New Zealand has a highly-diverse marine environment that is influenced by rapidly evolving weather systems and a complex coastline.  This makes predicting the movement of pollutants and organisms in our coastal waters a very challenging task.

By pairing high powered computer resources with a new advanced numerical code that can simulate ocean and atmospheric processes, MetOcean has developed effective tools that can help resolve this complexity.

The system architect, Dr David Johnson of MetOcean Solutions, says over the next year this innovative new system will be established with the collaborative support from fisheries scientists at MPI and researchers from the Cawthron Institute.

“The application will be designed within a GIS platform, and allow scientists and researchers with minimal numerical modeling experience to be able undertake meaningful simulations,” he says.

The core of the application will be a continuous ten-year hindcast of the ocean currents around New Zealand, including all the major ports and harbours.  A hindcast is essentially a recreation of the historical conditions.  Hour-by-hour currents throughout the water depth will be resolved for the tidal, oceanic and wind-driven flows.

“By creating a long term database which includes a range of climatic signals, we can undertake multi-year simulations of trajectory that provide robust statistics on things like connectivity and seasonality,” Dr Johnson says.

MPI Senior Advisor, Dr Daniel Kluza, sees many ways in which system will be used to help manage New Zealand’s marine resources and deal with potential biosecurity risks.

“We will be able to simulate the release of a foreign marine pest organism at one of our ports for example, and predict the likely spread into the adjacent coastal marine environment. Armed with this knowledge we may be able to prepare contingency plans to prevent or eradicate the pest.”

The system is expected to be fully operational by October 2013, and will be available to scientists for non-commercial research.

MetOcean Solutions Ltd have been providing oil spill trajectory prediction services to the offshore industry for many years, examining both the statistical likelihood of impacts to the coastal environment due to a spill, as well as the short range (1-7 day) predictions in the event a spill should occur. That system was adopted by Maritime New Zealand during the Rena incident in 2011, forming the primary guidance for oil trajectory and beaching predictions.

Check out the website The Great NZ Science Project.

Lagrangian Trajectory Modelling

Most Lagrangian Trajectory models have the ability to be driven off gridded
global weather models such as ECMWF and GFS. These would have horizontal resolutions of at most 16km and 30km respectively and where the full gridded fields
(required to drive the air quality models) are only available at 3 hourly
intervals. This approach is fine for regional scales, but for scales of around
10km or less, or where there is very complex terrain, higher resolution input
data is required. A number of Air Quality models, such as HYSPLIT and
FLEXPART have been adapted to use the higher resolution of models such as WRF
for these reasons. MSL’s hindcast was constructed using WRF and is available
down to a resolution of 4km in some areas to drive these Air Quality models.

Steady State Gaussian Modelling

For those wishing to run steady state models, the metorological data at the
site of interest is of key importance. However, there are not always
observations available for the researcher to use. Interpolating surrounding
observations to the site, or even using coarse model data may not give the
desired level of confidence. Interpolating the high resolution hindcast data
to the site may be the best option in some areas.

• 12 km hindcast over all New Zealand
• 4 km hindcast over Hauraki Gulf and Central NZ
• Further resolution as  requested
• Hourly data from 1979-2011

Applications of the data

• Renewable energy
• Climate studies
• Ocean modelling
• Catchment modelling
• Extreme value analysis
• Forensic analysis

Senior Meteorologist Greg Pearson notes that the hindcast was a substantial undertaking, and the data set represents a significant advancement in the toolkit for New Zealand environmental scientists and engineers. “The choice has always been between coarse resolution data over long time periods, or high resolution data over short time periods. This hindcast is both high resolution and spans one third of a century. The goal is to provide ready access to this data set for scientists, engineers and researchers”.

Technique and methodology
The hindcast was constructed using the latest atmospheric modelling techniques (Weather Research and Forecasting model – WRF) and was undertaken by our scientists over a 4 month period using a dedicated computing cluster at the Sabalcore facility in the USA. All of New Zealand was modelled at 12 km resolution, with boundaries provided by Climate Forecast System Reanalysis (CFSR) data set. However, even 12 km resolution is not fine enough to capture some of the important physical processes over many parts of the country. So, the WRF nesting capability was used to increase resolution to 4 km over Auckland and central regions of the country. Further nesting to 1.3 km is available on request.

Data accuracy validated
The hindcast results have been rigorously validated against measured meteorological data from a range of locations around New Zealand.

For more information contact us.

The forecast system relies on high resolution atmospheric modelling together with statistical post-processing from data being assimilated from measured wind. There had been significant modelling, process testing efforts to find the correct balance between physical and statistical modelling for this wind farm, which has led
to a very competitive solution. This enables the wind farm operator to maximise
the return on investment from the wind farm.

SwellMap.com has been providing marine weather to boating and marine enthusiasts
for the last seven years, since 2005. The team at MetOcean Solutions is delighted to be
recognised by MetService and collaborate to present quality forecast to an even larger
segment of the New Zealand population.

MetService’s aim is to give New Zealanders the best quality weather information they
need, every day of the year, with a one-stop-shop approach. Getting information from
SwellMap and feeding into the MetService weather site will offer the user a full range of
weather information.

“Partnering with MetService to deliver up-to-date SwellMap information to the public is
an exciting opportunity. With a fantastic reach and growing community of users, we are
delighted to provide the public with high-quality data for planning their marine
activities.” Peter McComb, Managing Director of MetOcean Solutions commented.

Just in time for summer, MetService.com new Marine & Surf section combines the latest
surf and beach cams from around New Zealand with comprehensive swell forecasts
from specialist provider MetOcean Solutions.

Once summer kicks off, so will MetService’s beach and surf weekend video forecasts;
they’ll be hosted by forecasters who are not only experts in the field – they’re regular
surfers themselves.

“We never stop thinking about how we can add to or improve on the website. It’s all
about quality information that’s easy and fun to use.“  said MetService interactive
manager Craig Delany.

SwellMap presents weather data our own proprietary wave model that uses state-of-
the-art atmospheric and oceanographic models. Observational data along with numeric
(computer) models are used generate high-resolution and quality forecasts. The site is
updated 4 times a day providing 7 days of weather data at 6 hourly intervals.
With user-friendly formats; including charts/maps, graphs and tables, SwellMap
presents an excellent and simplified way to interpret the week’s weather patterns at a
glance.

For more information contact:
Rosalina Pang
Marketing Coordinator
MetOcean Solutions Ltd
Ph: +64 6 758 5035
rosalina@metocean.co.nz

Rena, a Liberia-flagged 235m vessel ran aground at Astrolabe Reef, off the Bay of Plenty, New Zealand shoreline at approximately 2.20 am on the 5th of October, 2011. Heavy fuel oil was seen to be leaking from the vessel on Wednesday morning. Following the initial release of oil it was observed that ongoing releases occurred on each low tide.

Maritime New Zealand maintains and operates an oil spill trajectory model. To complement those predictions, the MetOcean Solutions trajectory system has been used as an independent source of information, based on the use of MetOcean’s local-scale high-resolution wind, wave and current forecast data.

MetOcean run operational forecasts with a New Zealand scale atmospheric model (WRF) plus a regional current (POM) and wave (SWAN) model. Forecast cycles are run every 6 hours with predictions for up to 7-days in advance. Data include the tidal and residual currents all around NZ.

The forecast system updates automatically every 6 hours, and these high-resolution data are being used in the General NOAA Operational Modelling Environment (GNOME), which predicts the fate of the spilled oil.

To account for the variability of time release of the spill, GNOME has been setup to release oil 2-hours either side on low tide only. In the attached simulation, the black particles represent the expected plume while the red particles represent a “minimum regret” or worst case spreading of the plume.

The first oil arrived on Mt Maunganui Beach in the morning of 10th October, and by midday there were long lines of stranded oil. Later that evening there were reports of oil reaching as far south as Papamoa.

Initial feedback from onsite responders suggest that the predicted timing and location for the beaching of the plume was accurate.

For more information on MSL emergency response interface (ERI), contact us at +64 6 758 5035 or email: enquiries@metocean.co.nz.

We are very please to announce the addition of Greg Pearson and Frank Sonntag to our team of scientists. Greg has come from his previous role as head of Research and Development at New Zealand Metservice to head up a new atmospheric group within MSL. Greg has extensive experience with the configuration and implementation of state of the art numerical weather prediction models. His skills will be fully utilised to continue to provide the best possible marine forecasts and hindcasts, and also to develop new land based meteorological services. Frank is an atmospheric scientist with outstanding skills in the design and development of operational forecast and hindcasting systems.

The new group’s first job has been an upgrade of the core operational model used to generate forecasts for WeatherMap. Other ongoing development projects are investigating optimal NWP configurations for our marine forecast services and the development of new forecast and hindcasting tools to service the wind energy sector.

WXTiles

A free service called WXTiles offers forecast overlays for existing GoogleMaps or OpenLayers web pages. Anyone can sign-up and add weather forecasts to their own websites. The WXTiles service gives access to pre-rendered overlay tiles that sit on top of a web page map. A Javascript API means that anyone can easily add weather forecasts to their web site with just a few lines of Javascript. The forecast overlays can also be added to Google Earth through a KML link.

Check out WXTiles at www.wxtiles.com

Web services for GIS

In collaboration with Explorer Graphics Limited, we are developing ways to integrate weather data into existing GIS applications. As well as providing key baseline information relevant to many operations, weather data is also integrated into management decision tools through GIS based queries.