Data gathered in the field during the sample collection phase of the National Geochemical Survey of Australia (NGSA) has been used to compile the Preliminary Soil pH map of Australia. The map, which was completed in late 2009, offers a first-order estimate of where acid or alkaline soil conditions are likely to be expected. It provides fundamental datasets that can be used for mineral exploration and resource potential evaluation, environmental monitoring, landuse policy development, and geomedical studies into the health of humans, animals and plants.

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.

Explaining spatial variation and habitat complexity of benthic habitats from underwater video through the use of maps. Different methodologies currently used to process and analyse percent cover of benthic organisms from underwater video will be addressed and reviewed.

OzCoasts is a web-based database and information system managed by Geoscience Australia that draws together a diverse range of data and information on Australia's coasts and estuaries. Maps, images, reports and data can be downloaded and there are tools to assist with coastal science, monitoring, management and policy. A Tropical Rivers module is the newest major feature of the website and was developed in partnership with the Griffith University node of the Tropical Rivers and Coastal Knowledge (TRaCK) consortium and Boab Interactive. The module contains the Australian Riverine Landscape Classifier (AURICL) and provides links to the TRaCK Digital Atlas. AURICL will assist researchers and policy makers make better decisions about riverine landscapes. It is a dynamic and flexible system (i.e. can be updated as new data layers become available) for classifying and comparing tropical catchments and their rivers based on the similarity, or dissimilarity, of a wide range of parameters. Importantly, AURICL provides researchers with: (i) data-sets to link stream segments from the National Catchment Boundaries database to estuary point locations for north Australia; (ii) a collection of riverine attribute data that sum their upstream contributions to an estuary; and (iii) an amalgamation of inputs for estuaries with multiple contributing streams. To date, researchers have only had access to very general data on the catchments that feed estuaries (e.g. catchment areas). The Mangroves and Coastal Saltmarsh of Victoria: Distribution, Condition, Threats and Management report is new to the Habitat Mapping module, and constitutes the first State-wide assessment of Victoria's coastal wetlands. The 514 page report, led by Prof. Paul Boon (Victoria University), examines the diversity of wetland types and plant communities along the Victorian coast and provides analysis of the ecological condition and major threats to coastal wetlands in Victoria. OzCoasts will also soon deliver the Coastal Eutrophication Risk Assessment Tool (CERAT) for the NSW Office of Environment and Heritage. CERAT will help identify and prioritise land use planning decisions to protect and preserve the health of NSW estuaries. A partnership between OzCoasts and the coastal facility of the TERN (Terrestrial Ecosystem Research Network) is also currently under negotiation.

pH is one of the more fundamental soil properties governing nutrient availability, metal mobility, elemental toxicity, microbial activity and plant growth. The field pH of topsoil (0-10 cm depth) and subsoil (~60-80 cm depth) was measured on floodplain soils collected near the outlet of 1186 catchments covering over 6 M km2 or ~80% of Australia. Field pH duplicate data, obtained at 124 randomly selected sites, indicates a precision of 0.5 pH unit (or 7%) and mapped pH patterns are consistent and meaningful. The median topsoil pH is 6.5, while the subsoil pH has a median pH of 7 but is strongly bimodal (6-6.5 and 8-8.5). In most cases (64%) the topsoil and subsoil pH values are similar, whilst, among the sites exhibiting a pH contrast, those with more acidic topsoils are more common (28%) than those with more alkaline topsoils (7%). The distribution of soil pH at the national scale indicates the strong controls exerted by precipitation and ensuing leaching (e.g., low pH along the coastal fringe, high pH in the dry centre), aridity (e.g., high pH where calcrete is common in the regolith), vegetation (e.g., low pH reflecting abundant soil organic matter), and subsurface lithology (e.g., high pH over limestone bedrock). The new data, together with existing soil pH datasets, can support regional-scale decision-making relating to agricultural, environmental, infrastructural and mineral exploration decisions.

Spatial interpolation methods for generating spatially continuous data from point locations of environmental variables are essential for ecosystem management and biodiversity conservation. They can be classified into three groups (Li and Heap 2008): 1) non-geostatistical methods (e.g., inverse distance weighting), 2) geostatistical methods (e.g., ordinary kriging: OK) and 3) combined methods (e.g. regression kriging). Machine learning methods, like random forest (RF) and support vector machine (SVM), have shown their robustness in data mining fields. However, they have not been applied to the spatial prediction of environmental variables (Li and Heap 2008). Given that none of the existing spatial interpolation methods is superior to the others, several questions remain, namely: 1) could machine learning methods be applied to the spatial prediction of environmental variables; 2) how reliable are their predictions; 3) could the combination of these methods with the existing interpolation methods improve the predictions; and 4) what contributes to their accuracy? To address these questions, we conducted a simulation experiment to compare the predictions of several methods for mud content on the southwest Australian marine margin. In this study, we discuss results derived from this experiment, visually examine the spatial predictions, and compare the results with the findings in previous publications. The outcomes of this study have both practical and theoretical importance and can be applied to the spatial prediction of a range of environmental variables for informed decision making in environmental management. This study reveals a new direction in and provides alternative methods for spatial interpolation in environmental sciences.

Catchment outlet sediments (0-10 cm depth, sieved to <2 mm) collected at a very low density over most of the Australian continent have been analysed using the Mobile Metal Ion (MMI®) partial extraction technique. Of the 54 elements analysed, eight are generally regarded as essential nutrients for plant growth: Ca, Cu, Fe, K, Mg, Mn, P and Zn. For these, 'bioavailability', defined here as the ratio of the partial digest concentration to the total concentration, has been investigated. This estimation of 'bioavailability' gives results comparable with standard agricultural measurements. Average 'bioavailability' ranges from 15.0% for Ca to 0.1% for Fe. Smoothed (kriged) colour contour maps for continental Australia have been produced for these eight nutrients and interpreted in terms of lithology (e.g., presence of carbonates in the MMI® Ca map), mineralization (e.g., well known and possibly less known mineral districts in the Cu, P and Zn maps), environmental processes (e.g., salinity in K map, weathering and acid generation in Fe map) and agricultural practices (e.g., application of fertilizers in P and Zn maps). This first application of a partial extraction technique at the scale of a continent has yielded meaningful, coherent and interpretable results.

Inland sulfidic soils have recently formed throughout wetlands of the Murray River floodplain associated with increased salinity and river regulation (Lamontagne et al., 2006). Sulfides have the potential to cause widespread environmental degradation both within sulfidic soils and down stream depending on the amount of carbonate available to neutralise acidity (Dent, 1986). Sulfate reduction is facilitated by organic carbon decomposition, however, little is known about the sources of sedimentary organic carbon and carbonate or the process of sulfide accumulation within inland sulfidic wetlands. This investigation uses stable isotopes from organic carbon (13C and 15N), inorganic sulfur (34S) and carbonate (13C and 18O) to elucidate the sources and cycling of sulfur and carbon within sulfidic soils of the Loveday Disposal Basin.

Geoscience Australia and the CO2CRC have constructed a greenhouse gas controlled release facility at an experimental agricultural station maintained by CSIRO Plant Industry at Ginninderra, Canberra. The facility is designed to simulate surface emissions of CO2 (and other greenhouse gases) from the soil into the atmosphere. CO2 is injected into the soil is via a 120m long slotted HDPE pipe installed horizontally 2m underground. This is fitted with a straddle packer system to partition the well into six CO2 injection chambers with each chamber connected to its own CO2 injection line. CO2 was injected into 5 of the chambers during the first sub-surface release experiment (March - May 2012) and the total daily injection rate was 100 kg/d. A krypton tracer was injected into one of the 5 chambers at a rate of 10 mL/min. Monitoring methods trialled at the site include eddy covariance, atmospheric tomography using a wireless networked array of solar powered CO2 stations, soil flux, soil gas, frequency-domain electromagnetics (FDEM), soil community DNA analysis, and krypton tracer studies (soil gas and air). A summary of the findings will be presented. Paper presented at the 2012 CO2CRC Research Symposium, Sunshine Beach, 27-29 November 2012.

Random forest (RF) is one of the top performed methods in predictive modelling. Because of its high predictive accuracy, we introduced it into spatial statistics by combining it with the existing spatial interpolation methods, resulting a few hybrid methods and improved prediction accuracy when applied to marine environmental datasets (Li et al., 2011). The superior performance of these hybrid methods was partially attributed to the features of RF, one component of the hybrids. One of these features inherited from its trees is to be able to deal with irrelevant inputs. It is also argued that the performance of RF is not much influenced by parameter choices, so the hybrids presumably also share this feature. However, these assumptions have not been tested for the spatial interpolation of environmental variables. In this study, we experimentally examined these assumptions using seabed sand and gravel content datasets on the northwest Australian marine margin. Four sets of input variables and two choices of 'number of variables randomly sampled as candidates at each split' were tested in terms of predictive accuracy. The input variables vary from six predictors only to combinations of these predictors and derived variables including the second and third orders and/or possible two-way interactions of these six predictors. However, these derived predictors were regarded as redundant and irrelevant variables because they are correlated with these six predictors and because RF can do implicit variable selection and can model complex interactions among predictors. The results derived from this experiment are analysed, discussed and compared with previous findings. The outcomes of this study have both practical and theoretical importance for predicting environmental variables.