(c) Uncertainty and variability of models: Up to date, a large part of research findings and our knowledge on reflective materials is derived based simulation studies. It is important to bear in mind that significant variability exists in models? setup and assumptions, which implies that different models could yield quantitatively different results. Take thermal comfort for an example, without consideration of AM580 effect, model predicts an improvement in outdoor thermal comfort of pedestrians by installation of reflective materials. After accounting for reflected radiation, model suggests a negative net impact on thermal comfort in outdoor urban environment. In addition, the driver of built environment simulations, viz. forcing from climate models also subjects to considerable uncertainties, to the extent that different global models can disagree on the future trend of climate changes expected in particular regions  and . As part of the IPCC Fourth Assessment (IPCC AR4) experiment, the ability of 22 models to predict South Asian summer monsoon precipitation has been examined. While 19 out of 22 models are able to capture the maximum precipitation, predicted total rainfall varies from 500 mm to 900 mm .