Homogenizing surface pressure time-series from operational numerical weather prediction models for geodetic applications
Homogenizing surface pressure time-series from operational numerical weather prediction models for geodetic applications
复制标题
DOI：
10.1515/jogs-2016-0004
复制
发表时间：
2016-01-01
影响因子：
1.3
通讯作者：
Dobslaw, H.
中科院分区：
文献类型：
--
作者：
Dobslaw, H.
通过ECMWF的平均海平面气压诊断，将ECMWF业务天气预报模式和ERA-Interim的14.5年6小时分析的全球表面气压网格映射到一个共同的参考地形。该方法减少了约一个数量级的相对偏差和残差变异，从而实现了约1百帕水平的两个数据集之间的一致性。其余的差异，而反映温度偏差和分辨率的再分析数据集的限制，但不再涉及到地形的局部粗糙度或业务模式的空间分辨率的变化。因此，所提出的减少方法允许获得表面压力的时间序列与长期的一致性的再分析从业务数值天气模型具有更高的分辨率和更短的延迟，使得结果适合于大地测量的近实时应用程序，需要连续更新的时间序列是均匀的多年。
Global surface pressure grids from 14.5 years of 6-hourly analyses out of both the operational ECMWF weather prediction model and ERA-Interim are mapped to a common reference orography by means of ECMWF's mean sea-level pressure diagnostic. The approach reduces both relative biases and residual variability by about one order of magnitude and thereby achieves a consistency among both data sets at the level of about 1 hPa. Remaining differences rather reflect temperature biases and also resolution limitations of the reanalysis data set, but are not anymore related to the local roughness in orography or to changes in the spatial resolution of the operational model. The presented reduction method therefore allows to obtain surface pressure time series with the long-time consistency of a reanalysis from an operational numerical weather model with much higher resolution and much shorter latency, making the results suitable for geodetic near real-time applications requiring continuously updated time series that are homogeneous over many years.