A mediator has been appointed to smooth out the stormy relationship between MetService and the National Institute for Water and Atmospheric Research (NIWA), following allegations that farmers have suffered because the two don’t cooperate enough.
Tensions over the relative capabilities of the organisations’ two different climate forecasting systems underlie the conflict. But neither Peter Uddstrom, who is responsible for the NIWA climate model, nor the MetService’s CIO, Russell Turner, will comment publicly on the re-emergence of a conflict that dates back to the mid-1990s. Instead, comment has come predominantly from Peter Hargreaves, former chief executive of NIWA, who left the organisation in 2002. He says changes to MetService are at the root of the problem.
At issue is the allegation that closer co-operation between NIWA, which has an international climate model based on a Cray “supercomputer”, and the MetService, with its weather data, could have seen people warned earlier of last year’s floods — early enough to minimise stock losses and other damage.
Hargreaves says that in the 1990s, during his tenure as chief executive, it was suggested that NIWA and the then Meteorological Office merge. But the government took the Met Office in another direction, turning it into a state-owned enterprise. Initially, there was an arrangement whereby the MetService bought the results of climate research from NIWA, but this collapsed in 1993, says Hargreaves.NIWA’s software program consists of a large-scale environmental model, and the researchers working on it look at weather forecasting as a research project, says Hargreaves, who is now with Antarctica New Zealand. Because it is a continuous real-time simulation, it is better at forecasting short-term events than the MetService’s systems, which depend on periodic updates, he says.
The MetService is working on what it calls “nearcasting” — the forecasting of near-term events, covering everything from floods to determining if the weather is suitable for painting road-markings. However, the MetService’s Turner would not be drawn on making comparisons between this work and the modelling that NIWA does.
Because the NIWA model is an international one, it might be assumed it would be too coarse-grained for predictions covering small areas of New Zealand. But NIWA sources say the model scales down successfully.
Uddstrom, the man most closely associated with the model, declined comment on this.
Friction between the two organisations has now resulted in the appointment of a mediator whose job is to improve relations.