Semi Automating Forecastsfor Canadian Airports in the

Great Lakes Areaby

 

George A. Isaac1,

With contributions from

Monika Bailey, Faisal S. Boudala, Stewart G. Cober,

Robert W. Crawford, Bjarne Hansen, Ivan Heckman,

Laura X. Huang, Alister Ling, and Janti Reid 

Cloud Physics and Severe Weather Research Section, and

Environment Canada

Great Lakes Operational Meteorology Workshop 2013Webinar – May 14, 2013

Acknowledgements

Funds from • Transport Canada • Search and Rescue New Initiatives Fund • NAV CANADA• Environment Canada

Also operations and research colleagues at CMC/RPN, others at CMAC-East (e.g. Stephen Kerr, Gilles Simard) and CMAC-West (e.g. Tim Guezen, Bruno Larochelle) and others within our Section (e.g. Bill Burrows)

Canadian Airport Nowcasting (CAN-Now)

• To improve short term forecasts (0-6 hour) or Nowcasts of airport severe weather.

• Develop a forecast system which will include routinely gathered information (radar, satellite, surface based data, pilot reports), numerical weather prediction model outputs, and a limited suite of specialized sensors placed at the airport.

• Forecast/Nowcast products to be issued with 1-15 min resolution for most variables.

• Test this system, and its associated information delivery system, within an operational airport environment (e.g. Toronto and Vancouver International Airports ).

On-Site Sfc MeasurementsObserver Reports

Radar Data

On-Site Remote SensingMicrowave Radiometer

Vertically Pointing Radar

NWP ModelsGEM REG, GEM LAM, RUC

Satellite Data

Lightning Data

Aircraft DataPilot Reports

Terminal Area Forecasts

Scientific AlgorithmsVisibility, RVR, Ceiling, Gust, Precipitation, Wind Shear, Turbulence, Cross-Winds,

AAR, CAT Level, etc.

Nowcasting MethodsABOM, INTW, Raw NWP,

Radar & Lightning Forecast, Conditional Climatology

Integrated Forecast and

Warning Product (Situation Chart)

Spatial Products for Specific Users

Time Series Products for

Specific Users

Real-time Verification

Products

Web-based Delivery System

Canadian Airport Nowcasting Forecast System (CAN-Now)

Forecaster InputTAF+, Blog

Climatology Data

On-Site Sfc MeasurementsObserver Reports

Radar Data

On-Site Remote SensingMicrowave Radiometer

Vertically Pointing Radar

NWP ModelsGEM REG, GEM LAM, RUC

Satellite Data

Lightning Data

Aircraft DataPilot Reports

Terminal Area Forecasts

Scientific AlgorithmsVisibility, RVR, Ceiling, Gust, Precipitation, Wind Shear, Turbulence, Cross-Winds,

AAR, CAT Level, etc.

Nowcasting MethodsABOM, INTW, Raw NWP,

Radar & Lightning Forecast, Conditional Climatology

Integrated Forecast and

Warning Product (Situation Chart)

Spatial Products for Specific Users

Time Series Products for

Specific Users

Real-time Verification

Products

Web-based Delivery System

Canadian Airport Nowcasting Forecast System (CAN-Now)

Forecaster InputTAF+, Blog

Climatology Data

Isaac, G.A., Bailey, M., Boudala, F.S., Cober, S.G., Crawford, R.W., Donaldson, N., Gultepe, I., Hansen, B., Heckman, I., Huang, L.X., Ling, A., Mailhot, J., Milbrandt, J.A., Reid, J., and Fournier, M. (2012), The Canadian airport nowcasting system(CAN-Now). Accepted to Meteorological Applications.

Algorithm Development• Visibility/Fog … RVR• Ceiling • Blowing Snow • Turbulence • Winds/Gusts/Shear • Icing• Precipitation Type• Precipitation Intensity• Lightning/Convective Storm• Real Time Verification

Main equipment at Pearson at the old Test and Evaluation site near the existing Met compound

7

•21 instrument bases with power and data feeds.•10m apart; rows 15m apart

1

4

32

56

78

9

1011

1213

14

1516

1718

1920

21

GTAAanemometer

NAV Canada 78D anemometer

#

MeteorologicalObservation Building1. Present Weather Sensor (Vaisala FD12P)

2. Spare3. Camera# Power distribution box4. Present Weather Sensor (Parsivel)5. 3D Ultrasonic Wind Sensor (removed)6. Microwave Profiling Radiometer (Radiometrics)7. Precipitation Occurrence Sensor (POSS)8. Icing detector (Rosemount)9. Precipitation gauge (Belfort) with Nipher Shield

Ultrasonic snow depth 10. Hotplate (Yankee – removed)11. Tipping Bucket rain gauge TB312. Precipitation Switch13. Spinning arm, liquid/total water content probe --

proposed14. 10 m Tower, 2D ultrasonic wind sensor15. Ceilometer (Vaisala CT25K)16. Vertically Pointing 3 cm Radar (McGill)17. Hotplate Precipitation Meter (Yankee)18. Temp, humidity, pressure, solar radiation19. Precipitation gauge (Geonor) with Nipher Shield20. Spare21. 10m Tower Spare

(Proposed or removed equipment)

Pearson Instrument Site

CAN-Now Situation Chart

Crosswinds:Dry RWY (precipitation rate ≤ 0.2 mm/h and visibility ≥ 1 SM): x-wind (knots) < 15 : GREEN15 ≤ x-wind (knots) < 20: YELLOW20 ≤ x-wind (knots) < 25: ORANGE x-wind (knots) ≥ 25 : RED (NOT PERMITTED)Wet RWY (precipitation rate > 0.2 mm/h or visibility < 1 SM): x-wind (knots) < 5 : GREEN 5 ≤ x-wind (knots) < 10: YELLOW10 ≤ x-wind (knots) < 15: ORANGE x-wind (knots) ≥ 15 : RED (NOT PERMITTED)--------------------------------------------------------------------------------------Visibility: vis (SM) ≥ 6 : GREEN (VFR) 3 ≤ vis (SM) < 6 : BLUE (MVFR)½ ≤ vis (SM) < 3 : YELLOW (IFR)¼ ≤ vis (SM) < ½ : ORANGE (BLO ALTERNATE) vis (SM) < ¼ : RED (BLO LANDING)--------------------------------------------------------------------------------------

Thresholds as applied on Situation Chart

Ceiling: ceiling (ft) ≥ 2500: GREEN (VFR)1000 ≤ ceiling (ft) < 2500 : BLUE (MVFR) 400 ≤ ceiling (ft) < 1000: YELLOW (IFR) 150 ≤ ceiling (ft) < 400: ORANGE (BLO ALTERNATE) ceiling (ft) < 150 : RED (BLO LANDING)--------------------------------------------------------------------------------------Shear & Turbulence: momentum flux FQ (Pa) < 0.75 : GREEN (LGT)0.75 ≤ mom. flux FQ (Pa) < 1.5 : YELLOW (MOD) mom flux FQ (Pa) ≥ 1.5 : RED (SEV) eddy dissipation rate (m2/3/s) < 0.3 : GREEN (LGT) 0.3 ≤ EDR (m2/3/s) < 0.5 : YELLOW (MOD) EDR (m2/3/s) ≥ 0.5 : RED (SEV) eddy dissipation rate (m2/3/s) < 0.3 : GREEN (LGT) 0.3 ≤ EDR (m2/3/s) < 0.5 : YELLOW (MOD) EDR (m2/3/s) ≥ 0.5 : RED (SEV)If the windspeed (relative to surface wind direction) exceeds, any of the following: level[2] (~125m/410ft) - level[0] >= 25 kts level[4] (~325m/1060ft) - level[0] >= 40 kts : RED level[5] (~440m/1440ft) - level[0] >= 50 kts --------------------------------------------------------------------------------------

Precipitation: rate (mm/h) > 7.5 : RED (HEAVY)2.5 < rate (mm/h) ≤ 7.5 : ORANGE (MODERATE)0.2 < rate (mm/h) ≤ 2.5 : YELLOW (LIGHT) 0 < rate (mm/h) ≤ 0.2 : GREEN (TRACE) rate (mm/h) = 0 : GREEN (NO PRECIP)--------------------------------------------------------------------------------------TSTM & LTNG:Lightning Distance ≤ 6 SM RED (TS)Lightning Distance ≤ 10 SM ORANGE (VCTS)Lightning Distance ≤ 30 SM YELLOW (LTNG DIST)Lightning within area (> 30 SM) YELLOWLightning forecast map received GREEN (NO LTNG FCST)--------------------------------------------------------------------------------------ICING:TWC < 0.1 g/m3 or TT ≥ 0°C GREEN TWC ≥ 0.1 g/m3 where TT < 0°C YELLOW (POTENTIAL ICING)

CAT-level: RVR (ft) < 600 RED (NOT PERMITTED) 600 ≤ RVR (ft) < 1200 -or- ceiling (ft) < 100 : RED (CAT IIIa)1200 ≤ RVR (ft) < 2600 -or- 100 ≤ ceiling (ft) < 200 : ORANGE (CAT II)2600 ft ≤ RVR < 3 SM -or- 200 ≤ ceiling (ft) < 1000 : YELLOW (CAT I) 3 ≤ RVR (SM) < 6 -or- 1000 ≤ ceiling (ft) < 2500 : BLUE (MVFR) RVR (SM) ≥ 6 -and- ceiling (ft) ≥ 2500 : GREEN (VFR)

--------------------------------------------------------------------------------------RWY Condition:precipitation rate (mm/h) > 0.2 : ORANGE (Possible WET rwy)precipitation rate (mm/h) ≤ 0.2 : YELLOW (Possible DRY rwy)

--------------------------------------------------------------------------------------Wx Only AAR:Cell colour is based on meteorological conditions – same as CAT-levelMeteorologically-limited theoretical maximum AAR determined from look-up table of documented AAR values based on runway configuration and meteorological conditions (CAT-level).Runway configuration determined solely from crosswind thresholds for maximum potential capacity.

Thanks to Bill Burrows

Web Site

• A Web site has been created at: http://saguenay-1.ontario.int.ec.gc.ca/cannow/cyyz/wx/index_e.php?airport=1

• The site is accessible externally only with a user name and password. The site is currently active in a research mode to obtain feedback..

Conditions

Change

Rapidly

WINTER CYYZ MAE SUMMER CYYZ MAE Variable REG LAM RUC 6h CLI REG LAM RUC 6h CLI

Temperature (oC) 1.7 2.3 1.9 3.9 1.5 1.7 1.1 3.0 Relative Humidity (%) 10.5 9.0 12.3 11.0 8.5 7.9 7.9 10.3 Wind Speed (m s-1) 1.6 1.2 1.4 1.8 1.4 1.4 1.2 1.6 Wind Direction (deg) 19.4 20.6 23.3 75.4 28.8 34.9 28.3 76.8

Max Wind Speed (m s-1) 2.3 2.4 1.7 N/A 2.3 2.6 1.5 N/A

Crosswind Rwy 1 (m s-1) 1.9 2.0 1.7 N/A 2.0 2.2 1.6 N/A

Crosswind Rwy 2 (m s-1) 1.9 2.0 1.7 N/A 2.0 2.2 1.6 N/A

Crosswind Rwy 3 (m s-1) 1.9 2.0 1.5 N/A 1.9 2.3 1.5 N/A

The mean absolute error for continuous variables for CYYZ. CLI refers to the error if a climate average were used as the predictor.

WINTER SUMMER Model

All WS WS > 5 kts All WS WS > 5 kts

REG 19.4 14.9 28.8 23.3

LAM 20.6 16.2 34.9 28.6

RUC 6h 23.3 18.1 28.3 23.3

Mean absolute error wind direction at CYYZ calculated with all the data and then when wind speeds less than 5 knots are removed.

CYYZ 1 December 2009 - 31 March 2010

00.5

11.5

22.5

33.5

44.5

5

00 03 06 09 12 15 18 21 00

Time of Day [UTC]

Tem

pera

ture

MA

E [

ºC]

REG

LAM

RUC6h

PERSISTENCE

CLIMATE AVG

CYYZ 1 December 2009 - 31 March 2010

0

0.5

1

1.5

2

2.5

3

3.5

4

00 03 06 09 12 15 18 21 00

Time of Day [UTC]

Win

d G

ust S

peed

MA

E [

ms-1

]

REG B01-B

LAM B01-B

RUC6h

PERSISTENCE

Theoretical Limit

NWP Models

Nowcasting

From Golding (1998) Meteorol. Appl., 5, 1-16

The main idea behind Nowcasting is that extrapolation of observations, by simple or sophisticated means, shows better skill than numerical forecast models in the short term. For precipitation, Nowcasting techniques are usually better for 6 hours or more.

Adaptive Blending of Observations and Models (ABOM)

Change predictedby model

Forecast atlead time p

CurrentObservation

Change predictedby obs trend

INTWINTW combines predictions from several NWP models by weighting them based on past performance (6 hours) and doing a bias correction using the most recent observation. (SMOW-V10 used GEM 1, 2.5 and 15 km)

Nowcasting Techniques Which Combine Model(s) and Observations

Related PapersIsaac, G.A., P. Joe, J. Mailhot, M. Bailey, S. Bélair, F.S. Boudala, M. Brugman, E. Campos, R.L.Carpenter Jr., R.W.Crawford, S.G. Cober, B. Denis, C. Doyle, H.D. Reeves, I.Gultepe, T. Haiden, I. Heckman, L.X. Huang, J.A. Milbrandt, R. Mo, R.M. Rasmussen, T. Smith, R.E. Stewart, D. Wang and L.J. Wilson, 2012b: Science of Nowcasting Olympic Weather for Vancouver 2010 (SNOW-10): A World Weather Research Programme project. Pure and Applied Geophysics. (DOI: 10.1007/s00024-012-0579-0).

Bailey, M.E., G.A. Isaac, I. Gultepe, I. Heckman and J. Reid, 2012: Adaptive Blending of Model and Observations for Automated Short Range Forecasting: Examples from the Vancouver 2010 Olympic and Paralympic Winter Games. Pure and Applied Geophysics. DOI 10.1007/s00024-012-0553-x.

Huang, L.X., G. A. Isaac, and G. Sheng, 2012: Integrating NWP Forecasts and Observation Data to Improve Nowcasting Accuracy. Weather and Forecasting, 27, 938-953.

Huang, Laura X, George A. Isaac, and Grant Sheng, 2012: A New Integrated Weighted Model in SNOW-V10: Verification of Continuous Variables. Pure and Applied Geophysics. DOI 10.1007/s00024-012-0548-7. Huang, Laura X, George A. Isaac, and Grant Sheng, 2012: A New Integrated Weighted Model in SNOW-V10: Verification of Categorical Variables. Pure and Applied Geophysics. DOI 10.1007/s00024-012-0549-6.

NWP Model with Minimum MAE in CAN-Now for Winter Dec 1/09 – Mar 31/10 and

Summer June 1/10 to Aug 31/10 Periods

Based on First 6 Hours of Forecast

Winter period – Dec. 1, 2009 to Mar. 31, 2010

Summer period - June 1 to August 31, 2010

Variable LAM REG RUC INTW

CYYZ CYVR CYYZ CYVR CYYZ CYVR CYYZ CYVR

TEMP 6 3 4 3.5 4.5 5 2.5 0.5

RH 6 6 no 6 no no 3.5 3

WS 2.5 3.5 4.5 3.5 3 no 1 2.5

GUST no no no 5 3.5 no 1.5 1.5

Winter

Summer

Variable LAM REG RUC INTW

CYYZ CYVR CYYZ CYVR CYYZ CYVR CYYZ CYVR

TEMP 2.5 2.5 2.2 2.5 1.5 no 0.5 0.5

RH 3 3 3.2 4.5 3 no 1 1

WS 3 5 3.5 5 2.2 no 1.5 2.5

GUST no no 5.5 no 2.2 no 0.5 4

Time (h) for Model to Beat Persistence

Huang, L.X., G.A. Isaac and G. Sheng, 2012: Integrating NWP Forecasts and Observation Data to Improve Nowcasting Accuracy, Weather and Forecasting, 27, 938-953.

0 1 2 3 4 5 60

0.5

1

1.5

2

2.5T

empe

ratu

re M

AE

[o C

]

Forecast Lead Time [hours]

CYYZ 1 December 2009 - 31 March 2010

LAMREGPERSISTENCEABOM LAMABOM REG

0 1 2 3 4 5 60

2

4

6

8

10

12R

H M

AE

[%

]

Forecast Lead Time [Hours]

CYYZ 1 December 2009 - 31 March 2010

LAM REG PERSISTENCEABOM LAMABOM REG

Shows the mean absolute error (MAE) in temperature and RH at CYYZ for the winter of 2009/10 as a function of forecast lead time averaged over the whole season. Temperature and relative humidity ABOM REG and ABOM LAM are compared to the raw model output and persistence.

Variable Category 1 Category 2 Category 3 Category 4 Category 5 Category 6 Category 7 Category 8Winds < 5 kts 5 ≤ w < 10

kts10 ≤ w < 15

kts15 ≤ w < 20

kts 20 ≤ w < 25

kts w ≥ 25 kts - -

Wind Direction

d ≥ 339 & d < 24º (N)

24 ≤ d < 69º (NE)

69 ≤ d < 114º (E)

114 ≤ d < 159º (SE)

159 ≤ d < 204º (S)

204 ≤ d < 249º (SW)

249 ≤ d < 294º (W)

294 ≤ d < 339º (NW)

Visibility v < 1/4 SM 1/4 ≤ v < 1/2 SM

1/2 ≤ v < 3 SM

3 ≤ v < 6 SM v ≥ 6 SM - - -

Ceiling c < 150 ft 150 ≤ c< 400 ft

400 ≤ c< 1000 ft

1000 ≤ c< 2500 ft

2500 ≤ c< 10000 ft

c ≥ 10000 ft - -

Precip Rate r = 0 mm/hr (None)

0 < r ≤ 0.2 mm/hr (Trace)

0.2 < r ≤ 2.5 mm/hr (Light)

2.5 < r ≤ 7.5 mm/hr

(Moderate)

r > 7.5 mm/hr

(Heavy)

- - -

Precip Type No Precip Liquid Freezing Frozen Mixed (w/Liquid)

Unknown - -

Table 2 (From Bailey's CAWW Talk)

Categories Being Used in CAN-Now Analysis

Heidke Skill Score: Multi-Categories

1 2 3 . . . . . K total

1 N(F1)

2 N(F2)

3 N(F3)

. . . .

K N(Fk)

total N(O1) N(O2) N(O3) N(Ok) N

Observed category

ForecastCategory

j

i Using:

Calculate:

WINTER CYYZ HSS & ACC SUMMER CYYZ HSS & ACC

Variable REG LAM RUC REG LAM RUC

Ceiling 0.45 0.62

N/A N/A

0.24 0.47

0.36 0.63

N/A N/A

0.33 0.67

Precipitation Rate 0.30 0.70

0.29 0.73

0.40 0.84

0.23 0.86

0.18 0.91

0.18 0.90

Visibility N/A N/A

N/A N/A

0.22 0.66

N/A N/A

N/A N/A

0.16 0.74

Visibility (BI09) 0.28 0.75

0.24 0.75

N/A N/A

0.15 0.78

0.08 0.80

N/A N/A

Crosswind Rwy 1 0.27 0.44

0.27 0.44

0.30 0.48

0.21 0.44

0.17 0.40

0.31 0.52

Crosswind Rwy 2 0.27 0.44

0.27 0.44

0.30 0.48

0.21 0.44

0.17 0.40

0.31 0.52

Crosswind Rwy 3 0.29 0.47

0.26 0.44

0.32 0.51

0.22 0.47

0.18 0.42

0.30 0.55

Precipitation Type 0.46 0.78

0.47 0.81

N/A N/A

0.36 0.90

0.29 0.94

N/A N/A

Maximum Wind Speed 0.18 0.35

0.19 0.35

0.30 0.45

0.07 0.30

0.11 0.32

0.29 0.48

Wind Direction 0.57 0.64

0.54 0.62

0.47 0.56

0.41 0.49

0.36 0.45

0.41 0.50

CYYZ Winter CYYZ Summer

Model

Variable Original M-C

HSS / ACC Relaxed M-C HSS / ACC

Original M-C HSS / ACC

Relaxed M-C HSS / ACC

Ceiling 0.45 / 0.62 0.46 / 0.61 0.36 / 063 0.30 / 0.55

Precipitation Rate 0.30 / 0.70 0.26 / 0.62 0.23 / 0.86 0.19 / 0.78

REG

Visibility (BI09) 0.28 / 0.78 0.27 / 0.70 0.15 / 0.78 0.16 / 0.73

Precipitation Rate 0.29 / 0.73 0.27 / 0.67 0.18 / 0.91 0.18 / 0.85 LAM

Visibility (BI09) 0.24 / 0.75 0.25 / 0.71 0.08 / 0.80 0.11 / 0.77

Ceiling 0.24 / 0.47 0.25 / 0.45 0.33 / 0.67 0.33 / 0.63

Precipitation Rate 0.40 / 0.84 0.40 / 0.81 0.18 / 0.90 0.18 / 0.85

RUC 6h

Visibility 0.22 / 0.66 0.19 / 0.60 0.16 / 0.74 0.15 / 0.70

The HSS and ACC scores for the relaxed set of criteria.

Summary

• Progress is being made to forecast aviation related variables using numerical model output and nowcast schemes. We already have a system which uses climatology (WIND III).

• RH predictions are poor, barely beating climatology. (Impacts visibility forecasts)

• Visibility forecasts are poor from statistical point of view. (also require snow and rain rates)

• Cloud base forecasts, although showing some skill, could be improved with better model resolution in boundary layer.

• Wind direction either poorly forecast or measured. • There are many difficulties in measuring parameters, especially

precipitation amount and type. • Overall statistical scores do not show complete story. Need

emphasis on high impact events.• Selection of model point to best represent site is a critical

process.

Summary (continued)

• Weather changes rapidly, especially in complex terrain, and it is necessary to get good measurements at time resolutions of at least 1 -15 min. CAN-Now and SNOW-V10 attempted to get measurements at 1 min resolution where possible.

• Because of the rapidly changing nature of the weather, weather forecasts also must be given at high time resolution.

• Verification of mesoscale forecasts, and nowcasts, must be done with appropriate data (time and space). Data collected on hourly basis are not sufficient.

• Nowcast schemes which blend NWP models and observations at a site, outperform individual NWP models and persistence after 1-2 hours.

Summary (continued)

• Currently using products to develop a First Guess TAF (FGT).

• The FGT system is being tested at the Aviation Weather Centres (CMAC-East and West) and is showing considerable promise, especially for VFR conditions.

• A recent IRP (last week) suggested many things that need addressing, including the verification of FGT and comparison with what forecasters are now producing. The algorithms definitely need some improvement (e.g. Low cloud is often predicted in Arctic under cold conditions when skies are clear, and there are issues with precipitation type)

Questions?