In forensic death investigations, estimating the postmortem interval (PMI) is critical. An accurate PMI estimate increases the speed and accuracy of identifying the remains by narrowing the time frame in which the death occurred, thus reducing the pool of possible decedents. Cape Town, South Africa has a high level of unnatural death, and due to a burdened death investigation system, many remain unidentified. There has been a tendency to broadly apply quantitative models of decomposition across biogeographically unique circumstances. A prime example is the widespread application of the total body score (TBS)/accumulated degree day (ADD) model developed by Megyesi et al. (2005), later refined by Moffatt et al. (2016). However, the appropriateness of applying a single model to a wide range of locations with unique geography and climates remains in question. The aim of the study was to evaluate and compare the accuracy of Megyesi and Moffatt models for estimating PMI in Cape Town, South Africa. Using pig carcasses, Finaughty established baseline data on the rates and patterns of terrestrial decomposition in summer and winter in two different locations in a forensically significant area of Cape Town. Among the baseline data, Finaughty derived TBS values using the Megyesi criteria. The present study used these values to estimate the ADD per the Megyesi and Moffatt models, which would correspond to an estimated PMI. These estimated values were compared to actual ADD values. Estimates of ADD were inaccurate for both models in winter, and only partially in summer. The Moffatt model was more accurate in earlier decomposition stages, with the Megyesi model more accurate in later decomposition stages. These results indicate the Cape Town environments may contain factors that the two models do not consider, producing inaccurate PMI estimations at various TBS' values. ADD does not depict the entire taphonomic story; the decomposition process appears to be too complex for universal modelling based on a single or narrow suite of variables. Seasonality was an important factor in determining the accuracy of the models, primarily resulting in underestimations of the true PMI values. These findings show the impracticality of applying models developed for- or in one region to any other and support the need to establish regionally-specific equations for estimating PMI in a forensic context. Alternatively, more complex models employing "big data" from a more comprehensive suite of variables which influence the rate and pattern of decay could be developed.