浙江沿海基于浮标的有效波高分析和数值预报订正

Significant wave height analysis and numerical forecast correction based on the buoy abservations along the Zhejiang coast

基于2010年10月―2022年12月浙江沿海5个浮标站观测资料和2020—2022年欧洲中期天气预报中心细网格模式海浪和海面10 m风预报资料,对浮标站的风、浪相关性进行分析,对有效波高预报进行评估和订正方法研究。0~240 h预报偏差和均方根误差表明:模式对远岸浮标的预报比实况稍偏大,对近岸浮标预报则略偏小,预报不稳定性随预报时效的延长而增大。远离海岸的浮标站24 h预报与实况的相关性高于近岸站,两者的相关系数分别约为0.94和0.85。有效波高预报存在一定系统偏差,且是预报误差的主要来源。近岸浮标有效波高预报值小于观测值,远岸浮标的预报值偏差则随着有效波高的增大由正转负,负偏差的绝对值逐渐增大。对于远岸浮标,仅针对有效波高进行预报和实况的一元一次回归就有很好的订正效果,但近岸浮标受地形影响大,海浪预报的误差更大,同时引入有效波高和10 m风速作为预报因子,进行二元一次回归的订正效果更好。订正方程应用到同类型的周边海区浮标站仍然有很好的订正效果,可推广到浙江沿海同类型海区。

Based on the observations of five buoys along the Zhejiang coast from October 2010 to December 2022,and sea wave and 10-meter wind forecasts of the European Medium-Range Weather Prediction Center fine grid model,the observed correlation between wind and wave is analyzed,the significant wave height(SWH)forecasting is verified and correction method is studied as well.Deviations and root mean square error(RMSE)of the 0~240 hour forecasts indicate that the SWH is slightly overestimated for the buoys away from the coastline,while a little bit underestimated for the near shore buoys.The forecasting instability increases along with the prolong of lead time.The correlated coefficients between the 24-hour forecasts and the observations of the buoys far away from the coastline are higher than those of the near shore buoys,with coefficients of 0.94 and 0.85,respectively.There exists systematic deviations for the SWH forecasting,which dominates the error sources for the buoys.The predicted SWHs of the near shore buoys are lower than observations,while the forecast biases of the buoys far away from the coastline change from positive to negative with the SWH increasing,and the greater the SWH,the more absolute biases received.It basically follows a linear regression between the predicted SWHs and observations for the buoys away from the coastline.While larger biases are recorded for the near shore buoys due to the influence of terrain,and a two-variable linear regression can get a better SWH correction result,with the predicted SWH and 10-meter speed as regression variables.The correction equations can achieve good results and can be applied to the locations similar to the buoys along the Zhejiang coast.