecasting seems to touch almost every spatial industry vertical market. Emergency
responders and city planners can use radar-derived rainfall information to increase
the accuracy of water management and flash flood forecasting, estimating not only
how much water will accumulate, but where it will fall. Insurers can use forecasts
to analyze potential storm damage before a hurricane hits their customers’ region
(in some cases using historic records of storms as a qualifier). Some NEXRAD data
feeds calculate areal amounts of rainfall in one hour, three hour, and 24-hour time
periods. When input into hydrologic flood models as polygons, these data predict
which watersheds are at risk of flooding. Even the spatial industry itself can use
forecasts of cloud cover to schedule remote sensing activities more intelligently.
And there’s a payoff. For instance, cities in cold climates have considerable
budgets for snow removal, including putting chemicals on the roads and deploying
plows strategically around town. Defining a threshold, for instance, of 0.5 inches
of snow, allows their system to sound the alarm when snowplows should begin plowing
combining forecast, areal extent of fallen snow, and the amount and the timing of
snowfall with crew schedules and availability can help dispatchers decide when to
call more people in, and where to deploy the trucks. Inadequate response to an
unexpectedly heavy snowfall, on the other hand, can exact not only a financial, but
a political toll, as happened last year in a Chicago blizzard (see