Published: Jan 1975
| ||Format||Pages||Price|| |
|PDF ()||31||$25||  ADD TO CART|
|Complete Source PDF (9.9M)||31||$139||  ADD TO CART|
A large data system is one which is hard to turn off. This inertia is directly proportional to the amount of money invested in hardware and facilities, to the number of careers involved, and to the length of record.
Large data systems are concerned with the collection of time series. What eventually will become a data management problem in a large data system using manual techniques quickly becomes a crisis with automatic equipment. It is accordingly essential in the design of continuous monitoring networks that the utilization of the data be adequately considered and optimized.
The arguments for optimization of time-series collection, analysis, and utilization apply equally to space fields. In either case, the problem is to match the resolution of the data collection system to that of the response system. For example, the digitizing or averaging period used for pollution detection should be consistent with the time constant for pollution control or abatement measures.
The institutional value of monitoring programs may be equally important, but is more difficult to predict. In any event, the analysis and interpretation of the data must correspond to the real world use to which it will be put.
water quality, monitors, sampling, environmental tests, data processing, automatic control equipment, pollution, environmental data utilization
Director, Marine Exosystems Analysis Program, Environmental Research Laboratories, National Oceanic and Atmospheric Agency, Boulder, Colo.