High-quality spatial climate data-sets for Australia

High-quality spatial climate data-sets for Australia

December 2009 | David A. Jones, William Wang and Robert Fawcett
This paper presents a new high-quality set of historical and real-time climate analyses for Australia, covering rainfall, temperature (maximum and minimum), and vapour pressure at daily and monthly timescales. These analyses are developed to improve understanding of past climate variability and change, and to enhance recent climate estimates. The analyses are produced at a resolution of 0.05° × 0.05° (approximately 5 km × 5 km) using new topography-resolving methods. These methods are similar to those used internationally but have been adapted for Australia, leading to significant improvements over existing data sets. The analyses are based on in situ observations managed by the Australian Bureau of Meteorology, with real-time updates available through the Bureau's climate website. The data include station observations and error surfaces, and are used for monitoring climate variability and change. The analyses are complemented by remotely sensed and model-derived data. The climate data used in the analyses include rainfall, temperature, and vapour pressure, with the earliest temperature data dating back to 1844 and rainfall data to 1832. The analyses are generated using both daily and monthly data from the Australian Data Archive for Meteorology (ADAM), which is updated in real time. The number of stations contributing to the analyses has increased over time, with a slight decline in recent decades. The analyses use anomaly-based methods, which are consistent with long-term climatology and provide a more accurate representation of short-term climate. The analyses are generated using a combination of the Barnes successive-correction method and three-dimensional smoothing splines. These methods are robust and efficient, and have been found to be effective for climate analyses. The accuracy of the analyses has been verified against station observations, with cross-validated error statistics provided for different periods. The RMSE for monthly maximum and minimum temperatures is typically between 0.5 and 1°C, with improvements over current Bureau practice. The RMSE for daily rainfall is slightly higher, with improvements in recent years. The RMSE for vapour pressure is also improved, with analyses showing slightly lower RMSE and MAE compared to previous studies. The analyses are used for monitoring climate change and documenting long-term trends. They are compared with existing Bureau of Meteorology data sets, showing good agreement in annual rainfall averages and trends. The data are also used for climate change studies, with the potential for future improvements through the use of remote sensing and model-derived data.This paper presents a new high-quality set of historical and real-time climate analyses for Australia, covering rainfall, temperature (maximum and minimum), and vapour pressure at daily and monthly timescales. These analyses are developed to improve understanding of past climate variability and change, and to enhance recent climate estimates. The analyses are produced at a resolution of 0.05° × 0.05° (approximately 5 km × 5 km) using new topography-resolving methods. These methods are similar to those used internationally but have been adapted for Australia, leading to significant improvements over existing data sets. The analyses are based on in situ observations managed by the Australian Bureau of Meteorology, with real-time updates available through the Bureau's climate website. The data include station observations and error surfaces, and are used for monitoring climate variability and change. The analyses are complemented by remotely sensed and model-derived data. The climate data used in the analyses include rainfall, temperature, and vapour pressure, with the earliest temperature data dating back to 1844 and rainfall data to 1832. The analyses are generated using both daily and monthly data from the Australian Data Archive for Meteorology (ADAM), which is updated in real time. The number of stations contributing to the analyses has increased over time, with a slight decline in recent decades. The analyses use anomaly-based methods, which are consistent with long-term climatology and provide a more accurate representation of short-term climate. The analyses are generated using a combination of the Barnes successive-correction method and three-dimensional smoothing splines. These methods are robust and efficient, and have been found to be effective for climate analyses. The accuracy of the analyses has been verified against station observations, with cross-validated error statistics provided for different periods. The RMSE for monthly maximum and minimum temperatures is typically between 0.5 and 1°C, with improvements over current Bureau practice. The RMSE for daily rainfall is slightly higher, with improvements in recent years. The RMSE for vapour pressure is also improved, with analyses showing slightly lower RMSE and MAE compared to previous studies. The analyses are used for monitoring climate change and documenting long-term trends. They are compared with existing Bureau of Meteorology data sets, showing good agreement in annual rainfall averages and trends. The data are also used for climate change studies, with the potential for future improvements through the use of remote sensing and model-derived data.
Reach us at info@study.space