Using the probability of storm transposition for estimating the frequency of rare floods
Using the probability of storm transposition for estimating the frequency of rare floods
复制标题
利用风暴换位概率估计罕见洪水的频率
DOI：
10.1016/0022-1694(63)90032-5
复制
发表时间：
1963
影响因子：
6.4
通讯作者：
G. N. Alexander
中科院分区：
文献类型：
--
作者：
G. N. Alexander
将有关台站附近若干台站的记录结合起来，可称为“区域”分析。如前所述，这种区域分析应通过增加样本量来减少抽样误差，但我们当然必须避免得出无效结论，即记录的“台站年”数给出了最高观测值的重现期。不使用实际的台站年数，而是可以通过考虑它们之间的空间相关性来估计台站的“有效”数量，使用Yule公式4）。康沃尔语、希尔和埃文斯对澳大利亚阿德莱德周围面积约15万平方英里的55个雨量站进行了长达94年的详细分析，结果表明，根据1485对雨量站的6天降雨量计算的站间相关系数平均值高达0.6。如此高的相关性证明了通过使用所谓的“均匀”区域的概念，使用区域或风暴基础，而不是点或站基础来解决问题的方法是合理的。在这种技术中，在均匀区域内观察到的高值通常在转换到相关台站之前最大化。基于露点的最大化程序旨在为该地区的风暴提供“最大可能”值。
Combining the records of a number of stations within the vicinity of the station concerned may be termed'regional'analysis. As stated such regional analysis should, by increasing sample size, reduce the sampling errors, but we must of course avoid the invalid conclusion that the number of'stationyears' of record gives the return period of the highest observed value. Instead of using the actual number of station-years it is possible to estimate the'effective'number of stations by taking into account the spatial correlation between them using Yule's formula4). In an elaborate analysis of 55 rainfall stations, covering an area of about 150,000 square miles around Adelaide, Australia, over a period of 94 years, Cornish, Hill and Evans 5) showed that the interstation correlation coefficient based on 6-day rainfalls from the 1485 pairs had an average value as high as 0.6. Such high correlations justify an approach to the problem using an areal or storm basis, rather than a point or station basis, by using the concept of so-called'homogeneous' areas.A further use of this regional notion is inherent in the concept of transposing storms within such a'homogeneous' area. In this technique high values observed within a homogeneous area are usually maximized before transposition to the station concerned. The maximization procedure based on dew points purports to give a'maximum possible'value for the storm over the area.