Papers included in Watershed 96 proceedings reflect the opinions of the authors and do not necessarily represent official positions of the Environmental Protection Agency.
Alan M. Lumb,
U.S. Geological Survey, Reston, VA
The U.S. Geological Survey (USGS) is in the final phases of the development of a new hydrologic database system called the National Water Information System-II (NWIS-II). NWIS-II is a distributed, UNIX1-based, Ingres relational data base that will manage the hydrologic data collected by USGS offices, located in the 50 states and Puerto Rico, as well as data collected at major program offices in Regional centers. The system will be installed on up to 70 sites.
NWIS-II integrates two types of hydrologic data, discrete and time-series, into a single data base. Discrete data are those collected periodically, such as water-quality analyses, field measurements of ground-water levels and surface-water flows, information about data collection sites, and well information. Time-series data are collected and recorded by instruments either on a continuous basis or an event-driven basis, such as hourly stage or water-level measurements, water-quality monitor records, and rainfall records. These data are then processed to provide results such as stream discharge calculated from the stage data.
The NWIS-II system provides major improvements over existing systems including:
NWIS-II will replace two aging hydrologic data systems. NWIS-I is a minicomputer-based system distributed in 40 locations and provides local data processing capability to the USGS District Offices. Once the data are processed locally, they are uploaded to the Water Data Storage and Retrieval System (WATSTORE), a mainframe-based system maintained in Reston, Virginia. WATSTORE was designed in the late 1960s to provide data processing for District offices as well as a National data base. It has had only minor software upgrades and limited increases in the types of data that may be stored within the system. NWIS-I provided local data processing capability but processed and stored the same limited types of data as WATSTORE.
Both WATSTORE and NWIS-I store time-series and discrete data in separate data files only loosely integrated through yet another data file containing site information. Users wanting to retrieve and work simultaneously with data from the various data files were frustrated by the complexity required to bring the data together.
NWIS-II marks a major improvement in the amount and quality of information about a measurement or an analysis that can be stored. Commonly called metadata, this information includes the who, what, why, when, and how of data collection and analyses. Metadata increases the value of data for reuse. The original collectors of data selected techniques that met the needs of their intended use of the data. In some cases, subsequent users of that data are unable to determine what methods were used to collect or analyze the data and what accuracy was associated with the data. These factors can determine whether use of the data would be appropriate for a new study. By storing the metadata with the actual data results, future users of the data can make more accurate assessments of its appropriateness to their needs.
As an example of this increased ability to store metadata, consider a typical visit to a site on a river to collect water-quality samples. At the site, samples are routinely collected at two locations, 100 yards downstream from the gaging station at a site suitable for sampling by wading at low flow and 2000 yards upstream at a bridge for sampling at higher flow. Table 1 shows some examples of the additional data and metadata that may be stored in NWIS-II.
| Data type | WATSTORE / NWIS-I | NWIS-II |
|---|---|---|
| Sample location | Gage house location only. | Description of location where samples were collected. |
| Sample method | Occasionally stored as a fixed value code. | Procedure identified along with specific time, persons sampling, and comments to describe any deviations from standard procedures. |
| Sample preparation | No information stored. | Complete information including how composited, split, filtered, treated , and shipped. Individual samples can be traced back to sample method and location. |
| Measurement or analysis | Result value with a method code specific to each laboratory. | Result value, procedure used to make measurement or analysis, rerun values if any, date and time of the measurement or analysis. Each result can be traced back to an individual sample. Each procedure can be tracked to a bibliographic citation |
NWIS-II provides a graphical-user interface under the X Window System for data entry and retrieval and includes extensive reference lists. Users of the system specified that the system should be easy to use and intuitive for the user. Accessing the reference lists from the graphical-user interface should simplify or eliminate the need for users to memorize codes and instead allow them to pick from lists of easily understood information. The NWIS-II system uses 266 reference tables, 18 of which will be maintained locally; the remainder will be maintained nationally.
In processing time-series data, the graphical-user interface leads the user through the steps required to convert stream stages to the additional products of streamflow information. The system keeps track of the steps the user has taken in the process and leads the user to the appropriate next step.
One major advantage of using the graphical-user interface rather than a character-based interface is that time-series data are easily viewed and edited graphically on the screen. For example, periods of missing streamflow or water-level records can be estimated by superimposing the hydrograph from another site and tracing the portion that must be estimated. Hydra, as this portion of NWIS-II is called, gives the user the ability to quickly view and, if necessary, edit data on screen.
A geographic information system (GIS) is an integral part of the NWIS-II system. GIS can be used to establish or verify the location information of a data site. On the basis of available data coverages, information such as county, hydrologic unit, river reach, geologic unit, land use, and congressional district can be automatically entered into the data base. Well data from different sources often will have slightly different location information. The GIS system notifies users if there is an existing site close to one being established. This allows the user to determine if the data are for two separate sites or are actually from the same site.
GIS has increased the flexibility of selecting sites for retrieval of data. Without GIS, users are limited to spatial searches tied to political units stored in the data base, such as counties or within polygons or circles. GIS allows the user to specify far more complex retrieval specifications based on the spatial data layers available. For example, with GIS it is possible to select data for wells located within 1500 meters of a stream reach or data for surface-water sites located in forested areas.
Using GIS to display sites with data meeting specified criteria gives the hydrologist an additional interpretive tool or a quality-assurance tool. For example, in a given aquifer, wells that exceed user-supplied criteria could show a pattern of contamination from a point source.
Applications for interpretation of data are limited within the NWIS-II system. Applications focus on data entry and verification, processing time-series data such as converting continuous stage data to daily maximum, minimum, and mean discharge values, and output of data into reports or machine-readable format. This change from the older NWIS-I and WATSTORE systems reflects the growth of readily available commercial software, including spreadsheets and statistical packages. NWIS-II produces output easily imported into these applications.
Other specialized hydrologic application programs that can be used to analyze, review, display, or synthesize data (as in a model) are external to the NWIS-II system. A separate group within the USGS, the Hydrologic Analysis Support Section (HASS), writes, maintains, enhances, and distributes USGS-developed models and application programs. NWIS-II will produce output that can be easily imported into the HASS programs.
EPA Home | Office of Water | Search | Comments