CAJ | 학술논문

根据1989-2006年中国海洋灾害公报中公布的中国赤潮发生频率的数据资料,建立近18 a来中国赤潮发生频率的时间序列,采用平均差值法、最小二乘法和自回归滑动平均法这3种趋势分析方法,统计检验了其长期变化趋势.并基于分形理论利用重标极差分析 (R/S分析)的原理和方法,计算了Hurst指数,建立了与的关系模型,对中国赤潮多发年份进行了预测分析.结果表明,近18 a来中国赤潮发生频率变化的趋势值分别为bAV = 0.143 56,bLS = 0.128 75和bAR (1) = 0.128 61,均为正值,赤潮发生频率呈显著的增加趋势;赤潮多发年份的变化呈现分布式布朗运动的特征, 与之间存在的函数关系,具有可预测性.根据预测2007年和2008年仍然是赤潮的多发年份,这与实际统计结果一致.基于目前赤潮发生机制尚不完全清楚的现状,利用赤潮发生频率的时间序列自身进行赤潮多发年的预测可能是一条有效的途径.
근거1989-2006년중국해양재해공보중공포적중국적조발생빈솔적수거자료,건립근18 a래중국적조발생빈솔적시간서렬,채용평균차치법、최소이승법화자회귀활동평균법저3충추세분석방법,통계검험료기장기변화추세.병기우분형이론이용중표겁차분석 (R/S분석)적원리화방법,계산료Hurst지수,건립료여적관계모형,대중국적조다발년빈진행료예측분석.결과표명,근18 a래중국적조발생빈솔변화적추세치분별위bAV = 0.143 56,bLS = 0.128 75화bAR (1) = 0.128 61,균위정치,적조발생빈솔정현저적증가추세;적조다발년빈적변화정현분포식포랑운동적특정, 여지간존재적함수관계,구유가예측성.근거예측2007년화2008년잉연시적조적다발년빈,저여실제통계결과일치.기우목전적조발생궤제상불완전청초적현상,이용적조발생빈솔적시간서렬자신진행적조다발년적예측가능시일조유효적도경.
Basing on date of the happened frequency of red tide from the Bulletin of Marine Disaster in China in recent 18 years (from 1989 to 2006), the time sequence of happening frequency of red tide in China was established, and the long-term variation trend was tested by using three statistical methods which include the average difference method, the least square method, and self-regression slope AR (1) process. The Hurst exponent was calculated, and the relationship function model between and was established by using the Rescaled Range Analysis method on Fractal Theory, and the prediction analysis of frequently happened year of red tide disaster in China was conducted. The results showed that the variation trends value of the happened frequency of red tide in China were bAV=0.14356, bLS=0.12875, and bAR (1) =0.12861, respectively. That's mean, the happened frequency of red tide has significant increased in recent 18 years. The change of the frequently happened year of red tide presented the fractional Brownian Motion characteristic, and the relationship function between and was , and the frequently happened year can be predicated. 2007, 2008 were the frequently happened year of red tide based on prediction, and this was the same with the statistical result. Since the status about happening mechanism of red tide is not clear, it can be a more effective way to forecast the frequently happened year of red tide by using time sequence of the happened frequency for red tide.