Abstract:

Red tide caused by planktonic algal bloom has been causing damages to aquatic ecosystems as well as economy. It is necessary to make near future forecast of chlorophyll concentration and to understand its drivers and mechanism. Previously, several time series analyses have been conducted. However, the nature of chlorophyll dynamics, namely, whether linear and stochastic process or nonlinear and deterministic process dominates, and how effective dynamical theory-based, nonlinear time series methods have not been well investigated in Hong Kong waters., Here, two sets of monthly time series of chlorophyll concentration in Hong Kong waters from 2002 to 2022 were collected, including in situ measurements by Environment Protection Department (EPD) and remote sensing data from the MODIS sensor of NASA. First, we evaluated the effectiveness of two time series analysis methods with different assumptions: Auto Regressive Integrated Moving Average (ARIMA; a linear method) and Empirical Dynamic Modeling (EDM; a nonlinear method). The nonlinear method outperformed the linear method for both data, suggesting that nonlinear and deterministic mechanism dominates in Hong Kong waters. To further understand factors affecting the chlorophyll dynamics, we applied a nonlinear causal detection method to the chlorophyll time series and it reveals significant causal effect of temperature, salinity, etc.  Additionally, the south or southeast data points in the MODIS and the EPD data exert strong causal effects on neighboring areas, with this influence declining to low levels within a distance of 40 km, suggesting that a few monitoring stations are sufficient to represent the south/southeast areas. The consistent patterns observed across both EPD and MODIS data suggest that satellite data is relatively reliable, even in coastal areas. In summary, this study highlights the importance of using nonlinear methods when examining chlorophyll dynamics in Hong Kong waters. Future efforts will focus on turning parameters of nonlinear models and forecasting the chlorophyll dynamics in Hong Kong waters.

All Are Welcome!