Human activities at sea can produce pressures and cumulative effects on ecosystem components that need to be monitored and assessed in a cost-effective manner. Five Horizon European projects have joined forces to collaboratively increase our knowledge and skills to monitor and assess the ocean in an innovative way, assisting managers and policy-makers in taking decisions to maintain sustainable activities at sea. Here, we present and discuss the status of some methods revised during a summer school, aiming at better management of coasts and seas. We include novel methods to monitor the coastal and ocean waters (e.g. environmental DNA, drones, imaging and artificial intelligence, climate modelling and spatial planning) and innovative tools to assess the status (e.g. cumulative impacts assessment, multiple pressures, Nested Environmental status Assessment Tool (NEAT), ecosystem services assessment or a new unifying approach). As a concluding remark, some of the most important challenges ahead are assessing the pros and cons of novel methods, comparing them with benchmark technologies and integrating these into long-standing time series for data continuity. This requires transition periods and careful planning, which can be covered through an intense collaboration of current and future European projects on marine biodiversity and ecosystem health.

Marine bacterioplankton play a crucial role in the cycling of carbon, nitrogen, and phosphorus in coastal waters. And the impact of environmental factors on bacterial community structure and ecological functions is a dynamic ongoing process. To systematically assess the relationship between environmental changes and bacterioplankton communities, this study delved into the spatiotemporal distribution and predicted metabolic characteristics of bacterioplankton communities at two estuarine beaches in Northern China. Coastal water samples were collected regularly in spring, summer, and autumn, and were analyzed in combination with environmental parameters and bacterioplankton community. Results indicated significant seasonal variations in bacterioplankton communities as Bacteroidetes and Actinobacteria were enriched in spring, Cyanobacteria proliferated in summer. While Pseudomonadota and microorganisms associated with organic matter decomposition prevailed in autumn, closely linked to seasonal variation of temperature, light and nutrients such as nitrogen and phosphorus. Particularly in summer, increased tourism activities and riverine inputs significantly raised nutrient levels, promoting the proliferation of specific photosynthetic microorganisms, potentially linked to the occurrence of phytoplankton blooms. Spearman correlation analysis further revealed significant correlations between bacterioplankton communities and environmental factors such as salinity, chlorophyll a, and total dissolved phosphorus (TDP). Additionally, the metabolic features of the spring bacterioplankton community were primarily characterized by enhanced activities in the prokaryotic carbon fixation pathways, reflecting rapid adaptation to increased light and temperature, as well as significant contributions to primary productivity. In summer, the bacterial communities were involved in enhanced glycolysis and biosynthetic pathways, reflecting high energy metabolism and responses to increased light and biomass. In autumn, microorganisms adapted to the accelerated decomposition of organic matter and the seasonal changes in environmental conditions through enhanced amino acid metabolism and material cycling pathways. These findings demonstrate that seasonal changes and human activities significantly influence the structure and function of bacterioplankton communities by altering nutrient dynamics and physical environmental conditions. This study provides important scientific insights into the marine biological responses under global change.

The Taihangshan-Yanshan region (TYR) is an important ecological barrier area for Beijing-Tianjin-Hebei, and the effectiveness of its ecological restoration and protection is of great significance to the ecological security pattern of North China. Based on the FVC data from 2000 to 2021, residual analysis, parametric optimal geodetector technique (OPGD) and multi-scale geographically weighted regression analysis (MGWR) were used to clarify the the multivariate driving mechanism of the evolution of FVC in the TYR. Results show that: (1) FVC changes in the TYR show a slowly fluctuating upward trend, with an average growth rate of 0.02/10a, and a spatial pattern of \"high in the northwest and low in the southeast\"; more than half of the FVC increased during the 22-year period. (2) The results of residual analysis showed that the effects of temperature and precipitation on FVC were very limited, and a considerable proportion (80.80% and 76.78%) of the improved and degraded areas were influenced by other factors. (3) The results of OPGD showed that the main influencing factors of the spatial differentiation of FVC included evapotranspiration, surface temperature, land use type, nighttime light intensity, soil type, and vegetation type (q > 0.2); The explanatory rates of the two-factor interactions were greater than those of the single factor, which showed either nonlinear enhancement or bifactorial enhancement, among which, the interaction of evapotranspiration with mean air and surface temperature has the strongest effect on the spatial and temporal evolution of FVC (q = 0.75). Surface temperature between 4.98 and 10.4 °C, evapotranspiration between 638 and 762 mm/a, and nighttime light between 1.96 and 7.78 lm/m2 favoured an increase in vegetation cover, and vegetation developed on lysimetric soils was more inclined to be of high cover. (4) The correlation between each variable and FVC showed different performance, GDP, elevation, slope and FVC showed significant positive correlation in most regions, while population size, urban population proportion, GDP proportion of primary and secondary industries, and nighttime light intensity all showed negative correlation with FVC to different degrees. The results can provide data for formulating regional environmental protection and restoration policies.

Human activity recognition has a wide range of applications in various fields, such as video surveillance, virtual reality and human-computer intelligent interaction. It has emerged as a significant research area in computer vision. GCN (Graph Convolutional networks) have recently been widely used in these fields and have made great performance. However, there are still some challenges including over-smoothing problem caused by stack graph convolutions and deficient semantics correlation to capture the large movements between time sequences. Vision Transformer (ViT) is utilized in many 2D and 3D image fields and has surprised results. In our work, we propose a novel human activity recognition method based on ViT (HAR-ViT). We integrate enhanced AGCL (eAGCL) in 2s-AGCN to ViT to make it process spatio-temporal data (3D skeleton) and make full use of spatial features. The position encoder module orders the non-sequenced information while the transformer encoder efficiently compresses sequence data features to enhance calculation speed. Human activity recognition is accomplished through multi-layer perceptron (MLP) classifier. Experimental results demonstrate that the proposed method achieves SOTA performance on three extensively used datasets, NTU RGB+D 60, NTU RGB+D 120 and Kinetics-Skeleton 400.

Spatial and temporal associations between sympatric species underpin biotic interactions, structure ecological assemblages, and sustain ecosystem functioning and stability. However, the resilience of interspecific spatiotemporal associations to human activity remains poorly understood, particularly in mountain forests where anthropogenic impacts are often pervasive. Here, we applied context-dependent Joint Species Distribution Models to a systematic camera-trap survey dataset from a global biodiversity hotspot in eastern Himalayas to understand how prominent human activities in mountain forests influence species associations within terrestrial mammal communities. We obtained 10,388 independent detections of 17 focal species (12 carnivores and five ungulates) from 322 stations over 43,163 camera days of effort. We identified a higher incidence of positive associations in habitats with higher levels of human modification (87%) and human presence (83%) compared to those located in habitats with lower human modification (64%) and human presence (65%) levels. We also detected a significant reduction of pairwise encounter time at increasing levels of human disturbance, corresponding to more frequent encounters between pairs of species. Our findings indicate that human activities can push mammals together into more frequent encounters and associations, which likely influences the coexistence and persistence of wildlife, with potential far-ranging ecological consequences.

There has been a resurgence of applications focused on human activity recognition (HAR) in smart homes, especially in the field of ambient intelligence and assisted-living technologies. However, such applications present numerous significant challenges to any automated analysis system operating in the real world, such as variability, sparsity, and noise in sensor measurements. Although state-of-the-art HAR systems have made considerable strides in addressing some of these challenges, they suffer from a practical limitation: they require successful pre-segmentation of continuous sensor data streams prior to automated recognition, i.e., they assume that an oracle is present during deployment, and that it is capable of identifying time windows of interest across discrete sensor events. To overcome this limitation, we propose a novel graph-guided neural network approach that performs activity recognition by learning explicit co-firing relationships between sensors. We accomplish this by learning a more expressive graph structure representing the sensor network in a smart home in a data-driven manner. Our approach maps discrete input sensor measurements to a feature space through the application of attention mechanisms and hierarchical pooling of node embeddings. We demonstrate the effectiveness of our proposed approach by conducting several experiments on CASAS datasets, showing that the resulting graph-guided neural network outperforms the state-of-the-art method for HAR in smart homes across multiple datasets and by large margins. These results are promising because they push HAR for smart homes closer to real-world applications.

Monitoring activities of daily living (ADLs) plays an important role in measuring and responding to a person\'s ability to manage their basic physical needs. Effective recognition systems for monitoring ADLs must successfully recognize naturalistic activities that also realistically occur at infrequent intervals. However, existing systems primarily focus on either recognizing more separable, controlled activity types or are trained on balanced datasets where activities occur more frequently. In our work, we investigate the challenges associated with applying machine learning to an imbalanced dataset collected from a fully in-the-wild environment. This analysis shows that the combination of preprocessing techniques to increase recall and postprocessing techniques to increase precision can result in more desirable models for tasks such as ADL monitoring. In a user-independent evaluation using in-the-wild data, these techniques resulted in a model that achieved an event-based F1-score of over 0.9 for brushing teeth, combing hair, walking, and washing hands. This work tackles fundamental challenges in machine learning that will need to be addressed in order for these systems to be deployed and reliably work in the real world.

Human Activity Recognition (HAR) plays an important role in the automation of various tasks related to activity tracking in such areas as healthcare and eldercare (telerehabilitation, telemonitoring), security, ergonomics, entertainment (fitness, sports promotion, human-computer interaction, video games), and intelligent environments. This paper tackles the problem of real-time recognition and repetition counting of 12 types of exercises performed during athletic workouts. Our approach is based on the deep neural network model fed by the signal from a 9-axis motion sensor (IMU) placed on the chest. The model can be run on mobile platforms (iOS, Android). We discuss design requirements for the system and their impact on data collection protocols. We present architecture based on an encoder pretrained with contrastive learning. Compared to end-to-end training, the presented approach significantly improves the developed model\'s quality in terms of accuracy (F1 score, MAPE) and robustness (false-positive rate) during background activity. We make the AIDLAB-HAR dataset publicly available to encourage further research.

Understanding the interactions among anthropogenic stressors is critical for effective conservation and management of ecosystems. Freshwater scientists have invested considerable resources in conducting factorial experiments to disentangle stressor interactions by testing their individual and combined effects. However, the diversity of stressors and systems studied has hindered previous syntheses of this body of research. To overcome this challenge, we used a novel machine learning framework to identify relevant studies from over 235,000 publications. Our synthesis resulted in a new dataset of 2396 multiple-stressor experiments in freshwater systems. By summarizing the methods used in these studies, quantifying trends in the popularity of the investigated stressors, and performing co-occurrence analysis, we produce the most comprehensive overview of this diverse field of research to date. We provide both a taxonomy grouping the 909 investigated stressors into 31 classes and an open-source and interactive version of the dataset (https://jamesaorr.shinyapps.io/freshwater-multiple-stressors/). Inspired by our results, we provide a framework to help clarify whether statistical interactions detected by factorial experiments align with stressor interactions of interest, and we outline general guidelines for the design of multiple-stressor experiments relevant to any system. We conclude by highlighting the research directions required to better understand freshwater ecosystems facing multiple stressors.

BACKGROUND: The range of Glires is influenced by human activities and climate change. However, the extent to which human activities and environmental changes have contributed to this relationship remains unclear. We examined alterations in the distribution changes and driving factors of the Himalayan marmot, plateau pika, and plateau zokor on the Qinghai-Tibet Plateau (QTP) using the maximum entropy (MaxEnt) model and a geographical detector (Geodetector).
RESULTS: The MaxEnt model showed that the contribution rates of the human footprint index (HFI) to the distribution patterns of the three types of Glires were 46.70%, 58.70%, and 59.50%, respectively. The Geodetector results showed that the distribution pattern of the Himalayan marmot on the QTP was influenced by altitude and the normalized difference vegetation index (NDVI). The distribution patterns for plateau pikas and plateau zokors were driven by HFI and NDVI. Climate has played a substantial role in shaping suitable habitats for these three Glires on the QTP. Their suitable area is expected to decrease over the next 30-50 years, along with their niche breadth and overlap. Future suitable habitats for the three Glires tended to shift toward higher latitudes on the QTP.
CONCLUSIONS: These findings underscore the impacts of environmental and human factors on the distribution of the three Glires on the QTP. They have enhanced our understanding of the intricate relationships between Glires niches and environments. This can aid in identifying necessary interventions for developing effective early warning systems and prevention strategies to mitigate Glires infestations and plague epidemics on the QTP. © 2024 Society of Chemical Industry.