An estimation of the Ocean Wave Heights using terrestrially seismic data
Traditionally, there are different approaches to monitoring the ocean wave field consisting of 1) measurements using insitu buoys, 2) numerical ocean wave modelling using wind forecast, and 3) satellite altimetry. Each of these ocean wave monitoring techniques have their own advantages and disadvant...
| Main Authors: | , |
|---|---|
| Format: | Conference Object |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://dair.dias.ie/id/eprint/1400/ https://dair.dias.ie/id/eprint/1400/1/Poster-EGU.pdf |
| Summary: | Traditionally, there are different approaches to monitoring the ocean wave field consisting of 1) measurements using insitu buoys, 2) numerical ocean wave modelling using wind forecast, and 3) satellite altimetry. Each of these ocean wave monitoring techniques have their own advantages and disadvantages associated with their spatial and temporal resolution. For example, buoys are physical point measurements with excellent temporal resolution (e.g., sub-hourly), but their spatial resolution is very poor (e.g., single point in space). Buoys are also expensive to maintain; ‘Real-time’ wave height estimations from numerical wave modelling is based on forecast wind, hence the model accuracy is dependent on wind prediction accuracy. . Compare to buoys, the temporal resolution of numerical models is poor (e.g., every 3 hours), but the spatial resolution is much better (various resolutions depending on the grid size); Satellite altimetry looks over a large region so the spatial coverage is very good but the temporal resolution is very poor (e.g., once every four days). In this work we consider terrestrial seismic (microseism) data as a proxy for wave heights. Under certain conditions, it has the potential for combined good spatial and temporal resolution, in quasireal time. This technique is based on the relationship between secondary microseism amplitudes recorded on land and the ocean wave-wave interactions which excite the sea floor, generating these secondary microseisms. Here we take a data driven approach, implementing an Artificial Neural Network (ANN) to quantify the complex underlying relationship between ocean wave height and microseism amplitude. Thus far we trained the ANN using the available seismic and numerical simulated data and then used the trained ANN to estimate significant Ocean Wave Height (SWH) at particular location(s) in the Northeast Atlantic using amplitudes from seismic station distributed across Ireland. Our preliminary results look very promising and show relatively small residuals for ... |
|---|