Two shallow sub-surface CO2 controlled release experiments were conducted at the Ginninderra test site during 2012. The theme of the first experiment was CO2 detection in the soil and surface emissions quantification. The theme for the second experiment was investigating sub-surface migration and broad scale detection technologies. Our objective overall is to design cheaper monitoring technologies to evaluate leakage and environmental impact in the shallow sub-surface. Over 10 different monitoring techniques were evaluated at the site against a known CO2 release. These included soil gas, soil CO2 flux, soil analysis, eddy covariance, atmospheric tomography, noble gas tracers, ground penetrating radar, electromagnetic surveys, airborne hyperspectral, in-field phenotyping (thermal, hyperspectral and 3D imaging), and microbial soil genomics. Technique highlights and an assessment of the implications for large scale storage are presented in the following corresponding talks.

Geoscience Australia and the CO2CRC operate a controlled release facility in Canberra, Australia, designed for simulating subsurface emissions of CO2 by injecting gas into a horizontal well. Three controlled release experiments were conducted at this site during 2012-2013, over 7-9 week periods, to assess and develop near-surface monitoring technologies for application to carbon dioxide geological storage sites (Feitz et al., 2014). A key well-established technique for characterizing surface CO2 emission sources from controlled release sites or natural CO2 seeps is soil flux surveys. The technique is often considered as the benchmark technique for characterizing a site's emissions or as a baseline for comparing other measurement techniques, but has received less attention with regards to its absolute performance. The extensive soil gas surveys undertaken during Release 1 (Feb-May 2012) and Release 3 (Oct-Dec 2013) are the subject of this paper. Several studies have highlighted factors which can have an effect on soil flux measurements, including meteorological influences such as air pressure and wind speed, which can increase or suppress soil fluxes (Rinaldi et al. 2012). Work at the Canberra controlled release site has highlighted the influence groundwater has on the spatial distribution of fluxes.). In addition, there are several different methods available for inverting soil flux measurements to obtain the emission rate of a surveyed area. These range in complexity from planar averaging to geostatistical methods such as sequential Gaussian simulation (Lewicki et al. 2005). Each inversion technique relies on its own subset of assumptions or limitations, which can also impact the end emissions estimate. Thus deriving a realistic estimate of the total emission rate will depend on both environmental forcing as well as the applied inversion method. An in-house method for soil flux interpolation has been developed and is presented. A cubic interpolated surface is generated from all the measurement points (Figure 1), from which a background linear interpolated surface is subtracted off, leaving the net leakage flux. The background surface is prepared by identifying all background points matching a certain criteria (for this release experiment distance from release well was used) and interpolating only over those points. In these experiments, soil flux surveys were collected on a predefined grid, using an irregular sampling pattern with higher density of samples nearer to the leak hotspots to provide higher spatial resolution in the regions where flux changes most rapidly (Figure 2). The same release rate of 144 kgCO2/day was used for both experiments. It was observed that the surface flux distribution shifts markedly between experiments, most likely a function of seasonal differences (2012 was wet; 2013 was dry) and resulting differences in groundwater depth, soil saturation and the extent of the vadose zone.. The depth to the groundwater measured at monitoring wells in proximity to the release well was 0.85-1.2 m during the 2012 (wet) release whereas it ranged from 1.9-2.3 m during the 2013 (dry) release experiment. The horizontal well is located 2.0 m below the ground surface. This paper explores the performance of soil flux surveys for providing an accurate estimate of the release rate, using a series of soil flux surveys collected across both release experiments. Emission estimates are generated by applying several common inversion methods, which are then compared to the known release rate of CO2. An evaluation as to the relative suitability of different inversion methods will be provided based on their performance. Deviations from the measured release rate are also explored with respect to survey design, meteorological and groundwater factors, which can lead and inform the future deployment of soil flux surveys in a monitoring and verification program.

To provide the solar power industry with a data resource to allow them to assess the economic potential of a site for a solar power plant. Specifically under the Solar Flagship program.

This database contains the monthly mean and montly long term mean fields from the NCEP/NCAR Reanalysis 1960-2000. Files contain the following data: airsfc.mon.mean.nc - surface air temperature land.nc - land/sea mask slp.mon.mean.nc - sea level pressure sst.mnmean.nc - sea surface temperature (see SST_README for more details) uwnd.mon.mean.nc - U (eastward) component of wind vwnd.mon.mean.nc - V (northward) component of wind shum.mon.mean.nc - specific humidity (this file does not contain all vertical levels, unlike the other 3-d variables) For all the above, files with 'ltm' instead of 'mean' contain the long-term monthly mean data. Data were downloaded on 25/11/2009 from the Earth System Reseach Laboratory (ESRL) Physical Sciences Division (PSD) website. (http://www.esrl.noaa.gov/psd/data/gridded/reanalysis/)

Results from the first pass application of the tomography technique using low accuracy sensors is presented and limitations of the sensors and technique discussed. BUll. Seismol. Soc. AM.

Following the drilling of a shallow CO2 reservoir at the Qinghai research site, west of Haidong, China, it was discovered that CO2 was continuously leaking from the wellbore due to well-failure. The site has become a useful facility in China for studying CO2 leakage and monitoring technologies for application to geological storage sites of CO2. During an eight day period in 2014, soil gas and soil flux surveys were conducted to characterise the distribution, magnitude and likely source of the leaking CO2. Two different sampling patterns were utilised during soil flux surveys. A regular sampling grid was used to spatially map out the two high flux zones which were located 20-50 m away from the wellhead. An irregular sampling grid with higher sampling density in the high flux zones, allowed for more accurate mapping of the leak distribution and estimation of total field emission rate using cubic interpolation. The total CO2 emission rate for the site was estimated at 649-1015 kgCO2/d and there appeared to be some degree of spatial correlation between observed CO2 fluxes and elevated surface H2O fluxes. Sixteen soil gas wells were installed across the field to test the real-time application of Romanak et al.'s (2012) process-based approach for soil gas measurements (using ratios of major soil gas components to identify the CO2 source) using a portable multi-gas analyser. Results clearly identified CO2 as being derived from one exogenous source, and are consistent with gas samples collected for laboratory analysis. Carbon-13 isotopes in the centre of each leak zone (-0.21 and -0.22 ) indicate the underlying CO2 is likely sourced from the thermal decomposition of marine carbonates. Surface soil mineralisation (predominantly calcite) is used to infer prior distribution of the CO2 hotspots and as a consequence highlighted plume migration of 20 m in 11 years. Detachment of the plume from the wellbore at the Qinghai research site markedly increases the area that needs surveying at sufficient density to detect a leak. This challenges the role of soil flux and soil gas in a CCS monitoring and verification program for leak detection, whereas these techniques may best be applied for characterising source and emission rate of a CO2 leak.

Tropical cyclones are the most common disaster in the Pacific, and among the most destructive. In December 2012, Cyclone Evan caused over US$200 million damage in Samoa, nearly 30 percent of Samoan GDP. Niue suffered losses of US$85 million following Cyclone Heta in 2004-over five times its GDP. As recently as January 2014, Cyclone Ian caused significant damage throughout Tonga, resulting in the first payout of the Pacific Catastrophe Risk Insurance Pilot system operated by the World Bank (2014). According to the Intergovernmental Panel on Climate Change (IPCC), intense tropical cyclone activity in the Pacific basin will likely increase in the future (IPCC 2013). But such general statements about global tropical cyclone activity provide little guidance on how impacts may change locally or even regionally, and thus do little to help communities and nations prepare appropriate adaptation measures. This study assesses climate change in terms of impact on the human population and its assets, expressed in terms of financial loss. An impact focus is relevant to adaptation because changes in hazard do not necessarily result in a proportional change in impact. This is because impacts are driven by exposure and vulnerability as well as by hazard. For example, a small shift in hazard in a densely populated area may have more significant consequences than a bigger change in an unpopulated area. Analogously, a dense population that has a low vulnerability to a particular hazard might not need to adapt significantly to a change in hazard. Even in regions with high tropical cyclone risk and correspondingly stringent building codes, such as the state of Florida, a modest 1 percent increase in wind speeds can result in a 5 percent to 10 percent increase in loss to residential property. Quantifying the change impact thus supports evidence-based decision making on adaptation to future climate risk.

The world's first satellite-derived mineral maps of a continent, namely Australia, are now publicly available as digital, web-accessible products. The value of this spatially comprehensive mineral information is readily being captured by explorers at terrane to prospect scales. However, potentially even greater benefits can ensue for environmental applications, especially for the Earth's extensive drylands which generate nearly 50% of the world's agricultural production but are most at risk to climate change and poor land management. Here we show how these satellite mineral maps can be used to: characterise soil types; define the extent of deserts; fingerprint sources of dust; measure the REDOX of iron minerals as a potential marine input; and monitor the process of desertification. We propose a 'Mineral Desertification Index' that can be applied to all Earth's drylands where the agriculturally productive clay mineral component is being lost by erosion. Mineral information is fundamental to understanding geology and is important for resource applications1. Minerals are also a fundamental component of soils2 as well as dust eroded from the land surface, which can potentially impact on human health3, the marine environment4 and climate5. Importantly, minerals are well exposed in the world's 'drylands', which account for nearly 50% of Earth's land area6. Here, vegetation cover is sparse to non-existent as a result of low rainfall (P) and high evaporation (E) rates (P/E<0.65). However, drylands support 50% of the world's livestock production and almost half of all cultivated systems6. In Australia, drylands cover 85% of the continent and account for 50% of its beef, 80% of its sheep and 93% of its grain production7. Like other parts of the world, Australia is facing serious desertification of its drylands6. Wind, overgrazing and overstocking are major factors in the desertification process8. That is, the agriculturally productive clay-size fraction of soils (often includes organic carbon) is lost largely through wind erosion, which is acerbated by the loss of any vegetative groundcover (typically dry plant materials). Once clay (and carbon) loss begins, then the related break down of the soil structure and loss of its water holding capacity increases the rate of the degeneration process with the final end products being either exposed rock or quartz sands that often concentrate in deserts.

Modelling tropical cyclone Yasi using TCRM

We have developed an autonomous CO2 monitoring station, based around the Vaisala GMP343 sensor. The station is powered by a solar panel and incorporates a data logger and a directional antenna for line-of-sight wireless communication with a base station. The base station communicates via the Telstra mobile phone data network. The concept of atmospheric tomography was tested at the Ginninderra site and proved very successful as a method of locating and quantifying a spatially small release of CO2. In this case the sensors were separated from the source by 40 m. The opportunity to test the method over a larger distance arose during the controlled release of Buttress gas during the stage 2B experiment at the Otway site. Gas was released at 8 tpd during daylight hours, and an approximate ring of 8 monitoring stations was deployed around the release point; the ring was about 800 m in diameter. Gas was released on 12 occasions, chosen to match wind directions that would carry the plume in the direction of one of the sensors. The dataset was too limited to carry out the full-scale Bayesian inversion that was demonstrated in the Ginninderra test (which lasted two months) but a simple inversion was possible. This located the source of the release correctly to within 20 m. The test demonstrated that inexpensive sensors could achieve enough stability and sensitivity to work (in this particular application) at the few ppm level. Moreover quite simple dispersion models could be used to predict plume geometry up to 500 m from the release. Overall the experiment indicates the basis of an inexpensive method for remotely monitoring areas of around a km2 for spatially small leakages.