CDM Convective Forecast Planning Guidance Help

Experimental Collaborative Decision Making (CDM) Convective Forecast Planning Guidance

Part 1 – Mission Connection

1. Product Description:

The Experimental Collaborative Decision Making (CDM) Convective Forecast Planning guidance is a graphical representation of convection meeting specific criteria of coverage, intensity, echo height, and confidence. The Experimental CDM Convective Forecast Planning guidance graphics are produced every 2 hours and valid at 2-, 4-, 6-, and 8-hours after issuance time. The product issuance time is approximately the bottom of the hour preceding a strategic planning telecon (i.e. the 1500 UTC issuance would be available at ~1430 UTC for the 1515 UTC strategic planning telecon).

The Experimental CDM Convective Forecast Planning guidance will be automatically produced from the following equally weighted models:

  • NOAA Short Range Ensemble Forecast (SREF),
  • High Resolution Rapid Refresh (HRRR) (3 most recent),
  • HIRES ARW model

The product will share the same format and dissemination method as the operational, forecaster-produced Collaborative Convective Forecast Product (CCFP). The Experimental CDM Convective Forecast Planning guidance will be made available during this period for user feedback.

2. Purpose/Intended Use:

The purpose of the Experimental CDM Convective Forecast Planning guidance is to aid in the reduction of air traffic delays, reroutes, and cancellations influenced by significant convective events. From a user’s perspective the Experimental CDM Convective Forecast Planning guidance is designed for strategic planning of air traffic flow management during the en route phase of flight. It is not intended to be used for traffic flow control in the airport terminal environment, nor for tactical traffic flow decisions. Specifically, the Experimental CDM Convective Forecast Planning guidance is used to support Federal Aviation Administration (FAA)-Airline CDM planning teleconferences which occur every two hours. It is a general strategic planning forecast baseline, as consistent as possible, shared among all meteorological organizations responsible for providing forecasts of convection to the air traffic managers within the FAA and/or within commercial aviation organizations.

As part of a larger effort to improve FAA and CDM decision making, the FAA has asked the NWS to experimentally produce the CDM Convective Forecast Planning guidance on two points. The first is to evaluate the need for convective weather information during the Collaborative Convective Forecast Product “off months” of November – February. The second is to evaluate the suitability of the Experimental CDM Convective Forecast Planning guidance in support of NAS strategic planning. The FAA/CDM vision is to supplement CDM Convective Forecast Planning guidance with an event-driven, impact-based Collaborative Aviation Weather Statement and a continual meteorological collaboration between industry and NWS meteorologists.

3. Audience/Users:

The primary users of the Experimental CDM Convective Forecast Planning guidance are FAA Traffic Flow Management and its CDM airline industry partners. The Experimental CDM Convective Forecast Planning guidance is the primary convective weather forecast for collaboratively developing a Strategic Plan of Operations (SPO) during this experimental period. The SPO is finalized during the collaborative TELCONS hosted by the FAA Air Traffic Control System Command Center Strategic Planning Team and conducted approximately every 2 hours immediately after the issuance of the Experimental CDM Convective Forecast Planning guidance.

4. Presentation Format:

The Experimental CDM Convective Forecast Planning guidance is available via the National Weather Service Telecommunications Gateway (NWSTG) circuit in an ASCII coded text format. An example of the Experimental CDM Convective Forecast Planning guidance ASCII coded text message is shown in the following graphic:

The format of the fields in the above graphic are described below.

General Format


Forecast Header Format

CCFP	CCFP Forecast Header (UTC)	4 Characters
ISSUED	Forecast Issuance Time (UTC)	CCYYMMDD_hhmm
VALID	Forecast Valid Time (UTC)	CCYYMMDD_hhmm

Forecast Area Format

AREA	AREA Type Header		4 Characters
AREAL COVERAGE				Convective Coverage Code1

Medium		=2			40-74%
Low		=3			25-39%

CONFIDENCE				Confidence
High		=1			50-100%
Low		=3			25-49%

GROWTH					Convective Growth Code
This field will always = 3 (No Growth)	(Convective Growth Code is not used in this experimental guidance)

TOPS  					Storm Height Code

FL400		=1
FL350-FL390	=2
FL300-FL340	=3
FL250-FL290	=4

SPEED 		Speed = 0		Knots = 0

(Polygon movement including speed and direction is not depicted in this experimental guidance)

VERT#  					Number of LAT/LON Pairs	Integer

LAT[x]LON[x]  				Vertical Latitude and Longitude Coverage Polygon

Latitude = LAT * 10.0 degrees
Longitude = LON * -1 * 10.0 degrees

LATT LONT 				Longitude and Latitude of Left Center of Box

Latitude = LATT * 10.0 degrees
Longitude = LONT * - 1 * 10.0 degrees
CANADA_FLAG*  	CANADA ON when MSC sends forecaster drawn polygons during
April thru September 2015

* Canada is not participating in the production of the Experimental CDM Convective Forecast Planning guidance.

The Experimental CDM Convective Forecast Planning guidance is also made available on the Aviation Weather Center (AWC) web site as an image.

5.Feedback Method:

Feedback will typically be collected via the survey below: and via comments provided to the webmaster. Opportunities for face-to-face responses will occasionally occur in the context of media workshops, public outreach events, etc.

For further information please contact:

Debra Blondin
National Weather Service
Aviation Weather Center
7220 NW 101st Terrace
Kansas City, MO
Phone: 816.584.7239

Part 2 – Technical Description

1. Format and Science Basis:

Algorithm Design Basics

The new CCFP was designed to mimic the human forecaster issued polygons to allow for a smoother transition from looking at the forecaster produced product to the automated product. In this regard, a climatology of forecaster-produced polygons was used as the basis for the algorithm design. The climatology of forecaster-produced polygons was collected over a period of 5 years. Information from the climatology included average polygon sizes stratified by polygon type, and convective coverage of composite reflectivity and echo tops in each polygon type. Since composite reflectivity has a correlation with echo top, and the CCFP is primarily an echo top forecast, it was determined that echo top would be the primary predictor of automated product. From the climatology study it was determined that there are three main types of polygons that a forecaster issues during the convective season. These include: sparse coverage-low confidence, sparse coverage-high confidence, and medium coverage-high confidence polygon. Very rarely was a solid coverage polygon issued (approximately only 4 times a season). It was also found that the polygons drawn exceeded the minimum size criteria of 3000 sq mi. The automated product therefore will only draw polygons similar in size to the forecaster drawn polygons. It was found that the minimum size of a forecaster drawn polygon was more on the order of 5000 sq mi for a medium coverage polygon and closer to 9000 sq mi for a sparse coverage polygon. The climatology study also determined that there were distinct bins of convective coverage found in the three main polygon types with the sparse coverage-low confidence having the lowest coverage, the sparse coverage-high confidence having slightly more convective coverage, and the medium coverage polygons having the most. At this time broken and solid lines of convection are not represented in the new CCFP. The line-type information is reserved for the event-driven CAWS. The final consideration was to maintain a common operating picture consistent with other products available at the ATCSCC. To this end, the new CCFP contains 3 of the most recent HRRR runs, a HIRES-ARW solution, and a SREF solution. The SREF solution uses the mean convective precipitation from all SREF members while the high resolution models use the echo top field for the determination of polygon areas.

Algorithm Description

The 4 high-resolution deterministic models are equally weighted to produce potential CCFP areas and then combined with the slightly coarser SREF model to provide smoother structure. Each high resolution model is considered by first filtering out echo top regions less than FL240, some lower echo tops are desirable for smoothness and to account for the potential low bias in the modeling solutions. The solutions are overlaid on each other and areas where there is substantial model agreement are identified. The algorithm requires that at least (2/5 or 40% of the solutions) agree. The next step is to remove polygons that do not reach minimum criteria determined by the climatology study. From the initial group of polygons that reach minimum size criteria, the average coverage of convection (based on echo tops greater than or equal to FL250) is calculated for each forecast member. Only forecast members that reach the climatology minimum coverages are allowed to contribute to the final convective coverage calculation. This ensures that if a line of convection is present in 3 of the members and the other 2 members do not have any convection that the highest coverage polygon will be issued. The polygon is then assigned as a sparse-low, sparse-high, or medium coverage polygon based on the historical observed coverages of the various polygon types. It is important to note that there is a slightly different criterion for medium polygons. Medium coverage polygons must have high coverage and be in one of the top 3 echo top bins. Medium coverage polygons are not allowed to exist at the lowest echo top level as those are less likely to disrupt enroute traffic. Once the polygon type is selected, the echo tops are considered and tagged with the highest 75th percentile of echo tops from each forecast inside the polygon. The highest echo top is used out of the 4 high-resolution model outputs. Once the polygons are created and tagged with echo tops, the algorithm is run through an image processing step to create the CCFP text product with latitude and longitude points of the given polygon types that can be ingested by a variety of software. Finally, any input from the Meteorological Services of Canada (MSC) over southeastern Canadian airspace are added to the text product.

2. Training:

No additional training is required to generate or use the Experimental CDM Convective Forecast Planning guidance.

3. Availability:

The Experimental CDM Convective Forecast Planning guidance will be available 7 days a week during the period from November 1, 2014 through March 1, 2016. The Experimental CDM Convective Forecast Planning guidance is updated every 2 hours.

The Experimental CDM Convective Forecast Planning guidance will be available at:

The ASCII files will be available to users via National Weather Service Telecommunications Gateway Under the following WMO Headers:

FAUS27 KKCI - 2 Hour Forecast
FAUS28 KKCI - 4 Hour Forecast
FAUS29 KKCI - 6 Hour Forecast
FAUS30 KKCI - 8 Hour Forecast