NIWA’s Right of Reply: My Response After
NIWA statement about Historic Weather Events website
The website Ian Wishart has based his opinions on is not New Zealand’s national climate database.
Wishart has conflated the Historic Weather Events website with the New Zealand National Climate Database that is used by NIWA and climate scientists nationally and internationally to analyse weather events and trends in the climate.NIWA’s New Zealand National Climate Database includes official observations back to the 1850s, including some descriptions of significant weather events of the time.
The Historic Weather Events website that Wishart has based his study on is a catalogue that was created to collate descriptive details on impacts and hazards from weather events. It is not an official type of meteorological record. It has gaps because it contains information gathered from newspaper articles, journals, books and reports that periodically come to light. Gathering these sources and validating them takes significant time and is an ongoing process.
The New Zealand National Climate Database is a national electronic database of high-quality instrumental observations, including temperature, rainfall, wind, pressure and much more. NIWA holds data from approximately 6500 climate stations across the country, surrounding islands, some Pacific Islands and Antarctica. NIWA also retains paper and digital copies of original historical observations which are classified as a heritage asset.
NIWA holds 10-minute, hourly and daily observations from more than 600 stations currently operated by NIWA, MetService and other organisations. These data are freely available via CliFlo (https://cliflo.niwa.co.nz/), along with about 80 types of monthly and annual statistical summaries, and several 30-year climate normals. This resource has been freely available to the public and researchers externally since 2007 and is frequently accessed.
NIWA’s scientists use the database to rank and quantify how unusual observations are, and report on it. Summaries using these data are published monthly, seasonally and annually at https://niwa.co.nz/climate/summaries.
NIWA released their monthly climate summary with extensive analysis and discussion on Cyclone Gabrielle: https://niwa.co.nz/climate/monthly/climate-summary-for-february-2023. To produce this analysis, NIWA used data from the National Climate Database and our archived material as described above, as well as other tools from international organisations to provide some historical context. The comparisons to contemporary storms for which we have high-quality observations were done using standard practices.
NIWA did not claim that Gabrielle was the strongest ever storm to pass near the North Island, only one of the strongest storms. Similarly, NIWA did not say it was caused by climate change, only that climate change was likely to be a factor in its severity.
Examining the significance of a weather event like Gabrielle extends well beyond a single meteorological metric. Atmospheric pressure, which Wishart focused on, is only one way of measuring a storm’s intensity and wind strength. NIWA typically examines several factors to describe a cyclone’s intensity in its climate summary, including rain, wind and sea-related impacts.
Great to see a carefully worded NIWA statement that doesn’t answer the questions posed.
Firstly, I didn’t “conflate” the two databases, as a glance at the two Climate of Fear reports will confirm. What I have said is that there are many, many historic storms with deeper lows than Gabrielle which NIWA should have known before singling out Gabrielle as one of the biggest storms to hit NZ since 1850.
The COF1 report found Gabrielle only reached #8 in the top ten storms between 1868 and 1890 – a 22 year snapshot out of 183 years of records – and that was if we measured Gabrielle at NIWA’s initial projection of 963 hPa. Gabrielle has since been revised weaker by NIWA to 968, meaning it would no longer make the top ten.
In a Twitter exchange late February, NIWA scientists were genuinely surprised when I pointed out barometric lows deeper than 963 in the past. It was a reasonable inference that if their main database contained those deeper lows it would have warned them before they put out statements about Gabrielle being one of the biggest ever.
And when NIWA issued its first statement responding to COF1, it said many records had not been entered from the 1800s because they had not been “validated” yet. Again, there is a reasonable inference that those same events are not loaded in the main database either, because surely if they had already been validated for CliFlo they would already be valid data to load into HWE?
Turning to NIWA’s final paragraph, they suggest I only measured storms on one metric – barometric lows:
“Examining the significance of a weather event like Gabrielle extends well beyond a single meteorological metric. Atmospheric pressure, which Wishart focused on, is only one way of measuring a storm’s intensity and wind strength. NIWA typically examines several factors to describe a cyclone’s intensity in its climate summary, including rain, wind and sea-related impacts.”
Any careful reading of the COF1 report will reveal that nearly every barometer reading was corroborated with extensive eyewitness reports of rain, wind and sea-related impacts, so that attempted rebuttal also flies like a cement duck.
A possible explanation for why NIWA seems to be flying blind on historical climate is their archaic (in my view) main CliFlo database. Having now created a username instantly and logged in, it is clear that CliFlo is packed full of instrument data: pressure, temp, windspeed, rainfall etc. However the database bears a striking resemblance to an Excel interface from the 1990s – it will call up instrument data for the stations you select, but it won’t do a general text search.
This means that a user cannot do a search across the database for all storms below a certain barometric pressure, or above a certain rainfall or windspeed. Those options don’t exist, at least not in the version NIWA has linked to above.
Nor is the information easily accessible. Just as an exercise I went to locate the Auckland Port barometer readings from the Great Storm of February 1868. The records aren’t there. The Port is not even listed as a station, and the earliest records from any station were 1960.
Presumably NIWA staff have a few internal bells and whistles, but based on the database design it certainly isn’t the climate equivalent of Google.
Finally, can anyone tell me where in NIWA’s response to my questions they actually answered the questions? Here’s what I asked:
I am struggling to reconcile NIWA’s statement that the 1924 Hawke’s Bay flood where 20.14 inches of rain (512mm) fell in just ten hours (bigger than Gabrielle) was a 1 in 150 year event:
…with the revelation that an equally big flood hit the same region in 1897 and a rainfall intensity of 292mm in just 2.5hrs had been recorded then:
Clearly on NIWA’s own evidence massive rain dumps in the HB catchment are relatively common…maybe 1-in-30yr events in the 1800s and 1900s…how does NIWA explain its apparent failure to correctly assess the probability of extreme events like Gabrielle in the HB region?
How does NIWA respond to the inference that its advice to local councils on probability may be flawed because of a failure to fully document historical data and reports into an integrated intelligence database that allows any region to instantly know all the major events in their past and where pressure points were?
Given that fully entering all major event/system (sometimes systems don’t create much damage by sheer luck, but they should still be recorded) data from the beginning of published news reports in NZ would provide a much better baseline from which to measure system frequency and intensity, why has this work not been completed in 31 years, and particularly since 2013 when all newspaper reports up to 1945 were digitised and available online?
I have a 5pm deadline for these questions, and will note that they have been asked if replies are not available by then.