This thesis is a report of the study of a large-scale Numerical Weather Prediction Model. The study investigated the feasibility of applying parallel algorithms to the HIRLAM Model so that it could be implemented on a Transputer Network. A set of partial differential equations which describe the behaviour of the atmosphere are presented and numerical methods are explored. The investigations focused on time critical regions of the sequential programs. From these a criterion for task distribution was devised. Expressions for the computation of speedup and scaled speedup were derived. A special set of test data, extracted from the Analysis of Hurricane Charlie (August 1986) was used as Input data for a 24 hour forecast. We report on the Case Study and how the model predicted the storm over Ireland. For comparison purposes the Model was also run on a VAX 4200 and on a Dell 386-SX PC, with the same data. The execution of the critical program modules was monitored throughout and a table of results is presented.