Using synthetic climate datasets for geotechnical and geoenvironmental design problems

Abstract

Location-specific climate datasets are required for the design and evaluation of a number of civil engineering projects. It requires huge effort to compile a multi-year quality-controlled climate dataset. In this paper, a method of generating simulated daily climate variables of interest from readily available climate normal using the general-purpose weather generator SIMETAW is presented. The accuracy of this method is assessed by comparing the climate datasets generated using SIMETAW with the recorded historical climate datasets for nine different sites across Canada with climates ranging from semi-arid to pre-humid. This comparison was done using visual presentations as well as statistical analyses of the two datasets. It was found that the multi-year daily climate datasets generated by SIMETAW using just 12 monthly climate normal values are fairly similar to the recorded historical climate datasets. The usefulness of SIMETAW-generated climate datasets was demonstrated by using them in numerical simulations of three different design problems, namely, infiltration into soils, swelling potential of an expansive soil, and soil cover design. From the results of these numerical simulations, it is concluded that the SIMETAW-generated multi-year daily climate datasets are satisfactory for use in the geotechnical and geoenvironmental problems of the kind simulated herein.