The Effect of Visitors on Zoo Reptile Behaviour during the COVID-19 Pandemic

Visitors to zoos can have positive, neutral, or negative relationships with zoo animals. This makes human–animal interactions (HAIs) an essential component of welfare and an important consideration in species selection for zoo exhibits and in enclosure designs. We measured the effect of visitors on reptiles by comparing open and closed periods during the lockdowns in response to the COVID-19 pandemic in the UK in a low-resolution dataset for thirteen species of reptiles and a high-resolution dataset focussing on just one of these. Scan sampling on thirteen reptile species (two chelonians and eleven squamates) showed species-specific differences in response to the presence/absence of visitors, with most taxa being only weakly affected. High-resolution scan sampling via video footage of an off-show and on-show enclosure was carried out for tokay geckos (Gekko gecko) over the open and closed periods. In this part of the study, tokay geckos were significantly more visible during zoo closure than when visitors were present on-exhibit, but there was no change in off-show animals, indicating the effect of visitors as opposed to other factors, such as seasonality, which applied equally to both on- and off-show animals. The high-resolution study showed that a significant effect was present for tokay geckos, even though the low-resolution suggested that they were more weakly affected than other taxa. Our results indicate that, for cryptic species such as this, more intensive sampling may be required to properly understand visitor effects. Our data do not allow the interpretation of effects on welfare but show that such assessments require a species-specific approach.

Detection of Bark Beetle Disturbance at Tree Level Using UAS Multispectral Imagery and Deep Learning

This study aimed to examine the potential of convolutional neural networks (CNNs) for the detection of individual trees infested by bark beetles in a multispectral high-resolution dataset acquired by an unmanned aerial system (UAS). We compared the performance of three CNN architectures and the random forest (RF) model to classify the trees into four categories: pines, sbbd (longer infested trees when needles turn yellow), sbbg (trees under green attack) and non-infested trees (sh). The best performance was achieved by the Nez4c3b CNN (kappa 0.80) and Safaugu4c3b CNN (kappa 0.76) using only RGB bands. The main misclassifications were between sbbd and sbbg because of the similar spectral responses. Merging sbbd and sbbg into a more general class of infested trees made the selection of model type less important. All tested model types, including RF, were able to detect infested trees with an F-score of the class over 0.90. Nevertheless, the best overall metrics were achieved again by the Safaugu3c3b model (kappa 0.92) and Nez3cb model (kappa 0.87) using only RGB bands. The performance of both models is comparable, but the Nez model has a higher learning rate for this task. Based on our findings, we conclude that the Nez and Safaugu CNN models are superior to the RF models and transfer learning models for the identification of infested trees and for distinguishing between different infestation stages. Therefore, these models can be used not only for basic identification of infested trees but also for monitoring the development of bark beetle disturbance.

A Generalizable Sample Resolution Augmentation Method for Mechanical Fault Diagnosis Based on ESPCN

Data augmentation has become a hot topic in the field of mechanical intelligent fault diagnosis. It can expand the limited training dataset by generating simulated samples, but there is still no effective method augmenting the resolution of low resolution sample. In this paper, a simple algorithm, namely, efficient subpixel convolutional neural network (ESPCN), is proposed to solve this deficiency. The ESPCN model performs the arrange operation on the raw low resolution data through the subpixel layer and outputs the result of four-channel multifeature maps. Then, the sample resolution is increased to four times compared with the raw low resolution sample. Finally, the generated high resolution dataset is employed to train the stacked autoencoders (SAE) for fault classification, and the raw high resolution dataset is used for testing. Two fault diagnosis cases with different sample dimensions and rotating speeds are set up to simulate the low resolution situation, and the experimental results verify the feasibility of the proposed algorithm.

A Deep Learning Approach to Assist Sustainability of Demersal Trawling Operations

Bycatch in demersal trawl fisheries challenges their sustainability despite the implementation of the various gear technical regulations. A step towards extended control over the catch process can be established through a real-time catch monitoring tool that will allow fishers to react to unwanted catch compositions. In this study, for the first time in the commercial demersal trawl fishery sector, we introduce an automated catch description that leverages state-of-the-art region based convolutional neural network (Mask R-CNN) architecture and builds upon an in-trawl novel image acquisition system. The system is optimized for applications in Nephrops fishery and enables the classification and count of catch items during fishing operation. The detector robustness was improved with augmentation techniques applied during training on a custom high-resolution dataset obtained during extensive demersal trawling. The resulting algorithms were tested on video footage representing both the normal towing process and haul-back conditions. The algorithm obtained an F-score of 0.79. The resulting automated catch description was compared with the manual catch count showing low absolute error during towing. Current practices in demersal trawl fisheries are carried out without any indications of catch composition nor whether the catch enters the fishing gear. Hence, the proposed solution provides a substantial technical contribution to making this type of fishery more targeted, paving the way to further optimization of fishing activities aiming at increasing target catch while reducing unwanted bycatch.

Characteristic profile

The rapid uptake of rooftop solar photovoltaic systems is introducing many challenges in the management of distribution networks, energy markets, and energy storage systems. Many of these problems can be alleviated with accurate short term solar power forecasts. However, forecasting the power output of distributed rooftop solar PV systems can be challenging, since many complex local factors can affect solar output. A common approach when forecasting such systems is to extract the daily seasonality from the time series using some form of seasonality model, and then forecast only the residuals that remain after seasonality extraction. In this work, we explore in detail the effectiveness of three commonly used seasonality models, and we propose a new one, called the "characteristic profile". We find that when seasonality models are integrated into the forecasting process, significant gains in forecast accuracy may be obtained - particularly for machine learning based approaches, which have a reduction in forecast error of 5-25%. Among the seasonality models, the characteristic profile demonstrates the highest forecast accuracy, resulting in reductions in forecast error of 8% or more compared to forecasting models that do not take seasonality into account. The benefits of this approach are particularly pronounced when forecasting solar PV systems that are curtailed, suffer from local shading, or consist of multiple sets of panels having different orientations and tilts. Our results are demonstrated on a high resolution dataset obtained from 258 sites in Western Australia over the course of a full year.

Seagrass habitat suitability model for Redang Marine Park using multibeam echosounder data: Testing different spatial resolutions and analysis window sizes

Integrating Multibeam Echosounder (MBES) data (bathymetry and backscatter) and underwater video technology allows scientists to study marine habitats. However, use of such data in modeling suitable seagrass habitats in Malaysian coastal waters is still limited. This study tested multiple spatial resolutions (1 and 50 m) and analysis window sizes (3 × 3, 9 × 9, and 21 × 21 cells) probably suitable for seagrass-habitat relationships in Redang Marine Park, Terengganu, Malaysia. A maximum entropy algorithm was applied, using 12 bathymetric and backscatter predictors to develop a total of 6 seagrass habitat suitability models. The results indicated that both fine and coarse spatial resolution datasets could produce models with high accuracy (>90%). However, the models derived from the coarser resolution dataset displayed inconsistent habitat suitability maps for different analysis window sizes. In contrast, habitat models derived from the fine resolution dataset exhibited similar habitat distribution patterns for three different analysis window sizes. Bathymetry was found to be the most influential predictor in all the models. The backscatter predictors, such as angular range analysis inversion parameters (characterization and grain size), gray-level co-occurrence texture predictors, and backscatter intensity levels, were more important for coarse resolution models. Areas of highest habitat suitability for seagrass were predicted to be in shallower (<20 m) waters and scattered between fringing reefs (east to south). Some fragmented, highly suitable habitats were also identified in the shallower (<20 m) areas in the northwest of the prediction models and scattered between fringing reefs. This study highlighted the importance of investigating the suitable spatial resolution and analysis window size of predictors from MBES for modeling suitable seagrass habitats. The findings provide important insight on the use of remote acoustic sonar data to study and map seagrass distribution in Malaysia coastal water.

Improving a Rapid Alignment Method of Tomography Projections by a Parallel Approach

The high resolution of synchrotron cryo-nano tomography can be easily undermined by setup instabilities and sample stage deficiencies such as runout or backlash. At the cost of limiting the sample visibility, especially in the case of bio-specimens, high contrast nano-beads are often added to the solution to provide a set of landmarks for a manual alignment. However, the spatial distribution of these reference points within the sample is difficult to control, resulting in many datasets without a sufficient amount of such critical features for tracking. Fast automatic methods based on tomography consistency are thus desirable, especially for biological samples, where regular, high contrast features can be scarce. Current off-the-shelf implementations of such classes of algorithms are slow if used on a real-world high-resolution dataset. In this paper, we present a fast implementation of a consistency-based alignment algorithm especially tailored to a multi-GPU system. Our implementation is released as open-source.