of the strain me-
ter network for a volume source in the up-
per crust at Grímsvötn is 0.05–0.1 km3. The
absence of a strain signal, together with the
seismicity constraints, suggest that <0.1 km3 of
erupted magma came from below 4 km depth
under the southern caldera rim.
Summary and Outlook
IMO’s multidisciplinary monitoring approach,
employing seismic and geodetic methods,
combined with access
/media/jar/myndsafn/2005EO260001.pdf
(Sigurðsson et al., 2004). In our model configuration the maximum elevation
of the Hofsjökull ice cap is approximately 1540 metres, i.e. more than 250 me-
tres lower than in reality. Hence, we use area-integrated data from an elevation
range of approximately 1450–1650 metres along the three profiles HN, HSV
and HSA (Jóhannesson et al., 2006). The winter balance on Hofsjökull has
been modelled
/media/ces/Paper-Olafur-Rognvaldsson_91.pdf
relate to other me-
teorological factors and if the performance depends on the temporal resolution
of the data and geographical location. The work should shed a light on which
aspects need improving. Increased understanding of the limitations of the sim-
ulations on a short timescale will also be beneficial for their use in hydrological
purposes.
This paper begins with a short description of rain
/media/ces/Paper-Olafur-Rognvaldsson_92.pdf
Reanalysis
IDF – Intensity-Duration-Frequency
IMO – Icelandic Meteorological Office
ME – Mean Error
MLE – Maximum Likelihood Estimation
RMSE – Root Mean Squared Error
9
Abstract
In Iceland, extreme precipitation can often lead to flooding. In coastal regions, short-lived
torrents can develop on steep slopes close to inhabited areas. The purpose of this report is
twofold: to present an updated
/media/vedurstofan-utgafa-2020/VI_2020_008.pdf
of the analogue method is introduced. In Section 3 hydrological and me-
teorological data used in the analysis are presented. Section 4 describes the different strategies
considered for implementing the method and Section 5 presents some results. Some concluding
remarks are made in Section 6.
2 The analogue method
Let X(t) be a state of a dynamical system at time t, known through the observation of k variables
/media/vedurstofan/utgafa/skyrslur/2013/VI_2013_008.pdf
on the use of
the relocation software, which was written by him and used in this research. First of all, I
want to thank them for their guidance, but especially my main supervisor, Kristín, for all
the patience, countless hours spent explaining and discussing various problems and giving
compliments for deeds well done.
I thank Ragnar Stefánsson for employing me at IMO, but this work was mostly funded
/media/vedurstofan/utgafa/skyrslur/2010/2010_003rs.pdf
Koulouriotis et al., 2003); developing indices for analysing and
comparing Fuzzy Cognitive Maps (C¸oban and Sec¸me, 2005); and
analysing the dynamic output of a Fuzzy Cognitive Map (Stach
et al., 2005). Despite the diversity of applications, the aspect of
analysing the dynamic behaviour the Fuzzy Cognitive Map is
surprisingly underexposed. Because of its importance in the
context of scenario development
/media/loftslag/Kok_JGEC658_2009.pdf