Phytoplankton dynamics in the Mar Menor, a Mediterranean coastal lagoon strongly impacted by eutrophication.

The Mar Menor hypersaline coastal lagoon has suffered serious degradation in the last three decades attributable to nutrient pollution. In 2015, the lagoon experienced an intensive bloom of cyanobacteria that triggered a drastic change of its ecosystem. Our analyses indicate that phytoplankton in 2016-2021 did not present a seasonal variability pattern; the community was mainly dominated by diatoms and punctually reached abundance peaks above 107 cell L-1 along with chlorophyll a concentrations exceeding 20 µg L-1. The predominant diatom genera during these blooms were different as well as the nutrient conditions under which they were produced. These high diatom abundances are unprecedented in the lagoon; in fact, our data indicate that the taxonomic composition, time variation patterns and cell abundance of phytoplankton in 2016-2021 differ notably in comparison to the data published before 2015. Consequently, our results support the finding that the trophic status of the lagoon has changed profoundly.

ReinforSec: An Automatic Generator of Synthetic Malware Samples and Denial-of-Service Attacks through Reinforcement Learning.

In recent years, cybersecurity has been strengthened through the adoption of processes, mechanisms and rapid sources of indicators of compromise in critical areas. Among the most latent challenges are the detection, classification and eradication of malware and Denial of Service Cyber-Attacks (DoS). The literature has presented different ways to obtain and evaluate malware- and DoS-cyber-attack-related instances, either from a technical point of view or by offering ready-to-use datasets. However, acquiring fresh, up-to-date samples requires an arduous process of exploration, sandbox configuration and mass storage, which may ultimately result in an unbalanced or under-represented set. Synthetic sample generation has shown that the cost associated with setting up controlled environments and time spent on sample evaluation can be reduced. Nevertheless, the process is performed when the observations already belong to a characterized set, totally detached from a real environment. In order to solve the aforementioned, this work proposes a methodology for the generation of synthetic samples of malicious Portable Executable binaries and DoS cyber-attacks. The task is performed via a Reinforcement Learning engine, which learns from a baseline of different malware families and DoS cyber-attack network properties, resulting in new, mutated and highly functional samples. Experimental results demonstrate the high adaptability of the outputs as new input datasets for different Machine Learning algorithms.

Analysis of Real-Time Face-Verification Methods for Surveillance Applications.

In the last decade, face-recognition and -verification methods based on deep learning have increasingly used deeper and more complex architectures to obtain state-of-the-art (SOTA) accuracy. Hence, these architectures are limited to powerful devices that can handle heavy computational resources. Conversely, lightweight and efficient methods have recently been proposed to achieve real-time performance on limited devices and embedded systems. However, real-time face-verification methods struggle with problems usually solved by their heavy counterparts-for example, illumination changes, occlusions, face rotation, and distance to the subject. These challenges are strongly related to surveillance applications that deal with low-resolution face images under unconstrained conditions. Therefore, this paper compares three SOTA real-time face-verification methods for coping with specific problems in surveillance applications. To this end, we created an evaluation subset from two available datasets consisting of 3000 face images presenting face rotation and low-resolution problems. We defined five groups of face rotation with five levels of resolutions that can appear in common surveillance scenarios. With our evaluation subset, we methodically evaluated the face-verification accuracy of MobileFaceNet, EfficientNet-B0, and GhostNet. Furthermore, we also evaluated them with conventional datasets, such as Cross-Pose LFW and QMUL-SurvFace. When examining the experimental results of the three mentioned datasets, we found that EfficientNet-B0 could deal with both surveillance problems, but MobileFaceNet was better at handling extreme face rotation over 80 degrees.

FASSVid: Fast and Accurate Semantic Segmentation for Video Sequences.

Most of the methods for real-time semantic segmentation do not take into account temporal information when working with video sequences. This is counter-intuitive in real-world scenarios where the main application of such methods is, precisely, being able to process frame sequences as quickly and accurately as possible. In this paper, we address this problem by exploiting the temporal information provided by previous frames of the video stream. Our method leverages a previous input frame as well as the previous output of the network to enhance the prediction accuracy of the current input frame. We develop a module that obtains feature maps rich in change information. Additionally, we incorporate the previous output of the network into all the decoder stages as a way of increasing the attention given to relevant features. Finally, to properly train and evaluate our methods, we introduce CityscapesVid, a dataset specifically designed to benchmark semantic video segmentation networks. Our proposed network, entitled FASSVid improves the mIoU accuracy performance over a standard non-sequential baseline model. Moreover, FASSVid obtains state-of-the-art inference speed and competitive mIoU results compared to other state-of-the-art lightweight networks, with significantly lower number of computations. Specifically, we obtain 71% of mIoU in our CityscapesVid dataset, running at 114.9 FPS on a single NVIDIA GTX 1080Ti and 31 FPS on the NVIDIA Jetson Nano embedded board with images of size 1024×2048 and 512×1024, respectively.

Analyzing environmental factors that favor the growth of the invasive brown macroalga Rugulopteryx okamurae (Ochrophyta): The probable role of the nutrient excess.

Time series of temperature, salinity and nutrients in the Strait of Gibraltar (SoG) were researched to analyze which factors explain the invasive success of Rugulopteryx okamurare, which has colonized wide coastal areas at the Spanish and Moroccan coasts since 2016. Temperature and salinity were higher in the SoG compared to its native habitat, implying that the alga is active during the whole seasonal cycle and grows optimally at the high salinities occurring in the SoG. Nitrate removal experiments indicate that the alga is able to linearly increase its N uptake rates following boost in nitrate concentration. Furthermore, R. okamurae N content ranged from 1.4% to 4.5% suggesting that this species has high N storage capacity potentially usable when the external N concentration decreases. These physiological characteristics would explain sharp growth of the alga in the SoG where high N concentrations are registered occasionally.

Efficient Face Recognition System for Operating in Unconstrained Environments.

Facial recognition is fundamental for a wide variety of security systems operating in real-time applications. Recently, several deep neural networks algorithms have been developed to achieve state-of-the-art performance on this task. The present work was conceived due to the need for an efficient and low-cost processing system, so a real-time facial recognition system was proposed using a combination of deep learning algorithms like FaceNet and some traditional classifiers like SVM, KNN, and RF using moderate hardware to operate in an unconstrained environment. Generally, a facial recognition system involves two main tasks: face detection and recognition. The proposed scheme uses the YOLO-Face method for the face detection task which is a high-speed real-time detector based on YOLOv3, while, for the recognition stage, a combination of FaceNet with a supervised learning algorithm, such as the support vector machine (SVM), is proposed for classification. Extensive experiments on unconstrained datasets demonstrate that YOLO-Face provides better performance when the face under an analysis presents partial occlusion and pose variations; besides that, it can detect small faces. The face detector was able to achieve an accuracy of over 89.6% using the Honda/UCSD dataset which runs at 26 FPS with darknet-53 to VGA-resolution images for classification tasks. The experimental results have demonstrated that the FaceNet+SVM model was able to achieve an accuracy of 99.7% using the LFW dataset. On the same dataset, FaceNet+KNN and FaceNet+RF achieve 99.5% and 85.1%, respectively; on the other hand, the FaceNet was able to achieve 99.6%. Finally, the proposed system provides a recognition accuracy of 99.1% and 49 ms runtime when both the face detection and classifications stages operate together.

A Dense Neural Network Approach for Detecting Clone ID Attacks on the RPL Protocol of the IoT.

At present, new data sharing technologies, such as those used in the Internet of Things (IoT) paradigm, are being extensively adopted. For this reason, intelligent security controls have become imperative. According to good practices and security information standards, particularly those regarding security in depth, several defensive layers are required to protect information assets. Within the context of IoT cyber-attacks, it is fundamental to continuously adapt new detection mechanisms for growing IoT threats, specifically for those becoming more sophisticated within mesh networks, such as identity theft and cloning. Therefore, current applications, such as Intrusion Detection Systems (IDS), Intrusion Prevention Systems (IPS), and Security Information and Event Management Systems (SIEM), are becoming inadequate for accurately handling novel security incidents, due to their signature-based detection procedures using the matching and flagging of anomalous patterns. This project focuses on a seldom-investigated identity attack-the Clone ID attack-directed at the Routing Protocol for Low Power and Lossy Networks (RPL), the underlying technology for most IoT devices. Hence, a robust Artificial Intelligence-based protection framework is proposed, in order to tackle major identity impersonation attacks, which classical applications are prone to misidentifying. On this basis, unsupervised pre-training techniques are employed to select key characteristics from RPL network samples. Then, a Dense Neural Network (DNN) is trained to maximize deep feature engineering, with the aim of improving classification results to protect against malicious counterfeiting attempts.

Role of small-sized phytoplankton in triggering an ecosystem disruptive algal bloom in a Mediterranean hypersaline coastal lagoon.

Monthly samplings carried out in 2016-2019 and satellite color images from 2002 to 2019 have been combined to determine the onset and causative species of the ecosystem disruptive algal bloom (EDAB) that affects the Mar Menor coastal lagoon (Western Mediterranean Sea) since 2015. Substantial changes in satellite spectral reflectance attributable to increasing abundance of Synechococcus were registered in 2014. Furthermore, cell abundances of this species in 2016 were the largest ever obtained in the lagoon (6 106 cells mL-1), with values similar to those reported for other Mediterranean hypertrophic estuaries and coastal lagoons. These results suggest that the early changes leading to the EDAB started in 2014 and that Synechococcus played a relevant role in its development. Moreover, diatom and dinoflagellate abundances changed substantially in 2016-2019, ranging from 102 to more than 104 cells mL-1. Some of these changes were linked to flood, suggesting that EDAB has modified substantially the homeostatic capacity of the lagoon.

Improving Real-Time Hand Gesture Recognition with Semantic Segmentation.

Hand gesture recognition (HGR) takes a central role in human-computer interaction, covering a wide range of applications in the automotive sector, consumer electronics, home automation, and others. In recent years, accurate and efficient deep learning models have been proposed for real-time applications. However, the most accurate approaches tend to employ multiple modalities derived from RGB input frames, such as optical flow. This practice limits real-time performance due to intense extra computational cost. In this paper, we avoid the optical flow computation by proposing a real-time hand gesture recognition method based on RGB frames combined with hand segmentation masks. We employ a light-weight semantic segmentation method (FASSD-Net) to boost the accuracy of two efficient HGR methods: Temporal Segment Networks (TSN) and Temporal Shift Modules (TSM). We demonstrate the efficiency of the proposal on our IPN Hand dataset, which includes thirteen different gestures focused on interaction with touchless screens. The experimental results show that our approach significantly overcomes the accuracy of the original TSN and TSM algorithms by keeping real-time performance.

Early life trophodynamic influence on daily growth patterns of the Alboran Sea sardine (Sardina pilchardus) from two distinct nursery habitats (bays of málaga and almería) in the western Mediterranean Sea.

The Alborán Sea sardine (Sardina pilchardus) is the most abundant small pelagic fish in the Mediterranean Sea. Along the northern coast of the Alborán Sea, Málaga and Almería Bay are two important nursery grounds with contrasting oceanographic characteristics. Post-larval stages of S. pilchardus were collected during the late autumn and late winter along with the mesozooplankton community. We examined the influence of feeding behavior on larval growth using environmental variables, daily growth metrics, otolith biometry, stable isotope analysis and trophodynamics. Trophic positions were calculated using mesoozooplankton as potential prey for S. pilchardus. During both seasons, we observed faster growth in larger otoliths that also corresponded to higher trophic positions. Our results indicate that the isotopic signature of larvae is highly influenced by the mesozooplankton community structure. Moreover, the trophic variables showed a significant linear relationship with the estimated ages for all larvae indicating dietary changes throughout ontogenic development.

Shifts in the protist community associated with an anticyclonic gyre in the Alboran Sea (Mediterranean Sea).

The diversity of protists was researched in the Alboran Sea (SW Mediterranean Sea) by means of high-throughput sequencing technologies based on the amplification of the V9 region of 18S rRNA. Samples were collected at different depths in seven stations following an environmental gradient from a coastal upwelling zone to the core of an oligotrophic anticyclonic gyre (AG). Sampling was performed during summer, when the water column was stratified. The superphyla Alveolata, Stramenopila and Rhizaria accounted for 84% of the total operational taxonomic units (OTUs). The most diverse groups were Dinophyceae (21% of OTUs), Marine Alveolates-II (MALV-II; 20%), Ciliophora (9%) and MALV-I (6%). In terms of read abundance, the predominant groups were Dinophyceae (29%), Bacillariophyta (14%), MALV-II (11%) and Ciliophora (11%). Samples were clustered into three groups according to the sampling depth and position. The shallow community in coastal stations presented distinguishable patterns of diatoms and ciliates compared with AG stations. These results indicate that there was a strong horizontal coupling between phytoplankton and ciliate communities. Abundance of Radiolaria and Syndiniales increased with depth. Our analyses demonstrate that the stratification disruption produced by the AG caused shifts in the trophic ecology of the plankton assemblages inducing a transition from bottom-up to top-down control.

Vehicle Counting in Video Sequences: An Incremental Subspace Learning Approach.

The counting of vehicles plays an important role in measuring the behavior patterns of traffic flow in cities, as streets and avenues can get crowded easily. To address this problem, some Intelligent Transport Systems (ITSs) have been implemented in order to count vehicles with already established video surveillance infrastructure. With this in mind, in this paper, we present an on-line learning methodology for counting vehicles in video sequences based on Incremental Principal Component Analysis (Incremental PCA). This incremental learning method allows us to identify the maximum variability (i.e., motion detection) between a previous block of frames and the actual one by using only the first projected eigenvector. Once the projected image is obtained, we apply dynamic thresholding to perform image binarization. Then, a series of post-processing steps are applied to enhance the binary image containing the objects in motion. Finally, we count the number of vehicles by implementing a virtual detection line in each of the road lanes. These lines determine the instants where the vehicles pass completely through them. Results show that our proposed methodology is able to count vehicles with 96.6% accuracy at 26 frames per second on average-dealing with both camera jitter and sudden illumination changes caused by the environment and the camera auto exposure.

Social Sentiment Sensor in Twitter for Predicting Cyber-Attacks Using ℓ1 Regularization.

In recent years, online social media information has been the subject of study in several data science fields due to its impact on users as a communication and expression channel. Data gathered from online platforms such as Twitter has the potential to facilitate research over social phenomena based on sentiment analysis, which usually employs Natural Language Processing and Machine Learning techniques to interpret sentimental tendencies related to users’ opinions and make predictions about real events. Cyber-attacks are not isolated from opinion subjectivity on online social networks. Various security attacks are performed by hacker activists motivated by reactions from polemic social events. In this paper, a methodology for tracking social data that can trigger cyber-attacks is developed. Our main contribution lies in the monthly prediction of tweets with content related to security attacks and the incidents detected based on ℓ 1 regularization.

Cross View Gait Recognition Using Joint-Direct Linear Discriminant Analysis.

This paper proposes a view-invariant gait recognition framework that employs a unique view invariant model that profits from the dimensionality reduction provided by Direct Linear Discriminant Analysis (DLDA). The framework, which employs gait energy images (GEIs), creates a single joint model that accurately classifies GEIs captured at different angles. Moreover, the proposed framework also helps to reduce the under-sampling problem (USP) that usually appears when the number of training samples is much smaller than the dimension of the feature space. Evaluation experiments compare the proposed framework's computational complexity and recognition accuracy against those of other view-invariant methods. Results show improvements in both computational complexity and recognition accuracy.

Interactive Effect of UVR and Phosphorus on the Coastal Phytoplankton Community of the Western Mediterranean Sea: Unravelling Eco-Physiological Mechanisms.

Some of the most important effects of global change on coastal marine systems include increasing nutrient inputs and higher levels of ultraviolet radiation (UVR, 280-400 nm), which could affect primary producers, a key trophic link to the functioning of marine food webs. However, interactive effects of both factors on the phytoplankton community have not been assessed for the Mediterranean Sea. An in situ factorial experiment, with two levels of ultraviolet solar radiation (UVR+PAR vs. PAR) and nutrients (control vs. P-enriched), was performed to evaluate single and UVR×P effects on metabolic, enzymatic, stoichiometric and structural phytoplanktonic variables. While most phytoplankton variables were not affected by UVR, dissolved phosphatase (APAEX) and algal P content increased in the presence of UVR, which was interpreted as an acclimation mechanism of algae to oligotrophic marine waters. Synergistic UVR×P interactive effects were positive on photosynthetic variables (i.e., maximal electron transport rate, ETRmax), but negative on primary production and phytoplankton biomass because the pulse of P unmasked the inhibitory effect of UVR. This unmasking effect might be related to greater photodamage caused by an excess of electron flux after a P pulse (higher ETRmax) without an efficient release of carbon as the mechanism to dissipate the reducing power of photosynthetic electron transport.

Seasonal changes in the distribution and abundance of marine cladocerans of the Northwest Alboran Sea (Western Mediterranean), Spain

The annual cycle of marine cladocerans was studied over six years from 1994 to 1999 within the frame of the monitoring Project ECOMALAGA at a nine stations in the NW of the Alboran Sea, with the aim of assessing seasonal patterns and interannual trends in distribution and abundance of marine cladocerans. Seven species (Penilia avirostris, Evadne nordmanni, Evadne spinifera, Pseudoevadne tergestina, Pleopis polyphemoides, Podon leuckarti and Podon intermedius) were detected in the northwest Alboran Sea. Total cladoceran relative abundance varied from 0 to 89 percent of the total cladocerans. The abundance of cladocerans was higher in summer-autumn than in winter-spring (7012 - 4711.100 and 743 - 217.100m-3 , respectively). The species composition was very different in terms of seasonality. P. polyphemoides, P. leuckarti and P. intermedius appeared mostly during the spring. P. tergestina, E. spinifera and E. nordmanni predominantly occurred during the winter. P. avirostris occurred mostly during the summer and autumn.

Inorganic carbon acquisition in algal communities: are the laboratory data relevant to the natural ecosystems?

Most of the experimental work on the effects of ocean acidification on the photosynthesis of algae has been performed in the laboratory using monospecific cultures. It is frequently assumed that the information obtained from these cultures can be used to predict the acclimation response in the natural environment. CO(2) concentration is known to regulate the expression and functioning of the CCMs in the natural communities; however, ambient CO(2) can become quite variable in the marine ecosystems even in the short- to mid-term. We propose that the degree of saturation of the photosynthesis for a given algal community should be defined in relation to the particular characteristics of its habitat, and not only in relation to its taxonomic composition. The convenience of high CO(2) experiments to infer the degree of photosynthesis saturation by CO(2) in the natural algal communities under the present ocean conditions, as well as its trend in a coming future is discussed taking into account other factors such as the availability of light and nutrients, and seasonality.

EFFECT OF CARBONIC ANHYDRASE INHIBITORS ON THE INORGANIC CARBON UPTAKE BY PHYTOPLANKTON NATURAL ASSEMBLAGES(1).

The role of carbonic anhydrase (CA) in inorganic carbon acquisition (dissolved inorganic carbon, DIC) was examined in Alboran Sea phytoplankton assemblages. The study area was characterized by a relatively high variability in nutrient concentration and in abundance and taxonomic composition of phytoplankton. Therefore, the relationship between environmental variability and capacity for using HCO3 (-) via external CA (eCA) was examined. Acetazolamide (AZ, an inhibitor of eCA) inhibited the primary productivity (PP) in 50% of the samples, with inhibition percentages ranging from 13% to 60%. The AZ effect was more prominent in the samples that exhibited PP >1 mg C · m(-3) · h(-1) , indicating that the contribution of eCA to the DIC photosynthetic flux was irrelevant at low PP. The inhibition of primary productivity by AZ was significantly correlated to the abundance of diatoms. However, there was no a relationship between AZ effect and CO2 partial pressure (pCO2 ) or nutrient concentration, indicating that the variability in the PP percentage supported by eCA was mainly due to differences in taxonomic composition of the phytoplankton assemblages. Ethoxyzolamide (EZ, an inhibitor of both external and internal CA) affected 13 of 14 analyzed samples, with PP inhibition percentages varying from 50% to 95%. The effects of AZ and EZ were partially reversed by doubling DIC concentration. These results imply that CA activity (external and/or internal) was involved in inorganic carbon acquisition in most the samples. However, EZ effect was not correlated with pCO2 or taxonomic composition of the phytoplankton.

Diversity of inorganic carbon acquisition mechanisms by intact microbial mats of Microcoleus chthonoplastes (Cyanobacteriae, Oscillatoriaceae).

The dissolved inorganic carbon (DIC) acquisition mechanisms were researched in intact microbial mats dominated by the cyanobacteria Microcoleus chthonoplastes Thuret, by determining the effect on photosynthesis of different inhibitors. The microbial mats exhibited high affinity for DIC at alkaline pH, with K(m(DIC)) values similar to the ones described for pure cultures of cyanobacteria and algae in which carbon concentrating mechanisms have been researched. Besides, the photosynthesis was non-sensitive to pH changes within the range of 5.6-9.6, indicating that HCO(3)(-) was the main DIC source used for photosynthesis. The M. chthonoplastes mats featured external and internal carbonic anhydrase (CA) activity as measured in intact cells and cell extracts, respectively. Acetazolamide (AZ, which slowly enters the cell and then inhibits mainly the external CA) and ethoxyzolamide (EZ, which inhibits both external and internal CA) reduced significantly the oxygen evolution rates, demonstrating that the CA was implied in the DIC acquisition. Vanadate inhibited photosynthesis by 60% although its application, when CA being inhibited (i.e. after applying AZ + EZ), did not produce any additional effect. It could indicate that ATPase-dependent HCO(3)(-) use occurred and also that this putative mechanism was coupled with CA-like activity at the plasma membrane. The involvement of Na(+)-dependent HCO(3)(-) transporters in DIC acquisition was also inferred as monensin and 4-4'-diisothiocyanatostibilene-2,2'-disulfonate (DIDS) reduced photosynthesis by 70%. DIDS produced a strong inhibitory effect even after application of AZ + EZ + vanadate, indicating that this mechanism was not related to CA activity. The microbial mats become subject to very unfavourable conditions for Rubisco carboxylation at their natural habitats (e.g. external pH of 10.5 and O(2) concentration doubled with respect to saturation concentration); therefore, this putative diversity of DIC acquisition mechanisms could ensure their growth under these extreme conditions.

Evidence for a plasmalemma-based CO2 concentrating mechanism in Laminaria saccharina.

A kinetic analysis of the photosynthesis inhibition by buffers allowed quantification of some components of the carbon concentrating mechanism (CCM) of the brown macroalga Laminaria saccharina. The CCM was based on the presence of acid regions outside the plasma membrane that increased the CO(2) concentration available for photosynthesis by 10-20 times above that of the bulk medium at alkaline pH. Furthermore, the results suggested that the CCM is located mainly on the cell membrane and not in the chloroplast, as suggested for most macroalgae. The degree of dissipation of the acid regions by a buffer was related to the buffer anion concentration (B(-)), estimated from the titration of the buffer from bulk medium pH to a pH endpoint value close to the first pK (a) of the carbonic acid system. A kinetic model describing the relationship between inhibition of photosynthesis by a buffer and B(-) was developed suggesting that buffers act as competitive inhibitors with IC(50) (the concentration of the buffer anion which reduces the reaction velocity by half) of 5.0 mol m(-3). This model can be used to estimate the inhibitory effect of any buffer on the photosynthesis of L. saccharina. Nevertheless, some buffers tested showed a lower effect than that predicted from the hyperbolic model suggesting that their strength as inhibitors depended on: (1) the pK (a) in relation to the first pK (a) of the carbonic acid system and (2) its molecular weight (i.e. its mobility).