Chapter 5 Climate Research

WCRP Activities > Global Energy and Water Cycle Experiment (GEWEX)

The Global Energy and Water Cycle Experiment (GEWEX) provides the focus for international studies of processes deter-mining the global hydrological cycle. The goal of the program is to observe, understand and model variations in the global water cycle, including the impact of increasing atmospheric concentrations of greenhouse gases. Research in Australia related to GEWEX is aimed primarily at measuring and understanding the components of the climate system that drive the water cycle.

A major aspect of GEWEX is a series of regional studies in significant catchments of the world as contributions to the Continental Scale Experiments (CSEs). In 2002, the Murray Darling Basin (MDB) was accepted by the GEWEX Steering Committee as a CSE, in a national project involving the Bureau of Meteorology Research Centre (BMRC) and Melbourne University under the auspices of the CRC for Catchment Hydrology, as well as Macquarie University, ANSTO, CSIRO Land & Water and CSIRO Atmospheric Research. Under the MDB project, Melbourne University and BMRC are collaborating on a measurement and modelling project aimed at improving the representation of soil processes in the BMRC model used for both weather and climate predictions. A study of the 2002-03 drought for the Murray-Darling Basin (Figure 5.2) has shown that the high temperatures associated with the drought are consistent with the strong warming evident since the 1950s.

Figure 5.2. Time series of May-October means of maximum temperature, rainfall and minimum temperature for the Murray-Darling Basin for the period 1952-2002. The 2002 drought, while no drier than some previous droughts, was the warmest on record.

The GEWEX community is working with the International Atomic Energy Agency to use isotopic data to infer the history and sources of water in rainwater, rivers and groundwater. Within the MDB project, ANSTO, in co-operation with Macquarie University and University of Technology Sydney (UTS), have used iso-topic data and modelling to investigate the intensification of the hydrological cycle in the Amazon Basin, and to study the depletion of O18 in the water of the Darling River. Results from the WCRP Atmospheric Model Intercomparison Project (AMIP) have been used to generate evaporation-toprecipitation ratios for all the CSE basins.

Studies at the Australian National University (ANU) and the CRC for Greenhouse Accounting have shown that canopy-scale photosynthesis and hence carbon uptake are sensitive to changes in diffuse solar radiation. This result suggests that changes in clouds and aerosols can affect carbon uptake in the terrestrial biosphere. The work has been extended to provide an explanation for the observation that pan evaporation appears to be decreasing worldwide; in particular, it is found that a decrease in direct solar radiation can provide an explanation.

In collaboration with ANSTO, statistical methods are being investigated at the University of Newcastle to estimate the sources of air pollution, especially fine particulate matter, from meteorological back trajectories. The approach has been applied to data from Cape Grim and will be applied to data from the Asian Pacific Regional Aerosol Characterisation Experiment (Figure 5.3). The University has carried out further studies on aerosol emissions and transport, in relation to mining activity in the Hunter Valley. Modelling studies on the formation and development of the west coast trough in Western Australia have been used at Murdoch University to predict the dispersion of pollution across Perth.

Figure 5.3. Annual average concentrations of the major aerosol components for the fine fraction at three key ACE-Asia sites at Hong Kong, Kosan (Cheju Is., South Korea), and Sado Is. (Japan), and (b) Weekly averages of minimum diurnal radon concentration at three key ACE-Asia sites: Hong Kong, Kosan (Cheju Is., South Korea), and Sado Is. (Japan).

Modelling studies at CSIRO Atmospheric Research (CAR) suggest that sulphate particles from human activities may have affected drought in the Sahel region of Africa (Figure 5.4). It is found that the particles affect the radiative properties of clouds, which in turn can impact on the large-scale distribution of cloud and precipitation.

Figure 5.4. Changes to rainfall patterns over the Sahel in pre-industrial times (left) and during the 1980s (right).

Improved understanding of cloud and its interaction with radiation is a key objective of GEWEX. At the University of Tasmania, the effects of low-level stratocumulus cloud on solar radiation are being investigated through a measurement and modelling project. A fractal cloud model is being linked to a Monte Carlo radiative transfer model, and the results suggest that the accuracy of satellite estimates of surface solar radiation are dependent upon the temporal resolution of the satellite sampling.

Research in BMRC on climate feedbacks has included a study in which the feedbacks exhibited under 2xCO2 conditions were compared in different climate models. is found that the differences between models in their internal feedbacks are not limited to cloud feedback (Figure 5.5). A study with the BMRC climate model showed that there are strong seasonal variations in feedbacks. This work relates to the new international Cloud Feedback Model Intercomparison Project (CFMIP), for which BMRC is contributing to the design and planning.

Figure 5.5. Schematic showing the influence of climate feedbacks, as determined on the amount and sign of radiative forcing driving a climate model. The arrows are indicative of the magnitude and sign of individual feedbacks, as determined from a Bureau of Meteorology Research Centre (BMRC) climate model. The dominant positive feedback is due to water vapour. In the BMRC model, cloud feedback is positive, but this varies greatly between models. The range in surface temperature changes indicated resultsfrom the varying effect of all feedbacks, but particularly of clouds.

The representation of the land surface in climate models continues to be investigated by several research groups. The Project for Intercomparison of Landsurface Parameterization Schemes (PILPS) is part of the GEWEX Global Land Atmosphere System Study (GLASS), and it is jointly coordinated by Macquarie University and ANSTO. Through PILPS studies, Macquarie University, ANSTO and UTS have now found clustering among landsurface schemes used in global climate models, such that a chronological sequence of improving performance can be identified. Moreover, the Chameleon Surface Model (CHASM) has been used to demonstrate that more complex schemes can produce more accurate simulations particularly in relation to evapotranspiration.

There has been further modelling work at Macquarie University on the relative impact of greenhouse-induced climate change compared with land-use changes on global and regional scales, including the impacts on the occurrence of extreme temperature and rainfall events. Over the Australian region, the results suggest that any analysis of the cause of climate trends within Australia should consider land use changes, in addition to the direct effects of global warming.

Macquarie University, in collaboration with BMRC, has used the CHASM model in the BMRC global climate model to explore the sensitivity of changes in surface variables, such as precipitation and evaporation, to the complexity of the land surface scheme compared with changes associated with doubling of the concentration of carbon dioxide in the atmosphere. The results suggest that, while the changes in mean values are robust, changes in spatial and temporal variability are sensitive to the details of the land surface scheme.

There have been observational studies at Murdoch University to improve our understanding of the impact of land-use changes on the climate of southwest Western Australia. The work now includes the use of satellite data to determine the climatic effects of land-use changes, and in situ measurements to examine the role of surface heterogeneity on the fluxes of heat and moisture into the atmosphere.

The modelling studies carried out by the various groups in Australia depend upon continuing development of the global climate models and the systems used to analyse model results. Over the period of this report, there has been continuing work in the BMRC global model on improving the representation of cloud processes, radiative transfer, gravity wave drag, and land surface processes. A Model and Climatological Data Comparison System MACCS) has also been developed to facilitate the analysis of model runs.

The single-column model (SCM) of the National Center for Atmospheric Research NCAR) has been adapted to include the parameterisation scheme of the BMRC climate model to facilitate the development and evaluation of parameterisation schemes. In particular, data from the Atmospheric Radiation Cloud Station ARCS) in Darwin are being used in the SCM to improve the representation of convective processes in models. The ARCS is part of the USA Department of Energy Atmospheric Radiation Measurement (ARM) program, and it is co-located with the BMRC climate monitoring and research station in Darwin. The ARM project also includes a collaborative project with CAR on the measurement and modelling of atmospheric radiation.

The BMRC station in Darwin has been collecting rainfall and associated data since 1987, and it is a ground-validation site for the National Aeronautics and Space Administration (NASA) - National Space Development Agency (NASDA) Tropical Rainfall Measuring Mission (TRMM) involving satellite-based observations of rainfall. The facilities in Darwin were used to support the Darwin Waves Experiment (DAWEX) in 2001-02, which focused on the generation of stratospheric gravity waves by deep convection. In 2002-03, there was a collaborative project under the ARM program (EMERALD-2) aimed at improving our understanding of the formation and development of tropical cirrus clouds.