data sources and validation
Tidetech's scientists produce high resolution tidal models in-house and use their expertise and knowledge to select the best performing meteorological and oceanographic third party datasets from official and academic sources.
Information and examples of the validation work we perform on our own tidal models is shown below. We also perform validation work on the data products we source from third parties.
Tidetech proprietary models carefully validated with available observations
Tidetech constructs hydrodynamic models of tidal elevation and currents using techniques developed at the world's foremost tidal research institute, the UK National Oceanography Centre (formerly the Proudman Oceanographic Laboratory).
Model output is validated using available observations, with the preferred method being validation against tidal currents obtained from current meters. Below is an example of validation of Tidetech’s high resolution model for San Francisco Bay.
Only a very small amount of current meter data is available worldwide, due to the challenges of measuring currents in the open sea. Instead of current meter data, model output can be validated against tidal elevations obtained from analysis of tide gauge data. At least 12 months of observation is required to obtain sufficient tidal constituents to accurately predict future tidal elevations. Locations where this quality of information is available is limited to major (primary or standard) ports. Ideally, to obtain the full set of tidal constituents the observations would need to span 18.6 years (the full tidal cycle).
The image below shows validation of Tidetech’s high resolution model for San Francisco Bay against tide gauge data.
On a larger model grid, averaging model values over a grid square will make comparisons more general in nature, but they are useful nonetheless. The image below shows comparisons between Tidetech 2km resolution English Channel model and tidal elevation predictions produced from analysis of tide gauge data.English Channel Regional Tidal Model – Standard and Secondary Port Height Comparisons. Official tidal height Predictions (UK National Oceanography Centre – Tide Gauge Derived) vs Tidetech. (Model – Blue. Observations – Green.)
Tidal currents can also be validated against data from official sources, although this approach should be used with caution. Unfortunately, predicted tidal currents from official sources can vary considerably in accuracy, unlike predictions of tidal elevations at major ports. Much of the data from official sources comes from observations taken a long time ago, before modern instrumentation. At least one month’s worth of observations is required to reproduce the primary constituents at a single location and, in plenty of cases, only a couple of day’s observations have been actually taken. Observations taken when the meteorological conditions were not calm are also questionable, since wind driven surface currents can distort results.
Tidal currents from Tidetech’s 2km English Channel model validated against predictions obtained from UKHO tidal diamonds.
the best performing ocean model data selected from multiple sources
Models include Mercator (NEMO) model run by the MyOcean consortium, and HYCOM run by the US Naval Research Laboratory, FOAM run by the UK Met Office and GSLA run by Australian Commonwealth Scientific and Industrial Research Organisation (CSIRO).
Validation for Mercator:
Metrics, comparing the model against in-situ or satellite observations, are routinely calculated for several model parameters (salinity, temperature, sea level anomaly, sea surface temperature). The table below shows the sea level anomaly (SLA) comparison for the north Pacific region between satellite altimeter and the model over a 12-month period.
In the top figure it is seen that, on average, there is a consistent 8cm Root Mean Square Error, which varies little with the forecast lead time (i.e. the forecast at 132 hours contains the same level of uncertainty as the forecast at 36 hours). The bias is always positive, meaning that the model SLA is always larger than the satellite SLA by approximately 2cm. The lower plot shows the number of altimeter measurements per day, a function of the number of satellites available. With the addition of satellites after November 2013, the bias appears to show less variability, indicating the value of having more satellites in the sky to reduce variability in the altimeter signal.
Sudre, J & Morrow, RA (2008) Global surface currents: a high-resolution product for investigating ocean dynamics. Ocean Dynamics DOI 10.1007/s10236-008-0134-9.
numerical weather model outputs sourced from official government sources and private weather providers
Models include GFS – Global Forecast System (US National Centre for Environmental Prediction), ECMWF (European Centre for Medium Range Weather Forecasting) and CMC (Canadian Meteorological Centre). The primary operational product provided is the GFS, a model in common use by weather routing companies.
Validation statistics, comparing operational forecast weather models, are routinely provided by NOAA. Statistical metrics averaged over the latest month’s data compare the performance of eight operational models (including GFS, CMC and ECMWF) at the global and regional scale. Murphy’s Skill Score is used as a comparative measure of forecasting skill for atmospheric models, where a value of 1 equals a perfect score. In the figure below, the performance of the GFS model (black line) can be compared against other models.
Skill comparison – Multiple atmospheric models
wave model outputs sourced from official government sources and private weather providers
Tidetech supplies wave data from the NCEP WW3 model (NCP).
The WMO (World Meteorological Organization) routinely produces monthly summary statistics for wave parameters from all operational wave model forecasts (16 models, 5-day forecasts) against a common set of wave buoys. These buoys are concentrated in the areas of the northwest European shelf, the northeast USA, the northwest USA, the Caribbean and Hawaii, and a handful of buoys elsewhere.
These summaries (the latest of which is for May to July 2014) show that the RMSE of significant wave height (Hs) increases with forecast time (from 0.2m to 0.6m, with a consistent spread of 0.2m) in a similar manner to the wind speed forecasts above. Most models tend to underestimate Hs consistently through the forecast window by 0.1–0.2m.
high quality observational data from satellites
Tidetech uses a 1km global blended product to produce daily sea surface temperature (SST) data. This product arises from a combination of different satellites.
Comparison of the global product with in-situ data is carried out daily. The figure below shows a scatter plot of blended SST against ship measurements for the 2nd May 2014, indicating the use of more than 3000 observations, with a mean bias of -0.12 degrees celsius and a RMS of 0.79 degrees. This is a day taken at random and appears typical.