Climate Modelling IV: GCMs Study Case

In parts I to III of previous climate modelling posts, we commented general aspects of climate models, how they work and also the main challenges and difficulties facing modellers.  In this fourth and final part of this series of posts, I want to show you an interesting application of Global Circulation Models (GCMs).

In Asia, the Mekong River (see Figure below), which is the 12th longest river in the world with a length of 4,350 kilometres, represents a fundamental source of resources for different economic activities in Southeast Asia, such as agriculture and fishing. Another fundamental aspect to consider is the strong intervention of the river with large hydraulic projects, such as dams for hydropower generation. Due to the river represents a fundamental source of resources for the different cities and human settlements located in the basin, interesting studies have been carried out in order to assess the potential effects of climate change in the Mekong River.

Mekong River 

Kingstonet al. (2011) explored the potential effects of global environmental change on water resources in the Mekong river basin and they also proposed a method to estimate the uncertainty of this prediction. For this purpose, they integrated GCMs with hydrological modelling. This method represents a really interesting approach to predict potential effect of climate change on the environment; different disciplines working together in order to create more sophisticated tools.

In addition, another significant contribution was to consider different climate scenarios given by a GCMs structure. Basically, they built different GCM scenarios working as a base for a hydrological model (SLURP) of the study area. The fundamental strength of this methodology was to observe modelled responses of the freshwater behaviour due to different climate scenarios. Specifically, they generated different global warming scenarios between 0.5 and 6.0°C. For the 2°C scenario, they used seven different GCMs. Although there were clear similarities among the GCMs, it was useful to use different models structures to analyse the influence of these structures in the final projection uncertainty.

Mekong River 

According to the main purpose of the study, the effort of researchers was mainly focused on the understanding and measurement of the model uncertainty. An interesting result was to find a significant effect on the uncertainty due to the differences in precipitation projections given by the different GCMs, even for the same scenario. As we could see in the previous post, one of the most relevant challenges facing GCMs modellers is to improve precipitation and water cycle simulation. This was also emphasized by other researchers in a study developed with multiple GCMSs in India, where were found severe uncertainties in future rainfall estimations. Conversely, a proper consistency was found in GCMs projection for both snowmelt season and evapotranspiration in the Mekong river basin.

Finally, the most significant outcome was related to the strong dependence of hydrological behaviour in the basin (discharge) with potential changes in precipitation patterns. Due to the seven GCMs applied and considering the different precipitation estimations of these, for the 2°C scenario the parameters of the hydrological model were numerically estimated with an uncertainty between -2.0 – 2.0% and the discharge pattern was suitably represented by the model (see Figures below).

As we can see, the obtained results ratify the proposed methodology of applying different GCMs in order to obtain a suitable estimation of the final uncertainty, minimizing this value and finally obtaining better projections. 

Mekong River, Parameter Uncertainty for HadCM3 (GCM) Model, 2°C Scenario

Mekong River, Monthly Discharge for Base – Line and Seven GCMs Applied, 2°C Scenario