Understanding the long range forecast

Resolution counts

Weather forecasting is basically an ‘initial value problem.’ This means that if weather models could capture the current weather perfectly i.e. correctly initialise the current state of the atmosphere, then in theory, they could forecast the future weather accurately well beyond the present limit of a week or so. After all, the atmosphere is a fluid, and follows strict laws of physics.

The difference between weather and climate

It may seem obvious. Weather is what we experience locally day-to-day and week-to-week, while climate is a background signal sitting ‘behind’ the weather, observed at the seasonal, annual or even decadal time scale. But the interplay between what is dominate in the short-term and what prevails in the longer run, is not so straight forward. For example, 2015 was a very dry year for regions in the north and east of both islands of New Zealand. But there were rain events in these areas, and some of them were very heavy.

Christmas past

Christmas Day in New Zealand can be a hit and miss affair, weather wise. Some years, a High arrives in time for Santa. Christmas Day 2009 was beautifully fine and warm across the country, with Dunedin hitting 32°C, and many other centres approaching  25°C.

Midday analysis for 25th December 2009. Midday analysis for 25th December 2009.

Radar and mariners – a long relationship

The development of modern radar started in 1886, when German physicist Heinrich Hertz showed that radio waves could be reflected from solid objects. In 1904, another German, Huelsmeyer, was the first to build a simple ship detection device, successfully detecting a ship in dense fog. In the pre-war period (the 1930s), many nations were working on radar devices. A major advance occurred once systems were developed that allowed short ‘pulses’ of radio energy to be generated, allowing the range of the object to be determined by timing the pulses.