HR: 1330h
AN: AE22A-1109 [PDF]
TI: Accuracy of the Lightning Mapping Array
AU: * Thomas, R J
EM: thomas@nmt.edu
AF: Geophysical Research Center, New Mexico Institute of Mining and Technology, Socorro, NM 87801
AU: Krehbiel, P
EM: krehbiel@ibus.nmt.edu
AF: Geophysical Research Center, New Mexico Institute of Mining and Technology, Socorro, NM 87801
AU: Rison, W
EM: rison@ee.nmt.edu
AF: Geophysical Research Center, New Mexico Institute of Mining and Technology, Socorro, NM 87801
AU: Hunyady, S
AF: Geophysical Research Center, New Mexico Institute of Mining and Technology, Socorro, NM 87801
AU: Winn, W
AF: Geophysical Research Center, New Mexico Institute of Mining and Technology, Socorro, NM 87801
AU: Hamlin, T
AF: Geophysical Research Center, New Mexico Institute of Mining and Technology, Socorro, NM 87801
AU: Harlin, J
AF: Geophysical Research Center, New Mexico Institute of Mining and Technology, Socorro, NM 87801
AB:
We have conducted a detailed study of the accuracy of the Lightning
Mapping Array in which experimentally determined values of the
location uncertainties are compared with simple geometric formulations
of the errors and with linear covariance estimates of the
uncertainties from the least-square solution procedure. Observations
of a sounding balloon that carried a GPS receiver and pulsed VHF
transmitter show that the mapping system is able to locate sources
inside the periphery of the network with a horizontal accuracy of
about 10 m rms and, when the sources are higher than 1-2 km above
ground, with a vertical accuracy of 30-50 m rms. The vertical
accuracy is determined primarily by the nearest station contributing
to the solution; the accuracy approaches that of the horizontal
locations when the source is directly above the close station and the
uncertainty increases by a factor of 2-3 or more for sources located
between stations or at low altitude.
For sources outside the network, the locations and their uncertainties
are best described in a network-centered spherical coordinate frame.
The radial distance $R$ of the source is determined primarily by
stations transverse to the arrival direction, which measure the radius
of curvature of the incoming wavefront. A simple geometric model
shows that the radial location error increases as $(R/D)^2$, where $D$
is the network diameter. The vertical location error is dominated by
the uncertainty of the source's elevation angle, which is determined
primarily by stations along the arrival direction, and increases as
$R^2/(zD)$ where $z$ is the source height above the network tangent
plane. The azimuthal or transverse position of the source is most
accurately determined, with the uncertainties increasing only as
$(R/D)$.
The location uncertainties are all proportional to $c\Delta t$, where
$\Delta t$ is the rms uncertainty of the time-of-arrival measurements.
$\Delta t$ can be accurately determined from the statistics of the
reduced chi-squared values of the solutions; the effective value of
$\Delta t$ was 50 ns rms for the LMA network operated during STEPS 2000.
Most of the lightning sources (typically 80%) are located by
only 6 or 7 stations of the network (6 being the minimum number
required by the processing) and therefore have somewhat greater
uncertainties than if they were located by the full complement of
measurement stations.
The uncertainties of the system were also investigated by determining
the scatter of airplane tracks detected by the LMA when the airplanes
were flying through ice crystal clouds (anvils or cirrus). For both
the aircraft and sounding balloon data, the measured uncertainties
agreed well with those predicted by the simple geometric models and by
the covariance error estimates. When zoomed in on, the aircraft and
balloon tracks exhibited characteristic `sawtooth' patterns that we
believe are caused by the gradual accumulation and successive
correction of systematic timing errors at each measurement station.
The effect of these errors is comparable to the other timing
uncertainties and are included in the overall error determinations.
We should be able to reduce this contribution to the timing errors
with small improvements to the system.
DE: 3304 Atmospheric electricity
DE: 3324 Lightning
DE: 3360 Remote sensing
DE: 3394 Instruments and techniques
SC: Atmospheric and Space Electricity [AE]
MN: 2003 Fall Meeting