Abstract
ABSTRACT Continued global warming leads to an increase in the frequency and intensity of marine heatwave events. Monitoring of spatial-temporal variations in sea surface temperature (SST) is crucial for mitigating the environmental and economic damage caused by marine heatwaves to ecosystems and fisheries. While geostationary satellites can provide hourly SST data at local and large scales, cloud cover poses a fundamental limitation due to its shielding of the infrared radiation from the ocean surface. To address this limitation, we applied Data Interpolating Empirical Orthogonal Functions (DINEOF) to hourly GEO-KOMPSAT-2A/Advanced Meteorological Imager (GK2A/AMI) SST data to produce gap-filled SST datasets. Validation of the gap-free AMI SST data was performed using in situ buoy data. Our results showed that the errors in the gap-free AMI SST did not significantly differ from those in the original satellite data. Notably, the gap-free SST exhibited lower errors even in the high-SST range (above 28 ~ 31 °C), confirming its ability to effectively monitor extreme ocean warming events. Seamless SST data played a crucial role in identifying and comprehensively analyzing the record-breaking East Asian marine heatwave event in 2021, tracing its evolution from the formation to decaying phases. These advanced data not only provide essential insights into the characteristics and impacts of marine heatwaves but also enable a detailed examination of their diurnal cycle by restoring data with an average of 75% missing values (79.34%). Moreover, these seamless SST data serve as inputs for numerical modeling and the deep learning algorithms and has immense potential for both forecasting marine heatwaves and implementing a 24-h continuous monitoring technique. These benefits extend beyond historical event detection, enabling the anticipation and mitigation of oceanic phenomena through proactive measures.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have