Proceedings & Presentations

You can access abstractspresentations and posters via the links below. Alternatively you can download the book of abstracts (pdf, 41 pages)

You can also read the workshop summary report.

DAY 1 (Tuesday, 28 October 2014)

Welcome and Introduction – Albert van Dijk

Plenary session: Information needs for Tomorrow

Keynote: Rob Vertessy, Bureau of Meteorology (presentation)

Warwick McDonald, CSIRO Land and Water (presentation)

Mike Makin, Murray-Darling Basin Authority (presentation)

Mike Roderick, Australian National University (presentation)

Session 1:

Are we getting close to integrating water and climate information?

session program

Session 2:

With ground networks in decline, can satellites meet our needs?

session program

Session 3:

How well can we trust our models, and how can we be sure?

session program

Poster session

DAY 2 (Wednesday, 29 October 2014)

Plenary session: Creating an environment for innovation

Keynote: Andy Pitman, ARC Centre of Excellence on Climate System Science, UNSW (presentation)

Helen Owens, Data Policy Branch, Dept. of Communications (presentation)

Brad Evans, Terrestrial Ecosystem Research Network (presentation)

Session 4:

What new water and climate information should we be developing?

session program

Session 5:

Is Australia’s data and model infrastructure ready for the future?

session program

Session 6:

Have we reached the limits of what can be forecast?

session program

Session 1: Are we getting close to integrating water and climate information?

Tuesday 28 October, 13:00-16:30

Session 1a: Information

Mohsin Hafeez (BoM) invited

Combining climate and water resources assessment information in decision making (presentation)

Darryl Lindner (Actew) invited

Actew’s use of water and climate information in operations and long-term planning (title tbc)

Session 1b: Modelling

Imtiaz Dharssi (BoM) invited

Are we improving weather forecasts through better initialisation of the land surface state? (abstract - presentation)

Dushmanta Dutta (CSIRO)

Australian Water Resources Assessment (AWRA) (abstract - presentation)

Jason Evans (UNSW)

Convection permitting regional climate modelling for short time scale precipitation extremes (abstract - presentation)

Propositions for discussion

  1. Water and climate information are already sufficiently well integrated for the purpose of decision making.
  2. Currently available water and climate information are inconsistent.
  3. Integration of water and climate information is best left to the users.
  4. Water and climate information cannot currently be interpreted together.
  5. There is no real need to integrate water and climate information.
  6. Water and climate information are too different to be integrated.
  7. Climate information needs to be better integrated into hydrological models.
  8. Climate models should be improved so they can replace hydrological models.
  9. There are some major scientific challenges in integrating water and climate information.
Session 2: With ground networks in decline, can satellites meet our needs?

Tuesday 28 October, 13:00-16:30

Session 2a

Leo Lymburner (Geoscience Australia) invited

Using a calibrated archive of Landsat data to characterise the distribution of water across the Australian continent between 1998 and 2012  (presentation)

Geoff Podger (CSIRO) invited

Ground and satellite observations in water resources management (presentation)

Belinda Medlyn (Macquarie Univ.) invited

Making good use of satellite data: one modeller’s perspective (presentation)

Session 2b

David Jones (BoM) invited

Monitoring Australia’s climate: current practices and some thoughts on future directions (presentation)

Eva van Gorsel (CSIRO) invited

The role of flux tower observations in water and climate information (presentation)

Propositions for discussion

  1. All models are wrong. Satellite-derived information is essentially modelled, and therefore wrong.
  2. Satellite data is only a valid observation if interpreted by experts.
  3. Satellites need to be calibrated with ground observations to be more than pretty pictures.
  4. Currently, satellite information is ‘validated’ by remote sensing scientists misinterpreting a handful of spatially inadequate samples.
  5. Satellite data should only ever be used to interpolate ground measurements.
  6. Satellite and ground data are only useful when both are available and carefully interpreted together.
  7. Fusion of satellite data, ground data and models is the only way to achieve the best possible information.
  8. Cheaper ground sensors, power supply and telemetry will reverse the decline in ground networks.
  9. The attention to satellites is contributing to the decline of field measurement and needs to be stopped.
Session 3: How well can we trust our models, and how can we be sure?

Tuesday 28 October, 13:00-16:30

Session 3a

Gab Abramowitz (UNSW)

Defining expectations: an approach to quantifying trust in modelling (abstract - presentation)

Beth Ebert (BoM) invited

Verification of numerical models – what are the biggest challenges? (abstract - presentation)

Bellie Sivakumar (UNSW)

Hydrologic systems as complex networks: structure, connections, and dynamics (abstract)

Session 3b

Dmitri Kavetski (Univ. Adelaide) invited

Hydrological modelling at the catchment scale:
Trusty Friend or Devious Foe? (abstract - presentation)

Andrew Frost (BoM)

Evaluation of the Australian Water Resource Assessment Landscape (AWRA-L) model, WaterDyn and CABLE (abstract presentation)

Lucy Marshall (UNSW)

Balancing the realities of environmental observations, model uncertainty and model truthfulness (abstract)

Gift Dumedah (Monash Univ.)

Diagnostic evaluation of land surface models from decision space –the hydrologic genome approach (abstract)

Propositions for discussion 

  1. The simplest model that explains the observations is necessarily the best model.
  2. All models are wrong, but some are still useful.
  3. The models are not the main problem, it is the quality of the data and assumptions that go into them.
  4. Much more effort is needed to objectively assess the performance of alternative models.
  5. We need to stop calibrating our models, it leads to a false sense of security.
  6. In circumstances where calibration is essential for a model to be useful, we should just use an empirical model (for example, based on data mining or Bayesian methods).
  7. We cannot know whether to trust our models. Therefore multi-model ensembles should be standard operational practice, not just a research endeavour.
  8. In the absence of quantitative knowledge of model inter-dependence, ensemble methods are meaningless.
  9. Inappropriate values for unconstrained parameters (through calibration or assumption) should remove any trust in predictive ability.
  10. Talking about ‘physically-based’ models is meaningless when there is not enough data to construct an empirical model.
Session 4: What new water and climate information should we be developing?

Wednesday 29 October, 13:00-16:30

Session 4a

Paul Dalby (In Fusion Consulting) invited

Overcoming the tyranny of climate: capturing the social and economic benefits of great climate research (abstract - presentation)

Peter Stone (CSIRO) invited

Information needs for the development of Northern Australia (abstract - presentation)

Bertrand Timbal (BoM)

Victorian Climate Initiative (abstract - presentation)

Session 4b

Seth Westra (Univ. Adelaide)

Challenges in attributing change in Australian natural hazards (abstract - presentation)

Jorge Peña-Arancibia (CSIRO)

Towards dynamic continental estimation of irrigated areas and water use (abstract - presentation)

Jay Larson (ANU)

Not Your Average Visualisation Project (abstract - presentation)

Propositions for discussion

  1. We do not need new information, just greater quality in existing information.
  2. We understand the most important water and climate processes well enough.
  3. Basic hydrological processes are still poorly understood and quantified.
  4. Basic climate processes are still poorly understood and quantified.
  5. We need more analysis of past records (e.g., historical and paleo-data) to interpret information.
  6. We do not need more data or information, but share and communicate the existing data better.
  7. We need better and faster access to information and data.
  8. The focus should be on making satellite data easier and cheaper to use.
  9. The main priority should be groundwater data, given the risk related to coal seam gas for example.
  10. With all this focus on predicting the future, we are neglecting understanding the present and past.
Session 5: Is Australia’s data and model infrastructure ready for the future?

Wednesday 29 October, 13:00-16:30

Session 5a: Applications

Luigi Renzullo (CSIRO)

Towards a hyper-resolution land data assimilation system for Australia (abstract presentation)

Mike Hutchinson (ANU)

Making the most of the ground based meteorological network using anomaly-based interpolation (abstract - presentation)

Adam Smith (BoM)

The Australian Water Resources Assessment (AWRA) Modelling System Implementation Project: Getting Australia’s data and model infrastructure ready for the future (abstract presentation)

 Session 5b: Data and Infrastructure

Tim Pugh (BoM) invited

Sharing data and infrastructure

Ben Evans (National Computing Infrastructure) invited

National Computing Infrastructure Environmental Data Collections, Management and Services Overview (presentation)

Propositions for discussion 

  1. Australia’s water and climate community data systems and models are out of date.
  2. The NWP and Astronomy communities should share their computing infrastructure more.
  3. Our challenges are rapidly being solved, provided we make use of global data sets and computing solutions, such as cloud computing, data services and the available infrastructure.
  4. The research community already has ready access to high performance computing infrastructure, but many do not realise it.
  5. The way operational and research data are shared in Australia is systematically biased and flawed.
  6. There is a lot of data storage and computing infrastructure, but it does not have the characteristics we need.
  7. We need better incentives for collaboration on large problems across institutions.
Session 6: Have we reached the limits of what can be forecast?

Wednesday 29 October, 13:00-16:30

 Session 6a: Flood forecasting

Tom Pagano (BoM)

Challenges of operational river forecasters (abstract - presentation)

Dongryeol Ryu (Melbourne Univ.) invited

Leveraging ground and remotely sensed observations for short-term streamflow forecasting (abstractpresentation)

Fiona Johnson (UNSW)

Driving through floodwaters: what’s the point of flood forecasting? (abstract presentation)

 Session 6b: Seasonal forecasting

Oscar Alves (BoM) invited

Improving seasonal climate forecasts

Narendra Tuteja (BoM) invited

Water availability forecasts for operational planning and management – business drivers, progress-to-date and key challenges (presentation)

Propositions for discussion 

  1. We are very close to the limits on predictability imposed by chaotic behaviour of the climate system.
  2. We don’t actually know how well we can forecast; past forecasts are not evaluated and forecasting methods change too fast.
  3. High quality and frequency ground and satellite radar precipitation measurement capability will transform our short-term hydrological forecasting capacity.
  4. There are still major gains and breakthroughs to be had in seasonal prediction.
  5. Our current forecasting methods are too fragmented over time scales; resolving that will cause a leap in the accuracy and utility of forecasts.
  6. We can’t forecasts well at all timescales, so we should we just focus on those that are both feasible and useful.
  7. The communication between the research and operational communities needs to be much better.
  8. Improvements should we focused on the measurement infrastructure, not the models.
  9. The priority should be on data quality control algorithms that synthesize data from different sources to identify outliers and infill missing values.
  10. Forecasters cannot use new data until the record is long enough to know its properties and biases.
  11. We need to transition from hydrograph mimicry to better representation of hydrologic processes.
  12. Scientists need to field test new methods under the supervision and on the terms of operational agencies.
  13. The influence of human interferences (farming, urbanization, deforestation) on forecasts cannot be predicted.

Poster session

Tuesday 28 October, 13:00-16:30

(* denotes students in contention for MSSANZ Student Poster Award)

Emetc, V. (ANU) *

Improving hydrological models focusing not on hydrology (abstract)

Gevaert, A. (VU Univ. Amsterdam) *

Evaluation of downscaled soil moisture and vegetation optical depth derived from the Land Parameter Retrieval Model (abstract)

Gharun, M. (Univ. Sydney)

Spatio-temporal controls on catchment ecohydrology: lessons learnt from eucalypt forests (abstract)

Guerschman, J., J. Pena-Arancibia (CSIRO)

Towards dynamic high resolution mapping of cropped areas in Australia (abstract)

Haughton, N., et al. (UNSW) *

Dissecting PLUMBER: Why are land surface models performing so poorly? (abstract)

Kala, J., M. De Kauwe (UNSW)

Influence of an optimal stomatal conductance scheme in Australian Community Climate and Earth System Simulator (ACCESSv1.3) (abstract - poster)

Kim, S. (UNSW) *

Improvement of soil moisture dataset combining AMSR2 soil moisture products (abstract)

Lievens, H., et al. (Ghent Univ.)

Impact of different data assimilation strategies for SMOS observations on flood forecasting accuracy (abstract)

Lin, Y.-S. (Macquarie Univ.)

A synthesis of a global stomatal conductance database under an optimal stomatal behaviour framework: patterns from leaf to ecosystem (abstract)

Liu, Y., et al. (UNSW)

Observing water availability impacts on vegetation using an enhanced passive microwave remote sensing method (abstract)

Lopez Lopez, P., et al. (Utrecht Univ.) *

Alternative configurations of quantile regression for estimating predictive uncertainty in water level forecasts for the upper Severn River: a comparison (abstract - poster)

Olson, R., et al. (UNSW)

The NARCliM Project: Model Evaluation and Climate Projections for Temperature and Precipitation for South-East Australia (abstract)

Pathiraja, S., et al. (UNSW) *

Hydrologic Modelling in Non-Stationary Catchments:A Data Assimilation Approach (abstract)

Summers, D., A. van Dijk (ANU)

Interpreting vegetation condition from satellite observations: accounting for the influence of water availability (abstract)

Bishop, T., A. Horta (Univ. Sydney)

Space-time monitoring of (sub) soil moisture for agricultural management: a case study (abstract)

Tian, S., et al. (ANU) *

The potential for improving terrestrial water storage estimates through assimilation of GRACE data into a hydrological model (abstract)

Ukkola, A. et al. (Macquarie Univ.) *

CO2-induced greening reduces streamflow in water-stressed climates in Australia (abstract)

Yebra, M., A. van Dijk (ANU)

Coupling gross primary production and transpiration for a consistent estimate of apparent water use efficiency (abstract)

Hasan, M. (UNSW) *

An approach to estimate rainfall at ungauged location by merging the radar and gauge estimates (abstract)