Modeling Sea Surface Temperature Variability with Meteorological and Water Quality Indicators Using VAR and Prophet Forecasting Models
Sea Surface Temperature (SST) is a crucial indicator of global climate change and oceanic conditions, significantly affecting marine and coastal ecosystems. This study investigates the dynamics of SST and its variations by utilizing ground observations and remote sensing data for the southeastern Ar...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2025-07-01
|
| Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://isprs-annals.copernicus.org/articles/X-G-2025/239/2025/isprs-annals-X-G-2025-239-2025.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Sea Surface Temperature (SST) is a crucial indicator of global climate change and oceanic conditions, significantly affecting marine and coastal ecosystems. This study investigates the dynamics of SST and its variations by utilizing ground observations and remote sensing data for the southeastern Arabian Gulf, particularly near Dubai’s coastline. Correlation analysis uncovered relationships between SST and a selected set of water quality indicators and meteorological parameters. The Facebook Prophet model was applied to forecast SST and was proven capable of capturing seasonal variations and irregular spikes. Vector Autoregression (VAR) model was employed to analyze the influence of meteorological parameters on SST forecasting results and their interrelationships. The results demonstrate a significant impact of previous SST lags, particularly the third lag (p-value = 0.008), along with the notable influence of air temperature and wind speed on SST forecasting. The Prophet and VAR models yielded Root Mean Square Error (RMSE) values of 0.82 and 0.93, respectively. Furthermore, the study revealed an underlying relationship between SST, salinity, and nitrate concentrations, providing deeper insights into SST dynamics and water quality indices. The proposed analysis approach can be applied to study and understand other climatic applications with seasonal and limited time-series data, expanding its relevance to broader environmental studies. |
|---|---|
| ISSN: | 2194-9042 2194-9050 |