Volcanic ash modeling with the NMMB-MONARCH-ASH model: quantification of offline modeling errors
Volcanic ash modeling with the NMMB-MONARCH-ASH model: quantification of offline modeling errors
复制标题
使用 NMMB-MONARCH-ASH 模型进行火山灰建模：离线建模误差的量化
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
A. Folch
中科院分区：
文献类型：
--
作者：
A. Marti;A. Folch
摘要。火山灰模拟系统用于模拟大气
Abstract. Volcanic ash modeling systems are used to simulate the atmospheric
dispersion of volcanic ash and to generate forecasts that quantify the
impacts from volcanic eruptions on infrastructures, air quality, aviation,
and climate. The efficiency of response and mitigation actions is directly
associated with the accuracy of the volcanic ash cloud detection and
modeling systems. Operational forecasts build on offline coupled modeling
systems in which meteorological variables are updated at the specified coupling
intervals. Despite the concerns from other communities regarding the
accuracy of this strategy, the quantification of the systematic errors and
shortcomings associated with the offline modeling systems has received no
attention. This paper employs the NMMB-MONARCH-ASH model to quantify these
errors by employing different quantitative and categorical evaluation
scores. The skills of the offline coupling strategy are compared against
those from an online forecast considered to be the best estimate of the
true outcome. Case studies are considered for a synthetic eruption with
constant eruption source parameters and for two historical events, which
suitably illustrate the severe aviation disruptive effects of European (2010
Eyjafjallajokull) and South American (2011 Cordon Caulle) volcanic
eruptions. Evaluation scores indicate that systematic errors due to the
offline modeling are of the same order of magnitude as those associated
with the source term uncertainties. In particular, traditional offline
forecasts employed in operational model setups can result in significant
uncertainties, failing to reproduce, in the worst cases, up to 45–70 % of
the ash cloud of an online forecast. These inconsistencies are anticipated
to be even more relevant in scenarios in which the meteorological conditions
change rapidly in time. The outcome of this paper encourages operational
groups responsible for real-time advisories for aviation to consider
employing computationally efficient online dispersal models.
2010 年埃亚菲亚德拉冰盖喷发远端火山云中火山灰浓度的业务预测
DOI：
10.1029/2011jd016790
发表时间：
2012
期刊：
Atmospheres
影响因子：
--
作者：
Webster H
通讯作者：
Webster H
影响因子：
3.5
作者：
Bonadonna, Costanza;Folch, Arnau;Puempel, Herbert
通讯作者：
Puempel, Herbert