Abstract:
This paper applies fractional integration methods to obtain evidence on ocean acidification, namely the decrease in the pH level in the Earth's oceans, using the annual Hawaii Ocean Time-series Station ALOHA series as well as the logged one for the period 1985-2024. The chosen modelling framework is more general than standard ones based on the I(0) versus I(1) dichotomy and sheds light on the long memory and persistence properties, as well as on the possible presence of trends, in the pH Level in the Earth's oceans. The results indicate that the series exhibit a negative and significant time trend; however, whether or not the null hypothesis of a unit root is rejected depends on the assumption made about the errors. The key finding (when the errors are not incorrectly specified as I(0) processes) is the presence of long memory, which implies that the effects of shocks are long-lived, regardless of whether or not mean reversion occurs. Moreover, the recursive analysis indicates that both the degree of persistence and the downward trend in the pH level have increased over time. This evidence points to the urgent need for decisive policies to address the issue of ocean acidification and protect marine life and biodiversity.