Satellite remote sensing is becoming a widely used monitoring technique in coastal sciences. Yet, no benchmarking studies exist that compare the performance of popular satellite-derived shoreline mapping algorithms against standardized sets of inputs and validation data. Here we present a new benchmarking framework to evaluate the accuracy of shoreline change observations extracted from publicly available satellite imagery (Landsat and Sentinel-2). Accuracy and precision of five established shoreline mapping algorithms are evaluated at four sandy beaches with varying geologic and oceanographic conditions. Comparisons against long-term in situ beach surveys reveal that all algorithms provide horizontal accuracy on the order of 10 m at microtidal sites. However, accuracy deteriorates as the tidal range increases, to more than 20 m for a high-energy macrotidal beach (Truc Vert, France) with complex foreshore morphology. The goal of this open-source, collaborative benchmarking framework is to identify areas of improvement for present algorithms, while providing a stepping stone for testing future developments, and ensuring reproducibility of methods across various research groups and applications.

Beaches around the world continuously adjust to daily and seasonal changes in wave and tide conditions, which are themselves changing over longer time-scales. Different approaches to predict multi-year shoreline evolution have been implemented; however, robust and reliable predictions of shoreline evolution are still problematic even in short-term scenarios (shorter than decadal). Here we show results of a modelling competition, where 19 numerical models (a mix of established shoreline models and machine learning techniques) were tested using data collected for Tairua beach, New Zealand with 18 years of daily averaged alongshore shoreline position and beach rotation (orientation) data obtained from a camera system. In general, traditional shoreline models and machine learning techniques were able to reproduce shoreline changes during the calibration period (1999–2014) for normal conditions but some of the model struggled to predict extreme and fast oscillations. During the forecast period (unseen data, 2014–2017), both approaches showed a decrease in models’ capability to predict the shoreline position. This was more evident for some of the machine learning algorithms. A model ensemble performed better than individual models and enables assessment of uncertainties in model architecture. Research-coordinated approaches (e.g., modelling competitions) can fuel advances in predictive capabilities and provide a forum for the discussion about the advantages/disadvantages of available models.