Eddies in the Arctic Ocean Revealed from MODIS Optical Imagery

Abstract

Here we investigate properties of ocean eddies in the key Arctic region of the northern Greenland Sea and the Fram Strait using visible and infrared Moderate Resolution Imaging Spectroradiometer (MODIS) Aqua data acquired from April to September in 2007 and 2018–2020. We infer eddy properties using visual identification and automated processing of their signatures in sea surface temperature (SST) and chlorophyll-a (chl-a) maps, and their gradients. Altogether, 450 (721) eddies were identified in SST (chl-a) data. Their radii span from 2 to 40 km (mean value 12 km). Most eddies are elliptical with a mean aspect ratio (eccentricity) of their axes equal 0.77 (0.64). Cyclones are smaller than anticyclones and prevail in both data sources. Cyclones tend to be more prevalent over shallow shelves, and anticyclones over deep water regions. Peak eddy activity is registered in June, while chl-a data also possess a second peak in April. In SST, the highest eddy probability is found along the East Greenland Current in the Nordbukta region at 76–78°N and along the West Spitsbergen Current at 78–80°N. In chl-a, most of them are observed in the central Fram Strait. The overall number of eddies with a positive chl-a anomaly, dominated by cyclones, is larger (62%) than that with a negative one (~38%). The number of eddies with positive and negative SST anomalies is nearly equal. Eddy translation velocities are 0.9–9.6 km/day (mean value 4.2 km/day). Despite frequent cloud and ice cover, MODIS data is a rich source of information on eddy generation hot-spots, their spatial properties, dynamics and associated SST and chl-a anomalies in the Arctic Ocean.

Variations and Depth of Formation of Submesoscale Eddy Structures in Satellite Ocean Color Data in the Southwestern Region of the Peter the Great Bay

“…The weakest manifestation of the investigated eddies was observed in temperature contrasts compared to almost any optical characteristics. This can be considered as a regional and seasonal result, because the opposite examples are also widespread, when the eddy manifests itself more strongly in temperature contrasts than in bio-optical contrasts [23]. During the study period, the temperature of the upper quasi-homogeneous layer is well warmed and mixed in the analyzed region.…”

“…In the satellite IR images, eddies are distinguished due to sea surface temperature contrasts, which is a basic oceanographic parameter [21,22]. However, for some areas, in a certain season, the manifestations of eddies in thermal contrasts may be weak or entirely absent [23]. In cases when satellite IR sensing fails to identify eddies due to small contrasts in the sea surface temperature fields, data in the visible range can be used, where the contrast can be significant due to different concentrations of chlorophyll-a (chl-a) [23][24][25], suspended solids (SSs) [26], colored dissolved organic matter (CDOM) [27], and even due to the distribution of broken ice [28].…”

“…However, for some areas, in a certain season, the manifestations of eddies in thermal contrasts may be weak or entirely absent [23]. In cases when satellite IR sensing fails to identify eddies due to small contrasts in the sea surface temperature fields, data in the visible range can be used, where the contrast can be significant due to different concentrations of chlorophyll-a (chl-a) [23][24][25], suspended solids (SSs) [26], colored dissolved organic matter (CDOM) [27], and even due to the distribution of broken ice [28]. These characteristics from satellite ocean color data are calculated from measurements of remotely sensed reflectance (Rrs) and can be attributed to optical characteristics of seawater.…”

“…It is important to note that the availability of high-quality satellite images in the optical range is not consistent from day to day due to cloud cover variability, both in the visible and IR parts of the spectrum. In this regard, it is not always possible to track the trajectories of movement and the number of eddies in the sea [23]. This is especially aggravated by the fact that the study of submesoscale eddies mainly requires high spatial resolution satellite data (pixels less than 100 m), which are not available for every day.…”

“…Previous studies on this topic typically focus on analyzing the amplitudes of changes in characteristics inside and outside of eddies [23,27,32], without considering their significance in relation to statistical noise. The knowledge of contrast/noise characteristics is necessary to investigate the quality of detection of submesoscale eddies in satellite images.…”