Description: Increased pollution in the coastal areas may cause changes in the biodiversity of marine organisms depending upon their physiological capacity and resilience to thrive under stressing environmental conditions. The present research evaluates the heavy metals pollution degree of coastal waters using the macroalgae Ericaria selaginoides as bioindicator along the Atlantic coast of Morocco. Eight stations were chosen: two located near Eljadida city, three nearby Safi city and three around the city of Essaouira. Results showed that the heavy metal content in the thalli of E. selaginoides, in seawater and sediment varied seasonally. At the same time, it was negatively correlated with algal biodiversity onsite. However, the Chemical Oxygen Demand was significantly higher at the polluted station S5 than at other stations, while Dissolved Oxygen and Biological Oxygen Demand were lower. E. selaginoides accumulated metals in the following order Fe 〉 Zn 〉 Mn 〉 Cu 〉 Ni 〉 Pb 〉 Cr 〉 Cd. In conclusion, E. selaginoides is overall more resilient to heavy metal pollution than other marine organisms in the Atlantic coast of Morocco, as indicated by substantially elevated concentrations of heavy metals in some sites. Our results support that E. selaginoides would be a suitable bioindicator for monitoring of heavy metals in polluted coastal areas.

Description: High-quality data is necessary for modern machine learning. However, the acquisition of such data is difficult due to noisy and ambiguous annotations of humans. The aggregation of such annotations to determine the label of an image leads to a lower data quality. We propose a data-centric image classification benchmark with ten real-world datasets and multiple annotations per image to allow researchers to investigate and quantify the impact of such data quality issues. With the benchmark we can study the impact of annotation costs and (semi-)supervised methods on the data quality for image classification by applying a novel methodology to a range of different algorithms and diverse datasets. Our benchmark uses a two-phase approach via a data label improvement method in the first phase and a fixed evaluation model in the second phase. Thereby, we give a measure for the relation between the input labeling effort and the performance of (semi-)supervised algorithms to enable a deeper insight into how labels should be created for effective model training. Across thousands of experiments, we show that one annotation is not enough and that the inclusion of multiple annotations allows for a better approximation of the real underlying class distribution. We identify that hard labels can not capture the ambiguity of the data and this might lead to the common issue of overconfident models. Based on the presented datasets, benchmarked methods, and analysis, we create multiple research opportunities for the future directed at the improvement of label noise estimation approaches, data annotation schemes, realistic (semi-)supervised learning, or more reliable image collection.