Uncertainties in wave energy potential estimation: case study at MARELAB

Abstract

Accurate and reliable wave resource characterization is crucial for assessing the feasibility and energy potential of wave energy technologies. Reanalysis models, like ERA5, are often used in preliminary site assessments, but the uncertainties inherent in applying these models are frequently overlooked. Although ERA5 is a valuable tool, its accuracy for wave energy assessments in specific regions, such as the Mediterranean, is still being explored [1]. The region's unique meteo-climatic conditions, characterized by calmer seas and more localized wave patterns compared to open oceans or northern seas, may lead to discrepancies between model predictions and observed data [2].

This study quantifies the divergences between freely available wave data from the ERA5 reanalysis model and in situ measurements, focusing on their impact on wave energy potential estimations in the Mediterranean Sea. The analysis is based on data collected at MaRELab, a field laboratory located in the Gulf of Naples, Italy, in the Middle Tyrrhenian Sea, which provides an ideal site for testing marine energy systems [3].

The Marine Renewable Energy Laboraroty (MaRELab) site is equipped with various environmental measurement instruments, including a DWSD-type wave buoy that provides significant wave height, peak period, and wave direction data, as well as an Acoustic Doppler Current Profiler (ADCP), which also measures the vertical wave profile. These in situ measurements are essential for understanding local wave dynamics and predicting the performance of wave energy devices [4].

A statistical analysis is here conducted to quantify the level of agreement between ERA5 model outputs and in-situ observed data, with a particular focus on key wave parameters, such as significant wave height and peak wave period. The analysis utilized common metrics, including mean error, root mean square error (RMSE), and correlation coefficients, to assess the magnitude and frequency of differences between the two datasets.

The primary objective of this study is to highlight the potential errors in wave energy assessments when using ERA5 reanalysis data, particularly in regions like the Mediterranean, where local conditions may not fully align with the model's assumptions. By examining these differences, the research aims to improve the accuracy and reliability of wave energy resource assessments. The findings emphasize the importance of accounting for these uncertainties in preliminary feasibility studies, particularly in areas where only reanalysis data is available. Ultimately, this study aims to support more informed decision-making in the planning and development of wave energy projects.