Dr. Shimon Wdowinski

Division of Marine and Life Sciences

University of Miami

Multi-temporal InSAR monitoring of wetland water levels: Case study of Water Conservation Area 1 in the Everglades, Southern Florida


Shimon Wdowinski1 and Sang-Hoon1,2


1) Division of Marine Geology and Geophysics, University of Miami, Miami, Florida

33149-1098, USA.

2) Department of Civil and Environmental Engineering, Florida International University, Miami, Florida 33174, USA.


Interferometric Synthetic Aperture Radar (InSAR) techniques can successfully detect phase variations related to the water level changes in wetlands and produce spatially detailed high-resolution maps of water level changes. Despite the vast details, the usefulness of the wetland InSAR observations are rather limited, because hydrologists and water resources managers need information on absolute water level values and not on relative water level changes.

In order to move from ‘relative’ to ‘absolute’ water levels and to include multi-temporal observations, we developed a new technique called Small Temporal Baseline Subset Analysis (STBAS). Our approach follows the small baseline subset (SBAS) algorithm used for calculating displacement time series of solid surfaces. However because interferometric coherence over wetland is mainly sensitive to the temporal baselines, we modified the interferogram selection criterion in SBAS from ‘small geometrical baseline’ to ‘small temporal baseline’. Another important change with respect to SBAS is ground truthing calibration, which is needed because wetland water levels can vary daily by a significant amount. The STBAS method consist of the following steps: (i) Selection of small temporal baseline interferometric pairs, (ii) Interferogram generation including unwrapping, (iii) Calibrations of water level changes using stage (water level) data, (iv) Estimation of relative water level time series using Singular Value Decomposition inversion, and (v) Calibration of absolute water levels time series by tying the relative water levels to reference water levels.

We tested the technique with two-year long Radarsat-1 data acquired over the Water Conservation Area 1 (WCA1) in the Everglades wetlands, south Florida (USA). The fine beam SAR data were acquired successively every 24 days during the years 2006-2007 except two missing acquisitions. The InSAR derived water level data were calibrated using 13 stage (water level) stations located in the study area. We evaluated the quality of our technique using a root mean square error (RMSE) analysis of the difference between InSAR observations and stage measurements. The average RMSE is 6.6 cm, which reflects the sum of two major contributions. The first contribution is uncertainty of the InSAR measurement in detecting water level change, which is estimated as 3-4 cm. The additional uncertainty of 2-3 cm reflects error propagation due to the conversion from relative to absolute water levels. The STBAS technique generates both detailed maps of water levels and water elevation time series for almost all pixels (50 meters) in the studied wetland. These products are very useful for constraining high-resolution flow models used as decision support tools for water resources management.


PLEASE CLICK HERE FOR DR. WDOWINSKI’S ENTIRE PRESENTATION