367ef086fa40de243f2cd368a2691681.ppt
- Количество слайдов: 17
COASTAL DATA INFORMATION PROGRAM 1975 - present Quality Assurance of Real-Time Data February 2004
Real-Time Web Products Southern California Summary Map Nowcast Model Products
1995: Wave Data Standards Analysis – US Army Corps of Engineers “… provide specifications on the analysis techniques that must be applied to wave data …. to ensure quality and uniformity of the final product. … The standard has evolved through a consensual approach involving the principal organizations which are responsible for the vast majority of wave measurements for this country. “ 2003: Handbook of Automated Data Quality Control Checks and Procedure of the National Data Buoy Center
QUALITY ASSURANCE Pressure Sensor Calibration COLD BOX • Constant Temperature • Standard Paros • Battery of statistical test of dynamic pressure responses over several weeks to establish noise floor. DEEP TANK Soak complete package (sensors & cable) – Establish noise floor
Generated from Aceh, Indonesia earthquake on Dec 26, 2005 @ 00: 59 UTC.
Quality Control Specifications Pressure Sensors: Time Domain • • • Max Wave Height Test Flat episodes Test Spike Test Max value Test: successive trough to crest Min Value Test: successive trough to crest Mean Shift Test Equal Peaks Test Acceleration Test Mean crossing Test Period Distribution Test
Quality Control Specifications Wave Buoys: Frequency Domain • Check Factors Test – ratio of vertical to horizontal displacement • Inclination Test • Period exceedence Test Quality of Telemetery: • Discontinuities in transmission • Data Corruption
Central Computer Automated Real-Time Processing Time Domain Time Series Quality Control Frequency Domain PRODUCTS WEB UPDATE Real-Time/Historical NWS/NDBC Transmission Pub/Non-Public STORAGE raw spectral parameters Station Specific Requests Error/Diagnostic Statistics Post Processing Inter-comparison of last 24 hours
Waimea Buoy, HA – Frequency Domain QC 106: First indicated bad by TP errors > 30 s -106 01 07/11/2003 20: 19: 30 hrs UTC code=551: (bdd_editor. f) Tp out of bounds: Flagged as NON-PUB 7/11/03 33. 33 BAD – Energy in Low Frequency for all records
Scripps Canyon North. West Buoy Frequency Domain QC Check factor values 114 01 08/16/2003 00: 39: 30 hrs UTC code=500: (bdd_editor. f) Abnormally high check factors (relationship between vertical and horizontal cross-spectra) Calibration Test – Bad displacement Direction & Period erratic. Retrieved buoy and performed calibration tests.
Wave Data Torrey Pines Inner – 20 m deep Torrey Pines Outer – 549 m deep Buoys approximately 6 miles apart
Dana Point Buoy, CA First indicated bad by inclination errors (> 5°) • 096 01 09/13/2003 18: 59: 30 hrs UTC code=702: (buoy_stats. f) Inclination off, arch: 58. 52 file: 52. 73 GOOD DISPLACEMENT PLOT BAD – Exagerrated Horizontal Amplitude
Modeling Waves Across the Basin Headland shadowing Refractive effects Directional wave buoy to initiate model Island shadowing Diffraction effects Amplification over shoals
Pacific Gas & Electric Company Diablo Canyon Nuclear Power Plant 2004
CDIP In-House Diagnostics • Paged: – off site (GPS & ARGOS) – data is > 1 hr old, – computer system is down, – computer partitions are > 90% full. • Programmers emailed : – Offsite – Data > 1 hr old, power outages, data transfer failed. System messages
CDIP Diagnostics 2004 - Data Collection Coverage Percentages ----------------La Jolla – 99. 89% Torrey Pines O – 95. 29% San Pedro – 95. 24 Santa Monica – 99. 64% Pt Dume – 95. 88% Goleta – 97. 06% Harvest – 94. 70% Pt. Reyes – 91. 88% Grays Harbor – 99. 31% Guam – 99. 72%
http: //cdip. ucsd. edu