NZ Weather Forum

March 2019 stats –

Mean daily temp = 20.5c (beat last year’s local record by 0.1c)
Rainfall = 40.8mm (60mm below local average)
Prevailing wind NNE
Zero thunder days. Thunder drought currently 97 days.
YTD daily mean temp = 21.2c (Around 1.5c above local average)
YTD rainfall = 77.4mm (139mm behind local normal accum. Driest start to a year since 2014)

March stats from my urban sited station. Average temp max 26.1 min 14.0 mean 20.06

22 days 25.1C or above

Rain recorded a few km away at a rural site - 38.4mm

Sent from my SM-N950F using Tapatalk

March 2019 was one of NZ's warmest on record. It was also generally dry apart from very heavy falls in the west and north of the South Island, and mostly sunny except for the northern South Island and Wellington region.


Looks like Dunedin and Hamilton Aero's warmest March on record, Ruakura's 2nd equal-warmest, and Tauranga Aero and Christchurch Aero's 3rd equal-warmest.
Dunedin also recorded its highest March mean T-max, Mangere and Auckland Aero their 2nd highest March mean T-max, and Ruakura its 3rd equal-highest March mean T-max.
(Edit: Updated some rainfall & sunshine figures)

Nev wrote: ↑Mon 01/04/2019 17:56 March 2019 was one of NZ's warmest on record. It was also generally dry apart from very heavy falls in the west and north of the South Island, and mostly sunny except for the northern South Island and Wellington region.

Tekapo logged 252 hours to the 30th. Nelson Aero was 30-40 hours behind the other sites in Nelson/Tasman and its value is highly suspect - on one day the EWSs logged around 10 hours and it had less than an hour ...

The Musselburgh value looks like a real stretch and I have always thought it looks to be one of the most "optimistic" of the EWSs.

Ch'ch Aero sun reading similar to Rangiora (172) given 31st not entered yet.

I see Auckland Airports average maximum in March (24.5C) was slightly higher than the January average max (24.4C).

In January there was persistent onshore freash SWesters while March lighter E/NE prevailed.

NZstorm wrote: ↑Tue 02/04/2019 07:25 I see Auckland Airports average maximum in March (24.5C) was slightly higher than the January average max (24.4C).

In January there was persistent onshore freash SWesters while March lighter E/NE prevailed.

One of the interesting things about being located on an isthmus.
Had the reverse here - with more NE winds in March moderating daytime temperatures more than January. 25.5C mean daily max in Jan vs 24.4C for March.

Rainfall 16mm
Highest max temp 31.2deg 6th
Lowest 1.8deg 1th
Ave max 19.7 deg

Richard wrote: ↑Tue 02/04/2019 08:09 Rainfall 16mm
Highest max temp 31.2deg 6th
Lowest 1.8deg 1th
Ave max 19.7 deg

Bit nippy compared to NIWA's Medbury mean T-max of 23.0C this March…

46.3mm of rain here for the month but no real weather max/min figures as I do mine manually.
It was 32C on March the 6th and 5.5C the minumum of the morning of the 29th.
When I was away the maximum did not exceed 30C or drop below 8C.

Nev wrote: ↑Tue 02/04/2019 12:27

Richard wrote: ↑Tue 02/04/2019 08:09 Rainfall 16mm
Highest max temp 31.2deg 6th
Lowest 1.8deg 1th
Ave max 19.7 deg

Bit nippy compared to NIWA's Medbury mean T-max of 23.0C this March…

I did it the old fashion way with a calculator, either chuck it away or I need to go back to school.

Richard wrote: ↑Tue 02/04/2019 17:34
I did it the old fashion way with a calculator, either chuck it away or I need to go back to school.

I believe in the old traditional way, where a hunam being checks the weather rather than an unmanned system where bugs and other things can affect readings, like spiders/bird nest building especially in dry/remote areas in a rainfall bucket tipping system.

NIWA has just released it's March figures and it's stating that NZ has just had it's 2nd warmest March on record and that it's been 26 months since NZ has had a below average month in terms of temperatures. Does anyone have any idea why this is? Are there any factors driving this string of above average temperatures (apart from climate change)? There must be something else at play here since places like the US, Australia and the UK have had numerous colder then average months in that same time period, why would NZ be different?

My 2 cents worth on the long stretch of above average temps... I've observed the Hadley cell during this period to be further south than normal (sometimes below 40S) during the warmer months. The flow on affects having been major blocking highs in the Tasman / South Pacific, creating 2 powerful marine heat waves in the region for the last 2 summers. Sea surface temperatures have then played a major role in both sides of the Tasman (Victoria / Tasmania) being well above normal on land. Sea surface water take months to cool, so cooler months have also had increased temperatures due to the sea surface temps in the region.

I'm sure only a few years ago NIWA was saying that even under a 2-3C AGW scenario, under some models NZ may only warm about 1C, there would be a greater temperature difference between the tropics/subtropics and Antarctica and so the Westerlies would increase. That's not what we have seen last two Summers. Last year was weird in that there was almost no "spring westerlies" period (at least not North of Taupo). It went straight into persistent anti-cyclones from October. Yes there we some Westerlies in January but not cold and nothing like normal Spring ones.
If this shifting of the Hadley Cell South is a new normal, we may be in a region that sees about 2C warming right away. Something to ponder.

Bradley wrote: ↑Wed 03/04/2019 11:55 NIWA has just released it's March figures and it's stating that NZ has just had it's 2nd warmest March on record and that it's been 26 months since NZ has had a below average month in terms of temperatures. Does anyone have any idea why this is? Are there any factors driving this string of above average temperatures (apart from climate change)? There must be something else at play here since places like the US, Australia and the UK have had numerous colder then average months in that same time period, why would NZ be different?

Technical issues to be considered would be whether the long term average used for comparison has been artificially lowered due to "corrections" made to raw data; whether the stations chosen to calculate the NZ-wide figure are representative; and whether those stations are suffering from urban heat island affects which are increasing with time. However, I checked the satellite lower tropospheric monthly anomaly maps at https://www.nsstc.uah.edu/climate/archives.html and the conclusion of 26 months since Jan 2017 not below normal for more than 50% of NZ seems to be correct, so we can probably dismiss those concerns.

A lot of those 26 months were near normal, and I dont know what criteria Niwa would use to allocate a month as "below", "normal", or "above". Its not clear if all the months of the reference period are split equally into thirds; or if 50% of the months are considered nomal and you need to get into the bottom or top 25% to be colder/warmer; or if they use an arbitrary value (eg +/- 0.5deg) to determine the status of a month. Therefore its not possible to calculate the mathematical probability.

That period has been characterised by warmer than usual SSTs in the Tasman Sea, and I dont think any periods of below normal SSTs. So thats probably the driver. Warmer than normal SSTs usually accompany high pressure and light wind anomalies, so my thinking would be weaker westerlies and more anticyclones closer to NZ than the long term normal would be driving the SST anomalies. That 26 month period corresponds with weak SOI activity and no strong El Nino or La Nina periods, which is probably why the atmospheric circulation anomalies have turned out the way they have. That, in turn, I would put down to being at solar minimum. When we start to climb out of solar minimum we will see a significant El Nino event develop (my prediction is not the 19/20 spring/summer, but the 20/21 one) and then we will have greater potential for below normal months.

Bear in mind that the circulation anomalies we experience related to the SOI are different to other parts of the world, and the cold months in the Northern Hemisphere have often been due to SSW events and leakage from the polar vortex (even down to quite low latitudes - eg snow in the Sahara last winter), which is a very different dynamic from what happens in the Southern Hemisphere.

NIWA wrote:New Zealand has not experienced a nationwide monthly mean temperature that was below average (0.51 ̊C to 1.20 ̊C below the 1981-2010 average) in 26 months or since January 2017.

Strange way of defining 'below average', i.e. "0.51 ̊C to 1.20 ̊C below the 1981-2010 average".
So May 2017 and Feb 2017, which were 0.3C and 0.1C repectively below their 1981-2010 normals are not considered 'below average'…

If we take the 30 values which are the departures from the (30 value) mean for the 1981 to 2010 Marchs, they range from -1.85 (1992) to +1.89 (1999). If we rank them in order from lowest to highest, the first ten (i.e. the coldest 10 of the 30) range from -1.85 to -0.52. The middle third (what we might reasonably call the "normal" seasons in the set) range from -0.34 to +0.39. The last ten (i.e. the warmest 10 of the 30) range from +0.49 to +1.89. [ All of this is based on the "adjusted" Niwa 7 station series, and if you want to play with the data yourself you can download it here https://www.niwa.co.nz/our-science/clim ... ature-data ]

So, defining normal as +/-0.5 seems reasonable (at least for the month of March, and the 1981-2010 period), because it matches closely the distribution if we split the series into three equal parts. If we were to take my other suggested approach, and define "normal" as the middle 50%, then that range would be from -0.85 to +0.83.

IF we assume that the other months are similar, and assume a normal distribution (which is probably not the case), then on the basis of chance we could expect one third of months to be below normal, or two thirds to be normal or above normal. I will try to get my head around how to calculate the probabilities involved in this case.

Edit - too hard, because the events are not independent (i.e. if a month is warm, then there is an increased chance the next month will be warm not cold, etc) conventional probability theory doesnt work. I'm sure the probability of a run of 26 "not colder" months could be calculated, but it needs a whole heap of work.

According to an answer from NIWA lower down the commentary in the tweet linked below, mean temperatures between 0.0°C and 0.5°C above and below average are deemed to 'near average'. To be warmer or colder than average, the anomaly has to be greater than 0.5°C.

https://twitter.com/NiwaWeather/status/ ... 8498265088

Tony T, thanks for the link to the NIWA seven stations data.

It would be interesting to know what range of anomalies the "middle third" covers for other months.

Simon Culling wrote: ↑Wed 03/04/2019 18:54 According to an answer from NIWA lower down the commentary in the tweet linked below, mean temperatures between 0.0°C and 0.5°C above and below average are deemed to 'near average'. To be warmer or colder than average, the anomaly has to be greater than 0.5°C.

https://twitter.com/NiwaWeather/status/ ... 8498265088

Tony T, thanks for the link to the NIWA seven stations data.

Yes, hence my post above (which the responder on that tweet looks to have more or less copied).

Interesting that the BoM in Australia, which use 1961–1990 normals as a comparison for their national mean temps, doesn't seem to make quite the same distinction, e.g. ...

• Aug 2018, with a national T-mean 0.07C below normal, is described as "close to average".

• June 2018, with a national T-mean 0.33C above normal, is described as "above average".

• May 2018, with a national T-mean 0.19C above normal, is described as "somewhat above average".

• April 2017, with a national T-mean 0.09C above normal, is described as, "A warmer than average month overall".

P.S. Note that Aug 2018 was Australia's only month in the a last 26 months with a T-mean below its 1961–1990 normal.

Interestingly though it's different on BOM's decile maps:

90+ is "much above average", 70-90 is "above average", 40-70 is "average", 10-40 is "below average", and 10- is "much below average". I'm going to check NIWA's 7SS monthly 1981 2010 values for the middle third ranges.

Nev wrote: ↑Wed 03/04/2019 21:36 P.S. Note that Aug 2018 was Australia's only month in the a last 26 months with a T-mean below its 1961–1990 normal.

Where do you get that figure from? I'm getting +0.03 for Aug 2018 and the last colder than normal month was Oct 2016 (-0.38°C).

http://www.bom.gov.au/climate/change/in ... Dallmonths

Willoughby wrote: ↑Wed 03/04/2019 23:00

Nev wrote: ↑Wed 03/04/2019 21:36 P.S. Note that Aug 2018 was Australia's only month in the a last 26 months with a T-mean below its 1961–1990 normal.

Where do you get that figure from? I'm getting +0.03 for Aug 2018 and the last colder than normal month was Oct 2016 (-0.38°C).

http://www.bom.gov.au/climate/change/in ... Dallmonths

From BoM's Monthly Climate Summaries…

http://www.bom.gov.au/climate/current/s ... ives.shtml

No need to wonder too much about why March was so warm.

TonyT wrote: ↑Thu 04/04/2019 09:49 s_wpac_cdas1_anom-1.pngNo need to wonder too much about why March was so warm.

Thanks Tony, I guess we can say goodbye to any sea level snow events in Canterbury this year, those SST's aren't shifting anytime soon, I can't think of any major reason winter this year will be any different to the warm winter last year?