NOAA's Next Generation Tide Gauges


Steve Gill
Chief Tidal Analysis Section
SSMC4, Sta. No. 7109
1305 East-West Highway
Silver Spring, MD 20910, USA

Introduction

The USA National Oceanic and Atmospheric Administration (NOAA) manages a National Water Level Observation Network (NWLON) of approximately 180 permanent stations, including the Great Lakes and US ocean island territories and possessions. NOAA initiated a fundamental modernization of water level measurement technology in the mid-1980's with the Next Generation Water Level Measurement System (NGWLMS). NGWLMS is as complete modernization of water level measurement sensors; data collection, transmission, and processing; and data base management, and data dissemination (Beaumariage and Scherer, 1987). The NGWLMS systems are being operated simultaneously with the older technology float/stilling well ADR gauges for approximately one year to establish data and datum continuity (Gill and Mero, 1990).

With many time series approaching/or over a century in length, a subset of the NWLON has actively supported the international GLOSS community. NOAA has also been involved in a parallel effort of data rescue of historical analog and manual tabulated data through digitization and quality control of historical data sets and putting them into the modernized data base system.

Starting in 1989, a NOAA Global Sea Level Program (GSLP) was initiated as an element of the NOAA Climate and Global Change Program. The program was designed as an integrated program of in-situ water level and ancillary measurements, analysis of satellite altimetry data from GEOSAT, ERS-1, and TOPEX/POSEIDON, and monitoring land movement using VLBI, GPS and absolute gravity measurements (Parker et al, 1992). NOAA has installed approximately 23 water level stations since 1989 using the NGWLMS system, concentrating in the southern Pacific and mid- and southern Atlantic Oceans where existing sea level measurement networks were quite sparse. Two of the stations are also located in the Indian Ocean and one at Esperanza, Antarctica. Although up to 50 stations were originally proposed, additional station installations have been stopped due to funding restrictions.

NGWLMS Gauge and Sensor Configuration

The NGWLMS systems relay data every three hours from the Data Collection Platform (DCP) to NOAA's Geostationary Operational Environmental Satellite (GOES) system. NOAA's NGWLMS Data Processing and Analysis System (DPAS) retrieves available data from the downlink on an hourly basis, decodes the data, and performs automated QC checks. The micro-processor based DCP's also have on-site data management and storage and have the capability of real-time dissemination via telephone. Battery power can be driven via AC chargers or using solar panels. Field teams interact with the systems on-site using lap-top computers or remotely via telephone. Gaps in satellite-transmitted data are recovered automatically by DPAS using scheduled telephone interrogations every two-weeks.

The primary sensor used at most locations is a self-calibrating downward looking acoustic system that transmits a short acoustic pulse through a 1.3 cm diameter sound tube to the water surface and to a calibration point. The calibration point is a known distance to the system leveling point so that station datum reference is maintained. Because the major potential source for error is due to unknown changes in propagation velocity due to vertical air temperature gradients from the calibration point to the water surface, sound tube air temperatures are monitored above and below the calibration point and are transmitted with each 6-minute interval measurement. The unwanted effects of high frequency waves and high velocity currents are mitigated using a combination of a high sampling rate and a protective well configuration that has a 3:1 well to orifice diameter ratio and a set of parallel plates below the orifice. (For a figure of the setup, see the references below).

The sensor takes 181 one-second samples in a three-minute period centered every 6-minutes. The sample mean and standard deviation are computed and outliers are determined and rejected. A new mean value is computed and reported every 6-minutes along with the standard deviation and the number of outliers.

A remote locations and other locations where the installation of an acoustic system protective well is not possible (e.g., Easter Island, Diego Ramirez), the NGWLMS primary sensor for water level measurement is a "digibub" system. The digibub uses a bubbler nitrogen driven pressure system integrated with a Paroscientific pressure transducer. Only the bubbler orifice and tubing are located under the water. The Paroscientific transducer is located in the DCP housing and is vented to the atmosphere. High rate sampling is employed, although not every one-second, and the computed means standard deviations and outliers are reported with every 6-minute measurement. At some locations where the digibub is deployed (e.g. Anchorage, AK), the unwanted effects of significant seasonal or tidal cycle changes in water density are mitigated by using a dual- orifice system (vertically separated) in which the pressure differences are used to infer a correction for density.

Each of the NGWLMS stations also has a backup water level measurement system that use a digibub configuration, however the pressure transducer is a less-accurate strain-gauge type sensor. The system uses an independent data logger with independent power and internal memory. Half-hourly interval data are routinely transmitted over satellite for redundancy with the primary sensor and filling of small gaps. If required, the higher rate 6-minute interval data can be obtained via telephone call. The data from the independent backup system have also proven to be extremely valuable, for instance, when the acoustic sensor high limits have been exceeded during storms and when the acoustic sensors have been partially frozen during extended extreme cold periods in the winter at northern locations.

The NGWLMS systems are capable of having up to 11 ancillary sensors configured measurement of hourly meteorological and oceanographic parameters. Most NOAA GSLP stations have a suite of ancillary sensors.

Data Base and Data Dissemination

Using automated and semi-automated procedures, the incoming data are processed and standard output products are generated and verified. Products include edited 6-minute time series, hourly heights, times and heights of high and low waters, monthly means of sea level and various tidal parameters, and monthly extremes. Engineering parameters are also tracked and stored as to the operational health of the DCP's, along with sensor parameters and leveling information.

NOAA is in the process of implementing on-line access to the historical time series and recent data in various stages of processing through TELNET and MOSAIC NOAA Home Page interfaces over Internet. Data availability will include observed raw 6-minute water level data and preliminary hourly data from the ancillary sensors from the most recent several days. Preliminary edited and verified 6- minute data, hourly heights, and monthly means are generally available within one month of the end of each month of measurement and will also be put on-line. Accepted data that have gone through the longer time-step review of vertical stability checks of platform and bench marks should be used to replace the preliminary data and will be put on-line within one-year of data collection. Data requests and updates on Internet access can be obtained through NOAA's Ocean and Lake Levels Division at gill@wlnet1.nos.noaa.gov.

References

Beaumariage, D.C. and W.D. Scherer, "New Technology Enhances Water Level Measurement," Sea Technology:28(5), pp. 29-32 (1987).

Parker, B.B., R.E. Cheney, and W.E. Carter, "NOAA Global Sea Level Program," Sea Technology, pp. 55-62 (June 1992).

Gill, S.K., and T.N.Mero, "Next Generation Water Level Measurement System: Implementation into the NOAA National Water Level Observation Network," Towards an Integrated System for Measuring Long Term Changes in Global Sea Level, Report of a Workshop held at Woods Hole Oceanographic Institution, pp. 133- 146 (May 1990).