Geoscience Australia is supporting the exploration and development of offshore oil and gas resources and establishment of Australia's national representative system of marine protected areas through provision of spatial information about the physical and biological character of the seabed. Central to this approach is prediction of Australia's seabed biodiversity from spatially continuous data of physical seabed properties. However, information for these properties is usually collected at sparsely-distributed discrete locations, particularly in the deep ocean. Thus, methods for generating spatially continuous information from point samples become essential tools. Such methods are, however, often data- or even variable- specific and it is difficult to select an appropriate method for any given dataset. Improving the accuracy of these physical data for biodiversity prediction, by searching for the most robust spatial interpolation methods to predict physical seabed properties, is essential to better inform resource management practises. In this regard, we conducted a simulation experiment to compare the performance of statistical and mathematical methods for spatial interpolation using samples of seabed mud content across the Australian margin. Five factors that affect the accuracy of spatial interpolation were considered: 1) region; 2) statistical method; 3) sample density; 4) searching neighbourhood; and 5) sample stratification by geomorphic provinces. Bathymetry, distance-to-coast and slope were used as secondary variables. In this study, we only report the results of the comparison of 14 methods (37 sub-methods) using samples of seabed mud content with five levels of sample density across the southwest Australian margin. The results of the simulation experiment can be applied to spatial data modelling of various physical parameters in different disciplines and have application to a variety of resource management applications for Australia's marine region.

A key component of Geoscience Australia's marine program involves developing products that contain spatial information about the seabed for Australia's marine jurisdiction. This spatial information is derived from sparse or unevenly distributed samples collected over a number of years using many different sampling methods. Spatial interpolation methods are used for generating spatially continuous information from the point samples. These methods are, however, often data- or even variable- specific and it is difficult to select an appropriate method for any given dataset. Machine learning methods, like random forest (RF) and support vector machine (SVM), have proven to be among the most accurate methods in disciplines such as bioinformatics and terrestrial ecology. However, they have been rarely previously applied to the spatial interpolation of environmental variables using point samples. To improve the accuracy of spatial interpolations to better represent the seabed environment for a variety of applications, including prediction of biodiversity and surrogacy research, Geoscience Australia has conducted two simulation experiments to compare the performance of 14 mathematical and statistical methods to predict seabed mud content for three regions (i.e., Southwest, North, Northeast) of Australia's marine jurisdiction Since 2008. This study confirms the effectiveness of applying machine learning methods to spatial data interpolation, especially in combination with OK or IDS, and also confirms the effectiveness of averaging the predictions of these combined methods. Moreover, an alternative source of methods for spatial interpolation of both marine and terrestrial environmental properties using point survey samples has been identified, with associated improvements in accuracy over commonly used methods.

Geoscience Australia is supporting the exploration and development of offshore oil and gas resources and establishment of Australia's national representative system of marine protected areas through provision of spatial information about the physical and biological character of the seabed. Central to this approach is prediction of Australia's seabed biodiversity from spatially continuous data of physical seabed properties. However, information for these properties is usually collected at sparsely-distributed discrete locations, particularly in the deep ocean. Thus, methods for generating spatially continuous information from point samples become essential tools. Such methods are, however, often data- or even variable- specific and it is difficult to select an appropriate method for any given dataset. Improving the accuracy of these physical data for biodiversity prediction, by searching for the most robust spatial interpolation methods to predict physical seabed properties, is essential to better inform resource management practises. In this regard, we conducted a simulation experiment to compare the performance of statistical and mathematical methods for spatial interpolation using samples of seabed mud content across the Australian margin. Five factors that affect the accuracy of spatial interpolation were considered: 1) region; 2) statistical method; 3) sample density; 4) searching neighbourhood; and 5) sample stratification by geomorphic provinces. Bathymetry, distance-to-coast and slope were used as secondary variables. In this study, we only report the results of the comparison of 14 methods (37 sub-methods) using samples of seabed mud content with five levels of sample density across the southwest Australian margin. The results of the simulation experiment can be applied to spatial data modelling of various physical parameters in different disciplines and have application to a variety of resource management applications for Australia's marine region.

Stations on the Australian continent receive a rich mixture of ambient seismic noise from the surrounding oceans and the numerous small earthquakes in the earthquake belts to the north in Indonesia, and east in Tonga-Kermadec, as well as more distant source zones. The noise field at a seismic station contains information about the structure in the vicinity of the site, and this can be exploited by applying an autocorrelation procedure to the continuous records. By creating stacked autocorrelograms of the ground motion at a single station, information on crust properties can be extracted in the form of a signal that includes the crustal reflection response convolved with the autocorrelation of the combined effect of source excitation and the instrument response. After applying suitable high pass filtering the reflection component can be extracted to reveal the most prominent reflectors in the lower crust, which often correspond to the reflection at the Moho. Because the reflection signal is stacked from arrivals from a wide range of slownesses, the reflection response is somewhat diffuse, but still sufficient to provide useful constraints on the local crust beneath a seismic station. Continuous vertical component records from 223 stations (permanent and temporary) across the continent have been processed using autocorrelograms of running windows 6 hours long with subsequent stacking. A distinctive pulse with a time offset between 8 and 30 s from zero is found in the autocorrelation results, with frequency content between 1.5 and 4 Hz suggesting P-wave multiples trapped in the crust. Synthetic modelling, with control of multiple phases, shows that a local Ppmp phase can be recovered with the autocorrelation approach. This approach can be used for crustal property extraction using just vertical component records, and effective results can be obtained with temporary deployments of just a few months.

Geoscience Australia has developed a number of open source risk models to estimate hazard, damage or financial loss to residential communities from natural hazards and is used to underpin disaster risk reduction activities. Two of these models will be discussed here: the Earthquake Risk Model (EQRM) and a hydrodynamic model call ANUGA, developed in collaboratoin with the ANU. Both models have been developed in Python using scientific and GIS packages such as Shapely, Numeric and SciPy. This presentation will outline key lessons learnt in developing scientific software in Python. Methods of maintaining and accessing code quality will be discussed (1) what makes a good unit test (2) how defects in the code were discovered quickly by being able to visualise the output data; and (3) how characterisation tests, which describe the actual behaviour of a system, are useful for finding unintended system changes. The challenges involved in optimising and parallelising Python code will also be presented. This is particularly important in scientific simulations as they use considerable computational resources and involve large data sets. This will be focus on: profiling; NumPyl using C code; and parallelisation of applications to run on clusters. Reduction of memory use by using a class to represent a group of items instead of a single item will also be discussed.

One of the important inputs to a probabilistic seismic hazard assessment is the expected rate at which earthquakes within the study region. The rate of earthquakes is a function of the rate at which the crust is being deformed, mostly by tectonic stresses. This paper will present two contrasting methods of estimating the strain rate at the scale of the Australian continent. The first method is based on statistically analysing the recently updated national earthquake catalogue, while the second uses a geodynamic model of the Australian plate and the forces that act upon it. For the first method, we show a couple of examples of the strain rates predicted across Australia using different statistical techniques. However no matter what method is used, the measurable seismic strain rates are typically in the range of 10-16s-1 to around 10-18s-1 depending on location. By contrast, the geodynamic model predicts a much more uniform strain rate of around 10-17s-1 across the continent. The level of uniformity of the true distribution of long term strain rate in Australia is likely to be somewhere between these two extremes. Neither estimate is consistent with the Australian plate being completely rigid and free from internal deformation (i.e. a strain rate of exactly zero). This paper will also give an overview of how this kind of work affects the national earthquake hazard map and how future high precision geodetic estimates of strain rate should help to reduce the uncertainty in this important parameter for probabilistic seismic hazard assessments.

The major tsunamis of the last few years have dramatically raised awareness of the possibility of potentially damaging tsunami reaching the shores of Australia and to the other countries in the region. Here we present three probabilistic hazard assessments for tsunami generated by megathrust earthquakes in the Indian, Pacific and southern Atlantic Oceans. One of the assessments was done for Australia, one covered the island nations in the Southwest Pacific and one was for all the countries surrounding the Indian Ocean Basin

This study tested the performance of 16 species models in predicting the distribution of sponges on the Australian continental shelf using a common set of environmental variables. The models included traditional regression and more recently developed machine learning models. The results demonstrate that the spatial distributions of sponge as a species group can be successfully predicted. A new method of deriving pseudo-absence data (weighted pseudo-absence) was compared with random pseudo-absence data - the new data were able to improve modelling performance for all the models both in terms of statistics (~10%) and in the predicted spatial distributions. Overall, machine learning models achieved the best prediction performance. The direct variable of bottom water temperature and the resource variables that describe bottom water nutrient status were found to be useful surrogates for sponge distribution at the broad regional scale. This study demonstrates that predictive modelling techniques can enhance our understanding of processes that influence spatial patterns of benthic marine biodiversity. Ecological Informatics

Obtaining reliable predictions of the subsurface will provide a critical advantage for explorers seeking mineral deposits at depth and beneath cover. A common approach in achieving this goal is to use deterministic property-based inversion of potential field data to predict a 3D subsurface distribution of physical properties that explain measured gravity or magnetic data. Including all prior geological knowledge as constraints on the inversion ensures that the recovered predictions are consistent with both the geophysical data and the geological knowledge. Physical property models recovered from such geologically-constrained inversion of gravity and magnetic data provide a more reliable prediction of the subsurface than can be obtained without constraints. The non-uniqueness of inversions of potential field data mandates careful and consistent parameterization of the problem to ensure realistic solutions.

The tragic events of the Indian Ocean tsunami on 26 December 2004 highlighted the need for reliable and effective alert and response sysems for tsunami threat to Australian communities. Geoscience Australia has established collaborative partnerships with state and federal emergency management agencies to support better preparedness and to improve community awareness of tsunami risks.