Australian Geoscience Data Cube Research Articles

Cloud Data and Computing Services Allow Regional Environmental Assessment: A Case Study of Macquarie-Castlereagh Basin, Australia

Large amounts of data at various temporal and spatial scales require terabyte (TB) level storage and computation, both of which are not easy for researchers to access. Cloud data and computing services provide another solution to store, process, share and explore environmental data with low costs, stronger computation capacity and easy access. The purpose of this paper is to examine the benefits and challenges of using freely available satellite data products from Australian Geoscience DataCube and Google Earth Engine (GEE) online data with time series for integrative environmental analysis of the Macquarie-Castlereagh Basin in the last 15 years as a case study. Results revealed that the cloud platform simplifies the procedure of traditional catalog data processing and analysis. The integrated analysis based on the cloud computing and traditional methods represents a great potential as a low-cost, efficient and user-friendly method for global and regional environmental study. The user can save considerable time and cost on data integration. The research shows that there is an excellent promise in performing regional environmental analysis by using a cloud platform. The incoming challenge of the cloud platform is that not all kinds of data are available on the cloud platform. How data are integrated into a single platform while protecting or recognizing the data property, or how one portal can be used to explore data archived on different platforms represent considerable challenges.

A novel framework for the use of remote sensing for monitoring catchments at continental scales

Historical ecology can teach us valuable lessons on the processes and drivers of environmental change that can inform future monitoring priorities and management strategies. Environmental data to study environmental history, however, is often absent or of low quality. Even when studying changes occurring during the last few decades, monitoring efforts are scarce due to logistical and cost limitations, leaving large areas unassessed. The aim of this study is to evaluate the use of estuarine water colour as an indicator of historical environmental change in catchments. Water colour change was assessed in estuaries in Australia from 1987 to 2015 using satellite remote sensing. Random points were selected for each estuary and applied to the Australian Geoscience Data Cube (based on Landsat images) to obtain reflectance data through time. We propose a framework where (i) water colour is used to detect historical changes in catchments using generalised additive models, (ii) possible stressors and pressures driving those changes are evaluated using other available historical data, and (iii) lessons learned inform appropriate monitoring and management actions. This framework represents a novel approach to generate historical data for large-scale assessments of environmental change at catchment level, even in poorly studied areas.

Coastal dynamics of Northern Australia – Insights from the Landsat Data Cube

We demonstrate the capacity of detailed time-series of remotely sensed data to identify notable changes and trends in coastal morphology and vegetation over the last three decades, and illustrate how these data can help identify drivers of change, and areas that require further investigation. Time-series analysis (1987–2016) of the Australian Landsat archive is enabled by the Australian Geoscience Data Cube (AGDC), in which the data has been processed and restructured. We use Hovmoller plots to graphically display this time series for three test sites on the tropical coast of northern Australia. Initial results reveal distinct differences between sites: 1. Abrupt loss of mangrove cover (2004, 2006), which clearly relates to the passage of cyclones (Junction Bay); 2. Gradual seawards expansion of mangrove (1987–1996), and subsequent rapid narrowing of the mangrove zone (2013–2016; Burketown); 3. Minor seawards expansion of mangrove (1987–2004) in a stable embayment (Darwin Harbour). The density and duration of the time series enables clear links to be made to recorded events (e.g. cyclones), and identifies focus areas that exhibit unprecedented patterns of change (e.g. loss of mangrove) where regional or global environmental processes may be driving local change.

Mapping of mangrove extent and zonation using high and low tide composites of Landsat data

Monitoring mangrove health and distribution requires reliable methods that can be undertaken rapidly and at a resolution that optimises costs and accuracy. The Landsat record has been used for this purpose, but its application has been limited by the capacity to provide accurate results that distinguish mangrove from adjoining communities. The Australian Geoscience Data Cube provides a framework for exploring the Landsat record from 1987 onwards, and as pre-processing has already been undertaken there are efficiencies gained using this resource. Using the Data Cube, we exploited the differential spectral signature of mangrove under high tide and low tide conditions at Darwin Harbour, Australia, a relatively stable mangrove ecosystem, using image composites that combined imagery corresponding to the highest 10% and lowest 10% of tides. By applying the automated RandomForest classification technique, we demonstrate the capacity to accurately determine the extent of mangrove zones. Classification identified five mangrove zones: (1) seaward margin dominated by Sonneratia alba, (2) Rhizophora zone dominated by Rhizophora stylosa, (3) tidal flat dominated by Ceriops tagal, (4) landward salt flat and (5) marginal hinterland. Image composites that included high and low tide images achieved the best outcomes with kappa co-efficient of 0.81 and overall accuracy of 82%.

Extracting the intertidal extent and topography of the Australian coastline from a 28 year time series of Landsat observations

The physical environment of the intertidal zone presents a number of challenges which make traditional surveying methods difficult to implement; whilst the acquisition cost of airborne elevation data can be prohibitive beyond localised applications. Remote sensing data from earth observation satellites has been well established as a method for mapping the extent and elevation profile of the intertidal zone; however the majority of these applications are local in nature, and based on analysis of a limited number of individual image observations.In this paper we outline an automated methodology to model the intertidal extent and topography of the Australian coastline, that leverages a full time series of Landsat observations from 1987 to 2015 managed in the Australian Geoscience Data Cube (AGDC). Our approach reframes the analysis from time of acquisition into the tidal domain through the use of a continental scale tidal model, combined with a median pixel compositing of Normalised Difference Water Index (NDWI) stacks, to robustly estimate the tidal extent and elevation profile across the observed tidal range. We demonstrate the ability of a median-based compositing approach to deal effectively with data quality, modelling outliers and other issues that commonly affect single scene methodologies. Our new framework enables us to produce a continental-scale intertidal model, detailing intertidal zone topography at a 25m spatial scale.We then demonstrate the potential of the modelling framework to derive digital elevation data based on tidal model attribution, validating our models against in-situ GPS observations, before discussing issues that must be addressed to move this capability to a continental scale. The ability of the model to examine coastal stability is introduced, in the context of the conceptual reframing of the data into the tidal domain, with a view to furthering this application in future versions of the model.

The Australian Geoscience Data Cube — Foundations and lessons learned

The Australian Geoscience Data Cube (AGDC) aims to realise the full potential of Earth observation data holdings by addressing the Big Data challenges of volume, velocity, and variety that otherwise limit the usefulness of Earth observation data. There have been several iterations and AGDC version 2 is a major advance on previous work. The foundations and core components of the AGDC are: (1) data preparation, including geometric and radiometric corrections to Earth observation data to produce standardised surface reflectance measurements that support time-series analysis, and collection management systems which track the provenance of each Data Cube product and formalise re-processing decisions; (2) the software environment used to manage and interact with the data; and (3) the supporting high performance computing environment provided by the Australian National Computational Infrastructure (NCI).A growing number of examples demonstrate that our data cube approach allows analysts to extract rich new information from Earth observation time series, including through new methods that draw on the full spatial and temporal coverage of the Earth observation archives. To enable easy-uptake of the AGDC, and to facilitate future cooperative development, our code is developed under an open-source, Apache License, Version 2.0. This open-source approach is enabling other organisations, including the Committee on Earth Observing Satellites (CEOS), to explore the use of similar data cubes in developing countries.

An inter-disciplinary approach to airborne electromagnetics (AEM) survey design for groundwater exploration using the Australian Geoscience Data Cube and Morphotectonics

Over the past decade, advances in new satellite and airborne sensor technologies provide an opportunity for rapid multi-scale mapping, measurement and monitoring of the physical state of the crust, including resolution of key elements of surface and sub-surface hydrological systems. These advances have been mirrored by the development in advanced computational research infrastructure which is now giving the groundwater research community access to high-resolution (spatial and temporal) biophysical datasets (e.g. climate, ecology, geoscience and geospatial) relevant to broader hydrological systems understanding. This infrastructure facilitates integration of multiple datasets and rapid and improved signal processing, inversion, and sophisticated analysis. These datasets provide a catalyst for collaboration, with inter-disciplinary approaches enabling new discovery science in a 'big data' environment, and enabling the qualitative and quantitative analysis and modelling of landscape and hydrological system processes. In Australian landscapes, airborne electromagnetics (AEM) is widely used in near-surface (<200m) groundwater investigations due to the ability to acquire consistent, spatially coherent information of high quality using calibrated systems, in very short timeframes. This study reports on an evolving inter-disciplinary approach to AEM survey design for groundwater exploration. Recent investigations have employed time series analysis of surface water availability (using the Australian Geoscience Data Cube (AGDC)) combined with morphotectonic analysis of digital elevation datasets, tectonic analysis, and geomorphic analysis of satellite optical data, to help predict preferential recharge zones and shallow groundwater resources. This novel approach has been used successfully for groundwater exploration in the western Murray Basin and Kimberley Region of northern Australia.

Frontier groundwater investigations in the west Kimberley (Fitzroy) Region: preliminary assessment of groundwater resource potential and the salinity hazard to proposed irrigation developments

A regional reconnaissance AEM survey in the west Kimberley region (lower Fitzroy River valley and May-Meda Catchments) has successfully mapped the extent of regional Canning Basin aquifers (e.g. Pool Sandstone, Grant Group and Erskine Sandstone) confirming these as significant potential groundwater resources. The survey has successfully mapped a multi-layered hydrostratigraphy to depths of ~400m (in resistive areas), also revealing significant tectonics manifested as large scale basin-scale tilting of stratigraphy and more localised folding and faulting. The survey also mapped other key objectives including: the seawater intrusion (SWI) interface for the lower Fitzroy River valley and the May and Meda River Catchments; river flush zones along the Fitzroy, May and Meda Rivers; groundwater salinity variations and potential salt stores in the shallow alluvium of these rivers and proposed irrigation areas. River-parallel surveys also identified reaches of the May and Meda Rivers with direct contact between the rivers and underlying regional aquifers of the Pool Sandstone and Grant Group. These reaches are areas of potential recharge, and/or groundwater discharge, with the AEM used to target hydrochemical investigations, drilling and aquifer testing. The AEM data reveal more complex relationships between perennial in-stream pools along the Fitzroy River, underlying aquifers, tectonics, and river alluvium. AEM survey design incorporated time series analysis of surface water availability through time (Water Observations from Space (WOfS)) derived using the Australian Geoscience Data Cube (AGDC). This has facilitated investigations of surface-groundwater interaction through ensuring the AEM transects coincide with permanent water holes and river reaches considered to be potentially sustained by groundwater discharge. The AEM data provide a framework for hydrogeological process understanding, while the knowledge generated in this project will inform water resource allocation planning, help assess risks to culturally and environmentally sensitive riverine ecosystems, and de-risk investment in agriculture and water infrastructure more broadly.