Rapid flotation of Microcystis wesenbergii mediated by high light exposure: implications for surface scum formation and cyanobacterial species succession.

Increasing occurrences of Microcystis surface scum have been observed in the context of global climate change and the increase in anthropogenic pollution, causing deteriorating water quality in aquatic ecosystems. Previous studies on scum formation mainly focus on the buoyancy-driven floating process of larger Microcystis colonies, neglecting other potential mechanisms. To study the non-buoyancy-driven rapid flotation of Microcystis, we here investigate the floating processes of two strains of single-cell species (Microcystis aeruginosa and Microcystis wesenbergii), which are typically buoyant, under light conditions (150 µmol photons s-1 m-2). Our results showed that M. wesenbergii exhibited fast upward migration and formed surface scum within 4 hours, while M. aeruginosa did not form visible scum throughout the experiments. To further explore the underlying mechanism of these processes, we compared the dissolved oxygen (DO), extracellular polymeric substance (EPS) content, and colony size of Microcystis in different treatments. We found supersaturated DO and the formation of micro-bubbles (50-200 µm in diameter) in M. wesenbergii treatments. M. aeruginosa produces bubbles in small quantities and small sizes. Additionally, M. wesenbergii produced more EPS and tended to aggregate into larger colonies. M. wesenbergii had much more derived-soluble extracellular proteins and polysaccharides compared to M. aeruginosa. At the same time, M. wesenbergii contains abundant functional groups, which was beneficial to the formation of agglomerates. The surface scum observed in M. wesenbergii is likely due to micro-bubbles attaching to the surface of cell aggregates or becoming trapped within the colony. Our study reveals a species-specific mechanism for the rapid floatation of Microcystis, providing novel insights into surface scum formation as well as succession of cyanobacterial species.

Contrastive representation learning on dynamic networks.

Representation learning for dynamic networks is designed to learn the low-dimensional embeddings of nodes that can well preserve the snapshot structure, properties and temporal evolution of dynamic networks. However, current dynamic network representation learning methods tend to focus on estimating or generating observed snapshot structures, paying excessive attention to network details, and disregarding distinctions between snapshots with larger time intervals, resulting in less robustness for sparse or noisy networks. To alleviate these challenges, this paper proposes a contrastive mechanism for temporal representation learning on dynamic networks, inspired by the success of contrastive learning in visual and static network representation learning. This paper proposes a novel Dynamic Network Contrastive representation Learning (DNCL) model. Specifically, contrast objective functions are constructed using intra-snapshot and inter-snapshot contrasts to capture the network topology, node feature information, and network evolution information, respectively. Rather than estimating or generating ground-truth network features, the proposed approach maximizes mutual information between nodes from different time steps and views generated. The experimental results of link prediction, node classification, and clustering on several real-world and synthetic networks demonstrate the superiority of DNCL over state-of-the-art methods, indicating the effectiveness of the proposed approach for dynamic network representation learning.

Dynamics of Microcystis surface scum formation under different wind conditions: the role of hydrodynamic processes at the air-water interface.

Due to climate change, Microcystis blooms occur at increasing frequencies in aquatic ecosystems worldwide. Wind-generated turbulence is a crucial environmental stressor that can vertically disperse the Microcystis surface scum, reducing its light availability. Yet, the interactions of Microcystis scum with the wind-generated hydrodynamic processes, particularly those at the air-water interface, remain poorly understood. Here, we explore the response of Microcystis (including colony size and migration dynamics) to varying magnitudes and durations of intermittent wind disturbances in a mesocosm system. The flow velocities, size of Microcystis colonies, and the areal coverage of the water surface by scum were measured through video observations. Our results demonstrate that low wind speeds increase colony size by providing a stable condition where Microcystis forms a scum layer and aggregates into large colonies at the air-water interface. In contrast, wind disturbances disperse scum and generate turbulence, resulting in smaller colonies with higher magnitudes of wind disturbance. We observed that surface scum can form rapidly following a long period (6 h) of high-magnitude (4.5 m s-1) wind disturbance. Furthermore, our results indicate reduced water surface tension caused by the presence of Microcystis, which can decrease surface flow velocity and counteract wind-driven mixing. The reduced surface tension may also drive lateral convection at the water surface. These findings suggest that Microcystis reduces surface tension, likely by releasing surface-active materials, as an adaptive response to various wind conditions. This could result in an increased rate of surface scum re-formation under wind conditions and potentially facilitate the lateral expansion of scum patches during weak wind periods. This study reveals new insights into how Microcystis copes with different wind conditions and highlights the importance of the air-water interface for Microcystis scum dynamics.

Neural Networks based Smart e-Health Application for the Prediction of Tuberculosis using Serverless Computing.

The convergence of the Internet of Things (IoT) with e-health records is creating a new era of advancements in the diagnosis and treatment of disease, which is reshaping the modern landscape of healthcare. In this paper, we propose a neural networks-based smart e-health application for the prediction of Tuberculosis (TB) using serverless computing. The performance of various Convolution Neural Network (CNN) architectures using transfer learning is evaluated to prove that this technique holds promise for enhancing the capabilities of IoT and e-health systems in the future for predicting the manifestation of TB in the lungs. The work involves training, validating, and comparing Densenet-201, VGG-19, and Mobilenet-V3-Small architectures based on performance metrics such as test binary accuracy, test loss, intersection over union, precision, recall, and F1 score. The findings hint at the potential of integrating these advanced Machine Learning (ML) models within IoT and e-health frameworks, thereby paving the way for more comprehensive and data-driven approaches to enable smart healthcare. The best-performing model, VGG-19, is selected for different deployment strategies using server and serless-based environments. We used JMeter to measure the performance of the deployed model, including the average response rate, throughput, and error rate. This study provides valuable insights into the selection and deployment of ML models in healthcare, highlighting the advantages and challenges of different deployment options. Furthermore, it also allows future studies to integrate such models into IoT and e-health systems, which could enhance healthcare outcomes through more informed and timely treatments.

Selection of photosynthetic traits by turbulent mixing governs formation of cyanobacterial blooms in shallow eutrophic lakes.

Prediction of the complex cyanobacteria-environment interactions is vital for understanding harmful bloom formation. Most previous studies on these interactions considered specific properties of cyanobacterial cells as representative for the entire population (e.g. growth rate, mortality, and photosynthetic capacity (Pmax)), and assumed that they remained spatiotemporally unchanged. Although, at the population level, the alteration of such traits can be driven by intraspecific competition, little is known about how traits and their plasticity change in response to environmental conditions and affect the bloom formation. Here we test the hypothesis that intraspecific variations in Pmax of cyanobacteria (Microcystis spp.) play an important role in its population dynamics. We coupled a one-dimensional hydrodynamic model with a trait-based phytoplankton model to simulate the effects of physical drivers (turbulence and turbidity) on the Pmax of Microcystis populations for a range of dynamic conditions typical for shallow eutrophic lakes. Our results revealed that turbulence acts as a directional selective driver for changes in Pmax. Depending on the intensity of daily-periodic turbulence, representing wind-driven mixing, a shift in population-averaged phenotypes occurred toward either low Pmax, allowing the population to capture additional light in the upper layers, or high Pmax, enhancing the efficiency of light utilization. Moreover, we observed that a high intraspecific diversity in Pmax accelerated the formation of surface scum by up to more than four times compared to a lower diversity. This study offers insights into mechanisms by which cyanobacteria populations respond to turbulence and underscores the significance of intraspecific variations in cyanobacterial bloom formation.

A Novel Sequence-to-Sequence-Based Deep Learning Model for Multistep Load Forecasting.

Load forecasting is critical to the task of energy management in power systems, for example, balancing supply and demand and minimizing energy transaction costs. There are many approaches used for load forecasting such as the support vector regression (SVR), the autoregressive integrated moving average (ARIMA), and neural networks, but most of these methods focus on single-step load forecasting, whereas multistep load forecasting can provide better insights for optimizing the energy resource allocation and assisting the decision-making process. In this work, a novel sequence-to-sequence (Seq2Seq)-based deep learning model based on a time series decomposition strategy for multistep load forecasting is proposed. The model consists of a series of basic blocks, each of which includes one encoder and two decoders; and all basic blocks are connected by residuals. In the inner of each basic block, the encoder is realized by temporal convolution network (TCN) for its benefit of parallel computing, and the decoder is implemented by long short-term memory (LSTM) neural network to predict and estimate time series. During the forecasting process, each basic block is forecasted individually. The final forecasted result is the aggregation of the predicted results in all basic blocks. Several cases within multiple real-world datasets are conducted to evaluate the performance of the proposed model. The results demonstrate that the proposed model achieves the best accuracy compared with several benchmark models.

On secondary structure avoidance of codes for DNA storage.

A secondary structure in single-stranded DNA refers to its propensity to undergo self-folding, leading to functional inactivity and irreparable failures within DNA storage systems. Consequently, the property of secondary structure avoidance (SSA) becomes a crucial criterion in the design of single-stranded DNA sequences for DNA storage, as it prohibits the inclusion of reverse-complement subsequences that contribute to such structures. This work is specifically focused on addressing the avoidance of secondary structures in single-stranded DNA sequences. We propose a novel sequence replacement approach, which successfully resolves the SSA problem under conditions where the stem exceeds a length of 2log2⁡n+2, and the loop is of length k≥4. These parameters have been carefully chosen to closely resemble the real-world scenarios encountered in biochemical processes, enhancing the practical relevance of our study.

DRL-Based URLLC-Constraint and Energy-Efficient Task Offloading for Internet of Health Things.

Internet of Health Things (IoHT) is a promising e-Health paradigm that involves offloading numerous computational-intensive and delay-sensitive tasks from locally limited IoHT points to edge servers (ESs) with abundant computational resources in close proximity. However, existing computation offloading techniques struggle to meet the burgeoning health demands in ultra-reliable and low-latency communication (URLLC), one of the 5G application scenarios. This paper proposes a Multi-Agent Soft-Actor-Critic-discrete based URLLC-constrained task offloading and resource allocation (MASACDUA) scheme to maximize throughput while minimizing power consumption on the remote side, considering the long-term URLLC constraints. The URLLC constraint conditions are formulated using extreme value theory, and Lyapunov optimization is employed to divide the problem into task offloading and computation resource allocation. MASAC-discrete and a queue backlog-aware algorithm are utilized to approach task offloading and computation resource allocation, respectively. Extensive simulation results demonstrate that MASACDUA outperforms traditional DRL algorithms under different IoHT points and data arrival rate intervals and achieves superior performance in delay, bound violation probability, and other characteristics related to URLLC.

Effect of light-mediated variations of colony morphology on the buoyancy regulation of Microcystis colonies.

Light is an important driver of algal growth and for the formation of surface blooms. Long-term buoyancy maintenance of Microcystis colonies is crucial for their aggregation at the water surface and the following algal bloom development. However, the effect of light-mediated variations of colony morphology on the buoyancy regulation of Microcystis colonies remains unclear. In this study, growth parameters, colony morphology and floatation/sinking performance of Microcystis colonies were determined to explore how variations in colony morphology influence the buoyancy of colonies under different light conditions. We quantified colony compactness through the cell volume to colony volume ratio (VR) and found different responses of colony size and VR under different light intensities. Microcystis colonies with higher VR could stay longer at the water surface under low light conditions, which was beneficial for the long-term growth and buoyancy maintenance. However, increased colony size and decreased compactness were observed at a later growth stage under relatively higher light intensity (i.e., >108 µmol photons m-2 s-1). Interestingly, we found a counterintuitive negative correlation between colony size and buoyancy of Microcystis under high light intensity. Additionally, we found that the influence of colony morphology on buoyancy was stronger at high light intensity. These results indicate that light could regulate the buoyancy via colonial morphology and that the role of colony morphology in buoyancy regulation needs to be accounted for in further studies under variable environmental conditions.

Multiple errors correction for position-limited DNA sequences with GC balance and no homopolymer for DNA-based data storage.

Deoxyribonucleic acid (DNA) is an attractive medium for long-term digital data storage due to its extremely high storage density, low maintenance cost and longevity. However, during the process of synthesis, amplification and sequencing of DNA sequences with homopolymers of large run-length, three different types of errors, namely, insertion, deletion and substitution errors frequently occur. Meanwhile, DNA sequences with large imbalances between GC and AT content exhibit high dropout rates and are prone to errors. These limitations severely hinder the widespread use of DNA-based data storage. In order to reduce and correct these errors in DNA storage, this paper proposes a novel coding schema called DNA-LC, which converts binary sequences into DNA base sequences that satisfy both the GC balance and run-length constraints. Furthermore, our coding mode is able to detect and correct multiple errors with a higher error correction capability than the other methods targeting single error correction within a single strand. The decoding algorithm has been implemented in practice. Simulation results indicate that our proposed coding scheme can offer outstanding error protection to DNA sequences. The source code is freely accessible at https://github.com/XiayangLi2301/DNA.

HB-DSBM: Modeling the Dynamic Complex Networks From Community Level to Node Level.

A variety of methods have been proposed for modeling and mining dynamic complex networks, in which the topological structure varies with time. As the most popular and successful network model, the stochastic block model (SBM) has been extended and applied to community detection, link prediction, anomaly detection, and evolution analysis of dynamic networks. However, all current models based on the SBM for modeling dynamic networks are designed at the community level, assuming that nodes in each community have the same dynamic behavior, which usually results in poor performance on temporal community detection and loses the modeling of node abnormal behavior. To solve the above-mentioned problem, this article proposes a hierarchical Bayesian dynamic SBM (HB-DSBM) for modeling the node-level and community-level dynamic behavior in a dynamic network synchronously. Based on the SBM, we introduce a hierarchical Dirichlet generative mechanism to associate the global community evolution with the microscopic transition behavior of nodes near-perfectly and generate the observed links across the dynamic networks. Meanwhile, an effective variational inference algorithm is developed and we can easy to infer the communities and dynamic behaviors of the nodes. Furthermore, with the two-level evolution behaviors, it can identify nodes or communities with abnormal behavior. Experiments on simulated and real-world networks demonstrate that HB-DSBM has achieved state-of-the-art performance on community detection and evolution. In addition, abnormal evolutionary behavior and events on dynamic networks can be effectively identified by our model.

Role Discovery-Guided Network Embedding Based on Autoencoder and Attention Mechanism.

Recently, network embedding (NE) is an amazing research point in complex networks and devoted to a variety of tasks. Nearly, all the methods and models of NE are based on the local, high-order, or global similarity of the networks, and few studies have focused on the role discovery or structural similarity, which is of great significance in spreading dynamics and network theory. Meanwhile, existing NE models for role discovery suffer from two limitations, that is: 1) they fail to model the varying dependencies between each node and its neighbor nodes and 2) they cannot capture the effective node features which are helpful to role discovery, which makes these methods ineffective when applied to the role discovery task. To solve the above problems of NE for role discovery or structural similarity, we propose a unified deep learning framework, called RDAA, which can effectively represent features of nodes and benefit the Role Discovery-guided NE with a deep autoencoder, while modeling the local links with an Attention mechanism. In addition, we design an elaborately binding technique to combine both parts and optimize the framework in a unified way. We conduct different experiments, including visualization, role classification, role discovery, and running time compared to popular NE methods for both proximity and structural similarity. The RDAA has better performance on all the datasets and achieves good tradeoffs.

Exploring Temporal Community Structure via Network Embedding.

Temporal community detection is helpful to discover and analyze significant groups or clusters hidden in dynamic networks in the real world. A variety of methods, such as modularity optimization, spectral method, and statistical network model, has been developed from diversified perspectives. Recently, network embedding-based technologies have made significant progress, and one can exploit deep learning superiority to network tasks. Although some methods for static networks have shown promising results in boosting community detection by integrating community embedding, they are not suitable for temporal networks and unable to capture their dynamics. Furthermore, the dynamic embedding methods only model network varying without considering community structures. Hence, in this article, we propose a novel unsupervised dynamic community detection model, which is based on network embedding and can effectively discover temporal communities and model dynamic networks. More specifically, we propose the community prior by introducing the Gaussian mixture model (GMM) in the variational autoencoder, which can obtain community information and better model the evolutionary characteristics of community structure and node embedding by utilizing the variant of gated recurrent unit (GRU). Extensive experiments conducted in real-world and artificial networks demonstrate that our proposed model has a better effect on improving the accuracy of dynamic community detection.

Expectation and Complaint: Online Consumer Complaint Behavior in COVID-19 Isolation.

Purpose: This study aims to articulate the nature of consumer complaining behavior (CCB) by analyzing the mechanism and characteristics of online CCB in COVID-19 isolated environment. Patients and Methods: For the purpose, this study collected data via a web-based questionnaire survey from 408 consumers in Shanghai, China during COVID-19 isolation. Through building and analyzing a structural equation model that consists of six latent variables such as perceived service quality, perceived product quality, customer satisfaction, negative emotion, customer complaint; the study analyzed the basic characteristics of CCB, and focused on the moderation test of consumer expectation to validate its important role in consumer decision-making behavior. Results: First, compared to perceived service quality, perceived product quality has a stronger influence on customer satisfaction and has a weaker influence on negative emotions in the COVID-19 isolated environment. Second, the total influence of perceived product quality on customer complaints is stronger than that of perceived service quality. Third, the direct impact of negative emotions on customer complaints was much stronger than the effect of customer satisfaction on customer complaints. Meanwhile, it can also act as a mediating variable to make customer satisfaction have an additional indirect effect on complaints. Finally, the study also found that consumer expectation can reinforce the influences of customer satisfaction on negative emotions and customer complaints, while it weakens the effect of negative emotions on customer complaints. Conclusion: This study suggests that the classic CCB factors still exert a stable influence on customer complaints through cognitive and emotional response pathways, but the influence difference is obvious in the COVID-19 isolated environment. And the influence processes are significantly moderated by consumer expectation level. Enterprises should conduct more targeted marketing interactions, according to these CCB characteristics.

A Secure High-Order Gene Interaction Detection Algorithm Based On Deep Neural Network.

Identifying high-order Single Nucleotide Polymorphism (SNP) interactions of additive genetic model is crucial for detecting complex disease gene-type and predicting pathogenic genes of various disorders. We present a novel framework for high-order gene interactions detection, not directly identifying individual site, but based on Deep Learning (DL) method with Differential Privacy (DP), termed as Deep-DPGI. Firstly, integrate loss functions including cross-entropy and focal loss function to train the model parameters that minimize the value of loss. Secondly, use the layer-wise relevance analysis method to measure relevance difference between neurons weight and outputting results. Deep-DPGI disturbs neuron weight by adaptive noising mechanism, protecting the safety of high-order gene interactions and balancing the privacy and utility. Specifically, more noise is added to gradients of neurons that is less relevance with the outputs, less noise to gradients that more relevance. Finally, Experiments on simulated and real datasets demonstrate that Deep-DPGI not only improve the power of high-order gene interactions detection in with marginal and without marginal effect of complex disease models, but also prevent the disclosure of sensitive information effectively.

Clover: tree structure-based efficient DNA clustering for DNA-based data storage.

Deoxyribonucleic acid (DNA)-based data storage is a promising new storage technology which has the advantage of high storage capacity and long storage time compared with traditional storage media. However, the synthesis and sequencing process of DNA can randomly generate many types of errors, which makes it more difficult to cluster DNA sequences to recover DNA information. Currently, the available DNA clustering algorithms are targeted at DNA sequences in the biological domain, which not only cannot adapt to the characteristics of sequences in DNA storage, but also tend to be unacceptably time-consuming for billions of DNA sequences in DNA storage. In this paper, we propose an efficient DNA clustering method termed Clover for DNA storage with linear computational complexity and low memory. Clover avoids the computation of the Levenshtein distance by using a tree structure for interval-specific retrieval. We argue through theoretical proofs that Clover has standard linear computational complexity, low space complexity, etc. Experiments show that our method can cluster 10 million DNA sequences into 50 000 classes in 10 s and meet an accuracy rate of over 99%. Furthermore, we have successfully completed an unprecedented clustering of 10 billion DNA data on a single home computer and the time consumption still satisfies the linear relationship. Clover is freely available at https://github.com/Guanjinqu/Clover.

Temporal Network Embedding for Link Prediction via VAE Joint Attention Mechanism.

Network representation learning or embedding aims to project the network into a low-dimensional space that can be devoted to different network tasks. Temporal networks are an important type of network whose topological structure changes over time. Compared with methods on static networks, temporal network embedding (TNE) methods are facing three challenges: 1) it cannot describe the temporal dependence across network snapshots; 2) the node embedding in the latent space fails to indicate changes in the network topology; and 3) it cannot avoid a lot of redundant computation via parameter inheritance on a series of snapshots. To overcome these problems, we propose a novel TNE method named temporal network embedding method based on the VAE framework (TVAE), which is based on a variational autoencoder (VAE) to capture the evolution of temporal networks for link prediction. It not only generates low-dimensional embedding vectors for nodes but also preserves the dynamic nonlinear features of temporal networks. Through the combination of a self-attention mechanism and recurrent neural networks, TVAE can update node representations and keep the temporal dependence of vectors over time. We utilize parameter inheritance to keep the new embedding close to the previous one, rather than explicitly using regularization, and thus, it is effective for large-scale networks. We evaluate our model and several baselines on synthetic data sets and real-world networks. The experimental results demonstrate that TVAE has superior performance and lower time cost compared with the baselines.

IoU Regression with H+L-Sampling for Accurate Detection Confidence.

It is a common paradigm in object detection frameworks that the samples in training and testing have consistent distributions for the two main tasks: Classification and bounding box regression. This paradigm is popular in sampling strategy for training an object detector due to its intuition and practicability. For the task of localization quality estimation, there exist two ways of sampling: The same sampling with the main tasks and the uniform sampling by manually augmenting the ground-truth. The first method of sampling is simple but inconsistent for the task of quality estimation. The second method of uniform sampling contains all IoU level distributions but is more complex and difficult for training. In this paper, we propose an H+L-Sampling strategy, selecting the high and low IoU samples simultaneously, to effectively and simply train the branch of quality estimation. This strategy inherits the effectiveness of consistent sampling and reduces the training difficulty of uniform sampling. Finally, we introduce accurate detection confidence, which combines the classification probability and the localization accuracy, as the ranking keyword of NMS. Extensive experiments show the effectiveness of our method in solving the misalignment between classification confidence and localization accuracy and improving the detection performance.

Bioflocculation effect of Glyptotendipes tokunagai on different Microcystis species: Interactions between secreted silk and extracellular polymeric substances.

Cyanobacterial blooms are a major problem in many lakes and can negatively impact public health and ecosystem services. The bioflocculation technique has proven to be a cost-effective, environmentally friendly technique with no secondary pollution to harvest multiple microalgae; however, few studies have focused on its effect on and potential for controlling cyanobacterial blooms in eutrophic lakes. In this study, the bioflocculation efficiencies of different Microcystis species under Glyptotendipes tokunagai (Diptera, Chironomidae) stress conditions and the interactions between secreted silk from Chironomid larvae and extracellular polymeric substances (EPS) from Microcystis were compared. The results indicated that G. tokunagai presented better bioflocculation efficiency on M. wesenbergii than on M. aeruginosa. The formation of "Large Algal Aggregate" flocs was promoted by the derived-soluble extracellular polymeric substances (i.e., proteins and polysaccharides, sEPS) from M. wesenbergii and silk from G. tokunagai. Both M. wesenbergii and midge silk had abundant functional groups, which was beneficial to the formation of the large aggregate. G. tokunagai secreted a large amount of silk to bridge with the sEPS of M. wesenbergii, forming a network structure via interaction between filamentous substance (i.e., complex of sEPS and silk) that plays an important role in the aggregation of Microcystis and the removal of the Microcystis biomass in the water column. The findings provide further insights that will benefit the existing efforts of combating Microcystis blooms in the water column via bioflocculation and will provide a new sustainable approach for inhibiting early bloom formation from the perspective of its provenance in the sediment-water interface.

Feedback regulation of surface scum formation and persistence by self-shading of Microcystis colonies: Numerical simulations and laboratory experiments.

Light availability is an important driver of algal growth and for the formation of surface blooms. The formation of Microcystis surface scum decreases the transparency of the water column and influences the vertical distribution of light intensity. Only few studies analysed the interactions between the dynamics of surface blooms and the light distribution in the water column. Particularly the effect of light attenuation caused by Microcystis colonies (self-shading) on the formation of surface scum has not been explored. In the present study, we simulate the effect of variable cell concentration of Microcystis colonies on the vertical distribution of light in the water column based on experimental estimates of the extinction coefficient of Microcystis colonies. The laboratory observations indicated that higher cell concentration of Microcystis enhance the light attenuation in water column and promotes surface scum formation. We extended an existing model for the light-driven migration of Microcystis by introducing the effect of self-shading and simulated the dynamics of vertical migration for different cell concentrations and different colonial morphologies. The simulation results show that high cell concentrations of Microcystis promote surface scum formation, as well as its persistence throughout diel photoperiods. Large and tight Microcystis colonies facilitate scum formation, while small and loose colonies increase scum stability and persistence. This study reveals a positive feedback regulation of Microcystis surface scum formation and stability by self-shading and provides novel insights into the underlying mechanisms.