Satellite-derived estimates of aerosol optical depth (AOD) are key predictors in particulate air pollution models. The multi-step retrieval algorithms that estimate AOD also produce quality control variables but these have not been systematically used to address the measurement error in AOD. We compare three machine-learning methods: random forests, gradient boosting, and extreme gradient boosting (XGBoost) to characterize and correct measurement error in the Multi-Angle Implementation of Atmospheric Correction (MAIAC) 1 × 1 km AOD product for Aqua and Terra satellites across the Northeastern/Mid-Atlantic USA versus collocated measures from 79 ground-based AERONET stations over 14 years. Models included 52 quality control, land use, meteorology, and spatially-derived features. Variable importance measures suggest relative azimuth, AOD uncertainty, and the AOD difference in 30–210 km moving windows are among the most important features for predicting measurement error. XGBoost outperformed the other machine-learning approaches, decreasing the root mean squared error in withheld testing data by 43% and 44% for Aqua and Terra. After correction using XGBoost, the correlation of collocated AOD and daily PM_2.5 monitors across the region increased by 10 and 9 percentage points for Aqua and Terra. We demonstrate how machine learning with quality control and spatial features substantially improves satellite-derived AOD products for air pollution modeling.

A previous multi-locus lineage (MLL) analysis of SSR-microsatellite data of old olive trees in the southeast Mediterranean area had shown the predominance of the Souri cultivar (MLL1) among grafted trees. The MLL analysis had also identified an MLL (MLL7) that was more common among rootstocks than other MLLs. We here present a comparison of the MLL combinations MLL1 (scion)/MLL7 (rootstock) and MLL1/MLL1 in order to investigate the possible influence of rootstock on scion phenotype. A linear regression analysis demonstrated that the abundance of MLL1/MLL7 trees decreases and of MLL1/MLL1 trees increases along a gradient of increasing aridity. Hypothesizing that grafting on MLL7 provides an advantage under certain conditions, Akaike information criterion (AIC) model selection procedure was used to assess the influence of different environmental conditions on phenotypic characteristics of the fruits and oil of the two MLL combinations. The most parsimonious models indicated differential influences of environmental conditions on parameters of olive oil quality in trees belonging to the MLL1/MLL7 and MLL1/MLL1 combinations, but a similar influence on fruit characteristics and oil content. These results suggest that in certain environments grafting of the local Souri cultivar on MLL7 rootstocks and the MLL1/MLL1 combination result in improved oil quality. The decreasing number of MLL1/MLL7 trees along an aridity gradient suggests that use of this genotype combination in arid sites was not favoured because of sensitivity of MLL7 to drought. Our results thus suggest that MLL1/MLL7 and MLL1/MLL1 combinations were selected by growers in traditional rain-fed cultivation under Mediterranean climate conditions in the southeast Mediterranean area.

This study introduces a laboratory experiment designed to assess the influence of annotation strategies, levels of class imbalance, and prior experience, on the performance of human annotators. The experiment focuses on labeling aerial imagery, using ArcGIS Pro, to detect and segment small-scale PVs, selected as a case study for rectangular objects. The experiment is conducted using images with a pixel size of 0.15m, involving both expert and non-expert participants, across different setup strategies and target-background ratio datasets. Our findings indicate that annotators generally perform more effectively in object detection than in segmentation tasks. A marked tendency to commit more Type II errors (False Negatives, i.e., undetected objects) than Type I errors (False Positives, i.e. falsely detecting objects that do not exist) was observed across all experimental setups and conditions, suggesting a consistent bias in annotation processes. Performance was better in tasks with higher target-background ratios (i.e., more objects per unit area). Prior experience did not significantly impact performance and may, in some cases, even lead to overestimation in segmentation. These results provide evidence that annotators are relatively cautious and tend to identify objects only when they are confident about them, prioritizing underestimation over overestimation. Annotators’ performance is also influenced by object scarcity, showing a decline in areas with extremely imbalanced class datasets and a low ratio of target-to-background. These findings may enhance annotation strategies for remote sensing research while efficient human annotators are crucial in an era characterized by growing demands for high-quality training data to improve segmentation and detection models.