Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 67
OCR for page 67
OCR for page 67
Appendix A Site Visits Members of the committee made site visits to several field offices in which the Advanced Hydrologic Prediction Service (AHPS) is used in one form or another. These were not all National Weather Service (NWS) field offices, but rather represent a range of agencies and organizations that use or generate hydrologic models and forecasts. All were quite familiar with AHPS and each has used either AHPS data, models, and/or products. Some serve other users who, in turn, use AHPS data or products, either directly or through the organization visited. Each group of professionals interviewed voiced strong support for AHPS and encouraged its continued development. Each pointed to different aspects that they believe are essential elements of AHPS, reflecting their specific needs and interests, and all had suggestions for improving AHPS. Despite the differences among the offices in responsibilities and needs, their responses to the interviews can be grouped into three categories: the science, interaction and connection, and identifying needs. Though each is discussed separately below, they are not mutually exclusive. THE SCIENCE Several people interviewed expressed concerns about the data and models that are the foundation of AHPS. For instance, many hydrologists who use AHPS products remain unconvinced about the adequacy of precipitation and temperature data as they affect quantitative precipitation forecasts (QPFs) and quantitative precipitation estimations (QPEs). Advances are needed to ascertain the correct usages of QPE, QPF, and probabilistic quantitative precipitation forecasts (PQPFs). Also needed is a long-term strategy to address systematic deficiencies such as these. Other examples were given, but most modeling and scientific concerns centered around the need for shorter-term forecasts. Shorter-term forecasts require more frequent updating. They are based on sufficiently sensitive models, and model input needs to be verified and validated for accuracy. These elements of the shorter-term forecast models were seen as critical to the success of AHPS. It appears to some in the field that short-term pressures to produce results are a stronger focus than the development of an end-to-end hydrologic prediction system. Others interviewed suggested that too much of AHPS seems to offer discrete solutions to localized problems, rather than an approach that addresses system-wide issues such as data availability and model accuracy. Therefore, those interviewed were clear that high priority must be given to advancing the science associated with the forecasting that is at the root of AHPS. Another theme that surfaced was the need to shorten the time between the inception of ideas and their implementation. Distributed modeling has been of concern for almost two decades, yet it has still not been used in operational forecasts. Similarly, QPE from radar have been discussed as a priority for sometime, but sufficient advances have not been made. These are examples of elements that are critical to the ultimate success of AHPS. That we have not gotten to where we need to be is a function of a number of factors, many of which are not easily controlled by the NWS. Because much of the data collection and some of the modeling are done through field offices and other agencies, there is a need for close interaction between NWS field offices and cutting-edge research. 69
OCR for page 67
70 Toward a New Advanced Hydrologic Prediction Service (AHPS) INTERACTION AND CONNECTION The need for communication among all players, those in various field offices and related organizations and those at headquarters, is clearly evident from the data sharing and model building that take place in various locations. A great deal of modeling that has direct application to AHPS is done in the field, and much of the necessary data are collected there. Conversely, models that are developed at headquarters can be most easily tested and implemented by regional and local offices. Clearly, improved communication can benefit all: ideas can get into operational settings more quickly and models developed in the field can more easily be incorporated into the AHPS process at headquarters. Absence of a clear, integrated chain of command and communication not only leads to difficulty in developing and implementing an end-to-end system, but it also confuses users who may need immediate answers and face uncertainty as to where they can go for those answers. For instance, the River Forecast Centers (RFCs) produce forecasts but do not officially issue them. Users, then, cannot contact the RFCs directly regarding feedback or problems associated with the forecasts. Connections and cooperation are also necessary among the numerous agencies that develop and use hydrologic data and models. Frustration at the time it has taken to see some ideas become operational may be lessened by agencies working together rather than by each doing its own modeling. This is probably a lower priority than the "in-house" cooperation that is discussed above, but it may be a way to leverage resources to better advantage and to avoid duplication of efforts. IDENTIFYING NEEDS One fact that became abundantly clear during the site visits is that hydrologic information is used by a wide range of users. The needs of these users vary significantly, as do their levels of sophistication. An important concern of those interviewed relates to the move to probabilistic forecasts in which a suite of values is provided, each with a given probability of exceedance. While supportive of this approach, all recognized the implications of this, which will end up shifting managerial decisions to users. Some will find it very useful, some will not. A current problem with AHPS that was cited by several of those interviewed is the lack of consistent data and products available at all locations. It was recognized that such consistency will take time as necessary data are collected and models developed and calibrated at new locations, but this is not always as easily recognized by different users and user groups. Some locations have flood inundation maps available; others don't. Some users look for long term forecasts; others require short term forecasts with frequent updates. For some users, the 6 hour time step of hydrologic models works; for others, a 3 hour time step would be more appropriate. Working in close collaboration and interacting regularly along the chain of command would help identify and prioritize the varying needs.