Week one of the 2017 Summer Experiment wrapped up on Friday with each desk having a breakout session to discuss takeaways from the week. Participants were able to give feedback from their perspective on the TCF forecast creation process, the use of the UK-Met model to predict clearing times at SFO, and using the national grids of clouds and visibility at the GFE desk.
UK-MET / SFO
Overall, it seemed the UK Met model did in fact add some value when trying to forecast the clearing time at SFO, particularly with how it treated the effects of the terrain to the west. The range of clearance times among the various high-resolution models analyzed was quite varied. For instance, on Thursday the 10th, the models were showing clearing times between 18-21Z for the forecast produced at 15Z. However, the order that the models cleared the stratus was pretty consistent from day to day. The HRRR was the earliest, then the UK-Met 330m model, then the LAMP. Typically the 330m and LAMP forecasts were more accurate, but varied between days.
Clouds & Visibility / GFE
The feedback that seemed to come back day after day was the "speckly" nature of the high resolution grids, which use the 3-d cloud fraction, and how that needs to be addressed going forward. From a user perspective, the detailed nature of the grids could be mistaken as more accurate, which is not necessarily the case. Discussions as whether the smoothing should be done on the national level or the local, and exactly how much smoothing as to not loose any detail that may be desired by other end-users. The possibility of the TAF formatter being expanded to take this into account was also brought up. Giving the WFO forecaster the ability to define whether to average around the TAF area, or give it more of a detailed point forecast. It was also noted that more detail is actually desired early in the TAF forecast period, and then less detail is fine for longer term TAF times.
The big question of the week was whether the extended TCF would be feasible and if so, is there added value to having a long-range forecast. From a forecaster perspective, it was noted that the additional workload of adding the extended TCF to the normal cycles was challenging. To this end, AWDE and customers suggested that it may not be necessary to generate a new long-range every cycle as traffic flow managers and planners aren't going to be looking beyond 24-hours particularly often. Instead, having an extended TCF issued along with the morning cycle and evening cycle, providing a 24 and 36-hour look ahead for the follow morning and evening as a first glance to the following day's issues. Or issuing an extended TCF at other certain specified times of day when a long-range look would be prudent. This would not only reduce the workload but also provide value to customers.