Past

Tales of the Lorax: Climate Change Prior to Human Arrival

Glacial-Interglacial Cycles

About 2.6 million years ago a cooling Earth passed a thermal threshold, triggering an ice age and the start of the Quaternary Period. Fifty glacials later, we are now 12,000 years into the most recent interglacial (the Holocene). Somewhat similar interglacials separated each of the glacials. Some were cooler, others warmer, but I suspect that Auckland’s present climate would have been recognisable to us in each of them (dominated by the westerlies but with notable influences from the Southern Ocean and the tropics). In contrast, the magnitude of the glacials generally increased as the Quaternary progressed with the most recent, peaking about 20,000 years ago, one of the largest. It is this glacial that we know most about in terms of New Zealand’s climate, but it is a reasonable assumption that prior glacial climates were qualitatively similar, at least in terms of the direction of climate change, and all very different to the interglacials.

Auckland during glacials and interglacials

At the height of the last glacial Auckland was cold, dry, and windy. The expansion of sea ice around Antarctica pushed latitudinal climate zones northward, increasing the influence of the Southern Ocean and weakening connections to the tropics. Weather systems now spawned over colder oceans, southerly winds were warmed less by those oceans, and warm northerlies were cooled more. Increased windiness would have been a consequence of the northward shunting of the westerlies (more like the current “furious fifties” than the “roaring forties”). And those colder winds would have carried less moisture.

Local pollen evidence (below) suggests that Auckland was perhaps 4.5–6.5°C colder through the final 10,000 years of the last glacial than during the mid to late Holocene (prior to human-caused global warming). Summer temperatures were perhaps like what we now experience in spring and autumn, and only about a degree warmer than winter for the top end of that 4.5–6.5°C range – so a bit like Invercargill’s climate today. Compared to only a few rare ground frosts today, the Invercargill analogy suggests that we may have had frosts even in summer and on perhaps half of winter days. In short, it was a tad chilly, and it is hardly surprising that the sub-tropical forests abandoned ship. Bear in mind though that the temperature depression relates to currently coastal sites which, at the last glacial maximum, were about 100 km from the ocean and 130 m above sea level. It would have been warmer at the glacial maximum coastline near Great Barrier Island (Figure 1a), so perhaps the sub-tropical forests clung on there, on favourable north-easterly slopes (maximum sunshine and sheltered from the biting southwesterlies).

The descent into the depths of the last glacial took tens of thousands of years, but the subsequent glacial-interglacial transition (to the start of the Holocene) was an order of magnitude faster. The Auckland pollen record suggests that warming got underway in earnest from 17,000 years ago, was 80% complete by 13,000 years ago, and reached a Holocene plateau 9,000 years ago. Warming was accompanied by increased wetness through the glacial-interglacial transition, presumably caused by the poleward retreat of the westerlies and the strengthening of the tropical connections that are such a key component of Auckland’s current climate. However, it wasn’t a monotonic rise, with some multi-century pauses and even reversals, probably linked to events in Antarctica, notably the Antarctic Cold Reversal 14,500–12,500 years ago.

The increasing moisture trend continued through the early Holocene, peaking about 8,000 years ago. This coincided with relatively cool summers and (perhaps) warm winters. Evidence for an early Holocene warm period, often reported in climate reconstructions for other New Zealand sites, is mixed. From about 7,500 years ago a multimillennial drying trend set in. This peaked about 3,000 years ago then recovered to something like the present. The drying trend coincided with increasing seasonality, but similar average annual temperature. Changes in seasonal radiation receipt are responsible for at least some of the reconstructed changes in seasonality.

Sometime after 1,000 years ago, all Auckland land-based sedimentary records become heavily influenced by human disturbance of the landscape and most of the climate proxies become invalid. The sedimentary story is still fascinating in terms of the evolving vegetation and lake environments, but of little relevance to reconstructing past climates. The climate story of the last millennium therefore needs to be progressed using alternative climate proxies, until we reach the era of scientific observations. That leaves us with a few centuries to bridge and leads us to a date with the whispering giants.

Kupe to Cook: Climate Change 1000–1853

The last section took us to 1,000 years ago, prior to the arrival of people. The subsequent transformation of the vegetation of much of New Zealand makes pollen of limited value as a climate proxy, but we can pick up our story using several others. Here we check out if Auckland had a Medieval Warm Period and a Little Ice Age, then what kauri tree rings have to tell us about climate variability.

The Medieval Warm Period

The term “Medieval Warm Period” derives from Europe’s climate history. There is abundant evidence that much of Europe and the North Atlantic experienced relatively warm conditions for a few centuries about 1,000 years ago. Amongst other tell-tale signs, European glaciers retreated, cultivation was feasible at higher elevations, and the Vikings were able to reach Greenland and even North America (pipping Christopher Columbus by half a millennium).

Several reconstructions of temperature for the Northern Hemisphere, mainly based on tree rings, also show a Medieval Warm Period, with peak warmth close to 1000 CE (Mann et al 2009). However, regional reconstructions show considerable diversity in timing and in some cases no evidence of exceptional medieval warmth. And a global-scale reconstruction published in 2009 showed no evidence for the Southern Hemisphere being warmer overall at the same time.  Because of this, the term Medieval Warm Period has fallen out of favour amongst palaeoclimatologists, replaced by Medieval Climate Anomaly, which recognises dynamic but inconsistent regional tempera-ture changes.

Auckland was probably warm during the Medieval Climate Anomaly, centred on about 1100–1200. The evidence is mostly indirect and relates to specific seasons, so a bit of inferring is required, based on temperatures around the country being spatially coherent. What I mean by this is that trends in temperature (mainly caused by changes in wind direction) are similar across sites, so information even from distant sites can be used to cautiously infer what was happening in Auckland. The first piece of evidence is the previously mentioned global-scale temperature reconstruction which suggests that for about 300 years centred on 1100, northern New Zealand had temperatures similar to the late 1900s (about 15.2°C). Second, we have a summer temperature reconstruction for Westland from silver pine tree rings (Cook et al 2002). Two notable warm periods were identified in the 1100s and 1200s and the researchers deduced that summer temperatures were 0.3–0.5°C above the twentieth century average, so again like the late 1900s. A third piece of evidence is a reconstruction of winter temperatures from oxygen isotopes captured in Waikato speleothems which shows peak temperatures in the 1100s.  Based on this, my best guess is that Auckland experienced temperatures at times during the 1100s and 1200s similar to the late 1900s and possibly even as warm as today.

The Little Ice Age

The European Medieval Warm Period probably got its name because of the contrast with the much colder centuries that followed: the Little Ice Age. The Greenland colonies were abandoned, glaciers advanced, the Baltic Sea completely froze a couple of times, and there were occasional frost fairs in London on the frozen Thames (with skittles, bear baiting, and much drinking). The year 1816 was par-ticularly cold. Known colloquially as “the year without a summer” and “eighteen hundred and froze to death”, summer temperatures were exceptionally cold across much of Europe, crops failed, and food riots were common. A recent study of continental-scale temperature changes found evidence of a cooling trend through the second millennium for all seven regions analysed, but with variation in the timing of when cooling started and when it peaked. So, unlike the Medieval Climate Anomaly, the Little Ice Age was indeed a global-scale phenomenon (PAGES 2k Consortium 2013).

The cold of 1816 was likely caused by the massive Tambora eruption of April 1815 in the Dutch East Indies (modern-day Indonesia). The largest volcanic eruption in recorded history, it ejected about 140 billion tonnes of magma, including fine aerosols that persisted in the stratosphere for many months.  The aerosols reflected some solar radiation back to space which temporarily cooled the planet. Interestingly, the Little Ice Age coincided with several phases of increased volcanic activity and with a couple of periods of slightly reduced solar radiation.

Westland tree rings and Waitomo speleothems both indicate that New Zealand temperatures dropped abruptly into the Little Ice Age. The trees tell us that Westland summer temperatures bottomed out in the early 1500s but winter temperatures in the Waikato didn’t reach their minimum until about 1700. When the Little Ice Age ended is arguable, but a date as late as the mid-1800s is plausible based on observations by early European explorers that glaciers had not yet retreated much from the moraines marking their farthest Little Ice Age advance (Lorrey et al 2014).

So how cold was the Little Ice Age? The silver pine tree rings and analysis of evidence of the Little Ice Age advance of 22 Southern Alps cirque glaciers suggest that summer temperatures were something like 0.5°C cooler than the late 1900s.  Unfortunately, the Waikato speleothems can’t help us because, although they reveal the temporal pattern of temperature change, reliable quantification has not yet been achieved. But there is another proxy that is highly informative – borehole temperatures.  Temperature profiles from two boreholes on the eastern side of the Southern Alps indicate that Little Ice Age average annual temperatures were about 0.9°C colder than the late 1900s, with a minimum at about 1800 (Whiteford et al 1996). If all three sources of evidence are correct, then winters across New Zealand must have been well over 1°C colder.

The mid-1800s observations of South Island glaciers close to their Little Ice Age limits were within a few decades of the start of reliable systematic observations of air temperature. The Auckland data show declining temperatures through the second half of the 1800s, which presumably reversed some warming coming out of the Little Ice Age. The low point was reached in the first 20 years of the 1900s with average annual temperatures of about 14.2°C and with a couple of years not reaching 13.5°C. My hunch is that this represented a return to the cold typical of the Little Ice Age. If I am right, that would make Little Ice Age Auckland about 1°C colder than the late 1900s (1961–1990) and 1.5°C colder than the first two decades of this century. You would have to go back to the 1940s to find a year this cold. So, Auckland was notably cooler – perhaps a bit like New Plymouth today.

Past Climate from Kauri Tree Rings

Kauri (Agathis australis) is a large endemic canopy emergent, occurring naturally and abundantly in New Zealand north of 38°S, often in relatively undisturbed environments. Ages in excess of 600 years are common and individual trees older than 1000 years are not unusual. Abundant material is also available from ninetieth and early twentieth century logging, as relic material in museums and timbers in colonial-era buildings. Large quantities of sub-fossil wood are also preserved in swamps, including much late-Holocene material (Boswijk et al 2006).

Throughout their lives, each kauri tree records its experience of life is in the accumulation of wood in the annual tree ring. Some years are good (wide rings), some typical (average), and some very dissapointing (narrow). Some of the variability is specific to the tree (especially if a storm tears off a branch!) and some is shared with neighbouring trees (local drought). But some is shared by trees across pretty much the enire growth range of kauri. If kauri are having a good year near the Bay of Islands, they probably are too south of Auckland too as far south as Kawhia and Katikati, respectively on the west and east coasts.

This large-scale common response is one of the whispers about past climate that kauri has left over thousands of years for us to disentagle (others include the chemistry of the wood, its density, and the characteristics of the cells). The fact that the "common signal" covers all of Northland and Auckland, and part of the Waikato, suggests that the whispers relate to regional-scale climate forcing. El Niño – Southern Oscillation (ENSO) has been identified as the most important contributor, with wide and narrow rings tending to be associated with El Niño and La Niña events respectively (Fowler et al 2008).

The top panel of the figure below shows a kauri master chronology derived by pooling data across multiple sites for the last seven centuries. Some decadal-scale features are apparent, indicating a few years of wider or narrow rings, but what the trees mostly have to tell us is that the growing conditions they experienced varied considerably from year to year. Note how that the scatter (year-to-year variability) of the series evolves over centuries. The two same-size shaded boxes highlight this. In the mid 1600s the scatter covers only a third of the vertical range of the box compared to all of it in the early 1900s. The lower panel shows the standard deviation (a scatter statistic) for each year, calculated for the 31 indices centred on that year.  Note how the scatter is lowest in the 1500s and 1600s, then increases over the next 300 years, but with some multi-decadal peaks and troughs.

The decrease in scatter as the Little Ice Age bit and the increase after it ended suggests that global warmth and ENSO activity are closely entwined. And because this entwining has held for the last seven centuries, then increased ENSO activity with continuing global warming seems plausible. Perhaps those kauri whispers have become shouts.

Additional Information

What causes ice ages?

Palynology basics

Meet the tree ring

Whispers of a 1000-year-old giant