Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in - - PowerPoint PPT Presentation

▶
uncertainty analysis of thunderstorm nowcasts for
SMART_READER_LITE
LIVE PREVIEW

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in - - PowerPoint PPT Presentation

Mar 05, 2024 234 likes •514 views

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer 1 , Thomas Hauf 1 , Caroline Forster 2 1 Leibniz Universitt Hannover, Institute of Meteorology and Climatology 2 DLR, Institute of Physics of the

slide-1
SLIDE 1

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing

Manuela Sauer1, Thomas Hauf1, Caroline Forster2

1Leibniz Universität Hannover, Institute of Meteorology and Climatology 2DLR, Institute of Physics of the Atmosphere

  • 25. November 2014 – 4th SESAR Innovation Days
slide-2
SLIDE 2

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

2

Outline

  • Thunderstorms, motivation and nowcast data used
  • Methodology and results of the uncertainty analysis
  • Application in weather avoidance routing
slide-3
SLIDE 3

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

3

Thunderstorms

  • Small scale phenomena

– cell diameter: 5 - 50 km – life time: 20 - 60 min

  • After generation: further life cycle and movement

is in principal predictable

  • Prediction of onset?

– generation mechanisms need to be forecasted:

  • deterministic components: orography, surface convergence lines
  • stochastic components: gravity waves, temperature inhomogeneities

not predictable! UNCERTAINTY

Source: http://www.bom.gov.au/storm_spotters/handbook/thunderstorms.shtml

slide-4
SLIDE 4

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

4

Motivation

event approaching an event in time and space lead time

slide-5
SLIDE 5

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

5

Rad-TRAM nowcast system

Radar Tracking and Monitoring

  • Based on radar composit

with timely resolution of 5 minutes

  • bject-based (> 37 dBZ)
  • Pyramidal image matcher
  • displacement
  • life cycle: growth and decay
  • f cells

Source: http://www.pa.op.dlr.de/MUCSOMMER/2012/ Further information: Tafferner, A., C. Forster, 2012: Weather Nowcasting and Short Term Forecasting, in Atmospheric Physics,

  • U. Schumann (ed.), Springer Verlag
slide-6
SLIDE 6

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

6

Rad-TRAM nowcasts

time lead time

t0 t5 t10 t15 t t+5 t+10 t+15 t+20 t+60 t+55 t55 t65

no further detection of this cell

…

timely resolution: 5 minutes horizontal resolution: 2 km

t60

slide-7
SLIDE 7

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

7

METHODOLOGY & RESULTS

Nowcast uncertainty determination

slide-8
SLIDE 8

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

8

Methodology

Δdistright Δdistleft Δdistforward Δdistback

N

  • 0 +

t30,right

Δdistright

N

  • 0 +

t30,forward

Δdistforward

N

  • 0 +

t30,left

N

  • 0 +

t30,backward

Δdistback

Δ = obs - nowcast Frequency distributions

slide-9
SLIDE 9

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

9

Results

Δdistright Δdistforward

N

t30,forward

Δdistforward

N

t30,back

Δdistback

N

t30,right

Δdistright Δdistleft

t30,left

N

90th percentile Cumulative distribution function

20120715

slide-10
SLIDE 10

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

10

Increase of uncertainty with lead time

LEFT RIGHT lead time

60 min 5 min 35 min

20120715

  • bservation is

always larger than nowcast in direction right 17.14 km P90,abs,60min 18.24 km 1.98 km P90,abs,60min 3.79 km

slide-11
SLIDE 11

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

11

17.14 km 18.24 km 1.98 km 3.79 km

Increase of uncertainty with lead time

LEFT RIGHT lead time

60 min 5 min 35 min

20120715

slide-12
SLIDE 12

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

12

80% intervals in all directions

100 km

nowcast time: 35 min

moving direction

RIGHT, 35 min BACKWARD, 35 min LEFT, 35 min FORWARD, 35 min

nowcast mean uncertainty 10th and 90th percentile

20120715

slide-13
SLIDE 13

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

13

80 % intervals in all directions

moving direction 5 min 60 min 35 min

20120715

slide-14
SLIDE 14

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

14

Differing results for different days/situations

July 15th, 2012 July 15th & 10th, 2012

No characteristic distribution

  • n right extent!
slide-15
SLIDE 15

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

15

Separate uncertainty analysis – propagation error

gravity center moving direction

5 min 60 min 35 min

distance lateral to mov direction 50 km Gravity center displacement (rel. to obs) distance in mov direction 50 km 50 km 50 km 50 km 50 km 50 km 50 km 0 km

No systematic misplacement!

20120715

slide-16
SLIDE 16

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

16

Δdistleft

Separate uncertainty analysis – extension error

gravity center

Δdistright Δdistforward Δdistback

5 min 60 min 35 min

distance 30 km Extension difference in direction RIGHT relative counts

  • 30 km

1 1 1 30 km

  • 30 km

30 km

  • 30 km

Relative small uncertainties (P90,rel,right,60min = 5.87 km)

20120715

slide-17
SLIDE 17

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

17

Correlated distances in separate analysis

Regular growth or shrinkage on both sides Equivalent correlation between FORWARD and BACKWARD deviations

20120715

slide-18
SLIDE 18

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

18

Regressions:

Integrated uncertainty development

20120715

slide-19
SLIDE 19

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

19

Integrated uncertainty development

Larger 90th percentiles for directions BACKWARD and RIGHT

20120715

slide-20
SLIDE 20

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

20

APPLICATION IN WEATHER AVOIDANCE ROUTING

Nowcast uncertainty determination

slide-21
SLIDE 21

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

21

Add uncertainty margin around a nowcast

movement direction

35 minutes nowcast

  • f a point cell

90th percentile uncertainty margin

20120715

slide-22
SLIDE 22

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

22

movement direction

~ 1.3° 525 km at 480kt 12 km

20120715

Add uncertainty margin around a nowcast

Additional heading change to that forced by the nowcasted cells extent.

slide-23
SLIDE 23

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

23

  • Spatial uncertainty determination method of nowcast cells

considering extent and displacement

– Isotropic behavior for both components – BUT: integrated uncertainty shows characteristic features

Conclusion I

Thank you for your attention! sauer@muk.uni-hannover.de Integral view Separate view

slide-24
SLIDE 24

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

24

  • Spatial uncertainty determination method of nowcast cells

considering extent and displacement

– Isotropic behavior for both components – BUT: integrated uncertainty shows characteristic features

  • Quantified rate of increasing uncertainty with lead time

.

Conclusion II

Thank you for your attention! sauer@muk.uni-hannover.de

slide-25
SLIDE 25

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

25

  • Spatial uncertainty determination method of nowcast cells

considering extent and displacement

– Isotropic behavior for both components – BUT: integrated uncertainty shows characteristic features

  • Quantified rate of increasing uncertainty with lead time
  • Uncertainty is even higher due to

– new, not yet nowcasted cells – Other not yet considered statistics

  • dissipating cells
  • merged/splitted cells
  • Further analyses on different thunderstorm situations

Conclusion

slide-26
SLIDE 26

Uncertainty Analysis of Thunderstorm Nowcasts for Utilization in Aircraft Routing Manuela Sauer et al. 4th SESAR Innovation Days, Madrid

26

THANK YOU

Contact information: Manuela Sauer sauer@muk.uni-hannover.de Thomas Hauf hauf@muk.uni-hannover.de Caroline Forster caroline.forster@dlr.de