Natural catastrophes are increasing worldwide. They are becoming more frequent but also more severe and impactful on our built environment leading to extensive damage and losses. Earthquake events account for the smallest part of natural events; nevertheless seismic damage led to the most fatalities and significant losses over the period 1981-2016 (Munich Re). Damage prediction is helpful for emergency management and the development of earthquake risk mitigation projects. Recent design efforts focused on the application of performance-based design engineering where damage estimation methodologies use fragility and vulnerability functions. However, the approach does not explicitly specify the essential criteria leading to economic losses. There is thus a need for an improved methodology that finds the critical building elements related to significant losses. The here presented methodology uses data science techniques to identify key building features that contribute to the bulk of losses. It uses empirical data collected on site during earthquake reconnaissance mission to train a machine learning model that can further be used for the estimation of building damage post-earthquake. The first model is developed for Christchurch. Empirical building damage data from the 2010-2011 earthquake events is analysed to find the building features that contributed the most to damage. Once processed, the data is used to train a machine-learning model that can be applied to estimate losses in future earthquake events.
Numerous studies have shown that urban soils can contain elevated concentrations of heavy metals (HMs). Christchurch, New Zealand, is a relatively young city (150 years old) with a population of 390,000. Most soils in Christchurch are sub-urban, with food production in residential gardens a popular activity. Earthquakes in 2010 and 2011 have resulted in the re-zoning of 630 ha of Christchurch, with suggestions that some of this land could be used for community gardens. We aimed to determine the HM concentrations in a selection of suburban gardens in Christchurch as well as in soils identified as being at risk of HM contamination due to hazardous former land uses or nearby activities. Heavy metal concentrations in suburban Christchurch garden soils were higher than normal background soil concentrations. Some 46% of the urban garden samples had Pb concentrations higher than the residential land use national standard of 210 mg kg⁻¹, with the most contaminated soil containing 2615 mg kg⁻¹ Pb. Concentrations of As and Zn exceeded the residential land use national standards (20 mg kg⁻¹ As and 400 mg kg⁻¹ Zn) in 20% of the soils. Older neighbourhoods had significantly higher soil HM concentrations than younger neighbourhoods. Neighbourhoods developed pre-1950s had a mean Pb concentration of 282 mg kg⁻¹ in their garden soils. Soil HM concentrations should be key criteria when determining the future land use of former residential areas that have been demolished because of the earthquakes in 2010 and 2011. Redeveloping these areas as parklands or forests would result in less human HM exposure than agriculture or community gardens where food is produced and bare soil is exposed.
Our poster will present on-going QuakeCoRE-founded work on strong motion seismology for Dunedin-Mosgiel area, focusing on ground motion simulations for Dunedin Central Business District (CBD). Source modelling and ground motion simulations are being carried out using the SCEC (Southern California Earthquakes Center) Broad Band simulation Platform (BBP). The platform computes broadband (0-10 Hz) seismograms for earthquakes and was first implemented at the University of Otago in 2016. As large earthquakes has not been experienced in Dunedin in the time of period of instrumental recording, user-specified scenario simulations are of great value. The Akatore Fault, the most active fault in Otago and closest major fault to Dunedin, is the source focused on in the present study. Simulations for various Akatore Fault source scenarios are run and presented. Path and site effects are key components considered in the simulation process. A 1D shear wave velocity profile is required by SCEC BBP, and this is being generated to represent the Akatore-to-CBD path and site within the BBP. A 3D shear velocity model, with high resolution within Dunedin CBD, is being developed in parallel with this study (see Sangster et al. poster). This model will be the basis for developing a 3D shear wave velocity model for greater Dunedin-Mosgiel area for future ground motion simulations, using Canterbury software (currently under development).
