The Storm in the Ecosystems Teacup

A one hundred year event......  by John Ogden


On the night of June 10th, 2014, a massive storm hit the northern part of Aotea, doing enormous damage to roads, tracks, bridges, property and the natural environment. According to newspaper reports more than 300mm of rain fell in 6 hours, accompanied by strong winds. There were many slips on the steep slopes of the Conservation Park, and masses of boulders, trees and debris were channelled down water-courses, which became rag-ing torrents. Once the damage to infrastructure and habitations had been assessed, it became apparent that effects on the natural environment were equally dramatic. This was unquestionably a “100-year event”; the 100 year old Kaiarara dam was obliterated – nothing similar has ever been recorded on the Barrier.

Well, not recorded by human inhabitants; many natural process do record such events and forest ecologists have ways to read these records, allowing them to reconstruct past storm events, much as geologists seek ways to put dates on past earthquakes. One of the keys to this is that trees produce annual growth rings, so that the age of surviving trees can be found, and pulses of regrowth and regeneration can be dated. Kauri trees (Agathis australis), and others, live for up to a thousand years, so clearly many trees in New Zealand forests experience several such events during their lifetimes, and often record them in their growth rings: typically a narrow ring during the year of damage, followed by wider rings as the survivor takes advantage of the newly opened canopy and reduced competition from neighbouring trees.

Caroline Ogden and Jim Kerr crossing the big slip on the way to the Mt. Heale hut. Slip created 10th April, 2014; Photo 11 Feb. 2015.

When storm damage is severe and widespread, such as after Cyclones Bernie and Bola in the 1980’s, extensive forest blow-down, slips and erosion occur over wide areas. Numerous canopy gaps may be created by tree falls even where the whole forest is not destroyed. The result is that many well-lit situations are created, in which forest seedlings can become established, initiating a new cohort of more or less synchronous regeneration. Often these events can be recognised in the past if many trees in an area can be shown to have established within a short period. This is particularly so for certain ‘light demanding’ , long-lived species such as kaikawaka (Libocedrus bidwillii) or kauri. If the tree-ring signature of survivors coincides with widespread recruitment of new seedlings of light-demanding species, and this is also backed up by other evidence, such as a pulse of sedimentation in lakes or estuaries, pre-historical events can be dated quite precisely. The pulse of sedimentation in the Fitzroy Harbour from the 2014 event is sure to be preserved for thousands of years. Most kauri rickers on Great Barrier, and the mature kanuka (Kunzea ericoides) stands, date from the period of European logging, which mimicked a natural disaster.

These sorts of studies have clearly demonstrated that most natural forests in New Zealand are composed of tree cohorts initiated following storm events, or other major disturbances. For example, as in 1983 Willie Shaw 1 demonstrated that tropical cyclones causing severe forest damage somewhere in the North Island, occur about every ten years, and are likely to be a major factor in moulding stand structure and composition.

In a study made on Mt. Hauhungatahi in the Tongariro National Park 2 I demonstrated that in a four-day period in March 1988 Cyclone Bola made as many gaps as had been created by ‘normal back-ground’ tree mortality in the previous ten years.

The long-lasting imprint of major disturbances is illustrated by the kaikawaka stands on Mt Hauhungatahi. These timberline stands owe their origin to the major disturbance caused by the eruption from Lake Taupo c.1700 years ago; the current trees represent the third (and a few of the second) generation since that event 3 . Meanwhile many understory tree species and tree-ferns have colonised gaps created by the demise of each cohort, and kamahi (Weinmannia racemosa) has invaded from below. This whole changing ecosystem composition and structure was initiated by a huge disturbance in about 230 AD.

Likewise in Westland, over 70% of forest stands studied were initiated during two periods, 250-350, and 550-600 years ago, coinciding with two of the last three major movements of the Alpine Fault. There seems no doubt that major slips, erosion and flooding at those times created conditions for forest destruction and regeneration.

We know that there have been shifts in climate in the past, with periods of increased and decreased storminess. Also, storms are not the only factors killing forest: fires, droughts, floods, insect outbreaks and diseases all play their roles – and all of these tend to be episodic. Moreover, one form of disturbance can render a forest more vulnerable to another, and a relatively small ‘trigger’ can thus initiate massive canopy die-back.

The clear conclusion is that what to us are terrible events, are also creating heterogeneity and initiating regeneration for many forest species. Although such events are almost annual, in any one place they ‘return’ on a timescale of three or four centuries, which agrees with the longevity of our canopy trees. The forest ecosystems of New Zealand are well adapted to cope with such events, they have a much longer memory than mere mortals.


1 Shaw, W. B. (1983). Tropical cyclones: determinants of pattern and structure in New Zealand’s indigenous forests. Pacific Science 37 (4): 405-414.
2 Ogden, J. et al. (1993). Episodic mortality, forest decline and diversity in a dynamic landscape: Tongariro National Park, New Zealand. In: Huettl/Mueller Dombois (Eds.) “Forest Decline in the Atlantic and Pacific Regions”. Springer-Verlag Berlin Heidelberg. Pp. 261-274.
3 Ogden, J. et. al. (1991). Forest gap formation and closure along an altitudinal gradient in Tongariro National Park, New Zealand. Journal of Vegetation Science 2: 165-172. See also: Ogden, J. et al. (2005). Long-term dynamics of the long-lived conifer Libocedrus bidwillii after a volcanic eruption 2000 years ago. Journal of Vegetation Science 16: 321-330.
4 Wells, A. et al. (1998). Evidence of widespread, synchronous, disturbance-initiated forest establishment in Westland, New Zealand. Journal of the Royal Society of New Zealand 28 (2): 333-345.


Environmental News Issue 34 2015