The combination of anthropogenic activity and climate variability has resulted in changes to hydrologic regimes across the globe. Changes in water availability impact on vegetation structure and function, particularly in semi-arid landscapes. Riparian and floodplain vegetation communities are sensitive to changes to surface-water and groundwater availability in these water-limited landscapes. Remote-sensing multi-temporal methods can be used to detect changes in vegetation at a regional to local scale. In this study, a `best-available pixel' approach was used to represent dry-season, woody-vegetation-canopy characteristics inferred from Normalised Difference Vegetation Index (NDVI). This paper describes a method in which Landsat 5 TM and Landsat 7 ETM+ data from 1987 to 2011 were processed using object-based image-analysis techniques to generate annual minimum NDVI values for vegetation communities in the Lower-Darling floodplain The changes detected in riparian and floodplain canopies over time can then be integrated with other spatial data to identify water-source dependence and infer a relationship between changes to the hydrologic characteristics of specific water sources and vegetation dynamics.

Geoscience Australia (GA) was invited by Murray-Darling Basin Authority (MDBA) in 2010 to participate in an evaluation of the Intermap IFSAR (Interferometric Synthetic Aperture RADAR) data that was acquired as part of the Murray-Darling Basin Information Infrastructure Project Stage 1 (MDBIIP1) in 2009. This evaluation will feed into the business case for Stage 2 of the project. As part of the evaluation GA undertook the following: 1. A comparison of the IFSAR Digital Surface Model (DSM) and Digital Terrain Model (DTM) with a recent LiDAR acquisition, covering approximately 9000Km2 of the Lower Darling Region. It focused on assessment of the data over various land cover and terrain types and identified opportunities and issues with integrating IFSAR with LiDAR. 2. A comparison of the IFSAR Vegetation Canopy Surface (DSM minus DTM) with the Lower Darling LiDAR Canopy Elevation Model (CEM). 3. A comparison between currently mapped man-made and natural water bodies over the Murray-Darling Basin with the IFSAR derived products (water mask). 4. Application of the National Catchment Boundaries (NCBs) methodology to the IFSAR data and comparison with the delineated watersheds from PBS&J (Intermap's sub-contractor). This report outlines the findings of this evaluation based on the 4 items above MDBA requested.

How much easier it would be to map and quantify the key elements of the hydrological cycle if the Earth's surface was transparent! Unfortunately, this is not the case and it is this very inability to penetrate to sufficient depths to map and quantify groundwater components of the hydrological cycle that currently necessitates the integration of satellite- airborne- and ground observations. In Australia, important advances have been made in the last 3 years in quantifying key elements of the hydrological cycle. This has been achieved in part through the increased use of Landsat, MODIS, SPOT, hyperspectral, NOAA and LiDAR datasets to improve the mapping and quantification of surface water, evapotranspiration, soil moisture and recharge and discharge. However, significant limitations remain in using satellite-based platforms alone for quantifying catchment water balances, surface-groundwater interactions, groundwater resource estimation and managing groundwater dependent ecosystems. Increasingly, the need to map the key elements of the hydrological cycle to calibrate water balance models and for environmental management, is leading to the development of more holistic systems approaches, involving the integration of satellite-, airborne and ground-based techniques and measurements. One example is in the River Murray Corridor (RMC) in SE Australia, where previous attempts to assess the water needs for iconic floodplain wetland ecosystems, based largely on satellite-based measurements, did not adequately take into account sub-surface soil conditions and groundwater quality and processes. In floodplain environments such as the River Murray Floodplain, the factors that govern tree health are invariably complex, and include a wide range of biophysical and biogeochemical factors.

Identification of groundwater-dependent terrestrial vegetation, and assessment of the relative importance of different water sources to vegetation dynamics, typically requires detailed ecophysiological studies over a number of seasons or years. However, even when groundwater dependence can be quantified, results are often difficult to upscale beyond the plot scale. Quicker, more regional approaches to mapping groundwater-dependent vegetation have consequently evolved with technological advancements in remote sensing techniques. These approaches however often fail to incorporate sub-surface hydrogeological processes in their interpretation of groundwater dependence. This study, undertaken in the semi-arid Darling River Floodplain in NSW, Australia, innovatively combines Landsat Normalised Difference Vegetation Index (NDVI) time series data with hydrogeological, hydrogeochemical and hydrogeophysical data to assess the relative importance of hydrological processes and groundwater characteristics. Central to the approach is the use of airborne electromagnetics which provides a 3-dimensional context to otherwise point-based borehole data. This approach has resulted in an improved understanding of vegetation dynamics including the spatial distribution of vegetation utilising groundwater, timing and duration of groundwater use, and response to different hydrologic regimes (e.g. rainfall, lateral bank recharge, and overbank flooding). In particular, the study has established that the deeper (>25m), semi-confined aquifer is only rarely important to vegetation dynamics, with the shallow unconfined aquifer and river flush zones being of greater importance. These findings are being used to assess the suitability of proposed groundwater-development schemes in the study area, and have implications for riparian vegetation management more broadly.

The Integrated Vegetation Cover (2003), hereafter referred to as the IVC03 dataset represents vegetation cover across Australia and was compiled by integrating a number of recent vegetation-related datasets.This dataset was developed to assess and report on the type and extent of native, non-native and non-vegetated cover types across the whole landscape. This dataset has been used to describe vegetation types found in each National Action Plan and Natural Heritage Trust regions across Australia. Data are stored as a raster of 100m resolution and are projected in Albers conic equal-area coordinates.Version 1 incorporates a selection of the latest available vegetation data as at July 2003.Vegetation cover in the IVC03 dataset is described using a 12-class attribute schema that was developed to meet vegetation-related information needs of the Commonwealth Government natural resource management arena.The IVC03 dataset has 12 attribute classes:1Native forests and woodlands 2Native shrublands and heathlands 3Native grasslands and minimally modified pastures4Horticultural trees and shrubs5Perennial crops6Annual crops and highly modified pastures7Plantation (hardwood)8Plantation (softwood/mixed)9Bare10Ephemeral and Permanent Water Features11Built-up99Unknown/not reportable. Five datasets were used as inputs to create the IVC03 dataset.A national ruleset was developed to assign orders of precedence to each attribute value from each input dataset.The ruleset was used to integrate the inputs into a single vegetation cover dataset. The five input datasets were:-Agricultural Land Cover Change (ALCC95);-Forests of Australia 2003 (FOA03);-1996/97 Land Use of Australia, Version 2 (LUA97);-Land Use Mapping at the Catchment Scale (LUMCS03); and-National Vegetation Information System 2000 (NVIS00).

National vegetation cover derived from: - Values 1, 7, and 8 from the 2007 forests dataset (BRS) - Values 2 and 3 from the NVIS 3.1 dataset (ERIN) - Values 1-6 and 9-11 from the catchment scale land use dataset (as at April 2009, BRS) - Any remaining no data areas filled from the Integrated Vegetation 2008 dataset (BRS) The datasets were resampled to 100 metre grids and projected to Albers equal area if required. The integrated vegetation grid was derived using a conditional statement weighing each input grid in the order listed above. Bureau of Rural Sciences, Canberra are custodians of the dataset.

Atlas of Regolith Materials of Queensland. Companion to the 1:2,500,00 Queensland Regolith-Landform Map and GIS. Both broad and detailed regolith mapping and characterisation of materials was used to build an understanding of the regolith and its associated landforms. This state-wide overview contributes significantly to understanding the regolith and landform processes and regolith materials of Australian arid and coastal environments. This Queensland study extends seamlessly from the Northern Territory Regolith Landform Map and provides a broad-scale framework fro guiding geochemical prospecting for a wide range of minerals and materials.

Extended abstract detailing the use of MODIS Enhanced Vegetation Index time series data to map and monitor Groundwater Dependent Ecosystems in the Hat Head National Park.

Subtidal to intertidal deposits from Kaipara Harbour in Northland preserve a 23,000+ year incomplete sedimentary record of the transition from terrestrial to estuarine conditions in the Wairoa River arm of the harbour. Cores are used to reconstruct the depositional setting for this transition, which we interpret as a succession from dune and freshwater wetland to shallow estuarine environments. The fossil pollen record provides a proxy of Last Glacial Maximum and Late Glacial vegetation for the area. Stability of the palaeo-dune landscape during the postglacial marine transgression is interpreted on the basis of strong dominance of tall forest taxa (Dacrydium) in the pollen record and soil development in dune sands. Reworking of buried dune and wetland sediments has only reached to a depth of 1.5 m below the modern tidal flat. As such, the site provides a rare example of good preservation of Pleistocene deposits at the coast, where extensive reworking and loss of record are more typical.

Fresh groundwater resources are a highly valuable commodity, particularly in semi-arid to arid landscapes where annual precipitation is low and surface water is scarce. Water security, often achieved through the development of groundwater resources, is a high priority for rural communities within these water-limited landscapes. However this is often at the expense of the environment when alterations to the groundwater system, often in conjunction with drought conditions, can detrimentally impact floodplain and riparian vegetation structure and function. Remote-sensing methods can be used to detect such changes in vegetation. In this study, a multi-temporal Landsat Normalised Difference Vegetation Index (NDVI) approach was used to detect changes in riparian and floodplain vegetation in the Lower-Darling floodplain, NSW, Australia. When integrated with surface and subsurface data, these changes provided insight into how surface water availability and subsurface geological and hydrogeological characteristics influenced vegetation distribution and behaviour at multiple scales. It was found that while the availability of water resources was the primary driver of changes in vegetation canopy dynamics, this availability was strongly influenced by both tectonic and hydrogeological processes. These findings were of particular importance when considering the suitability of groundwater development options and they have implications for future groundwater assessment studies.