Predicting regional coastal sea level changes with machine learning

Abstract

All ocean basins have been experiencing significant warming and rising sea levels in recent decades. There are, however, important regional differences, resulting from distinct processes at different timescales (temperature-driven changes being a major contributor on multi-year timescales). In view of this complexity, it deems essential to move towards more sophisticated data-driven techniques as well as diagnostic and prognostic prediction models to interpret observations of ocean warming and sea level variations at local or regional sea basins. In this context, we present a machine learning approach that exploits key ocean temperature estimates (as proxies for the regional thermosteric sea level component) to model coastal sea level variability and associated uncertainty across a range of timescales (from months to several years). Our findings also demonstrate the utility of machine learning to estimate the possible tendency of near-future regional sea levels. When compared to actual sea-level records, our models perform particularly well in the coastal areas most influenced by internal climate variability. Yet, the models are widely applicable to evaluate the patterns of rising and falling sea levels across many places around the globe. Thus, our approach is a promising tool to model and anticipate sea level changes in the coming (1–3) years, which is crucial for near-term decision making and strategic planning about coastal protection measures.

Convolutional Neural Network and Optical Flow for the Assessment of Wave and Tide Parameters from Video Analysis (LEUCOTEA): An Innovative Tool for Coastal Monitoring

“…On the other hand, the LEU-COTEA system will also be improved for the Atlantic coast to increase the tide class and storm surge values for mesotidal and macrotidal coasts. Currently, applying machine learning models for climate changes and sea-level rise is of great interest [20,92,110]. Several machine learning models have already been implemented to predict sea-level rise through regression functions [111,112] or track Hurricane forecasting [113,114].…”

Artificial Intelligence for Earth System Predictability (AI4ESP) (2021 Workshop Report)

“…In this context, nearshore and short-timescale processes (e.g., extreme coastal sea levels) have been the primary target of ML efforts (Sztobryn 2003;Bajo and Umgiesser 2010;French et al 2017), as discussed in section 7.3. More recently, the impact of offshore (natural) processes, such as internal climate variability, was examined through observational ocean temperature data (Nieves, Marcos, and Willis 2017) and ML techniques (Nieves, Radin, and Camps-Valls 2021) to predict regional coastal sea levels on timescales from 1 to 3 years. Input for short-term predictions may be obtained as well through statistical ML downscaling of climate models (Sithara, Pramada, and Thampi 2021).…”

Machine‐Learning Based Reconstructions of Past Regional Sea Level Variability From Proxy Data

“…As a practical alternative for reconstructing the missing values, in this paper, we propose the use of region-specific sea level proxy data from upper-ocean temperature estimates. These estimates have previously proven to be successful in revealing observed coastal sea level changes associated with natural climate fluctuations (Nieves et al, 2021). Our further goal is to provide a novel modeling framework, developed with ML methods and physical knowledge (tide gauge observations and proxy data) to accommodate the different situations (described above and) generally encountered in the reconstruction of past sea levels on regional scales.…”

“…As an illustrative example, three regions (one for each ocean basin) are only shown here. Our models are, however, applicable to any coastal region on the globe most impacted by internal climate variability (Nieves et al., 2021). For each region, we chose one station with significant gaps (a station with at least a missing observation segment of several consecutive years) to reconstruct the incomplete record.…”

“…This knowledge is essential to understand current changes and improve future sea level predictions in response to climatic shifts. In fact, a recent study suggests that internal climate variability is still substantially impacting sea level in many coastal regions around the globe (Nieves et al, 2021). However, unfortunately, sea level records from tide gauges around the world's coastlines frequently include gaps (with over 50% missing data points at many stations) and biases (due to for example vertical land movements) (Holgate et al, 2013), which limits the study of past local and regional sea level variations.…”