Impacts of including an ocean component in a dynamically downscaled climate model over south-eastern Australia — Australian Meteorological and Oceanographic Society
  1. Schedule
  2. Afternoon Tea + Poster Session 2
  3. Impacts of including an ocean component in a dynamically downscaled climate model over south-eastern Australia

Impacts of including an ocean component in a dynamically downscaled climate model over south-eastern Australia (#2021)

Stephanie M Downes 1 , Fei Ji 1 , Marine Roge 2 , Alejandro Di Luca 3 , Guillaume SERAZIN 2 , Alex Sen Gupta 2 , Kathleen Beyer 1
  1. NSW Department of Planning, Industry and Environment, Lidcombe, NSW, Australia
  2. Climate Change Research Centre, University of NSW, Sydney, NSW, Australia
  3. Climate Change Research Centre and Centre of Excellence for Climate Extremes, University of NSW, Sydney, NSW, Australia

Whilst global climate models (GCMs) contain atmosphere, ocean, land and cryosphere components, the same is not necessarily true for the regional climate models for which they provide boundary input. Most dynamically downscaled projections for domains within and around Australia include only comprehensive land and atmospheric components and large scale GCM sea surface temperature. However, the role of the ocean in regulating land and atmospheric dynamics extends deep beyond the ocean surface. Here we compare two high-resolution model simulations to explore the impacts of coupling a 3-dimensional dynamic ocean model (Nucleus for European Modelling of the Ocean; NEMOv3.4) to a Weather Research Forecasting (WRF3.5.1) regional model for their representation of south-eastern Australia’s climate processes. The ocean and atmospheric components are simulated at 1/4o resolution for the 1990 – 2009 period. The lateral boundary conditions are driven by ERA-Interim reanalysis data every 6 hours, and the regional model components coupled hourly. We focus on the daily to interannual biases in surface precipitation, temperature and associated extremes when the ocean component is included and excluded, compared with available observations. We evaluate differences between the two experiments (with and without an ocean component), and between the experiments and observations, providing both a model evaluation and an estimate of the added value of including an ocean component. We find maximum temperatures can differ by up to 1oC on seasonal timescales between the two simulations types, and extreme precipitation differs by up to 20%. We provide a summary of key regions where inclusion of an ocean component may enhance biases compared to observations, but overall conclude that including an ocean component performs similarly as a stand-alone atmospheric downscaled configuration, with reduced biases in some regions. Since no tuning was done for reducing biases against observations, we would except the model to perform better with further development.

#amos2020