Tidal monitoring on sandy beaches using perpendicular time-lapse photography

Long term time-lapse photography has proven to be a key tool in monitoring changes in coastal environments, particularly in terms of morphology. The present study adapts and simplifies the approach of some precedents, such as the Argus and CoastSnap systems, to remotely monitor tidal inundation on a sandy beach at Magilligan on the north coast of Northern Ireland. Such a system could prove essential in the study of the effect of waves and tides on groundwater flow and saline intrusion in coastal aquifers, its consequences for sensitive subsurface infrastructure (such as water supply wells), and in the reconciliation of continuous data from same. Photographic data in this study have been gathered using a remote, solar powered time-lapse camera over a six-month period, capturing full neap and spring tidal cycles. Images are captured at hourly intervals and automatically uploaded to the cloud for remote access. The camera is located just 25 metres from the high water mark, overlooking the beach and perpendicular to the sea. This setup contrasts with previous studies where there is a need to find an elevated location at greater distance from the area of investigation. The extent to which a tide inundates up a sandy beach is governed primarily by astronomical effects, which are considered in this study, but also beach slope and atmospheric conditions. It is known that the beach at Magilligan has both a shallow grade (0.02 m) and a high tidal variation (> 150 m between spring and neap tides). Profiles of beach slope are gathered using a differential GPS, while a solar weather station on site, which also uploads data to the cloud, is used to gather atmospheric data. For tidal reference, a traditional tide gauge measuring tide levels at a pier 15 km east of the site is used. Captured images are post-processed using image analysis techniques based around characterising the tidal front against the visual contrast between pixels of sand and pixels of seawater using a routine in MATLAB^®. From this analysis, a numerical value for tidal inundation is extracted. Analysis of these data indicates that the tide times (timing of high and low tides) correspond well with those measured at the nearby tide gauge, however important differences exist in terms of magnitude. In comparing these differences with atmospheric data from the site, it is possible to align larger and smaller inundation events with shifts in wind direction and speed. The calibration process involved in digitising the captured images is time-consuming, however, it may be possible to predict tidal inundation from a site using only a remote weather station — knowing how a change in wind speed or direction will affect inundation on the beach. It has already been shown that such instrumentation can be used to detect changes in beach morphology (as a key element in tidal inundation), this research therefore represents an important development in the low-cost remote monitoring of tidal inundation, particularly in locations where regular ground surveying is challenging.