HISTORICAL SHORELINE MAPPING .1. IMPROVING TECHNIQUES AND REDUCING POSITIONING ERRORS

A critical need exists among coastal researchers and policy-makers for a precise method to obtain shoreline positions from historical maps a nd aerial photographs. A number of methods that vary widely in approac h and accuracy have been developed to meet this need. None of the exis ting methods, however, address the entire range of cartographic and ph otogrammetric techniques required for accurate coastal mapping. Thus, their application to many typical shoreline mapping problems is limite d. In addition, no shoreline mapping technique provides an adequate ba sis for quantifying the many errors inherent in shoreline mapping usin g maps and air photos. As a result, current assessments of errors in a il photo mapping techniques generally (and falsely) assume that errors in shoreline positions rue represented by the sum of a series of wors t-case assumptions about digitizer operator resolution and ground cont rol accuracy. These assessments also ignore altogether other errors th at commonly approach ground distances of 10 m. This paper provides a c onceptual and analytical framework for improved methods of extracting geographic data from maps and aerial photographs. We also present a ne w approach to shoreline mapping using air photos that revises and exte nds a number of photogrammetric techniques. These techniques include ( 1) developing spatially and temporally overlapping control networks fo r large groups of photos; (2) digitizing air photos for use in shoreli ne mapping; (3) preprocessing digitized photos to remove lens distorti on and film deformation effects; (4) simultaneous aerotriangulation of large groups of spatially and temporally overlapping photos; and (5) using a single-ray intersection technique to determine geographic shor eline coordinates and express the horizontal and vertical error associ ated with a given digitized shoreline. As long as historical maps and air photos are used in studies of shoreline change, there will be a co nsiderable amount of error (on the order of several meters) present in shoreline position and rate-of-change calculations. The techniques pr esented in this paper, however, provide a means to reduce and quantify these errors so that realistic assessments of the technological noise (as opposed to geological noise) in geographic shoreline positions ca n be made.