Detection of Shoreline Changes from ERTS-1 Data

DETECTION OF SHORELINE CHANGES FROM ERTS- 1 DATA Jacques L. G. Empfoincourt and Charles C. Wielchowsky* Geographers have always had a desire to reach high observation points in order to obtain a synoptic view of the earth's surface; however, their efforts until recently have been frustrated by the limitations of available tools. Areal coverage under uniform lighting conditions, for example, has always been restricted because observations have been confined to sub-orbital views of a few tens of square miles. Furthermore, there have been limitations in the timing of observations. A world-wide view of any geographic phenomena formerly had to be obtained by piecing together a great number of observations made at different times. In many cases the phenomena of interest are changing at such a rate that the big picture is obsolete before it is put together. (1) Another restriction has been that heretofore the earth could be seen only through the sensing of visible light with the naked eye or camera. Today, however , with new sensors capable of gathering environmental information in a much wider band of the electro-magnetic spectrum, earth scientists are able to observe phenomena in totally different "lights." With the launching of ERTS-I, NASA's first Earth Resources Technology Satellite and the first satellite dedicated entirely to the study of the earth, geographers were provided with an excellent opportunity to apply their expertise in making and updating maps and in analyzing the physical and cultural dynamics of the environment. To date, ERTS data have been regarded by some as "the universal geographic tool" and by others as nothing more than an interesting experiment with little practical geographic value. The authors feel, however, that the true geographic value of ERTS data lies somewhere between these two extremes. The purpose of this paper is to demonstrate how ERTS-I imagery can be used in a limited study of a relatively small but dynamic area. To this end, ERTS-I data were used to detect changes in the configuration of the land/water interface (shoreline) that have taken place in the last 15 to 20 years in the Alabama coastal zone. This was done by comparing ERTS-I images to 1953 and 1957 AMS 1:250,000 scale maps. Further investigation revealed that the configurational changes noted on the maps resulted from either cartographic error, from the addition or subtraction of land area by the continuation of the physical processes that shaped the shoreline, or from an interruption of the dynamic equilibrium that was present. The physical and cultural agents responsible •Mr. Emplaincourt is a geographer and Mr. Wielchowsky is a geologist at the Geological Survey of Alabama. This paper has been approved for publication by the State Geologist. This paper was accepted for publication in January 1974. Vol. XIV, No. 1 39 for addition or subtraction of land area over the last 15 to 20 years were also identified on a preliminary basis. METHODS. The ERTS-I Multispectral Scanner Subsystem (MSS) images the earth's surface in four different bands of the electromagnetic spectrum . (2) These bands are: Band 4 (0.5µ-0.6µ)-?Geß? Band 5 (0.6µ-0.7µ)-?ß? Band 6 (0.6µ-0.8µ)-?ߣ?G Infrared Band 7 (0.8µ-1.1µ)-?ß? Infrared Band 7 is perhaps the most useful for cartographic applications. It allows penetration of thin clouds, yields excellent land/water delineation, and gives superior natural feature definition. (3) Therefore, transparencies of band 7 MSS images acquired on four different dates over Mobile Bay, Alabama, were enlarged to a scale of 1:250,000 so that they could be compared with AMS maps of the same scale. This scale was selected because: 1) only two AMS 1:250,000 maps are required to cover the entire study area; 2) perceptional image quality is good on MSS bulk data to scales of 1:250,000; (4) and 3) 1:250,000 is a scale that is compatible with the magnitude of the major detected changes. Tidal fluctuations were ignored in this study due to the fact that mean tidal range is only about 1 foot (30 cm) at the lower end of the bay and 1.5 feet (45...