A Doppler Radar for Continuous Remote Measurement
of River Ice Velocity
MICHAEL G. FERRICK, NORBERT E. YANKIELUN, AND DAVID F. NELSON
River during the March 1993 ice breakup and
INTRODUCTION
again in December 1993 during a frazil ice run
Accurate ice velocity measurement during river
prior to ice cover formation. Ice velocity data
breakup in spring is needed for analysis of river
were continuously acquired, processed, dis-
and ice dynamics and for flood hazard or water
played, and stored. The system can be transport-
resource assessment. Ice velocity measurements
ed in a vehicle and rapidly redeployed to permit
presented by Ferrick et al. (1993) demonstrated the
data acquisition at a number of locations during a
spatial and temporal variability of the motion dur-
single event with minimal operator interaction.
ing a dynamic breakup of the Connecticut River.
Simultaneous ice velocity data were obtained at
Similar though generally slower and more grad-
the same site with video techniques during both
ually variable motion occurs during the freeze-up
of these events.
of a river. Ferrick et al. (1992) described the acqui-
In this paper we describe the Doppler radar
sition of ice velocity data by locating a length refer-
system, analyze the contributions to measure-
ence on the ice, recording the ice motion on video-
ment error inherent in the system, determine er-
tape, developing a computer-generated reference
ror bounds, and develop data processing meth-
grid that is dubbed over the videotape, and timing
ods to minimize the error in the velocity measure-
the motion of individual floes through the grid.
ment. Quantitative comparisons of measured ice
The shortcomings of this method are that it is
velocity are made between the radar and video
laborious and it requires visual observation of the
techniques for both the breakup and frazil events.
ice, which can be precluded by fog and darkness.
Methods for enhancing the accuracy of Doppler
Ice scour of the bed and banks frequently prevents
radar ice velocity measurement and its ease of
the use of measurement methods that require
use follow from the analysis.
sensor and cable placement in the river. Radar sys-
tems have been applied to hydrological investiga-
tions, including the measurement and mapping of
SYSTEM DESCRIPTION
ice thickness on rivers and lakes (Arcone and
The primary requirement for a successful
Delaney 1987, Arcone 1991, Yankielun et al. 1992,
radar measurement is that the target provide suf-
1993) and the continuous monitoring of river stage
ficient backscatter of the incoming wave. If the
(Yankielun and Ferrick 1993). Doppler radar tech-
surface area illuminated by the radar were per-
niques for velocity measurement are well estab-
fectly smooth, the incident energy would be re-
lished (Barton 1964, Skolnik 1980) and provide a
flected away from the radar antenna. However,
remotely mounted, readily installed, continuously
the inherent roughness of frazil, sheet, and rub-
recording alternative for ice velocity measurement
ble ice provides backscatter towards the antenna
that is not affected by light, fog, or ice motion con-
(Fig. 1). The fundamental relationship between
ditions.
A continuous wave (CW) Doppler radar with
transmitted and received power of a radar signal
real-time data acquisition and digital signal pro-
traveling through free space was described in
cessing capability was mounted on the Cornish
Skolnik (1980). Lewis et al. (1987) discussed the
(N.H.)Windsor (Vt.) bridge over the Connecticut
specific problem of radar detection of floating ice
Previous Page
