This thesis examines and compares measured and modeled ocean
current data at the Po River estuary. The measured data were collected using
an Acoustic Doppler Current Meter (AWAC), while the modeled data were
sourced from the CMEMS numerical model. The primary goal is to evaluate
the model’s accuracy and assess its effectiveness in predicting ocean currents
in a shallow coastal environment.
To ensure comparison between the datasets, the data were processed
through interpolation and averaging. Several statistical methods were used
in the analysis, including Root Mean Square Error (RMSE), Mean Absolute
Error (MAE), Euclidean distance, and cosine similarity. Additionally, we
trained a linear regression model and a Lasso regression model to determine
whether they could improve CMEMS model predictions.
The results reveal that the CMEMS numerical model does not accurately
replicate the observed ocean currents in the study area. Wind has the biggest
role in influencing currents at the measurement site, while the impact of river
discharge is comparatively less.
|
