National Airborne Field Experiments for Prediction in Ungauged Basins

Environmental remote sensing has matured significantly over the past two decades as a result of new satellites and intensive airborne campaigns. As such, current remote sensing technology has a huge potential for hydrologic prediction in ungauged basins, through an ability to measure many land surface states, fluxes and parameters that impact on basin prediction. For instance, it is now possible to measure evapotranspiration rates that impact on soil moisture and baseflow, near-surface soil moisture content that controls rainfall partitioning into infiltration and runoff, snow water equivalent of the snow pack that impacts springtime runoff, vegetation parameters such as leaf area index and greenness that impact on evapotranspiration, land surface elevation and canopy height that impact on runoff routing and evapotranspiration, and so on. However, there are still many unanswered questions that need to be addressed, such as validation of these data products from new sensors, maturing retrieval algorithms, developing techniques for downscaling, and merging remote sensing data with model predictions through the process of data assimilation. To answer these questions and more it is essential that modelling be undertaken in conjunction with field experimentation in well instrumented basins together with intensive ground-based and airborne measurements of the appropriate type and spatial/temporal resolution. This paper describes the National Airborne Field Experiments (NAFE) planned for 2005 and 2006 (see www.nafe.unimelb.edu.au) in several well instrumented catchments in south-eastern Australia with a range of climatic, land use, land cover and topographic conditions. While these experiments have a particular emphasis on the remote sensing of soil moisture, they are open for collaboration from interested scientists from all disciplines of environmental remote sensing and its application. Moreover, scientists will additionally be addressing questions on carbon budgets, ecohydrology and flood forecasting in 2005, and bushfire prediction, evapotranspiration and precipitation in 2006. The catchments to be studied in these experiments are i) the Goulburn River experimental catchment in the Upper Hunter during November 2005 and ii) the Yanco/Colleambally and Kyeamba creek experimental areas of the Murrumbidgee catchment during November 2006 (see www.oznet.unimelb.edu.au). Approximately 100 hours of airborne data will be collected across each of the 4-week field campaigns, together with an extensive amount of ground truth data. The airborne measurements will consist of vertical and horizontal polarisation passive microwave data together with thermal infrared, near infrared, visible and lidar data. Passive microwave data will be collected in both mapping and multi-incidence angle line modes. The mapping data will be collected at a range of spatial resolutions (62.5m to 1km for passive microwave, 1m to 20m for thermal to visible, and 1m for lidar) across spatial extents ranging from individual farms to 50km regions. The high resolution farm data will be collected twice a week at each of 8 farms while low resolution data will be collected once a week for the entire region. Additionally, the CoSMOS-2 experiment initially planned for Europe has been moved to Australia as part of NAFE '05 and it is anticipated that flux measurements will be made as part of NAFE '06 if there is sufficient interest. A trial campaign to evaluate the airborne and ground measurements planned for both NAFE campaigns has recently been conducted in the Waikerie region of South Australia. This is the first airborne data set with such high resolution passive microwave data. Moreover, it is unique in that there is such a complete suite of airborne measurements made from the same low-cost platform. This paper presents preliminary data obtained from that trial campaign, showing a clear soil moisture signal in the passive microwave data.