The current COVID-19 pandemic highlights the need for zoonotic infectious disease surveillance. Avian influenza virus (AIV) poses a significant threat to animal and public health due to its pandemic potential. Virus-contaminated water has been suggested as an important AIV spread mechanism among multiple species. Nevertheless, few studies have characterized the global AIV subtype diversity and distribution in environmental water. Therefore, this study aims to provide an updated descriptive and phylogenetic analysis of AIVs isolated in water samples from high risk-sites for influenza outbreaks (i.e., live bird markets, poultry farms, and wild bird habitats) on a global scale. A total of 234 hemagglutinin (HA) gene sequences of 21 subtypes were reported from nine countries between 2003 – 2020. Fourteen AIV subtypes were solely reported from Asian countries. Most of the viral sequences were obtained in China and Bangladesh with 47.44% and 23.93%, respectively. Likewise, the greatest global AIV subtype diversity was observed in China with twelve subtypes. Live bird markets represented the main sampling site for AIV detection in water samples (64.10%), mostly from poultry cage water. Nevertheless, the highest subtype diversity was observed in water samples from wild bird habitats, especially from the Izumi plain and the Dongting Lake located in Japan and China, respectively. Water from drinking poultry troughs evidenced the greatest subtype diversity in live bird markets, meanwhile, environmental water used by ducks had the highest number of different subtypes in poultry farms. The maximum-likelihood phylogenetic tree showed that some HA sequences were closely related among different poultry/wild bird-related environments from different geographic origins. Therefore, the results suggest that even though the availability of HA gene sequences in public-access databases varies greatly among countries, environmental AIV surveillance represents a useful tool to elucidate potential viral diversity in wild and domestic bird populations.