Regional Numerical Weather Modeling

The apparent paradox of independent researchers nonetheless dependent on complementary research in adjacent states is, in fact, a compelling rationale for collaborative research. Seen from this perspective, the seemingly random list of potential topics resolved themselves into two topical areas:

1. Coupling regional weather models with an array of submodels such as ground cover and soil moisture that are specific to the conditions of the Great Plains. Useful. submodels cannot be developed in isolation if they are to have maximum utility. For instance, questions of scalability, data format, and capability for real-time use in regional models need to be addressed early in model and submodel development.
2. Utilizing the regional models to develop other models as products to address the real-time needs of state and local policy and decision makers. For instance, models in transportation that integrate weather models to better predict storm impacts and to help route hazardous materials shipments, models that utilize weather information to predict the course of major natural disasters such as fire or flood, and models for use in precision agriculture. All such models must be understood by the user community as aids to improve insight and intuition, rather than as models that "make decisions."

A constant theme of group discussions emphasized the importance of the interfaces between regional model developers and submodel developers, between weather modelers and application modelers, and between application modelers and the intended user community. While it belabored the obvious, all discussants emphasized that it was insufficient to just "toss data or results over the wall" to the users of those results. The necessity to clarify mutual needs, expectations, and assumptions is always easy to say and tough to practice, especially at the interfaces between groups which historically have had few interactions.

The group conclude that the most valuable next steps are process oriented. Rather than moving immediately to a selection of one or more collaborative research topics early attention needed to be paid to ensuring that the full community of academic research relevant to the development and applications of regional weather modeling was identified and invited to join the effort. With the mapping of faculty, staff, and equipment resources in hand there would be a natural process of developing joint research initiatives.. A self-selected and ad hoc group was identified to work with EPSCoR state directors and to begin the following steps:

1. Distribute a 1-2 page description to the widest possible academic, governmental, and private sector research community that would outline the challenges and opportunities in regional weather modeling and its applications to societal needs. The document would invite all interested to indicate their capabilities and interests. This list will form the foundation for many later initiatives. There would also be prominent mention of the meetings described below at 2) and 3).
2. Set up an informal meeting for those interested in Great Plains weather modeling at the January 1998 national weather meeting in Phoenix. The meeting would serve to broaden the community of those interested and to recruit their participation in future activities.
3. Organize one or more meetings for spring(?) 1998 for regional model and submodel developers and for application model developers and their user community. The meeting(s) would be structured to include formal presentations by , for example, someone in hydrology modeling, to inform specialists in other areas of the current state of hydrology modeling, developing trends, what the field needs from others, and what it can contribute to other fields. The meeting would also have a major focus on poster sessions, with sufficient time allocated for participants to meet the authors and fully understand the science. Additional time would be allocated for informal discussions and the identification of potential collaborators. These sessions could well lead to the identification of research resources of people and/or equipment not currently found in the Great Plains region that are critical to a successful collaboration. The University of Oklahoma offered to host this first organizing meeting.