NDBC Real Time Wave Data Processing and QC/QA Chung-Chu Teng National Data Buoy Center Stennis Space Center, MS 39529, USA Chung-Chu.Teng@noaa.gov
NDBC Wave Stations 88 buoys (28 directional waves) and 4 C-MAN stations
C-MAN Wave Measurements Chesapeake Light St. Augustine Diamond Shoal
Buoy Wave Measurements 6-m NOMAD buoy 3-m discus buoy 10-m discus buoy
New Stand-alone Wave System/Buoy Directional Wave Measurement System: • - Small & light • Low power consumption • Low cost • Built-in transmitter • Standalone • Configurable Directional Wave buoy: • Small & light, deployable from small vessels • Low cost • Low maintenance 1.8 m COLOS buoy
NOMAD Directional Wave Measurements Use a small directional wave buoy (tether to a 6-m NOMAD buoy or stand-alone) GOES or Iridium RF/LOS communication
C-MAN Wave Measurements GOES or Iridium RF/LOS communication Deploy a stand-alone directional wave buoy Install a fixed directional wave sensor on the structure
Nondirectional wave data • Wave energy spectrum • Wave parameters: Peak (or dominant) wave period, Tp Mean (or average) wave period, Tz or Ta Significant wave height, Hs S(f) Frequency, f
Buoy Wave Measurement Ocean waves Buoy hull motions Measurements (Sensors) Onboard processing Transmission (or storage) Shoreside processing Wave data
Nondirectional wave data from buoys From buoy motion to wave data Sh(f): spectrum of buoy heave motion Sw(f): wave spectrum (acceleration) PTF: power transfer function From acceleration to displacement spectra
Noise Correction S(f) Low-frequency noise Frequency (f)
NDBC directional wave buoy systems • Based on the “slope following” principle • Can only use axis-symmetrical buoys • Buoy pitch and roll information are required to determine directional wave data
Directional wave algorithm (1) f : wave frequency; 2 : wave direction 1: vertical motion (heave) 2: N-S slope (pitch) 3: W-E slope (roll) K: the wave number C and Q : co- and quad-spectra
Directional wave algorithm (2) θ1 and θ2: mean and principal wave directions r1 and r2 : directional energy spreading
Configurations of NDBC directional wave systems • HIPPY – a gimbaled gyro system that measure pitch and roll directly • MO – use only magnetometer outputs to estimate buoy pitch and roll • ARS – derive buoy pitch and roll from angular rate sensors
NDBC Wave Systems • DACT Wave Analyzer (WA) • DACT Directional Wave Analyzer (DWA) • VEEP Wave Analyzer (WA) • Wave Processing Module (WPM) • Directional Wave Processing Module (DWPM) • Non-Directional Wave Processing Module (NDWPM) • Directional Wave Measurement System (DWMS)
General Principles for Data QA/QC • Reasonability: Data should be in reasonable ranges and intervals. • Continuity: Data should maintain the continuity in time and space. • Consistency (or correlation): Data should have proper time correlation, spatial correlation, and correlations with other measurements.
Some wave data QC techniques • Range check • Time continuity • Internal consistency (e.g., QMEAN) • Wind vs. wave energy (wind-wave algorithm) • Swell direction check • High frequency spectral spikes, • Wave height vs. average wave period • Wave direction vs. wind direction (> 0.35 Hz) • When needed, wave data from NCEP’s wave models and other organizations will be used to check its buoy wave data.
Hard Flags (by hierarchy - highest to lowest) • T Transmission parity error (Applies to continuous winds and non-WPM wave data) • M Missing sensor data (A result of a garbled or missing message). • W A WPM wave message is short, missing a checksum, or parity errors are detected. • E Calculation limits are exceeded or are in error (waves only, flags WVHGT). • D Delete measurement from release and archive (A Data Analyst or automated QC has failed the sensor). • S Invalid statistical parameter (in waves, QMEAN is not between QMIN and QMAX, flags WVHGT). • V Failed time continuity. • L Failed range limits. • H Hierarchy reversal has occurred (BARO, WSPD, WDIR only). • R A related measurement has failed a hard QC check.
Soft Flags (in alphabetical order) • a Measurement is above monthly, regional limit. • b Measurement is below monthly, regional limit. • c Measurement has been adjusted, or corrected (applies to DEWPT and WVHGT, DOMPD and AVGPD). • d Failed standard deviation test (continuous winds only). • f Measurement failed hourly time continuity. • g Failed gust-to-mean wind speed ratio (applies to standard and continuous winds). • i Continuous and hourly wind speeds don’t agree. • j One, and only one, transmission error detected in the continuous wind string of a GOES message (all continuous wind measurements are flagged, if more than one error detected, than flag is upgraded to a T flag). • k Difference between duplicate measurements is too high. • m High frequency spikes detected in the wave spectrum (C11), WVHGT is flagged. • n Measurement failed comparison with NCEP model fields. • p Failed wave height to wave period comparison test. • q Swell direction is from an improbable direction. • r Related measurement failed (continuous winds only). • s Stuck raw compass ( RCOMP and WDIR are flagged). • t Tendency difference between duplicate sensors is too high. • v Failed relative humidity verses visibility check. • w Failed wind direction verses wave direction check. • x Wind wave energy is too high for prevailing wind speed. • y Wind wave energy is too low for prevailing wind speed. • z Failed bow azimuth verses wind direction check.
NDBC Technical Document 03-02 Handbook of Automated Data Quality ControlChecks and Procedures of the National Data Buoy Center February 2003 National Data Buoy Center Stennis Space Center, Mississippi 39529-6000
Range check (wave height) Previous EQC limit: 5.2 m (Lakes Erie and Huron) New EQC limit: 7.5 m
We always improve our systems and DQC/QA process so we can get more accurate wave measurement