Tracing Shoreline Change in the Mouth of San Bernard River, Texas
Grace Chen and David Buzan, Texas Parks and Wildlife Department, Coastal Fisheries Division
Noticeable changes in the channel, especially the mouth, of the San Bernard River have been reported and confirmed. To estimate the extent to which the San Bernard River has shoaled over the past decades, historical imagery was analyzed using Geographic Information System (GIS) techniques. Specifically, heads-up digitizing was employed to trace the location of the mouth of San Bernard River at historical time points and distances of shoreline movement over the study period were computed.
The source data for this study are historical images that cover the mouth of San Bernard River and its surroundings. In particular, the 1974 topographic map prepared by the U.S. Geological Service, the 1995 digital orthophoto quadrangles (DOQs) maintained by the Texas Natural Resources Information System (TNRIS), and the 2002 infrared digital images taken by the General Land Office were used to delineate the river courses at respective times. The attempt to trace the river channel prior to the operation of the Intracoastal Waterway (ICWW) was abandoned because no data were available. The Texas Digital Aerial Photo Archive (TxDAPA) program has scanned and georeferenced historical aerial photography for the county of Brazos. However, the existing 1944 images found of the San Bernard River do not cover the study area (see Figure 1).
Procedures: Tracing River Courses
Since the primary interest of this study is to estimate changes in the shoreline around the San Bernard River, channel courses were traced 20 kilometers along the river upstream from the mouth and along the ICWW from Cedar Lakes to Jones Lake. Knowing that the scale of base maps used for channel delineation affects the details of shoreline behavior (Mandelbrot, 1967), we fixed the digitizing process at a scale of 1 to 5,000. Additionally, concerned with the fluctuation of shoreline location in response to varying sea levels, we defined shorelines at the outmost edges of the sand bars that are visible from the 1995 and 2002 aerial photos (Figure 2A and Figure 2B), even though part of the sand bars might be under water during high tides.
As illustrated in Figure 3, we extended the river banks to the Gulf and identified two inflection points from which the shoreline becomes straight and completely merged with the Gulf. A box with the width bounded by these inflection points and arbitrary height was drawn to denote the relative size of the mouth area at respective times. To facilitate the comparison of shoreline change over time, we delineated the precise location of the mouth of the San Bernard River by drawing a line from the right to the left bank of the river prior to their expansion to the Gulf. The right bank of the mouth was then chosen as the reference point for determining the magnitude of shoreline movement.
The result of geospatial analysis indicated that the mouth of the San Bernard River has moved westward significantly over the past 30 years. As shown in Figure 3, the right bank of the mouth of the San Bernard River moved 1100 meters from 1974 to 1995 and 1033 meters from 1995 to 2002. This is equal to an annual rate of 52.38 and 147.58 meters per year for the respective periods. Obviously, the shoreline change in the San Bernard Estuary has accelerated recently. What could have contributed to the accelerating movement of shoreline along the San Bernard River in addition to the westward Gulf current that constantly transports sediment deposited by the Brazos River toward the west? One may speculate that the Brazos River located just east of the study area might take more freshwater out of the San Bernard River via the ICWW canal recently than before. If this is the case, the reduced discharge of the San Bernard River could decrease its resistance to the influence of Gulf current and associated sedimentation.
To investigate the possibility of increasing diversion of water to the Brazos River through the ICWW, we measured the change in the width of the ICWW between the San Bernard River and Jones Lake using GIS (see Figure 4). Taking the left bank of the San Bernard River and the edge between ICWW and Johns lake observed in individual images as two endpoints, the canal's widths were estimated by dividing the areas bounded by the endpoints by their corresponding lengths. The resulting widths for years 1974, 1995, and 2002 are 83, 157, and 163 meters, respectively. This is equivalent to an annual increase of 3.57 meters for the period between 1974 and 1995, and 0.79 meters between 1995 and 2002. Unlike the trend of shoreline movement, the increase of ICWW's width has not accelerated lately. The speculation that the San Bernard River has been weakened by the Brazos River because of water diversion therefore is not strongly supported by the finding of this analysis, although simulation of flow circulation may prove it otherwise. Besides the evidence that sediment deposited by the Brazos River is transported westward by the Gulf current, mechanisms behind the accelerated movement of shoreline at the mouth of the San Bernard River require further study.
The authors would like to thank Dr. David Brock with the Texas Water Development Board for sharing his hypothesis concerning the relationship between water diversion through ICWW and shoreline changes. We also thank Dr. Charles Smith and Mr. Nathan Kuhn with the Texas Parks and Wildlife Department for their comments on the manuscript.
Mandelbrot, B.B. 1967. How Long is the Coast of Britain? Statistical Self-Similarity and Fractional Dimension. Science 156: 636 - 638.