A team of New Zealand climate researchers has gone back through more than 160 years of rainfall records and reached a sobering conclusion. The three worst meteorological droughts ever measured in this country all struck before 1950, and the way the modern economy plans for dry years may badly underestimate what nature is capable of producing.
The work, by Adam Brown of the University of Waikato and Dave Frame of the University of Canterbury, along with Luke Harrington also at Waikato, was published in The Conversation on 15 May 2026 and republished by RNZ. The team selected 97 high-quality weather stations across both islands, each with at least 70 years of continuous data and observations stretching back before 1914. They then compared each station’s historical rainfall against its own long-term average to identify periods of unusually severe deficit.
The most extreme single event in the instrumental record was the drought that began in the winter of 1914 and ran into February 1915. According to the researchers, more than half of all stations recorded one of their five driest periods on record between July and October 1914. The 1907 to 1908 summer drought was almost as widespread, with many South Island stations logging their driest conditions on record and more than twice as many stations recording extreme deficits as the next severe drought on the list. The third entry on the podium is the 1945 to 1946 drought, which hammered the north and east of the North Island. Rainfall shortages there were about 22 percent greater than during the better remembered 2012 to 2013 drought.
That last comparison is the one that should give farmers, councils and insurers pause. The 2012 to 2013 drought is the benchmark most New Zealanders still picture when they think of a hot, brown summer. It cost the national economy an estimated 0.7 percent of GDP, dried out pastures from Northland to the Wairarapa, and prompted emergency declarations across the upper North Island. The 1945 to 1946 event was meaningfully worse, but it sits outside living memory for almost everyone now making planning decisions.
The researchers point out that the 2007 to 2008 drought, which quadrupled silage prices and cost the economy several billion dollars, ranks well below all three of their early twentieth century events. In other words, the worst dry spells this country has experienced in the past 30 years are not the worst the climate system has produced. They are simply the worst the current generation has lived through.
That gap between memory and history matters because most modern risk assessments lean heavily on the recent past. Insurance models, council water supply plans and farm budgets are typically calibrated against the last 30 to 50 years of weather. If the early twentieth century holds the real upper bound on drought severity, then a great deal of New Zealand infrastructure may be sized for the wrong design event.
The authors are explicit about that gap. “Recent droughts may no longer provide a reliable guide for what is to come,” they write. They argue that warming temperatures layer an additional risk on top of natural variability, because warmer air pulls more moisture out of soils and intensifies evaporation during dry spells. A drought of 1914 magnitude, dropped on top of 2026 temperatures and 2026 land use, would not behave the same as the original.
There is some good news buried in the data. The 97 station network shows that severe droughts have not become more frequent in recent decades. What has changed is the temperature backdrop they unfold against. As Brown, Frame and Harrington put it, “climate change adds another layer of risk, with rising temperatures expected to further parch soils and increase evaporation.”
The implications cut across several parts of the economy. Dairy and sheep farmers in the eastern North Island and the Canterbury plains carry the most direct exposure, because pasture growth collapses quickly once soil moisture deficits open up. Hydro generators rely on reliable inflows into Lakes Pukaki, Tekapo and Hawea, and a multi month rainfall shortfall in the Mackenzie catchment can force the wholesale electricity market to lean on thermal generation at short notice. Town water supplies in places such as Hawke’s Bay, Tasman and Auckland have all been tested in the past decade by dry stretches well short of the 1945 standard.
The researchers stop short of recommending specific policy moves, but the direction of travel is clear enough. Risk plans built only on instrumental records since the 1970s probably need to be stress tested against the deeper history sitting in the rain gauges. The 1914 and 1945 events are not theoretical, they actually happened, and the same atmospheric ingredients that produced them are still in circulation.
You can read the original analysis at The Conversation and the RNZ writeup is here.
What do you reckon, is your council, your farm or your business planning around the worst drought in living memory, or against the worst this country has actually seen? Drop a comment below and let us know.
This article was written by AI, briefed to report the facts, hopefully without some of the bias people bring to the job 🙂
Leave a Reply