This seminar was targeted toward researchers and scientists working in the south San Francisco Bay to better understand NOAA CO-OPS vertical datums information. Marti Ikehara, the CA State Geodetic Advisor for NOAA/NOS/NGS presented the seminar. The seminar was 2 hours and interactive with questions and answers. The purpose of the seminar was for researchers and scientists to have a clear understanding of the vertical datums and water level data on the NOAA CO-OPS website, and to understand how to read, and how not to read, the CO-OPS materials available on their website.
The South Bay Salt Pond Restoration project has installed 2 tidal gauges in the south bay; a short-term gauge (3 months) at Dumbarton Bridge (NOS Station 9414509), and a long-term gauge (indefinite time, at least a year) at the confluence of Coyote Creek and Alviso Slough (NOS Station 9414575). By the end of July we hope to have a link to these data posted on this website.
There were a few key ideas that came out of this seminar:
- Determining vertical datums can be very tricky
- Use of RTK technology to determine vertical datums is not always accurate, though it can be very precise. It is best to increase accuracy by taking redundant readings at least 3 hours apart in order to get readings from different satellites to average. One should always tie readings from GPS units to known horizontal and vertical controls that are nearby.
- Tide gauge data was not necessarily tied to geodetic benchmarks, so one must carefully review the metadata associated with any datum dataset to see if the geodetic vertical datum was computed.
- The NOAA program VDATUM, which models conversion of multiple vertical datums, may not be accurate enough for construction purposes, and should not be relied upon. In particular, VDATUM is poor in the South Bay for conversion of NADV88 to tidal datums and can result in significant (0.5 FT) errors. The U.S. Geological Survey Open File Report (see link below) has conversion factors specific to the South Bay based upon surveys conducted in 2005 and reviewed by NOAA.
- The new tide gauge data from the Dumbarton Bridge and Coyote/Alviso Stations have had bench marks leveled during installation, so the new data, when available, will provide the most accurate conversion of station or tidal datums to NADV88.
Meeting materials and supporting documents
- Seminar Agenda (PDF)
- Presentation by Marti Ikehara, part 1—Geodetic Vertical Datums (PPTX)
- Presentation by Marti Ikehara, part 2—Tidal Datums (PPTX)
- 2005 Hydrographic Survey of South San Francisco Bay (Foxgrover et al., 2007) (PDF)
- Tide Gauge Data and Plant Communities (Dave Finlaysen) (coming soon)
- Bob Battalio paper (coming soon)
- NOAA Index of Tide Stations, 1990 (PDF)
This is a couple of pages from the out-of-print and no-longer-updated but very-useful and more-useful-than-the-internet indexes which list all tide stations and key data. This information is very useful when determining the need to do new data collection and establishing judgment about confidence in tidal datum information. This type of information could be useful to assessing the accuracy of Vdatum as well.
- Estimated Historic Sea Level Change Components (PDF)
This is a table of relative sea level rise Bob Battalio of ESA-PWA, worked on for the 1990 BCDC SLR study. They used tidal benchmark re-leveling to estimate vertical land motion. The idea was to use the Presidio-based “regional” sea level rise signal, add local vertical land motion at nearby tide gauges, and interpolate between tide stations. Then add local subsidence if any due to fill placement etc. The re-leveling data were sparse then, and these are the type of data we need still.
- Sea Level Rise—Annotated Chart (PDF)
This is a scan of a graph from Bob Battalio (ESA-PWA) and the thought process that went into the sea level rise (SLR) rate in the 1990 BCDC SLR report. Basically, they were all trying to figure out how to establish a short term trend (to be updated) with a very noisy signal, including the 1982-83 El Nino. They ended up using a triangular weighted average.