Bruce ZRV. 2022. Version 1 of the New Zealand landslide fatalities database, 1760–2020. Lower Hutt (NZ): GNS Science. 114 p. (GNS Science report; 2021/38). doi:10.21420/SW43-TJ28.
Abstract
A detailed survey of online newspaper archives from 1840 to 1950 has expanded the known catalogue of historical landslide fatalities in New Zealand from 1760 to the present day. Searches using a wider array of keywords have found more minor events with low- and single-fatality counts and expanded the trend of anthropogenic fatalities related to mining, rural land-use and infrastructure development during the 19th and early 20th century. With the inclusion of this new survey data, we estimate that 1508 fatalities have occurred from 918 landslide and slope instability events in New Zealand since 1760. Of these, 743 fatalities are attributed to natural or undetermined landslide causes, while 765 fatalities are derived from anthropogenic cut-slope activity, including mining, land clearance, infrastructure development and resource extraction activities. Natural landslide fatalities are dominated by four large hydrothermally induced slope failure events (three hydrothermally weakened landslide / debris flow events at Little Waihi and one hydrothermally weakened lakeshore landslide at Ohinemutu, together comprising 231 fatalities) and two major lahars (162 fatalities from the White Island and Tangiwai events, respectively), as well as rainfall-induced landslides (157), snow avalanche (62 fatalities), earthquake-induced landslides (63 fatalities), landslides without any recorded direct contributing cause other than natural slope weathering processes (59 fatalities) and a few fatalities from other minor causes, such as coastal cliff collapse (8 fatalities). Anthropogenic fatalities are dominated by excavations and cut-slope activities (379), induced slope failures during opencast and alluvial mining (314) and slope failures within quarries (72). Although not strictly landslides, during the course of data compilation, a comparable dataset of 314 fatalities from 270 underground rockfall and collapse events was compiled from the same search terms, comprising mine tunnel failures (219), service tunnel failures (58) and mine-shaft and stope failures (37). From 1860, annual natural fatality rates per capita are high, beginning at 15/100,000 with the severe winter fatalities of 1863, descending rapidly through the remaining decade to below 1/100,000 by 1865, reducing further after 1900 to less than 0.5/100,000 in most years, excepting years with large multiple-fatality events. Anthropogenic annual fatality rates peak at around 4–5/100,000 in the 1870–1880s, which coincides with the economic demographic influx, particularly associated with the gold rushes. These fatalities drop steeply after 1880 to 2–3/100,000 and then descend again between 1900 and 1920 until, by 1940–1950, rates reach less than 1/100,000. Despite these reductions, anthropogenic landslide fatalities remain higher than natural fatalities in most years, excepting those that have multiple-fatality natural events. The falling rate of anthropogenic- and workplace-induced landslide fatalities has not been studied in detail but is likely due to improved working standards, mechanisation of slope cutting and excavation methods and urbanising demographic trends from 1900 away from high-risk rural and resource-extraction activities. This database (version 1) is a work in progress, and further work will now focus on completing the location of all incidents from 1900 to 2020, together with refinement of referencing documents for each event and collation of demographic data and other metadata concerning fatality victims. (The author)
