基于人工神经网络构建的赤潮短期预报模型及应用-海洋预报

基于人工神经网络构建的赤潮短期预报模型及应用

单位：1. 福建省海洋预报台, 福建福州 350003;
2. 厦门大学海洋与地球学院, 福建厦门 361102;
3. 南方海洋科学与工程广东省实验室, 广东珠海 519000

分类号：X55

出版年·卷·期（页码）：2023·40·第二期（67-76）

摘要：

利用大数据赤潮预报方法，基于福建沿岸24个生态浮标和5个大浮标历史数据及实时监测数据，采用人工神经网络实现福建沿岸赤潮的业务化预报。赤潮短期预报模型由误差反向传播网络（BP）和径向基神经网络（RBF）构成，结合福建沿岸所有赤潮事件的高频采样数据样本，每天可算出480个预报结果，最后对预报结果进行发生概率等级判断，最终实现福建沿岸10个赤潮监测区赤潮发生概率等级的业务化预报。赤潮短期预报模型成功预报出2019年5月下旬福建北部发生的多起赤潮事件，2019年和2020年24 h时效的赤潮预报结果正确率达到95%和99%，赤潮识别率达到60%和55%。

Based on the historical data and real-time monitoring data of 24 ecological buoys and 5 large buoys along the coast of Fujian, this study uses artificial neural network to investigate the operational forecasting of red tide along the coast of Fujian. The short-term red tide forecasting model is composed of Error Back-Propagation (BP) Neural Network and Radical Basis Function Neural Network (RBF). According to the high-frequency sampling data of all the red tide events along the coast of Fujian, 480 prediction results can be calculated every day. Finally, through judgments on the prediction results, the operational forecasting of red tide occurrence probability level can be obtained in 10 red tide monitoring areas along the coast of Fujian. The red tide short-term forecasting model successfully predicts many red tide events in northern Fujian in late May 2019. The 24-hour Probability of Correct Result (POCR) reaches to 95% (2019) and 99% (2020), and the 24-hour Probability of Detection (POD) reaches to 60% (2019) and 55% (2020).

参考文献：

[1] 赵冬至, 马志华, 关春江, 等. 中国典型海域赤潮灾害发生规律[M]. 北京: 海洋出版社, 2010. ZHAO D Z, MA Z H, GUAN C J, et al. The occurrence pattern of red tide in typical sea areas of China[M]. Beijing: China Ocean Press, 2010.
[2] 高波, 邵爱杰. 我国近海赤潮灾害发生特征、机理及防治对策研究[J]. 海洋预报, 2011, 28(2): 68-77. GAO B, SHAO A J. Study on characteristics, mechanisms and strategies of harmful algal blooms in China coastal waters[J]. Marine Forecasts, 2011, 28(2): 68-77.
[3] FLEMING L E, KIRKPATRICK B, BACKER L C, et al. Review of Florida red tide and human health effects[J]. Harmful Algae, 2011, 10(2): 224-233.
[4] WYATT T, ZINGONE A. Population dynamics of red tide dinoflagellates[J]. Deep Sea Research Part II: Topical Studies in Oceanography, 2014, 101: 231-236.
[5] 矫晓阳. 叶绿素α预报赤潮原理探索[J]. 海洋预报, 2004, 21(2): 56-63. JIAO X Y. The exploration on the principle of red tide forecasting with chlorophyll α[J]. Marine Forecasts, 2004, 21(2): 56-63.
[6] ZHANG J F, BAI Y P, YU J L, et al. Forecast of red tide in the South China Sea by using the variation trend of hydrological and meteorological factors[J]. Marine Science Bulletin, 2006, 8(2): 60- 74.
[7] 李星. 海表温度对连江黄岐赤潮影响的研究[J]. 海洋预报, 2021, 38(3): 95-103. LI X. The influence of SST on the red tide near Huangqi in Lianjiang[J]. Marine Forecasts, 2021, 38(3): 95-103.
[8] 张健, 杨翼, 杨璐, 等. 东海近岸海域赤潮与环境因子的关系[J]. 广东海洋大学学报, 2019, 39(1): 66-70. ZHANG J, YANG Y, YANG L, et al. Relationship between red tide occurrence and environmental factors in offshore waters of East China Sea[J]. Journal of Guangdong Ocean University, 2019, 39(1): 66-70.
[9] 吴玉芳. 赤潮高发期间厦门海域叶绿素值预报方程建立及应用于灾害性赤潮预报模式的研究[J]. 海洋预报, 2012, 29(2): 39-44. WU Y F. Establishment of a chlorophyll forecast equation and its application in red tide forecasting in Xiamen offshore area[J]. Marine Forecasts, 2012, 29(2): 39-44.
[10] 刘沛然, 黄先玉, 柯栋. 赤潮成因及预报方法[J]. 海洋预报, 1999, 16(4): 46-51. LIU P R, HUANG X Y, KE D. The general report of red tide mechanisms and prediction[J]. Marine Forecasts, 1999, 16(4): 46- 51.
[11] STUMPF R P, LITAKER R W, LANEROLLE L, et al. Hydrodynamic accumulation of Karenia off the west coast of Florida[J]. Continental Shelf Research, 2008, 28(1): 189-213.
[12] MCGILLICUDDY D J JR, TOWNSEND D W, HE R, et al. Suppression of the 2010 Alexandrium fundyense bloom by changes in physical, biological, and chemical properties of the Gulf of Maine[J]. Limnology and Oceanography, 2011, 56(6): 2411-2426.
[13] 张志锋, 贺欣, 张哲, 等. 渤海富营养化现状、机制及其与赤潮的时空耦合性[J]. 海洋环境科学, 2012, 31(4): 465-468, 483. ZHANG Z F, HE X, ZHANG Z, et al. Eutrophication status, mechanism and its coupling effect with algae blooming in Bohai [J]. Marine Environmental Science, 2012, 31(4): 465-468, 483.
[14] 王丹, 何恩业, 刘桂梅, 等. 秦皇岛北戴河赤潮生物与环境因子之间的关系[J]. 海洋预报, 2013, 30(5): 1-7. WANG D, HE E Y, LIU G M, et al. Relationship between red tide organisms and environmental factors in the Beidaihe waters of the Qinhuangdao[J]. Marine Forecasts, 2013, 30(5): 1-7.
[15] SUGIHARA G, MAY R, YE H, et al. Detecting causality in complex ecosystems[J]. Science, 2012, 338(6106): 496-500.
[16] 徐海龙, 谷德贤, 张文亮, 等. 基于时间序列的海洋赤潮灾害特征分析[J]. 海洋通报, 2014, 33(4): 469-474. XU H L, GU D X, ZHANG W L, et al. Analysis of the red tide features based on time series in the China Sea[J]. Marine Science Bulletin, 2014, 33(4): 469-474.
[17] 杨建强, 罗先香, 丁德文, 等. 赤潮预测的人工神经网络方法初步研究[J]. 海洋科学进展, 2003, 21(3): 318-324. YANG J Q, LUO X X, DING S W, et al. A preliminary study on artificial neural network method for red tide prediction predicting red tide[J]. Advances in Marine Science, 2003, 21(3): 318-324.
[18] LEE J H W, HUANG Y, DICKMAN M, et al. Neural network modelling of coastal algal blooms[J]. Ecological Modelling, 2003, 159(2-3): 179-201.
[19] RECKNAGEL F. ANNA-artificial neural network model for predicting species abundance and succession of blue-green algae [J]. Hydrobiologia, 1997, 349(1-3): 47-57.
[20] QIN M J, LI Z H, DU Z H. Red tide time series forecasting by combining ARIMA and deep belief network[J]. Knowledge-Based Systems, 2017, 125: 39-52.
[21] COAD P, CATHERS B, BALL J E, et al. Proactive management of estuarine algal blooms using an automated monitoring buoy coupled with an artificial neural network[J]. Environmental Modelling & Software, 2014, 61: 393-409.
[22] 厦门大学. 赤潮样本数据筛选方法及计算机可读存储介质: 中国, 202010902619.7[P]. 2020-12-25. Xiamen University. Screening method of red tide sample data and computer readable storage medium: CN, 202010902619.7[P]. 2020-12-25.
[23] 韩力群. 人工神经网络理论、设计及应用[M]. 北京: 化学工业出版社, 2002. HAN L Q. Theory, design and application of artificial neural network[M]. Beijing: Chemical Industry Press, 2002.
[24] WANG L, ZENG Y, CHEN T. Back propagation neural network with adaptive differential evolution algorithm for time series forecasting[J]. Expert Systems with Applications, 2015, 42(2): 855-863.
[25] ASLANARGUN A, MAMMADOV M, YAZICI B, et al. Comparison of ARIMA, neural networks and hybrid models in time series: tourist arrival forecasting[J]. Journal of Statistical Computation and Simulation, 2007, 77(1): 29-53.
[26] MUTTIL N, CHAU K W. Neural network and genetic programming for modelling coastal algal blooms[J]. International Journal of Environment and Pollution, 2006, 28(3-4): 223-238.
[27] LAKSHMANAN V. Using a genetic algorithm to tune a bounded weak echo region detection algorithm[J]. Journal of Applied Meteorology, 2000, 39(2): 222-230.
[28] SEDKI A, OUAZAR D, EL MAZOUDI E. Evolving neural network using real coded genetic algorithm for daily rainfall – runoff forecasting[J]. Expert Systems with Applications, 2009, 36(3): 4523-4527.
[29] ZHANG A S, ZHANG L. RBF neural networks for the prediction of building interference effects[J]. Computers & Structures, 2004, 82(27): 2333-2339.
[30] MORFIDIS K, KOSTINAKIS K. Comparative evaluation of MFP and RBF neural networks' ability for instant estimation of r / c buildings' seismic damage level[J]. Engineering Structures, 2019, 197: 109436.
[31] 福建省海洋预报台. 业务化赤潮预警方法及计算机可读存储介质: 中国, 202010902630.3[P]. 2021-01-15. Fujian Marine Forecasts. Operational red tide warning method and computer-readable storage medium: CN, 202010902630.3[P]. 2021-01-15.
[32] ANDERSON C R, SAPIANO M R P, PRASAD M B K, et al. Predicting potentially toxigenic Pseudo-nitzschia blooms in the Chesapeake Bay[J]. Journal of Marine Systems, 2010, 83(3-4): 127-140.
[33] ANDERSON C R, KUDELA R M, KAHRU M, et al. Initial skill assessment of the California harmful algae risk mapping(CHARM) system[J]. Harmful Algae, 2016, 59: 1-18.
[34] GLIBERT P M, BURKHOLDER J M, PARROW M W, et al. Direct uptake of nitrogen by Pfiesteria piscicida and Pfiesteria shumwayae, and nitrogen nutritional preferences[J]. Harmful Algae, 2006, 5(4): 380-394.
[35] LEWITUS A J, HORNER R A, CARON D A, et al. Harmful algal blooms along the north American west coast region: history, trends, causes, and impacts[J]. Harmful Algae, 2012, 19: 133-159.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】