Geoscience Reference
In-Depth Information
survivability experience, maintenance cost experience, model improvements, new
technology developments (even new DART designs), increasing international contributions,
and updated information on the entire suite of siting issues listed previously.
Reliability of the DART Network
Since the build-up of the DART network began in 2006, it has experienced signiicant out-
ages that can have adverse impacts on the capability of the TWCs to issue eficient warnings,
to use near-real-time forecasts, and to cancel warnings when a tsunami threat is over. The data
loss also reduces post-tsunami model validation capability. Figure 4.7 indicates how network
availability steadily declined to a low of 69 percent in February 2009. The number of DART
stations deployed grew from 10 in July 2006 (7 new DART II systems, along with 3 older DART I
systems) to 39 in March 2008 (including replacement of the original DART I systems with
DART II systems). By March 2009, only a year after the DART array was completely deployed,
100.0%
99.8%
97.2%
96.7%
95.4%
94.5%
90.8%
91.9%
91.1%
90.9%
90.5%
90.0%
90.3%
89.4%
87.9%
87.6%
87.0%
87.0%
85.7%
86.8%
86.2%
85.9%
85.0%
84.9%
85.1%
82.8%
82.5%
82.4%
81.6%
81.1%
81.8%
80.0%
79.4%
80.0%
79.5%
77.1%
77.5%
76.2%
72.9%
72.1%
70.6%
69.5%
70.0%
69.4%
66.0%
63.9%
60.0%
FIGURE 4.7
Chart of DART II network performance through December 2009, deined as the percentage
of hourly transmissions of water column heights received vs. expected. The peaks in performance occur
during Northern Hemisphere summer when maintenance is performed. Note, however, that the peak
values in performance are decreasing with time as well. SOURCE: National Data Buoy Center, NOAA.
