Step 5: Evaluate Climate Model Data
Checking the reliability of the model data for your region helps to establish confidence ratings in climate projections. This is based on the premise that the closer the model simulation is to the observed climate, the closer the enhanced greenhouse response of the simulation will be to the real-world response.
There are a number of ways to evaluate climate modelled data. In most cases, this may simply be a case of comparing the modelled data climatology against the observed data. The climatology metrics could include the long-term average, standard deviation (which represents year-to-year variability) or annual or seasonal cycle of the relevant climate variables (e.g. temperature and rainfall).
It is useful to conduct a literature review because climate modelling over the Asian region involves specific challenges. These include:
- representation of regional processes (such as tropical convection, monsoons, and tropical cyclones)
- complex land use
- significant orography (including the Himalayas and the Tibetan Plateau)
- extra effects from snow and snow melt.
Monsoon behavior in the region is difficult to simulate accurately in global climate models, partly because the topographic interactions with the atmospheric flow and the air-sea interaction process are not adequately represented in the simulations. The complex topography also affects the ability to properly represent relevant physical processes within the models, and global climate models still have problems fully capturing the year-to-year variability of the ocean temperatures related to the El Niño–Southern Oscillation.
When resources are not constrained and there is a real need to conduct a more sophisticated evaluation, it is also useful to assess how the model represents the observed characteristics of large-scale climate features such as the El Niño–Southern Oscillation and monsoon, or how the model reproduces the observed relationship between climate variables (e.g. rainfall) and large-scale climate features (e.g. monsoon).
CASE STUDY EXAMPLES