Hydrodynamic plots of tight formations in the Deep Basin of the Western CanadaSedimentary Basin (WCSB) indicate the lack of a water leg and the presence of water on top of gas; quite unconventional. Furthermore, experience indicates that in many instances properties such as matrix and fracture density in these tight formations cannot be modeled using conventional statistical distributions. Widely used approaches such as normal, geometric and fractal distributions do not reproduce accurately the behaviour of such properties in many cases. Thus it could be said that the unconventional fluids distribution meets the unconventional rocks distribution in unconventional formations. These occurrences have led to the development of the research presented in this work. Original contributions are as follows: 1) Reservoir simulation covering over 1.5 million years that corroborates the presence of a block of water on top of gas in the tight gas Nikanassin formation of the WCSB.2) Implementation of a new methodology to populate matrix and fracture density properties in reservoir simulation with the use of a Variable Shape Distribution (VSD) model.The question is if it is reasonable to think that gas can be trapped for hundreds of years and more by an updip water block. A reservoir simulation model has been created as indicated in item 1 above in order to answer this question. The water seal is shown to be the result of very low permeability and high capillary pressures, properties that are generally found in tight gas formations. The model is defined by a geometry that mimics the geologic interpretation of the Nikanassin Basin Center Gas Accumulation (BCGA) in the Western Canada Sedimentary Basin (WCSB) and rock properties that provide a good representation of the real behavior of the reservoir in the Deep Basin of Alberta. Reservoir simulation has been used historically to model in the order of 10 to 40 years of production. But to our iii knowledge this is the first reservoir simulation that goes beyond 1.5 million years. The results lead to the conclusion that updip water blocks provide good seals in the Deep Basin.The new methodology mentioned above in item 2 introduces an extension of the VSD approach for reservoir simulation purposes. In the original methodology when matching input information in the VSD equation, each data point is assigned an integer according to its ranking (Nt) in the overall sample population. If these integers are used in a normal score transformation for reservoir simulation, the hard data (for example permeability) will not be honoured. As a result, in order to honour the hard data, the integers have been replaced with real numbers found by calculating the values that produce the exact data points when input into the VSD equation. Since this equation has an exponential form it cannot be solved directly for Nt and it is necessary to perform an iterative process to find the exact values.The new methodology is demonstrated with data from the WCSB tight gas Nikanassin Group and Cadomin form...