Assessment and integration of conventional, RTK-GPS and image-derived beach survey methods for daily to decadal coastal monitoring

作者：Harley Mitchell D^*; Turner Ian L; Short Andrew D; Ranasinghe Roshanka

来源：Coastal Engineering, 2011, 58(2): 194-205.

DOI：10.1016/j.coastaleng.2010.09.006

• Beach surveys
• Semivariogram
• ARGUS coastal imaging
• RTK-GPS
• Collaroy-Narrabeen Beach

Coastal monitoring across a broad range of time-scales was recognized in the latest report by the Intergovernmental Panel on Climate Change as key to better understanding the likely impacts of climate change at the coast. A unique and historic coastal monitoring program undertaken at the Collaroy-Narrabeen embayment in south-eastern Australia comprises: 1) 30 years of monthly conventional (Emery method) surveys of five cross-shore profile lines; 2) three years of monthly three-dimensional surveys of the entire embayment using RTK-GPS mounted to an all-terrain vehicle (ATV); and 3) four years of hourly shoreline measurements using coastal imaging technology (ARGUS). This study evaluates the strengths and limitations of conventional, RTK-GPS and image-derived surveys for coastal monitoring at daily to decadal coastal time-scales. High-accuracy RTK-GPS was used to first assess the accuracy of the conventional and image-derived survey methods. The magnitude of daily to decadal coastal variability was then characterized by calculating the temporal semivariogram of the integrated survey dataset. With both measurement errors and the degree of beach variability quantified, the corresponding signal-to-noise ratios (SNR) of each survey method at different time-scales were determined. The value of the simple and cost-effective Emery method was verified by this analysis, with measurement error significantly smaller than the degree of overall beach variability (SNR = 8.4). The accuracy, speed and efficiency of ATV-mounted RTK-GPS meanwhile make it suitable for three-dimensional beach surveys. Image-derived surveys were found to be an effective means of remotely measuring the considerable degree of beach variability identified at time-scales of less than one month. These measurements however become indistinguishable from survey noise (i.e. SNR <= 1) when considering typical weekly (or smaller) variations at large distances from the cameras.

• 出版日期2011-2