International Journal of Applied Earth Observation and Geoinformation
An early forecasting method for the drift path of green tides: A case study in the Yellow Sea, China
Introduction
A new type of marine ecological disaster caused by the excessive and large-scale proliferation of marine green algae is known as a “green tide.” Green tides represent a widespread phenomenon in estuaries, lagoons, inner bays, and other waters worldwide. Since the 1970s, the frequency and regions affected by green tides have shown an increasing trend. Green tide disasters had been reported in the United States and Canada, while the worst disaster occurred on the coast of Brittany in France (Charlier et al. 2007; Teichberg et al. 2010). Since 2008, large-scale green tides have occurred in the Yellow Sea for eight consecutive years (Keesing et al. 2011; Liu et al. 2013a), while greatly affected aquaculture, coastal tourism, maritime transport, maritime events, and other related industries. Green tides also caused major economic losses and social effects on coastal cities in Jiangsu and Shandong provinces (Ye et al., 2011). Therefore, research on the sources of green tides, characteristics of the variation in the distribution, and the mechanism involved in their occurrence in the Yellow Sea is of great importance in disaster prevention, environmental protection, and economic development.
Many studies have been conducted on green tides in the Yellow Sea, but the location of green tide generation remains controversial. It is widely believed that South Yellow Sea serves as the source of green tides (Pang et al., 2010). Satellite imagery and model results have shown that green tides formed in the offshore region of Jiangsu Province and drifted into the North Yellow Sea under the influence of monsoons and ocean currents (Feng et al. 2010; Keesing et al. 2011; Cui et al. 2012; Qiao et al. 2009; Ho et al., 2011). Recent research studies have suggested that the laver culture in Rudong County, Jiangsu Province, provided the most direct and sufficient initial biomass for green tides, and also was the primary cause of the continuous blooming of green tides in the South Yellow Sea since 2007 (Li et al., 2014; Wang et al., 2015). During April and June, many algal colonies are often blown into the sea and grew rapidly in a suitable environment, and finally formed the green tides.
Based on remote sensing monitoring, many scientists have discussed the spatial distribution of green tides and their paths as they drift into the Yellow Sea on an interannual timescale (Wu et al., 2013; Huang et al., 2014). During the blooming and duration of green tides, satellite data can be used to estimate the distribution and cover area of green tides accurately. Since the usefulness of remote sensing inversion is limited on the ocean surface, most green tides have been in a striped distribution and started to gather and grow quickly soon after they were first discovered. However, no or only a little algae floats on the surface during the early growth stage of a green tide, so remote sensing inversion cannot capture the drift path of a green tide. Huang et al. (2014) and Guo et al. (2016) documented the distribution of green tides when they were first discovered with satellite remote sensing during 2008–2013 and 2008–2015, respectively. These studies showed that the initial discovery of green tides mainly occurred in mid- or late-May or in early June; the locations were mainly in the offshore regions in Yancheng city and Rudong County of Jiangsu Province. Significant interannual variation occurred in the timing and location of green tides.
The above studies mainly analyzed the features of the variation in green tide drift paths from the blooming to the extinction periods. Few studies have analyzed the early drift path of green tides from mid-April to mid- and late-May. Predictive studies on early drift path play an essential role in preventing and controlling green tide disasters. Therefore, this paper focused on predicting the early drift paths of green tides before they gathered and floated on the sea surface. We thought wind could not drag an underwater green tide directly, so we believed the early drift paths of green tides were mainly controlled by oceanic circulation. Because the local wind has the strongest effect on the summer circulation in the South Yellow Sea, this study was based on numerical simulation and artificial neural network algorithm methods. We used a climate variability index as the prediction factor and the interannual variation of the local wind field as the intermediary factor; then we established a prediction method for the early drift paths of green tides.
Section snippets
Methods and data
This study mainly consisted of two parts. First, numerical experiments were used to determine the most important dynamic factors that control the early drift path of green tides and their interannual variation. Second, based on an artificial neural network algorithm, the early drift paths of green tides were predicted with the climate variability indicators.
Based on the Regional Ocean Modeling System (ROMS), a 3-dimensional marine dynamical model was established to simulate the drifting of
Numerical experiments
A numerical experiment to test the sensitivity of parameters for the factors influencing the drift path of green tides was first conducted (Exp. A), including the tidal current, wind field, and sea temperature. Using the Lagrange method, a float drift that represents the path of green tide algae was applied to check the sensitivity of the drift path of green tides to factors that influence them such as tidal current, wind field, and sea temperature. The experiments mainly included four groups,
Discussion
According to the results of Exp. A and previous studies, the monsoon wind is the main controlling factor of current field in southern Yellow Sea and thus the main controlling factor of the early drift path of green tides during summer, so the prediction of green tides’ early drift path is in a sense the forecasting of the interannual variation of summer monsoon. There are reasons to believe that the long-term variations of local monsoon wind are controlled by and lag behind the climatological
Conclusions
Using numerical experiments, this study revealed the main factors controlling the early drift path of green tides and characteristics of their interannual variation in the Yellow Sea. Second, a nonlinear prediction model was established by the artificial neural network method. This model was used to predict the early drift path of green tides in the next year with six climate variability indicators such as Nino3.4 in the current year. The early drift paths of green tides during 2012–2015 were
Acknowledgements
This work was supported by the National Key Research and Development Program of China (Grant No. 2016YFC1402000), National Natural Science Foundation of China (Grant Nos. 41476018, 41421005, and U1406401), the Public Science and Technology Research Funds Projects of the Ocean (Grant No. 201205010) and the CAS Strategic Priority Project (Grant No. XDA19060202). This work was also supported by the High-Performance Computing Center, IOCAS.
References (19)
- et al.
Estimation of corn yield using multi-temporal optical and radar satellite data and artificial neural networks
Int. J. Appl. Earth Observ. Geoinf.
(2017) - et al.
Inter- and intra-annual patterns of Ulva prolifera green tides in the Yellow Sea during 2007–2009, their origin and relationship to the expansion of coastal seaweed aquaculture in China
Mar. Pollut. Bull.
(2011) - et al.
Tempo-spatial distribution and species diversity of green algae micro-propagules in the Yellow Sea during the large-scale green tide development
Harmful Algae
(2014) - et al.
The world’s largest macroalgal bloom in the Yellow Sea, China: formation and implications
Estuar. Coast. Shelf Sci.
(2013) - et al.
Tracking the algal origin of the Ulva, bloom in the Yellow Sea by a combination of molecular, morphological and physiological analyses
Mar. Environ. Res.
(2010) - et al.
Banded structure of drifting macroalgae
Mar. Pollut. Bull.
(2009) - et al.
Application of artificial neural networks for the soil moisture retrieval from active and passive microwave spaceborne sensors
Int. J. Appl. Earth Observ. Geoinf.
(2016) - et al.
Green tides on the brittany coasts
Us/Eu Baltic International Symposium. IEEE
(2007) - et al.
Satellite monitoring of massive green macroalgae bloom GMB: imaging ability comparison of multi-source data and drifting velocity estimation
Int. J. Remote Sens.
(2012)
Cited by (14)
Estimating Ulva prolifera green tides of the Yellow Sea through ConvLSTM data fusion
2023, Environmental PollutionMonitoring, modeling and projection of harmful algal blooms in China
2022, Harmful AlgaeCitation Excerpt :Some HAB events are believed to be associated with the ENSO (El Niño-Southern Oscillation). When forecasting the trajectory of green tides, the Nino3.4 index is used as one climate index to feed the artificial neural network algorithm (Hu et al., 2018). Observation and modeling should be tightly coupled to improve the forecast of HAB events.
Drift path of green tide and the impact of typhoon “Chan-hom” in the Chinese Yellow Sea based on GOCI images in 2015
2020, Ecological InformaticsCitation Excerpt :Therefore, the drift speed and direction tracking of green tide have always been the focus of green tide disaster prevention and mitigation. The outbreak of the Yellow Sea green tide has the characteristics of wide distribution area, large coverage and long duration(Hu et al., 2018; Ji et al., 2015). Traditional investigation, sampling and measurement methods can obtain the causes of green tides and changes in seawater biomass in detail, but require a lot of manpower and resources(Liu et al., 2015).
Characteristics and influence of green tide drift and dissipation in Shandong Rongcheng coastal water based on remote sensing
2019, Estuarine, Coastal and Shelf ScienceCitation Excerpt :Few detailed studies have been conducted on the drift trajectory, drift speed, dissipation characteristics, and distribution density of green tides in the Rongcheng area. As found in this study, after reaching the Rongcheng coastal water, green tides mainly drift towards the northeast, which is slightly different from the integral drift direction of green tides south of the Rongcheng coastal water (Xing et al., 2011; Bao et al., 2015), although the main driving factors are still the wind and sea surface circulation (Qiao et al., 2011; Hu et al., 2018). Xing et al. (2018) analyzed the daily average drift speed of green tides in the whole Yellow Sea area, between May and June 2016, and found a velocity of about 4.5 km/d (Xing et al., 2018), which is similar to the average northward drift speed of green tides in the Rongcheng coastal water.
Unprecedent green macroalgae bloom: mechanism and implication to disaster prediction and prevention
2023, International Journal of Digital Earth