Sector Occupancy Analysis with the Adverse Weather Diversion Model DIVMET

Contact information:

Manuela Sauer Email: sauer@muk.uni-hannover.de

Thomas Hauf Email: hauf@muk.uni-hannover.de

Motivation for model development Motivation & Objectives of the study Setup for feasibility analysis & Basic results Further results

M. Sauer1, P. Hupe1, L. Sakiew1, T. Hauf

1, C.-H. Rokitansky2, M. Kerschbaum3 1Leibniz Universität Hannover, Institute of Meteorology and Climatology 2Universität Salzburg, Computer Sciences 3Austro Control, Vienna

References & Additional information

[1] NAWPC, National Aviation Weather Program Strategic Plan. Prepared by the Joint Action

Group for Aviation Weather, for the National Aviation Weather Program Council. OFCM

Document FCM-P32-1997

[2] B. C. Bernstein, Integrated Icing Diagnostic Algorithm (WEB address:

http://www.rap.ucar.edu/largedrop/integrated)

[3] D. McNally, K. Shet, C. Gong, J. Love, C. H. Lee, S. Sahlman, and J.-H. Cheng, 2012,

Dynamic Weather Routes: A Weather Avoidance System for Near-Term Trajectory-Based

Operations (WEB address:

http://www.aviationsystemsdivision.arc.nasa.gov/publications/2012/

ICAS2012_McNally.pdf)

Adverse weather conditions, e.g. thunderstorms or icing, are

responsible for

• about 50% of all aircraft delays

• > 10% of all accidents and incidents in aviation

Main objective of any future adverse weather solution for aviation is

the reduction of delays and the increase of safety.

Understanding the interaction of the two complex systems

air traffic and adverse weather.

Investigation, exploration and development of an adverse

weather ATM solution model: DIVMET

Similar tools in the US developed at MIT-Lincoln Lab [1], at NCAR,

Boulder [2] and by NASA [3]. Necessity of these developments results

from increased weather related delays.

Note: Lack of equivalent tools to support ATM in adverse weather

conditions in Europe.

Development of the DIVMET algorithm

• Simulates the collaborative decision making

between pilot and ATC

• Proposes a short and safe route

through a field of storms

Needs

• Transformation of adverse weather into weather objects

• Weather objects are impenetrable for aircraft

• Account for motion, decay and generation of weather objects with

time

Squall line situation over Austria

Observations

• Nearly completely blocked air space over Czech Republic

• Highly blocked air space over Austria

• Unexpected shift of air traffic from Czech Republic to Austria

• Maximum workload for Austrian controllers due to increased

airspace sector occupancy

Key questions

Is it possible to predict the arising sector and work load using

weather forecast and/or nowcast models?

Can the shift of trajectories be simulated?

Is, especially, DIVMET suitable for sector load shift analyses?

Benefits

• Ability to schedule resources (personnel, airspace sector

distribution)

• Provision of deviation routes

• Avoidance of sector closing and holding patterns because of the

overall traffic situation and sector occupancy

longitude (°E)

latitu

de (

°N)

Lightning strikes from July 17th, 2010

Fig. 2: Lightning data over Austria from ALDIS

(received from Austro Control).

Fig. 3: Europe (Source:

http://de.wikipedia.org/wiki/Österreich).

Fig. 4: Radar composit over Austria and

parts of Czech Republic with overlayed flight

tracks of July 17th, 2010 between 07:00 and

07:05 PM (Source: Austro Control).

Radar composit of July 17th, 2010, 07:00-07:05 PM

Basic numerical experiment

• 1°x1° grid sectors

• Intended homogenous coverage of airspace by trajectories

• Approximated by 63 direct routes connecting all outer grid points

except those on the same border

• Flights only in one direction; no interaction and no conflicts

considered

• Main flow from the east and north

• Setting of route points (RP) every 15 s at a flight velocity of 280 m/s

Number of route points per sector, the sector route coverage

density, is assumed as a measure for sector load

Implementation of stationary weather objects

Three kinds of effects:

1. Crowding effect along the convex hull of a weather object

2. Blocked airspace by adverse weather with no routes

3. Compensating effects when considering a larger area

longitude (°E)

latitu

de (

°N)

Fig. 5: Generation of flight trajectories (total number of RP: 4439).

longitude (°E)

latitu

de (

°N)

planned routes new routes cloud

Fig. 6: Relative change of the number of route points (total number of RP: 4471).

1

2

3

• 0.5° x 0.5° grid sectors

Smaller sectors show larger effects

• Main flow

from west

and north

• Main

deviation to

the south due

to limited

knowledge

• Unlimited

knowledge

• Balanced

deviation to

the north and

south

• Shorter

detour

Conclusions

• Simulation of sector load shift and anticipated effects is possible

• More efficient routes and more balanced load of available sectors

in case of an increased radar field of view

• DIVMET is suitable for this application

Transfer to real conditions (airspace sectors, air traffic routes)

longitude (°E)

latitu

de (

°N)

planned routes new routes cloud

Fig. 7: Higher spatial resolution (0.5° x 0.5°) and sector load change.

planned routes new routes cloud

longitude (°E)

latitu

de (

°N)

longitude (°E)

latitu

de (

°N)

planned routes new routes cloud

Total number of RP:

4701 (+262)

Total number of RP:

4564 (+125)

Total number of RP:

4471 (+32)

Compared to the

undisturbed situation

Measure for the

detour length

Additional information

T. Hauf, L. Sakiew, and M. Sauer, Adverse weather diversion model DIVMET,

Submitted to Journal of Aerospace Operations

Adverse weather object, e.g. storm

Seen part of the weather object

Convex hull

Original, planned route

Deviation route

Flown route

Conical radar field of view

Fig. 1: Concept of the algorithm DIVMET. The

green line shows the given route of an aircraft.

The grey sectors represent the aircraft’s radar

field of view. If any weather object (blue) is

recognized in this field

of view and the aircraft will hit it or its safety

margin/convex hull (red) on the given route, a

decision is made and a rerouting via an

heading change will happen in the model.

Fig. 8: Artificial north-south oriented weather object with a limited

field of view (80 NM, 80°).

Fig. 9: Artificial north-south oriented weather object with an unlimited

field of view.

– – – –

– –

–

–

– –

–

–

– – –

356 393 393 355

382

348 389

359 360 378

388 338

- 5% - 2% - 22%

- 8%

- 4%

+ 5% + 6% +20%

+15%

+ 4%

+ 9%

+ 9%

+ 3% + 6% + 4% + 1%

+ 1% + 8%

+ 8% + 2% + 7%

+15%

+73% +11%

+11%

+14% +19%

+41%

+11% +18% +34%

+27% +49% + 6%

+ 6% +22% + 3%

+ 4% +17% + 6% +42% +10% +49%

+22% + 5% +13%

+35% + 4%

+31% + 5% +11% +65% +29%

+13% +20% +134% +15% +90% +15%

+ 9% +101% +10% +68% + 1%

+21% +27% +22% +23%

+17% + 1%

+ 8%

+ 3% +20%

+28% +38% +10%

+21% + 1% +35% +33% + 1% +13%

+53% +93%

- 7% - 8%

- 7% - 2%

- 2% - 2% - 5%

- 5% - 5% - 4% - 5% - 4% - 6%

-32%

-89% -26% -71% -22%

-96% -78%

-39% -17%

-67% -45%

-84% -64%

-67% -34%

-13% -13%

-17%

-28%

-14%

- 1% - 6%

- 5%

- 4% - 3% - 3% - 7%

- 6%

- 3% - 3% - 2%

- 9% - 9% - 9%

- 1% - 4% - 8%

- 1%

- 6%

-20% -98%

-46%

-12%

-11%

Flashes 4:00-4:29PM

Flashes 5:00-5:29PM

Flashes 6:00-6:29PM

Flashes 7:00-7:29PM

Flashes 8:00-8:29PM

Austria

Linz Innsbruck Vienna Salzburg Munich Zurich

Prague Klagenfurt Graz

This poster can be found on

http://www.muk.uni-hannover.de/ download/free/forschung/hauf/ AMS_2013_Poster_Sauer.pdf