NOTE: The galeon mailing list is no longer active. The list archives are made available for historical reasons.
All -I've composed a draft of a couple of use cases pertinent to aviation weather. These use cases are slightly different than the one that Ben Domenico submitted, though complementary. In part of Ben's use case, data used as input to the model run (ie: assimilation) is served up via WCS - this is 100% applicable to NNEW as well. However, the use cases below all deal with using the *results* or output of a model run, or observational data (eg: radar). Also, the use cases below all focus on gridded continuous
data sets, rather than feature-like discrete datasets.Lastly, Im not sure how well these descriptions conform to exactly the definition
of a 'Use Case', or 'User Story', but I did my best.... -------1. As part of airport operations, an air-traffic controller [or dispatcher or ??] needs to monitor the weather conditions existing near a specific airport. The actor is specifically interested in adverse weather conditions that exist, or are forecasted to exist, within an airway, where an airway is defined as "a volume, defined by a set of XYZ vertices - an approach volume
around an airport is considered an airway".The actor would be interested in a set of products that may be deemed as potentially hazardous to flight operations, including the flights themselves as well as ground-based
operations. These products would include those that Ben mentioned: -- point data from lightning strike observations -- "station" observations from fixed weather stations in the region -- vertical profiles from nearby balloon soundings and wind profilers-- trajectory data obtained from other aircraft which have taken off and landed recently
-- volumetric scans from ground-based radars-- visible, infrared, and water-vapor (and possibly other wavelength) satellite imagery -- gridded output from local, national, hemispheric weather forecasts (eg: icing potential,
wind speed, convection, wind shear, turbulence).The actor would need to become aware of adverse conditions as the conditions occur in realtime, or as the conditions become known (eg: forecasted), so that he/she could respond accordingly (eg: require deicing procedures, reroute air traffic, cease ground operations, etc). The actor would need to be able to ascertain the severity of the conditions (eg: the values), the location (in XYZ) and the time for which they existed or are
forecasted. -------2. As part of (pre-) flight planning, commercial flight planners need to understand the weather conditions that exist or are forecasted along planned flight paths. For a particular flight plan, the planner needs to ascertain whether or not potentially dangerous conditions exist or will exist, and if so, whether or not the flight plan (or flight path) can be
(preferably insignificantly) altered to avoid those conditions.The flight planner knows the waypoints along the flight path for the flight in question, and those waypoints would include latitude, longitude, altitude (above sea level) as well as time. In other words, the flight planner knows where the aircraft is flying from/to, the intended cruising altitude(s) along each leg of the flight, and the departure/arrival time for each waypoint along the flight path. A waypoint simply represents a point in XYZTime,
and not necessarily an airport.To ascertain the weather conditions along the flight path, the planner extracts (from 'The System') subsets of gridded data products along the path, where those products are deemed as potentially hazardous to flight. A few examples of products are:
* Model outputted 4D gridded/continuous Icing potential field (ie: likelihood of icing)
* Model outputted 4D gridded/continuous Wind Speed field (ie: m/sec) * Model outputted 4D convection * Observational 2D radar product product showing cloud-tops* Observational 4D Airmet/Sigmets (polygonal shapes at altitude with adverse conditions) * Observational feature-like PIREPs product [non-gridded, non-continuous, non-WCS]
The flight planner would specify the waypoints for the intended flight path, including the X, Y, Z and Time coordinates. In addition, the flight planner would specify the geometry to be extruded from the volumetric product(s) along the flight path. The geometry could represent a vertical range, a horizontal range, or a radial range (ie: radius). Extracting a vertical range along the flight path would result in a vertical cross-section grid, that provides the flight planner with conditions vertically above and below the flight path. Extracting a horizontal range along the path would result in a horizontal cross-section grid, that provides the flight planner with conditions extending normal to the flightpath in the XY plane. Extracting a radius along the path would result in a cylindrical extrusion or volume [this may/not be a "grid"], that provides the flight planner with conditions within a certain radius from the flight path. [Also, it may be necessary to specify the number of sample points to be used along the path, as well as the algorithm used for determining the actual
path between waypoints, eg: Linear geometry or Great Circle, or ?.]Time is a component of each of the waypoints specified, though time can be either variant or invariant. If time is invariant, then the system should use the best data product valid for that time (eg: the most recent forecast product valid for that time, or the most recent observational data). If the time is variant, then the system should use some sort of (specified?) interpolation technique at each sample point, to interpolate/combine forecast results valid at different times (eg: 12:00 Z and 13:00 Z) into a single gridded result. In addition, different interpolation techniques may be applicable at each sample point for the Z dimension as well as the XY plane dimension. Note that not all fields can be interpolated (eg: wind direction can not be linearly interpreted between 0 degree and 180 degree values without potentially
resulting in misleading and even dangerous results).Regardless of what geometry is specified for extraction, the flight planner could use the returned grids to see if adverse conditions exist in an area of interest, when they existed or when they are forecasted to exist, and where the flight could be rerouted to avoid those conditions. From the results, the flight planner would need to be able to ascertain the severity of the conditions (eg: the values), the location (in XYZ) and the time for which they existed or are
forecasted. --------------------- These user stories are also available at the following: NNEW Wiki: https://wiki.ucar.edu/display/NNEWD/NNEW+Requirements GALEON Wiki: http://galeon-wcs.jot.com/GALEON%202%20Key%20Documents Comments/suggestions welcome... - Thanks, Rob
galeon archives: