Pyrodinium bahamense (var. compressum) has been the only dinoflagellate species that has caused major public health and economic problems in the Southeast Asian region for more than 2 decades now. Blooms of the organism have been reported in Malaysia, Brunei Darussalam, the Philippines and Indonesia. The ASEAN-Canada Red Tide Network has recorded 31 blooms of the organism in 26 areas since 1976 when it first occurred in Sabah, Malaysia. As of 1999, the most hardly hit country has been the Philippines which has the greatest number of areas affected (16) and highest number of PSP cases (about 1,990). Cysts of the P. bahamense var. compressum have been reported in three bays in the Philippines and in one bay in Indonesia; but it is very likely, however, that they could be present in adjacent bodies of water/countries where sampling and analysis of sediments have not been done. Research on sediment-dated cores should be undertaken in addition to those that relate to the role of cyst beds in the organism\'s bloom dynamics.

Most of the data and information useful to understand Pyrodinium bloom dynamics have come from harmful/toxic algal monitoring and research that have developed to different degrees in the various countries in the region affected by the organism\'s bloom. Molecular genetic studies to distinguish population between and within the Atlantic and Indo-West Pacific areas need to be undertaken. Such studies will be helpful in elucidating questions relating to the distinction between P. bahamense var. compressum and P. bahamense var.bahamense; and the origin and spread of P. bahamense blooms, particularly in the Indo-West Pacific region.Regional collaborative research and monitoring efforts can help harmonize local data sets and ensure their quality and availability for comparative analysis and modeling. Temporal patterns of the blooms at local and regional scales, possible signals and trends in the occurrence/recurrence and spread of Pyrodinium blooms and possible role of climactic changes as El Niño event can then be detected; and predictive models developed.