Abstract
The multistep-ahead prediction of chlorophyll provides an effective means for early warning of red tide. However, since multistep-ahead forecasting presents challenges, such as vague interactive relationships among ocean factors, long-term dependence modeling, and accumulative errors, existing methods mostly concentrate on the current time or one-step-ahead forecasting. In this paper, a hierarchical multistep-ahead forecasting model spatial-temporal attention network(STAN), which integrates the spatial context extractor network(SCE-net), long short-term memory network(LSTM), and the temporal attention mechanism, is proposed for the prediction of chlorophyll. In STAN, the input layer utilizes SCE-net to excavate relationships among ocean factors and generate high-level semantic via embedding factors into a continuous low-dimensional space. The middle layer applies LSTM to build the long-term dependencies of corresponding semantic representations. The output layer uses another LSTM with temporal attention to reduce accumulative errors and maintain temporal continuity. The attention can assign different weights to the middle layer’s hidden state and generate a context vector. Then the context vector and the final predicted value are considered as the current input for better forecasting. The buoy observation data of the Xiamen coastal area monitored in 2009–2011 is used to verify the efficiency of STAN. Experimental results prove that STAN outperforms the state-of-the-art methods of multistep-ahead prediction. When using 7 observation steps to forecast 15 steps, the MAPE of STAN is 0.3209, and the MAE is 0.1 lower than the values of the baselines approaches.
Similar content being viewed by others
References
Diaz RE, Friedman MA, Jin D, Beet A, Kirkpatrick B, Reich A, Kirkpatrick G, Ullmann SG, Fleming LE, Hoagland P (2019) Neurological illnesses associated with Florida red tide (Karenia brevis) blooms. Harmful Algae 82:73–81
Davidson K, Anderson DM, Mateus M, Reguera B, Silke J, Sourisseau M, Maguire J (2016) Forecasting the risk of harmful algal blooms. Harmful Algae 53(mar.):1–7
Zohdi E, Abbaspour M (2019) Harmful algal blooms (red tide): a review of causes, impacts and approaches to monitoring and prediction. Int J Environ Sci Technol 16(3):1789–1806
Tian W, Liao Z, Zhang J (2017) An optimization of artificial neural network model for predicting chlorophyll dynamics. Ecol Model 364:42–52
Bui M-H, Pham T-L, Dao T-S (2017) Prediction of cyanobacterial blooms in the Dau Tieng reservoir using an artificial neural network. Mar Freshw Res 68(11):2070–2080
Guallar C, Delgado M, Diogene J, Fernandez-Tejedor M (2016) Artificial neural network approach to population dynamics of harmful algal blooms in Alfacs Bay (NW Mediterranean): case studies of Karlodinium and pseudo-nitzschia. Ecol Model 338:37–50
Chen Q, Guan T, Yun L, Li R, Recknagel F (2015) Online forecasting chlorophyll a concentrations by an auto-regressive integrated moving average model: feasibilities and potentials. Harmful Algae 43:58–65
Xiao X, He J, Huang H, Miller TR, Christakos G, Reichwaldt ES, Ghadouani A, Lin S, Xu X, Shi J (2017) A novel single-parameter approach for forecasting algal blooms. Water Res 108:222–231
Qin M, Li Z, Du Z (2017) Red tide time series forecasting by combining ARIMA and deep belief network. Knowl-Based Syst 125:39–52
Lu F, Chen Z, Liu W, Shao H (2016) Modeling chlorophyll-a concentrations using an artificial neural network for precisely eco-restoring lake basin. Ecol Eng 95:422–429
Taieb SB, Atiya AF (2015) A bias and variance analysis for multistep-ahead time series forecasting. IEEE Trans Neural Netw Learn Syst 27(1):62–76
Doucoure B, Agbossou K, Cardenas A (2016) Time series prediction using artificial wavelet neural network and multi-resolution analysis: application to wind speed data. Renew Energy 92:202–211
Zhou T, Gao S, Wang J, Chu C, Todo Y, Tang Z (2016) Financial time series prediction using a dendritic neuron model. Knowl-Based Syst 105:214–224
Bahdanau D, Cho K, Bengio Y (2014) Neural machine translation by jointly learning to align and translate. CoRR,abs/1409.0473
Zhang B, Xiong D, Su J (2018) Neural machine translation with deep attention. IEEE Trans Pattern Anal Mach Intell 42(1):154–163
Wu X, Du Z, Guo Y, Fujita H (2019) Hierarchical attention based long short-term memory for Chinese lyric generation. Appl Intell 49(1):44–52
Wen P, Yuan W, Qin Q, Sang S, Zhang Z (2020) Neural attention model for recommendation based on factorization machines. Appl Intell 1–16
Chang L, Chen W, Huang J, Bin C, Wang W (2020) Exploiting multi-attention network with contextual influence for point-of-interest recommendation. Appl Intell 1–14
Xie J, Zhang J, Yu J, Xu L (2019) An adaptive Scale Sea surface temperature predicting method based on deep learning with attention mechanism. IEEE Geosci Remote Sens Lett 17(5):740–744
Du S, Li T, Yang Y, Horng S-J (2020) Multivariate time series forecasting via attention-based encoder–decoder framework. Neurocomputing 388:269–279
Qin Y, Song D, Cheng H, Cheng W, & Cottrell G (2017) A Dual-stage attention-based recurrent neural network for time series prediction. In IJCAI 2627–2633
Shih S-Y, Sun F-K, Lee H-y (2019) Temporal pattern attention for multivariate time series forecasting. Mach Learn 108(8–9):1421–1441
Cui Q, Wu S, Huang Y, Wang L (2019) A hierarchical contextual attention-based network for sequential recommendation. Neurocomputing 358:141–149
Tao Y, Ma L, Zhang W, Liu J, Liu W, Du Q (2018) Hierarchical attention-based recurrent highway networks for time series prediction. arXiv preprint arXiv:180600685
Wang L, Zhang T, Wang X, Jin X, Xu J, Yu J, Zhang H, Zhao Z (2019) An approach of improved multivariate timing-random deep belief net modelling for algal bloom prediction. Biosyst Eng 177:130–138
Nazeer M, Wong MS, Nichol JE (2017) A new approach for the estimation of phytoplankton cell counts associated with algal blooms. Sci Total Environ 590:125–138
McGowan JA, Deyle ER, Ye H, Carter ML, Perretti CT, Seger KD, de Verneil A, Sugihara G (2017) Predicting coastal algal blooms in southern California. Ecology 98(5):1419–1433
Park Y, Pyo J, Kwon YS, Cha Y, Lee H, Kang T, Cho KH (2017) Evaluating physico-chemical influences on cyanobacterial blooms using hyperspectral images in inland water, Korea. Water Res 126:319–328
Liping W, Binghui Z (2013) Prediction of chlorophyll-a in the Daning River of three gorges reservoir by principal component scores in multiple linear regression models. Water Sci Technol 67(5):1150–1158
Wu N, Huang J, Schmalz B, Fohrer N (2014) Modeling daily chlorophyll a dynamics in a German lowland river using artificial neural networks and multiple linear regression approaches. Limnology 15(1):47–56
Park Y, Cho KH, Park J, Cha SM, Kim JH (2015) Development of early-warning protocol for predicting chlorophyll-a concentration using machine learning models in freshwater and estuarine reservoirs, Korea. Sci Total Environ 502:31–41
Xiaobo L, Fei D, Guojian H, Jingling L (2014) Use of PCA-RBF model for prediction of chlorophyll-a in Yuqiao reservoir in the Haihe River basin, China. Water Sci Technol Water Supply 14(1):73–80
Rajaee T, Boroumand A (2015) Forecasting of chlorophyll-a concentrations in South San Francisco Bay using five different models. Appl Ocean Res 53:208–217
Kim Y, Shin HS, Plummer JD (2014) A wavelet-based autoregressive fuzzy model for forecasting algal blooms. Environ Model Softw 62:1–10
Ye L, Cai Q, Zhang M, Tan L (2014) Real-time observation, early warning and forecasting phytoplankton blooms by integrating in situ automated online sondes and hybrid evolutionary algorithms. Ecol Inform 22:44–51
Yu J, Yan X (2019) Whole process monitoring based on unstable neuron output information in hidden layers of deep belief network. IEEE Trans Cyber PP(99):1–10
Zhang F, Wang Y, Cao M, Sun X, Du Z, Liu R, Ye X (2016) Deep-learning-based approach for prediction of algal blooms. Sustainability 8(10):1060
Yu J, Yan X (2019) Active features extracted by deep belief network for process monitoring. ISA Trans 84:247–261
Cho H, Choi U, Park H (2018) Deep learning application to time-series prediction of daily chlorophyll-a concentration. WIT Trans Ecol Environ 215:157–163
Shin Y, Kim T, Hong S, Lee S, Lee E, Hong S, Lee C, Kim T, Park MS, Park J (2020) Prediction of chlorophyll-a concentrations in the Nakdong River using machine learning methods. Water 12(6):1822
Kao I-F, Zhou Y, Chang L-C, Chang F-J (2020) Exploring a Long Short-Term Memory based Encoder-Decoder framework for multi-step-ahead flood forecasting. J Hydrol 583:124631
Acknowledgments
This work is supported by National Program on Key Research Project of China (2016YFC1401900) and Open fund of the Key Laboratory of Digital Ocean, State Oceanic Administration, China (B201801030).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
He, X., Shi, S., Geng, X. et al. Spatial-temporal attention network for multistep-ahead forecasting of chlorophyll. Appl Intell 51, 4381–4393 (2021). https://doi.org/10.1007/s10489-020-02143-y
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-020-02143-y