Reconstruction of Ocean Color Data Using Machine Learning Techniques in Polar Regions: Focusing on Off Cape Hallett, Ross Sea

Abstract

The most problematic issue in the ocean color application is the presence of heavy clouds, especially in polar regions. For that reason, the demand for the ocean color application in polar regions is increased. As a way to overcome such issues, we conducted the reconstruction of the chlorophyll-a concentration (CHL) data using the machine learning-based models to raise the usability of CHL data. This analysis was first conducted on a regional scale and focused on the biologically-valued Cape Hallett, Ross Sea, Antarctica. Environmental factors and geographical information associated with phytoplankton dynamics were considered as predictors for the CHL reconstruction, which were obtained from cloud-free microwave and reanalysis data. As the machine learning models used in the present study, the ensemble-based models such as Random forest (RF) and Extremely randomized tree (ET) were selected with 10-fold cross-validation. As a result, both CHL reconstructions from the two models showed significant agreement with the standard satellite-derived CHL data. In addition, the reconstructed CHLs were close to the actual CHL value even where it was not observed by the satellites. However, there is a slight difference between the CHL reconstruction results from the RF and the ET, which is likely caused by the difference in the contribution of each predictor. In addition, we examined the variable importance for the CHL reconstruction quantitatively. As such, the sea surface and atmospheric temperature, and the photosynthetically available radiation have high contributions to the model developments. Mostly, geographic information appears to have a lower contribution relative to environmental predictors. Lastly, we estimated the partial dependences for the predictors for further study on the variable contribution and investigated the contributions to the CHL reconstruction with changes in the predictors.

Evidence for the Impact of Climate Change on Primary Producers in the Southern Ocean

“…The Sen slope was also used to describe trends at the scale of MEASO zones, sectors and areas. Satellite-based estimates of chl-a and NPP fail at low solar elevations, and we have not attempted to "fill in" the winter gaps in satellite data (e.g., as Park et al, 2019). The proportion of missing data increases with latitude and so to avoid any potential bias in area-averages, trends were only calculated when more than half of potential observations for an area were present.…”

Phytoplankton Bloom Changes under Extreme Geophysical Conditions in the Northern Bering Sea and the Southern Chukchi Sea

“…The machine learning model used in this study is the random forest (RF) [23]. The overall framework is similar to previous works [21,24]. However, there were several different preprocessing steps for Chl-a images: (i) a 3 × 3 median filter was applied to Chl-a data, (ii) Chl-a images with only 0.1% pixels were excluded, and (iii) the pixels that were not observed 15 days before and after were also removed for continuity.…”

Sustainable Marine Ecosystems: Deep Learning for Water Quality Assessment and Forecasting

“…This information is critical to further our scientific understanding of biogeochemical oceanic processes, such as organic carbon production and export, phytoplankton dynamics, and responses to climatic disturbances [72]. Despite the current widespread availability of ocean color observations, mapping of ocean color is spatio-temporally limited and challenged by inconsistent information due to mainly cloud covers, particularly in polar regions [73]. These regions are usually covered by dense clouds throughout the year, limiting the valid range of satellite observations.…”