Click for next page ( 51


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 50
Implementation Process MANAGEMENT AND SYSTEM ENGINEERING APPROACH The Committee has reviewed the NWS plans, management strategies, and system engineering approach for implementing its modernization, includ- ing the three systems essential for improved weather services and planned restructuring in the field: Next Generation Weather Radar (NEXRAD), Automated Surface Observing System (ASOS), and Advanced Weather Interactive Processing System (AWIPS). Management The NWS has done a commendable job in planning its modernization. A new matrix organization is in place and top management staffing is com- plete. Under the NOAA Assistant Administrator for Weather Services (the head of NWS), deputies have been appointed for operations and for modern- ization. The Deputy Assistant Administrator for Modernization oversees the NEXRAD Joint System Program Office, Office of Systems Development (which includes the ASOS and AWIPS projects), Office of Systems Operation, Office of Hydrology, and the Transition Program Office. All of the other NWS headquarters and field offices are also involved in the modernization effort to varying degrees. Using a matrix approach to management, the Deputy Assistant Administrator for Modernization also oversees and coordi- nates the modernization roles and activities of these other NWS offices as well. The Transition Program Office supports the coordination function. Support functions, such as contracting, personnel management, external relations, and facilities construction, are provided by NOAA headquarters and the Department of Commerce. The NWS has developed a number of innovative procedures intended to facilitate effective implementation of the modernization concept. Techniques 50

OCR for page 50
such as risk reduction through prototyping and strong user involvement help ensure that the products of modernization will be useful and will accurately embody the original requirements. A new system was created recently to ensure adequate internal communications, reporting, and coordination; all offices of the NWS are now aware of the plans and status of the moderniza- tion. However, NOAA and the Department of Commerce appear to have a shortage of staff to provide administrative support, such as procurement and personnel, and to handle the external contacts with Congress, user groups, and the public that are essential to implementation of the modernization and associated restructuring. Moreover, the Committee is concerned that the project management, engineering, and support staff may not be as strong as required for an effort of this magnitude. System Engineering and Integration It appears to the Committee that the NWS lacks an overall policy for configuration control of large systems and for the development and mainte- nance of complex software. Even though government policy wisely expects contractors to provide their own well-understood and tested standards and methods, it is in the government's interest to monitor carefully and to manage development and maintenance contracts. An overall NWS guiding policy is needed to set forth minimum requirements to be met by contractors in the process of developing and maintaining software and in reporting progress through specific process-sensitive metrics. Such a policy would mitigate against problems in multicontractor development and maintenance, and would protect the government against undue cost and hardship should a contractor be unable to complete a contractual commitment or should a subsequent change in contractors occur. System engineering in the NWS environment is vital because of the phased development and because the NWS systems must remain operational during upgrading and modernization. It appears that some elements of the systems now being procured may be abandoned during subsequent phases of modernization. For example, the functions of the Principal User Processor, a part of the Next Generation Weather Radar system, eventually will be performed by the Advanced Weather Interactive Processing System. Although this change may be appropriate and necessary, it might have been avoided if an overall system design or configuration control process had been in place several years ago. Communications and interfacing standards or planned evolution toward such standards is currently lacking. In a related issue, the NWS may be reticent to apply resources to the development of formal standards and methods because of the lack of adequate funding. Whereas this may reduce near-term costs, it probably will increase 51

OCR for page 50
life-cycle costs by making long-term maintenance and enhancement more diffi- cult. Despite these near-term funding pressures, preserving the viability of the NWS systems suggests that a stronger commitment to formal methods is in the national interest. Recommendation: The National Weather Service should establish overall policies and procedures for the development of major systems, including consideration of the interaction between systems, and establish software development and maintenance standards. Hardware Status The NWS plans for its contractors to install as well as construct hard- ware. Both Next Generation Weather Radar (NEXRAD) site preparation and Weather Forecast Office (WFO) building construction are being handled by a special Department of Commerce activity and appear to be under control. The Advanced Weather Interactive Processing System (AWIPS) and Automated Surface Observing System (ASOS) contractors are also expected to handle site preparation for their systems. Overall, the Committee is impressed with the progress that has been made in developing hardware and preparing for field installation. Delays in procurement and funding for the AWIPS are the most serious concerns involving hardware. The NEXRAD, a joint program with the USAF and the FAA, is well along, with initial production units already being installed, although some trou- blesome delays have occurred in software and hardware delivery. Cooperation among agencies seems excellent. There is no question that the system will constitute a major step forward in severe weather warning. The ASOS is now in the production phase. Because at least 1000 ASOS units are to be built, installing the system will be a major effort involving the NWS, the FAA, and airport authorities. Indeed, at full production, one or two ASOS units will be installed every day for four years. The AWIPS and its associated communications, essential for the integra- tion and operation of the modernization systems, are not as far along in devel- opment because of external delays in the approval of contracting steps and continuing funding constraints. This situation poses a major problem in the Modernization and Associated Restructuring Demonstration (MARD) and certification process that must precede restructuring of the NWS. Present plans call for a two-step implementation of AWIPS, an initial configuration followed by a software upgrade after certification. Even with this approach, there is very little time to commission the initial operational configuration of AWIPS before performance confirmation, MARD, and certification. It is 52

OCR for page 50
essential that AWIPS move forward expeditiously in fiscal years 1991 and 1992 if the NWS schedule for restructuring is to be met.1 The systems discussed above must be integrated, which in itself is a major project. The NWS management is aware of this and has several support service contractors assisting in both planning and in the integration of NEXRAD, AWIPS, and ASOS. System Security In the broadest sense, system security embraces all elements that can influence overall system resiliency. These include such diverse factors as satellite continuity, availability of backup power, and access security of com- munications, computer, and software systems. The Committee's study to date has not reviewed the security issues involved in modernization. It is already apparent, however, that the NWS has concentrated on physical security and has not paid sufficient attention to the security of electronic access. Meeting the needs of public and private users of NWS data requires "connectivity." This must be balanced by the provision of adequate security to ensure the continuity, integrity, and accuracy of the data and information being distrib- uted. Recommendation: The National Weather Service should satisfy itself that the security of its data systems will be adequate to preclude a breakdown of critical services in the event of improper intervention, either intentional or inadvertent, in its data and communications systems. The Committee plans to examine in greater detail those aspects of mod- ernization related to system security and resiliency. HUMAN AND FINANCIAL RESOURCES Temporary Management and Project Staff Over the past several years the NWS has assumed an increasingly large workload associated with modernization and restructuring. Additional man- agement workload has been undertaken both at NWS headquarters and in See also Chapter 3, page 33. 53

OCR for page 50
NWS regions with essentially no increase in total staff. Several people have been shifted entirely to planning and implementing the modernization, whereas others have modernization duties in addition to their former work- loads. The NOAA and the Department of Commerce also provide adminis- trative support and assist with constituent and congressional affairs. Management of some operations has been curtailed or eliminated to pre- pare for the modernization and restructuring. The following are two exam- ples: • The branch at NWS headquarters concerned with the management of warning and coordination was eliminated and the work distributed to other offices; the individuals thus made available are now working on the modern- ization. However, the ability of NWS headquarters to manage properly an important ongoing warning operation and to formulate plans for carrying out this function under modernization has been impaired. For example, there now is no leadership in planning the way in which the warning and coor- dination functions will be conducted by the 115 Weather Forecast Offices whose responsibilities will cross state boundaries versus today's 50 offices that operate within state boundaries. • Management workload at NWS headquarters and in the field is increas- ing as maintenance problems increase with the aging of current equipment and facilities. At the same time, the installation and maintenance of new systems must be planned and managed. Other areas are understaffed, particularly at NWS headquarters, as indi- cated by the following examples: • The Committee, as of January 1991, had not received from the NWS a draft of its detailed plans for the certification process, even though NWS management had placed a high priority on early review of the plan by the Committee. This is indicative of the work overload in the Transition Program Office. • As mentioned in the previous section, the size of the project manage- ment, engineering, and support staff appears to be insufficient. This is of particular concern with regard to the length of time required to initiate procurements, the provision of appropriate oversight of contractors, and the ability to respond properly to the inevitable difficulties that arise during devel- opment and implementation of complex hardware and software systems. • There is rapid growth at NWS headquarters and in the field in the need to communicate with the external world—government officials at all levels, user groups, and the general public. This external community is very con- cerned about the restructuring associated with modernization; lack of ade- quate communication could jeopardize the plan to restructure the NWS. The 54

OCR for page 50
constituent affairs staff assisting the NWS headquarters should be increased1 and a Constituent Affairs Officer added to each of the four regional offices in the contiguous 48 states. • Development by NWS headquarters of plans for transition of service activities (e.g., agriculture, aviation, fire weather, and marine programs) to the era of the modernized NWS is behind schedule. Recommendation: A Constituent Affairs Officer should be assigned to each of the four National Weather Service Regional Offices in the contiguous 48 states until the completion of modernization and restructuring. Recommendation: The management and project staffs at National Weather Service headquarters, National Oceanic and Atmospheric Administration, and Department of Commerce administrative support should be increased temporarily during the implementation of mod- ernization by at least 20 to 40 well-qualified people. At the regional level, staffing is marginal to cover ongoing operations and the new work associated with modernization. The latter, for example, includes the logistics of acquiring property for new offices, placement and surveys for the installation of new equipment, analyzing the meteorological aspects of the workload, and planning staffing and training. The intensity of this activity is increasing rapidly to the point that serious problems may arise unless the staff is increased. Recommendation: The staff at each National Weather Service regional office should be increased temporarily during the imple- mentation of modernization by one to four people as required. Operational Staff The Committee's review has identified a number of areas in which staff- ing could be insufficient. Some of these are discussed elsewhere in this report: • Workloads at the River Forecast Centers (Chapter 4, page 42). See also discussion of Constituent Affairs Officer in Chapter 5, page 48. 55

OCR for page 50
• Weather Forecast Office (WFO) staffing to handle community pre- paredness activities (Chapter 5, page 48). In addition, the Committee is concerned about the plan to have only one meteorologist on duty during the night shift at each WFO. The weather is no less life-threatening and damaging at night than during the day and evening. It is doubtful that a single meteorologist can adequately monitor and forecast changes and issue timely warnings under bad weather conditions. The Committee understands that the concept of one meteorologist covering the night shift is based on the premises that • a single meteorologist can properly review, modify, and keep up to date a four-dimensional database in the Advanced Weather Interactive Processing System (AWIPS) which covers at least 48 hours and from which all opera- tional forecasts will be produced automatically and disseminated with no manual activity involved; and • the meteorologist's main role during bad weather situations will be to concentrate on only the first 12 hours of the forecast period using the AWIPS database, and relying on computer-generated forecasts beyond 12 hours. A major problem with this concept is that it is based on the untested and questionable idea of producing all operational weather forecasts by computer, a deferred capability of the AWIPS system. The concept of machine-pro- duced operational public weather forecasts based on a meteorologist modify- ing or accepting computer-generated Model Output Statistics is to be tested at Norman, Oklahoma in the near future. However, this will involve the shift meteorologist merely accepting or changing numerical values in the Model Output Statistics guidance. The NWS needs to determine whether restrictions inherent in the new concept will prevent shift meteorologists from incorporating their expertise and evaluation of the synoptic situation and its expected evolution into the final forecast. For example, the ability to add manually significant detail to automated analyses, by using additional data not employed in the automated analyses, appears to be eliminated. Antecedent weather conditions, current radar and satellite data, and locally acquired observations can be important in producing correct analyses. The Committee questions whether all of this can be done by one meteorologist who merely changes the geometry of some lines in guidance graphics or the numbers in a Model Output Statistics matrix. It is certainly advisable to have a minimum number of people working at night, especially when the weather is expected to be benign. The problem of minimum night staffing arises when bad weather is expected or develops unexpectedly. Perhaps flexibility can be incorporated into the one-person staffing when significant weather and associated forecast problems are antici- 56

OCR for page 50
pated during a night shift. For example, the Science and Operations Officer or the Warning Coordination meteorologist, who normally works in the daytime at Weather Forecast Offices, could be responsible for assisting the night shift meteorologist. Such persons may be needed to provide special interpretation and advice to emergency management centers regarding specific local warning and community response problems. Recommendation: As a part of the Modernization and Associated Restructuring Demonstration, the National Weather Service must thoroughly test the concept of forecasts being automatically produced at night by using a final four-dimensional database. Recommendation: The proposal to produce operational forecasts by computer that are equal to or better than current manually produced forecasts and warnings should be demonstrated for a variety of weather conditions and locations. The new procedures should be operational and their efficacy established before the meteorological staff at a Weather Forecast Office is reduced to the proposed one person on the night shift. Recommendation: An alternate operational plan for staffing the night shift should be formulated for use until the proposed concept has been fully developed and proven. Financial Resources The Committee recognizes that many of the suggestions made in this report have a potential impact on the budget for the NWS modernization and associated restructuring. For example, the Committee is aware that the limit- ing factor in the employment of additional personnel is money, not authorized positions. However, in most cases, the additional personnel required tempo- rarily to assist in modernization activities could save money in the long term by reducing future development, implementation, and maintenance problems and their attendant costs. Although the solution proposed for the problem of limited meteorological staff on the night shift at the Weather Forecast Offices may reduce the overall savings visualized from restructuring the NWS until the effectiveness of automation can be demonstrated satisfactorily, the Committee believes that this would be compensated by savings to the public and govern- ments from reduced loss of life and destruction of property. The Committee recommendation with the largest cost impact is undoubt- edly that concerning the phasing of funds for NOAA satellites to ensure their continuity (see discussion of environmental satellites in Chapter 2). This requires early funding rather than an increase in total cost over the life of the 57

OCR for page 50
program. Satellite data have become such an important part of the nation's weather forecasting and warning services that the continuity of observations must be ensured to avoid a major degradation or interruption of these serv- ices. Rectifying the current low incremental funding of the Advanced Weather Interactive Processing System program (discussed on pages 33, and 52) would also require an increase in near-term budgets, but would probably reduce the overall cost of implementation. The network of observation stations required in natural and undeveloped areas to preserve the viability and integrity of the climate record (see Chapter 2), will also lead to additional costs; these might properly be borne by the U.S. Global Change Research Program. CERTIFICATION REQUIREMENTS AND APPROACH Congressional concern about the impact of the changes in existing weather stations, as proposed in the restructuring of the NWS associated with modernization, resulted in the certification requirements in Title IV of Public Law 100-685 (U.S. Congress, 1988). The relevant parts are • Section 408 which requires the Secretary of Commerce "not to close, con- solidate, automate, or relocate any...office" unless the Secretary certifies to the Congress "that such action will not result in any degradation of weather services provided to the affected area." It further states, "Such certification shall include— "(1) a detailed comparison of the services provided to the affected area and the services to be provided after such action; "(2) any recent or expected modernization of National Weather Service operations which will enhance services in the affected area; and "(3) evidence, based upon operational demonstration of modernized National Weather Service operations, which supports the conclusion that no degradation in services will result from such action." • Section 407 (b) states that "...the National Implementation Plan shall include... (2) special measures to test, evaluate, and demonstrate key elements of the Modernized National Weather Service operations prior to national implementation, including a multistation operational demonstration which tests the performance of all components of the modernization in an integrated manner for a sustained period;..". The latter is the Modernization and Associated Restructuring Demonstration (MARD) that is being planned for a period of one year in the Midwest. The NWS considers MARD to be the cornerstone upon which the certification process will be based. 58

OCR for page 50
Clearly, the need to certify expected performance before each step hi the modernization and associated restructuring is implemented places a major burden and responsibility on the NWS, the NOAA, and the Secretary of Commerce. Because of the large temporal (daily, seasonal, and annual) and geographical variations in the weather, it will be difficult to establish certifi- cation procedures to demonstrate conclusively "that no degradation in services will result from [any] action." Clearly, the certification procedure must assess objectively the quality and timeliness of the forecasts and warnings. Although the Committee has not received the proposed detailed certifi- cation plans for review, its following initial views are offered for consideration. First, specific comparisons of the quantity and quality of weather information, forecasts, and warnings, and their prompt dissemination must be obtained, both during the MARD and during the process of certifying the capabilities of any Weather Forecast Office (WFO) to serve its area of responsibility. The comparison process must be designed to be an end-to-end verification of the capabilities of the NWS and the WFO to acquire information; convert it into useful analyses, forecasts, special weather statements, and warnings; and transmit these products in a timely manner to users in the public and private sectors. The Committee believes that carefully constructed and unbiased comparisons will demonstrate a noteworthy improvement in the quality and accuracy of service. Second, to increase the credibility of the certification process in the eyes of user groups and Congress, it may be appropriate, at some stage, to involve an independent evaluation of the statistical and analyti- cal measures developed during the initial operations of the WFOs as applied to each specific certification. The Committee intends to give careful and thorough attention to the proposed NWS certification plans as soon as they are received. 59