Among the members of the genus Flavivirus are several important human pathogens, including the dengue (DEN) and Japanese encephalitis (JE) viruses. From the analysis of gene sequence data of samples of these virus populations it is possible to infer phylogenetic relationships, which in turn can yield important epidemiological information, including their demographic history in humans. In this study, we use a recently developed method, based on coalescent theory, to infer the population dynamics of a variety of mosquito-borne flaviviruses. Our study involves the testing of alternative hypotheses, the estimation of confidence intervals around demographic model parameter values, and the placing of the maximum likelihood (ML) demographic model into a "real time" epidemiological history. We reveal that all the Flavivirus populations studied are growing at an exponential rate, with the rates of population growth of dengue virus serotypes 2 and 3 increasing rapidly in the recent past, and that of Japanese encephalitis virus changing from constant population size to exponential growth within the last century. We therefore demonstrate that the use of these coalescent methods may be extremely valuable in monitoring responses to interventions such as vaccination or vector control.