Use of NWP for Nowcasting
- An international Workshop sponsored by WMO WWRP

Background and motivation:
In recent years, numerical weather prediction (NWP) models have demonstrated to possess improved capability in providing guidance for high-impact weather prediction at km-scale for forecast ranges up to 24-36 hours. In the nowcasting (or very-short-term of 0-6 hours) range, however, the usefulness of NWP is limited due to a number of scientific issues (e.g. spin-up, analysis of highly complex, nonlinear physical processes, initialization techniques, etc.) as well as limited awareness of the needs of nowcasting in the NWP community. As the demand to improve very-short-term forecasts of high-impact-weather have been increasing, research on the use of high-resolution NWP for nowcasting has also expanded rapidly.

Traditionally, nowcasting systems use detection and simple extrapolation techniques to provide forecasts up to a couple of hours ahead based on the underlying conceptual models. The forecast skills of these systems usually drop rapidly after the first hour. New, advanced approaches to nowcasting depend on high-resolution NWP products to improve nowcasting products beyond the first hour. Blending of traditional extrapolation-based techniques with high-resolution NWP (the so-called heuristic approach) is gaining popularity in the nowcasting community. Further development of this approach will realize a truly integrated solution to nowcasting. The increased need of NWP products in nowcasting applications poses great challenges to the NWP community because the nowcasting application of high-resolution NWP requires special considerations that are different from the longer range NWP. Among them are, e.g., convective scale data assimilation, NWP with model resolution less than a few km to adequately resolve the physical processes impacting nowcasting, modeling of these physical processes within the model and as part of the data assimilation, dealing with the model spin-up problem, rapid error growth at the convective scale, etc.

Purpose and approach:
The Working Group on Mesoscale Weather Forecasting Research (WGMWFR) and the Working Group on Nowcasting Research (WGNR) within WMO/WWRP recognize the importance of NWP to nowcasting and consider it a common challenge to both groups. To address this challenge, a workshop on the Use of NWP for Nowcasting was proposed jointly by the two Working Groups and approved by WMO. The participants of the Workshop are by invitation only, including experts in both areas. The aim is to review the current status of the use of high-resolution NWP in nowcasting and identify mechanisms and directions that could be employed for advancement through international collaboration. The key scientific issues and limitations in the area will be discussed. Although the major theme of the workshop is to discuss the application and development of high-resolution NWP for nowcasting, the topics will also include reviews of common nowcasting techniques and physical processes that are most relevant to the very-short-term forecasting problems. Based on the outcome of the workshop, the WGMWFR and WGNR will formulate a set of recommendations to the JSC of WMO/WWRP on how to progress this activity further. These could include e.g. the development of a project plan for an RDP (Research Demonstration Project) or FDP (Forecast Demonstration Project) in this field, and search for funding and partnerships for supporting the project.

Specifically, the following topics will be covered in the proposed workshop:
- Use and potential of existing techniques and observation types for nowcasting
- Major physical processes that influence the forecast accuracy for the very-short-term
- Requirements for mesoscale NWP systems, in terms of products, quality, timeliness, etc., from the perspective of nowcasting applications
- Conversely, requirements from mesoscale models on observations, typical problems and limitations experienced on this account
- Experiences to date in, and plans for, application of mesoscale NWP for nowcasting in rapid update cycling systems, the challenges, and potential of various data assimilation techniques
- Different operational setups for nowcasting and longer range, including how to handle issues related to specific location and phenomenon
- Data types that have the most impact for nowcasting purposes
- Identification of topics where stronger interactions between the nowcasting and NWP communities should be fostered
- Ways to assess and evaluate the potential of mesoscale NWP models for nowcasting purposes
- Discussion on possible follow-up actions and experiments

The first two days of the workshop will cover presentations from the participants in four sessions on various NWP and nowcasting subjects. The morning of the last day will be spent on a short session on verification followed by small group discussions. A plenary session is arranged in the afternoon to develop plans for international collaboration. It is envisioned that the presentations and discussions will lead to a short-term action plan and a long-term road map to improve the application of high-resolution NWP for nowcasting and to enhance international collaboration in this important area.

Travel and lodging
The workshop will be held at the NCAR Center Green Campus, 3080 Center Green Drive, in Boulder, Colorado. WMO provides funds to cover the travel and lodging to limited number of participants. The hotel and transportation information will be sent to the confirmed participants in separate emails.




Call For Papers


Breakout Group Report