Content uploaded by Ahmed Abdalla
Author content
All content in this area was uploaded by Ahmed Abdalla on Jul 15, 2019
Content may be subject to copyright.
Content uploaded by Abdelali Fadil
Author content
All content in this area was uploaded by Abdelali Fadil on Jun 04, 2019
Content may be subject to copyright.
Preliminary Results of Land Subsidence Monitoring in Louisiana from
GNSS Observations
Ahmed Abdalla1, Abdelali Fadil2, Rui Fernandes3 , Machiel Bos3, Clifford Mugnier1, J. Anthony Cavell1,
1Center for GeoInformatics, Dept of Civil and Environmental Engineering, Louisiana State University, Baton Rouge, LA, USA, 2Cadi
Ayyad University, Faculty of Sciences Semlalia, Marrakech, Morocco, 3University of Beira Interior- IDL, Segal, Covilhã, Portugal.
Email addresses: {aabdalla1,cjmce,cavell}@lsu.edu; a.fadil@uca.ma; {rui,msbos}@segal.ubi.pt
IUGG19-4357, Montreal, Canada
Abstract
Land subsidence has a significant impact on land use management, public
safety, and industry. Various processes such as mining activities, over-
exploitation of groundwater and structural loading are claimed to have
effectively contributed to land subsidence in Louisiana during the past 30
years. Louisiana has one of the largest coastal wetlands in the United States of
which the coastal zone extends from the western border of Texas to Mississippi
in the Eastern border. In this study, we have selected several monitoring
stations based on global navigation satellite systems (GNSS) to estimate the
land subsidence rate in Louisiana since 2013. These stations are a part of
CORS (Continuously Operating Reference Stations) network in Louisiana. This
initial selection will be further densified using the entire network of 60+ CORS
operated by the Center for GeoInformatics (C4G) at Louisiana State University.
The GNSS data have been processed using the GipsyX software with JPL (Jet
Propulsion Laboratory) products by applying the Precise Point Positioning (PPP)
strategy. The vertical velocities aligned to ITRF2014 are computed using the
HECTOR software, which can estimate linear trends, offsets and seasonal
signals simultaneously. We investigate the stochastic properties of the time
series to ensure the correct choice of noise model to produce realistic
uncertainties for the estimated subsidence rates.
GULFNet
The Louisiana Spatial Reference Center (LRSC) was created by cooperation
between the Louisiana State University, Center for GeInformatics (C4G) and
the National Geodetic Survey (NGS) to establish a state-wide precise geodetic
network. This network is known as GULFNet and it consists of advanced and
modern Global Navigation Satellite Systems (GNSS) receivers.
GULFNet is continuously operated and it is established to be a positional
infrastructure network that is fundamental for the following purposes in
Louisiana:
❑Surveying work
❑GIS mapping
❑Geospatial applications
❑Precision farming
❑Navigation
GULFNet is over120 stations and it is primarily in Louisiana and the Gulf Coast
from Texas at the border of Mexico to Florida to Fernandina Beach in the east
coast of Florida. It is also aiming at determining and localizing the land
subsidence in Louisiana with high-precision. This is based on time-series
analysis of the CORS observations which helps for future coastal restoration
efforts.
Site 1: 1ULM
Stations
As a preliminary step, we analyzed the legacy of the archived RINEX files of
two site stations, namely, 1ULM and HAMM. The 6-year time-series of the two
selected stations (1ULM and HAMM) from Jan 2013 to December 2018 is
processed using GIPSY-X software, and the vertical and horizontal velocity
estimates are computed by HECTOR software
Site 2: HAMM
Acknowledgments
This work was funded by the Center for Geoinformatics (C4G) at Louisiana
State University. The GNSS data were analyzed using GipsyX software by
CalTech/JPL and the velocity estimates were computed by HECTOR software by
SEGAL.
1ULM (Northing)
Trend
: -1.044 +/-
0.150 mm/year
cos
yearly: 0.428 +/- 0.167 mm
sin
yearly: 0.306 +/- 0.171 mm
Amp
yearly: 0.555 +/-
0.164 mm
Pha
yearly: 35.591 degrees
1ULM (Easting)
Trend
: -13.456 +/-
0.295 mm/year
cos
yearly: -0.658 +/- 0.237 mm
sin
yearly: 0.038 +/- 0.244 mm
Amp
yearly: 0.705 +/- 0.230 mm
Pha
yearly: 176.683 degrees
1ULM (Up)
Trend
: -3.394 +/-
0.335 mm/year
cos
yearly: -2.622 +/-
0.563 mm
sin
yearly: -4.276 +/- 0.573 mm
Amp
yearly: 5.048 +/-
0.567 mm
Pha
yearly: -121.516 degrees
Figure 2: Time-series of changes in horizontal (a and b) and vertical (c) direction of 1ULM
station from January 2013 to December 2018 with the associated yearly velocity estimates
relative to ITRF2014 (tables at the right). Units for y-axis are in mm
Figure 1: GULFNet stations starting from the border of Mexico at gulf coast of Texas (TX), Louisiana
(LA), Mississippi (MS) Alabama (AL) and Florida (FL) including the east coast until Fernandina Beach
GULFNet including selected site stations
The GULFNet CORS stations are shown in blue circles in Fig 1 (lower left inset).
The red stars represent the selected cite stations in this study which are
located in the State of Louisiana. The two stations are 1ULM which is located in
the city of Monroe in the North of Louisiana and HAMM which is located in the
city of Hammond in the south-east of Louisiana.
(a
(b
(c
Figure 3: Time-series of changes in horizontal (a and b) and vertical (c) direction of HAMM
station from January 2013 to December 2018 and with the associated yearly velocity
estimates relative to ITRF2014 (tables at the right). Units for y-axis are in mm.
HAMM (Northing)
Trend
: -0.459 +/-
0.325 mm/year
cos
yearly: 0.868 +/- 0.253 mm
sin
yearly: 1.009 +/- 0.261 mm
Amp
yearly: 1.355 +/- 0.255 mm
Pha
yearly: 49.296 degrees
HAMM (Easting)
Trend
: -12.506 +/-
0.233 mm/year
cos
yearly: 1.374 +/- 0.197 mm
sin
yearly: 0.149 +/- 0.204 mm
Amp
yearly: 1.397 +/- 0.200 mm
Pha
yearly: 6.176 degrees
HAMM (Up)
Trend
: -1.050 ± 0.840 mm/year
cos
yearly: 1.374 ± 0.197 mm
sin
yearly: 0.149 ± 0.204 mm
Amp
yearly: 1.397 ± 0.200 mm
Pha
yearly: 6.176 degrees
The gap in fig 3c represents outliers due to a10 cm offset which was detected
in the vertical direction for year 2016 as seen in fig 4. the offset is simply
related to antenna height within GipsyX processing setup where we see it
entirely existing through the year 2016. Hence, we omit Year 2016 until the
offset is solved (later) to avoid unrealistic results.
Figure 4: An offset detected in year 2016. Units for y-axis are in mm
1ULM
HAMM
(a
(b
(c
