Model precipitation bias: ACCESS 1.3 AMIP run – GPCP observations (mm/day)
by Matt Wheeler
Centre for Australian Weather and Climate Research (CAWCR), Bureau of Meteorology
A strong scientific collaboration is continuing to develop between the ARCCSS, Bureau of Meteorology, CSIRO, and international partners on the unique weather and climate science issues of the Maritime Continent (MC).
The MC is the name given to the region of many islands to the north of Australia straddling the equator between about 100° to 160°E. It comprises a mix of both large and small islands located within the climatological Indo-Pacific warm pool of high sea surface temperatures – the ‘boiler box’ of the globe, and largest centre of atmospheric rising motion. It therefore has a strong influence on both local and global weather and climate, and is an essential component of our prediction models.
The issues of the MC region are its frequent convective storms, strong diurnal cycle, complex topography, multiple interacting scales of variability, and its location at the interface (in both the atmosphere and ocean) between the Pacific and Indian oceans. These issues require special attention for progress to be made in our weather and climate modelling and prediction efforts.
Of particular concern are biases in the MC region that are present in current models. For example, like many models, the Bureau’s ACCESS and POAMA models exhibit large precipitation biases across the MC. These biases directly impact predictions of tropical synoptic weather (e.g. tropical depressions and cyclones), multi-week predictions of the progression of the Madden-Julian Oscillation (MJO) across northern Australia, seasonal predictions of the Indian Ocean Dipole (IOD), and indirectly impact predictions over southern Australia through altered teleconnection patterns. Addressing these biases would improve the quality of our short range, multi-week, seasonal and climate change predictions.
To address the issues of the MC, a national joint initiative between CAWCR and ARCCSS began in 2013 with the aim to improve the models. A comprehensive plan was written that advocated the formation of an integrated research team and the harnessing of existing effort as well as additional resources to fill identified gaps. A kick-off workshop was held in December 2013, with regular video meetings held since that time. Early results have pointed to the sensitivity of the model bias to the details of the model convective parameterization and microphysics. Results have also shown that the many observation-based precipitation products (satellite, radar, in-situ) for the region exhibit significant differences.
In parallel with the national interest in the MC, international interest has grown with the development of the proposed Years of the Maritime Continent (YMC) field program, aiming for intensive land, ship and aircraft observations during 2017-19. Australian participation is sought in the form of our new research vessel, the RV Investigator, with its advanced dual polarisation meteorological radar[MCW1] . Of particular interest are detailed observations of clouds, precipitation, and associated dynamics in an oceanic location that experiences a strong diurnal cycle under the influence of a nearby island. Such observations should ultimately lead to an improved representation of the land and sea breeze circulations and other sub-grid scale phenomena in our global models, thereby reducing the mean biases, and improving predictions.
For more information, contact: Matt Wheeler