Explore the vast range of titles available.

Dust storms have many negative consequences, and affect all kinds of ecosystems, as well as climate and weather conditions. Therefore, classification of dust storm sources into different susceptibility categories can help us mitigate its negative effects. This study aimed to classify the susceptibility of dust sources in the Middle East (ME) by developing two novel deep learning (DL) hybrid models based on the convolutional neural network–gated recurrent unit (CNN-GRU) model, and the dense layer deep learning–random forest (DLDL-RF) model. The Dragonfly algorithm (DA) was used to identify the critical features controlling dust sources. Game theory was used for the interpretability of the DL model’s output. Predictive DL models were constructed by dividing datasets randomly into train (70%) and test (30%) groups, six statistical indicators being then applied to assess the DL hybrid model performance for both datasets (train and test). Among 13 potential features (or variables) controlling dust sources, seven variables were selected as important and six as non-important by DA, respectively. Based on the DLDL-RF hybrid model – a model with higher accuracy in comparison with CNN-GRU–23.1, 22.8, and 22.2% of the study area were classified as being of very low, low and moderate susceptibility, whereas 20.2 and 11.7% of the area were classified as representing high and very high susceptibility classes, respectively. Among seven important features selected by DA, clay content, silt content, and precipitation were identified as the three most important by game theory through permutation values. Overall, DL hybrid models were found to be efficient methods for prediction purposes on large spatial scales with no or incomplete datasets from ground-based measurements.

Invertebrates are central to the functioning of ecosystems, yet they are underappreciated and understudied. Recent work has shown that they are suffering from rapid decline. Here we call for a greater focus on invertebrates and make recommendations for future investigation.

Ecologists and space agencies must forge a global monitoring strategy, say Andrew K. Skidmore, Nathalie Pettorelli and colleagues.

Coral reef ecosystems are under increasing pressure from local and regional stressors and a changing climate. Current management focuses on reducing stressors to allow for natural recovery, but in many areas where coral reefs are damaged, natural recovery can be restricted, delayed or interrupted because of unstable, unconsolidated coral fragments, or rubble. Rubble fields are a natural component of coral reefs, but repeated or high-magnitude disturbances can prevent natural cementation and consolidation processes, so that coral recruits fail to survive. A suite of interventions have been used to target this issue globally, such as using mesh to stabilise rubble, removing the rubble to reveal hard substrate and deploying rocks or other hard substrates over the rubble to facilitate recruit survival. Small, modular structures can be used at multiple scales, with or without attached coral fragments, to create structural complexity and settlement surfaces. However, these can introduce foreign materials to the reef, and a limited understanding of natural recovery processes exists for the potential of this type of active intervention to successfully restore local coral reef structure. This review synthesises available knowledge about the ecological role of coral rubble, natural coral recolonisation and recovery rates and the potential benefits and risks associated with active interventions in this rapidly evolving field. Fundamental knowledge gaps include baseline levels of rubble, the structural complexity of reef habitats in space and time, natural rubble consolidation processes and the risks associated with each intervention method. Any restoration intervention needs to be underpinned by risk assessment, and the decision to repair rubble fields must arise from an understanding of when and where unconsolidated substrate and lack of structure impair natural reef recovery and ecological function. Monitoring is necessary to ascertain the success or failure of the intervention and impacts of potential risks, but there is a strong need to specify desired outcomes, the spatial and temporal context and indicators to be measured. With a focus on the Great Barrier Reef, we synthesise the techniques, successes and failures associated with rubble stabilisation and the use of small structures, review monitoring methods and indicators, and provide recommendations to ensure that we learn from past projects.

Changes in spring and autumn phenology of temperate plants in recent decades have become iconic bio-indicators of rapid climate change. These changes have substantial ecological and economic impacts. However, autumn phenology remains surprisingly little studied. Although the effects of unfavorable environmental conditions (e.g., frost, heat, wetness, and drought) on autumn phenology have been observed for over 60 y, how these factors interact to influence autumn phenological events remain poorly understood. Using remotely sensed phenology data from 2001 to 2012, this study identified and quantified significant effects of a suite of environmental factors on the timing of fall dormancy of deciduous forest communities in New England, United States. Cold, frost, and wet conditions, and high heat-stress tended to induce earlier dormancy of deciduous forests, whereas moderate heat- and drought-stress delayed dormancy. Deciduous forests in two eco-regions showed contrasting, nonlinear responses to variation in these explanatory factors. Based on future climate projection over two periods (2041–2050 and 2090–2099), later dormancy dates were predicted in northern areas. However, in coastal areas earlier dormancy dates were predicted. Our models suggest that besides warming in climate change, changes in frost and moisture conditions as well as extreme weather events (e.g., drought- and heat-stress, and flooding), should also be considered in future predictions of autumn phenology in temperate deciduous forests. This study improves our understanding of how multiple environmental variables interact to affect autumn phenology in temperate deciduous forest ecosystems, and points the way to building more mechanistic and predictive models.

Ecosystems are self-regulating systems that provide societies with food, water, timber, and other resources. As demands for resources increase, management decisions are replacing self-regulating properties. Counter to previous technical approaches that applied simple formulas to estimate sustainable yields of single species, current research recognizes the inherent complexity of ecosystems and the inability to foresee all consequences of interventions across different spatial, temporal, and administrative scales. Ecosystem management is thus more realistically seen as a “wicked problem” that has no clear-cut solution. Approaches for addressing such problems include multisector decision-making, institutions that enable management to span across administrative boundaries, adaptive management, markets that incorporate natural capital, and collaborative processes to engage diverse stakeholders and address inequalities. Ecosystem management must avoid two traps: falsely assuming a tame solution and inaction from overwhelming complexity. An incremental approach can help to avoid these traps.

Habitat loss and fragmentation due to human activities is the leading cause of the loss of biodiversity and ecosystem services. Protected areas are the primary response to this challenge and are the cornerstone of biodiversity conservation efforts. Roughly 15% of land is currently protected although there is momentum to dramatically raise protected area targets towards 50%. But, how much land remains in a natural state? We answer this critical question by using open-access, frequently updated data sets on terrestrial human impacts to create a new categorical map of global human influence (‘Low Impact Areas’) at a 1 km 2 resolution. We found that 56% of the terrestrial surface, minus permanent ice and snow, currently has low human impact. This suggests that increased protected area targets could be met in areas minimally impacted by people, although there is substantial variation across ecoregions and biomes. While habitat loss is well documented, habitat fragmentation and differences in fragmentation rates between biomes has received little attention. Low Impact Areas uniquely enabled us to calculate global fragmentation rates across biomes, and we compared these to an idealized globe with no human-caused fragmentation. The land in Low Impact Areas is heavily fragmented, compromised by reduced patch size and core area, and exposed to edge effects. Tropical dry forests and temperate grasslands are the world’s most impacted biomes. We demonstrate that when habitat fragmentation is considered in addition to habitat loss, the world’s species, ecosystems and associated services are in worse condition than previously reported.

Visual analysis of complex fish habitats is an important step towards sustainable fisheries for human consumption and environmental protection. Deep Learning methods have shown great promise for scene analysis when trained on large-scale datasets. However, current datasets for fish analysis tend to focus on the classification task within constrained, plain environments which do not capture the complexity of underwater fish habitats. To address this limitation, we present DeepFish as a benchmark suite with a large-scale dataset to train and test methods for several computer vision tasks. The dataset consists of approximately 40 thousand images collected underwater from 20 habitats in the marine-environments of tropical Australia. The dataset originally contained only classification labels. Thus, we collected point-level and segmentation labels to have a more comprehensive fish analysis benchmark. These labels enable models to learn to automatically monitor fish count, identify their locations, and estimate their sizes. Our experiments provide an in-depth analysis of the dataset characteristics, and the performance evaluation of several state-of-the-art approaches based on our benchmark. Although models pre-trained on ImageNet have successfully performed on this benchmark, there is still room for improvement. Therefore, this benchmark serves as a testbed to motivate further development in this challenging domain of underwater computer vision.

Assessing the seasonal patterns of the Amazon rainforests has been difficult because of the paucity of ground observations and persistent cloud cover over these forests obscuring optical remote sensing observations. Here, we use data from a new generation of geostationary satellites that carry the Advanced Baseline Imager (ABI) to study the Amazon canopy. ABI is similar to the widely used polar orbiting sensor, the Moderate Resolution Imaging Spectroradiometer (MODIS), but provides observations every 10–15 min. Our analysis of NDVI data collected over the Amazon during 2018–19 shows that ABI provides 21–35 times more cloud-free observations in a month than MODIS. The analyses show statistically significant changes in seasonality over 85% of Amazon forest pixels, an area about three times greater than previously reported using MODIS data. Though additional work is needed in converting the observed changes in seasonality into meaningful changes in canopy dynamics, our results highlight the potential of the new generation geostationary satellites to help us better understand tropical ecosystems, which has been a challenge with only polar orbiting satellites. Cloud cover and scarcity of ground-based validation hinder remote sensing of forest dynamics in the Amazon basin. Here, the authors analyse imagery from a high-frequency geostationary satellite sensor to study monthly NDVI patterns in the Amazon forest, finding support for spatially extensive seasonality.