A Tool for Visualising the output of a

DBN for fog forecasting

Information Technology

Tali Boneh, Xuhui Zhang,

Ann Nicholson & Kevin Korb

Faculty of Information Technology

Monash University

Overview

• The problem: fog forecasting (at airports)

• Previous BN modelling:

– “static” Bayesian decision network (no explicit time)

– Operational decision support tool since 2006

• DBN for fog forecasting

• Fog DBN Visualisation tool

Fog Formation

Fog Forecasting for Airlines

• Terminal Aerodrome Forecast (TAF) issued every 6 hours; valid for 30h

3pm ---------------------------------| ~ 9am (Perth) /11am (Melbourne)

9pm ------------------------| ~ 9am

3am ---------------| ~ 9am

• When fog probability ≥ 30%, fog included in the TAF

• If fog forecasted, aircrafts must carry enough fuel to

– reach an alternative airport

– maintain a holding pattern above the airport

• When chance of fog >5% but <30%, forecast is “Code Grey”

ARC Linkage 2002-2004

2005 2006- 2008 2010 2009 2004 2003 2011 2015 2012 2013

Boneh’s PhD 2003-2008

ARC Linkage 2012-2014 (2015) Boneh post-doc

Boneh @ BoM

2002 2014

1. ARC Linkage 2012-2014. Improving Meteorological Forecasting with Knowledge Management Systems.

2. ARC Linkage 2012-2014.Temporal and spatial Bayesian network modelling for improved fog forecasting ARC $224K, BoM $148K

Boneh , T. (2010) Ontology and Bayesian Decision Networks for Supporting the Meteorological Forecasting Process. PhD thesis, Monash University

Monash and Australian Bureau of Meteorology collaboration 2003-2015

BOFFIN decision support tool for forecasting fog at Melbourne airport

Phase 1: 2004 – 2008 (static Bayesian Decision Network)

Four priority airports were chosen for fog forecast improvement:

• Melbourne (~ 12-13 fogs)

– Networks for 3pm, 6/9pm, midnight

• Sydney (~ of 4 to 5 fogs per year, largest traffic volumes),

– Networks for midnight, 2am and 3am

• Canberra (~ 42 fogs)

– Networks for 6pm, midnight and 3am

• Perth (~ 12 fogs per year, large

distances to the nearest alternate airports

– A network for 3pm

Prediction: Fog Yes or No?

No prediction of onset and clearance

Month

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember

8.467.718.468.188.468.188.708.578.188.468.188.46

LengthOfNight

Nov to JanFeb and OctMarch and SeptApr and AugMay to July

25.116.216.616.825.3

RainNoRain

0 to 4.5>= 4.5

90.69.41

Fog

fognofog

3.3096.7

U

LapseRate9pmCont

< 2.052.05 to 2.752.75 to 3.25>= 3.25

26.218.117.737.9

2.74 ± 0.75

Moisture

Vfavfavunfav

26.419.254.4

SternParkyn

0 to 11 to 22 to 55 to 1010 to 1515 to 3030 to 100

46.815.117.59.794.124.402.28

4.8 ± 11

Decision

saynofogCodeGreyLessThan5CodeGrey5CodeGrey10CodeGrey20sayfog

28.704428.937828.029326.391524.670022.6587

Gradient

Vfavfavunfav

31.219.349.4

“Static” semi-causal BN for fog forecasting

Decision

saynofogCodeGreyLessThan5CodeGrey5CodeGrey10CodeGrey20sayfog

28.704428.937828.029326.391524.670022.6587

Month

JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember

8.467.718.468.188.468.188.708.578.188.468.188.46

LengthOfNight

Nov to JanFeb and OctMarch and SeptApr and AugMay to July

25.116.216.616.825.3

RainNoRain

0 to 4.5>= 4.5

90.69.41

Fog

fognofog

3.3096.7

U

LapseRate9pmCont

< 2.052.05 to 2.752.75 to 3.25>= 3.25

26.218.117.737.9

2.74 ± 0.75

Moisture

Vfavfavunfav

26.419.254.4

SternParkyn

0 to 11 to 22 to 55 to 1010 to 1515 to 3030 to 100

46.815.117.59.794.124.402.28

4.8 ± 11

Gradient

Vfavfavunfav

31.219.349.4

Background

Decision

Guidance

Predictors

Target

BOFFIN: deployed 2006

• Decision support tool in operational use for fog forecasting at Melbourne airport – BN as back-end reasoning engine

– Integrated with data stream of current weather observations

– Allows forecasters to adjust data entered as evidence (“what-if” functionality)

• Improved forecasting 2006-2014 (fewer missed fogs without increase in false alarms) Boneh et al. (To Appear). Weather and Forecasting. Fog Forecasting for Melbourne Airport using a Bayesian Decision Network.

• Limitation: no temporal aspect to prediction, i.e. time of onset and clearance

Dynamic Bayesian networks

• For explicit reasoning over time, a copy of each process variable X at each time T: X0, X1, … XT-1, XT, XT+1, …

• Typically: – Same structure within each time slice (intra-slice arcs)

– Stationary time series (i.e. parameters not time dependent), so summarised in two time slice DBN (with sliding window), used with “roll-up, roll-out” approximation

Weather

Fog

Weather

Fog

Weather

Fog

Clearance Onset times

Clearance Onset times

Weather Weather

Forecast time Now

Forecast time Future

Forecast time Future

NWP Predictions

Current Obs

NWP Predictions

Current Obs

NWP Predictions

DBN Framework for Fog

Weather

Fog

Weather

Fog

Weather

Fog

Clearance Onset times

Clearance Onset times

Weather Weather

Forecast time Now

Forecast time Future

Forecast time Future

NWP Predictions

Current Obs

NWP Predictions

Current Obs

NWP Predictions

DBN Prototype 1 Add observations directly

Non-causal (NB)

12

Melbourne Fog DBN: Prototype 1

14

5 Weather variables (discretised): children of both “length of night” and the Fog node

15

Melbourne Fog DBN: Prototype 1

16

Change nodes: If Fog, Clearance? Time of clearance? If NoFog, Onset? Time of Onset? If NoFog, then Onset, will it clear?

Melbourne Fog DBN: Prototype 1

17

Visualisation and understandability: 8 time slices (Netica demo)

ForecastTime Info.

Nodes midN 3am 6am

F_F_TmidN ~{fog} nofog1 nofog1

delta_F1_TmidN * onset onset

delta_F2_TmidN * nochange nochange

Onset_TmidN * 0.25 0.25

Clear_if_Clearance_TmidN * * *

Clear_if_Onset_TmidN * * *

Wvis_TmidN 1 1 1

F_F_T3am * fog fog

delta_F1_T3am * * clearance

delta_F2_T3am * * *

Onset_T3am * * *

Clear_if_Clearance_T3am * * 1

Clear_if_Onset_T3am * * *

Wvis_T3am * 1 1

F_F_T6am * * ~{fog}

delta_F1_T6am * * *

delta_F2_T6am * * *

Onset_T6am * * *

Clear_if_Clearance_T6am * * *

Clear_if_Onset_T6am * * *

Wvis_T6am * * 4

First attempt at presenting DBN outputs A fog case (fog 00:15 - 4:00)

midN

3am

6am

net slice

First attempt at presenting DBN outputs A fog case (fog 00:15 - 4:00)

probabilities of ‘fog now’ at different forecast times

forecast time 9pm

0.0

01

00

94

3

0.0

01

38

59

0.0

02

10

72

9

0.0

02

93

86

1

0.0

05

80

14

2

0.0

10

74

48

0.0

20

31

79

0.0

33

99

42

0 0.0

08

58

42

3

0.0

21

45

69

0.0

35

59

07

0.0

51

64

43

Pro

bab

iliti

es

Start interval Time

forecast time 3am

0.0

01

00

94

3

0.0

01

38

59

0.0

02

10

72

9

0.0

02

93

86

1

0.0

05

80

14

2

0.0

10

74

48

0.0

20

31

79

0.0

33

99

42

1

0.9

77

57

8

0.9

03

71

4

Pro

bab

iliti

es

Start interval Time

forecast time midN

0.0

01

00

94

3

0.0

01

38

59

0.0

02

10

72

9

0.0

02

93

86

1

0.0

05

80

14

2

0.0

10

74

48

0.0

20

31

79

0.0

33

99

42

0

0.0

39

37

63

0.0

47

93

43

0.0

53

80

41

Pro

bab

iliti

es

Start interval Time 0

.00

10

09

43

0.0

01

38

59

0.0

02

10

72

9

0.0

02

93

86

1

0.0

05

80

14

2

0.0

10

74

48

0.0

20

31

79

0.0

33

99

42

0

0.0

79

93

13

Pro

bab

iliti

es

Start interval Time

Prior 6am

Forecast Time forecast time 6am

forecast time 6pm 0

.00

10

09

43

0.0

01

38

59

0.0

02

10

72

9

0.0

02

93

86

1

0.0

05

80

14

2

0.0

10

74

48

0.0

20

31

79

0.0

33

99

42

0 0.0

00

69

47

23

0.0

04

68

56

8

0.0

12

91

31

0.0

27

53

23

0.0

48

80

81

Pro

bab

iliti

es

Start interval Time

• Our forecasters had no idea what the DBN was doing/saying • And we weren’t sure either!

Fog DBN Visualisation Tool

• Developed using D3, a JavaScript library for manipulating documents based on data within a web browser, uses a combination of HTML, SVG, and CSS

• Extends Matthew Weber’s “block” template

• Developed in collaboration with visualisation research group at Monash

Selected Cases for Demo

22

# Actual DBN date onset clearance Comment and forecast at 3pm

1 fog fog hit

28/06/2011 19:00 2:30 Always looked like fog (TAF)

2 no fog no fog correct

4/07/2003 Always looked like no fog (TAF)

3 fog no fog miss

3/07/2008 7:21 8:00 Looked like no fog until 6am (CG)

4 no fog

fog false alarm

21/05/2009 Fog on previous day cleared at 9:47am (TAF)

X-axis: time over night Y-axis: P(X=x|Observations to date)

30% chance of fog

Prior Select X=x To

display

Case “fog hit” (fog event successfully forecast): sequence of observations for weather variables and fog, from midday to following 9am

Forecast time – here midday

Visualisation of forecast

time

fogTonight: hovering over point brings up probability

Note: Fog prior not selected but is always shown “faded”

Displaying more info: e.g. probability of fogOnset

Displaying more info: e.g. probability of fogOnset

Overlay showing not only probability for fogOnset in next 3 hrs, but distribution of times for onset and clearance

Overlay showing not only probability for fogOnset in next 3 hrs, but distribution of times for onset and clearance

Probability of “fogNow”, option for overlay with distribution for time of fog clearance

Demo

No fog case, correctly predicted

No fog case, false alarm

Showing probabilities for single value, at different forecast times

No fog case, false alarm

No fog case, false alarm

Where are we at?

• Bur. of Met collaborators ‘like’ the DBN visualisation tool

– Understanding of the DBN model

– Assisting with ongoing development

• No usability testing as such

• Currently doing lots of experimental work on DBN – Selection of weather variables as predictors

– Discretisation of weather variables, onset and clearance times

• Specific research project outcomes:

– prototype DBN, papers

• Future: embed DBN in support tool for operational use (1-2 years?)