Search

Due to the rise in tornadoes in the United States, the built environment continues to suffer damage. Masonry buildings comprise a significant portion of the built heritage environment for most countries and are susceptible to natural hazards. However, research at the intersection of extreme wind loading and structural strengthening often neglects historic and aging infrastructure, particularly unreinforced masonry buildings, leading to their exclusion from the building code. This study aims to identify various factors that affect damages to unreinforced masonry structures during tornado loading. To begin, the structural performance of historic masonry constructions after the Mayfield, KY tornado in 2021 was evaluated, with multimodal data collected on-site, including aerial photographs, street-view imagery, and detailed field notes on observed damages. The on-site data included aerial photographs, street-view imagery, and detailed field notes for the observed damages. This data was supplemented by archival research and analyzed to identify the underlying relationships between building attributes and damage levels. The results indicate that damages in historic buildings were extensive and proportional to the year of construction and the distance from the tornado but not necessarily to the retrofit status. [ABSTRACT FROM AUTHOR]

The Taravu valley, located in the south-west of Corsica, is one of the most extensive watersheds on the island. Recent paleoenvironmental studies allowed to restitute the chronostratigraphy of the two ponds in the lower valley. However, the paleogeographic evolution of this sector still remains largely to be specified. In order to reconstruct the transformations of the landscape of the lower Taravu valley during the Late Holocene (2250 BCE), we have developed an innovative interdisciplinary approach combining the methods of historical geography (ancient maps and aerial photographs), geoprocessing of topographic data (slope map and drainage map) that we associated with the results of an electromagnetic survey (EMI) and previous stratigraphic data. This synthesis enables the creation of a geomorphological map of the lower Taravu valley, to identify several major landscape changes since the Neolithic period (5800-3000 BCE), and to follow the evolution of the alluvial forms of the Taravu and the shore line since the end of the 18th c. CE. [ABSTRACT FROM AUTHOR]

• Monospecific woodland groves were associated with soil forms and flooding regimes in a floodplain. • Extreme floods removed trees and created substrates for woodland regeneration. • Damage by elephants and meso-browsers affected regeneration dynamics. • Flooding and herbivory interactively generated a shifting woodland mosaic anchored by soil. Our study addressed the regeneration dynamics of the patch mosaic of monospecific woodland groves in the floodplains of the Luvuvhu and Limpopo rivers following the removal of human settlements to create the Makuleke Contractual Park falling within the northern Kruger National Park. We anticipated that subsequent increases by elephants and meso-browsers like impalas would deflect or even prevent the re-establishment of these groves through the damage that they inflicted on plants. We compared the responses exhibited in groves of fever trees (Vachellia xanthophloea) associated with layered clay alluvium with those of ana trees (Faidherbia albida) associated with deep sandy deposits. Multi-decadal changes in the woodland mosaic were identified using aerial photographs and Google Earth imagery supported by onsite inspections and related to flood events and soil features. Woodland regeneration following recent major floods was tracked using onsite inspections over five years recording sapling growth and disappearance along with the breakage, bark removal and defoliation imposed by elephants, medium-sized browsing ruminants and porcupines. Fever trees were concentrated near the confluence of the two rivers where floodwaters inundated this region. Extreme floods carried away flanking ana trees while leaving sandy deposits open for recolonization. Elephants pushed over tall fever trees but not the ana trees, while regenerating saplings of both tree species were broken or uprooted. Browsing impalas defoliated juvenile trees contributing to their demise. These impacts contributed to the contraction or even disappearance of some fever tree groves and suppressed the growth and establishment of ana tree saplings. While these herbivores contributed to the shifting patch mosaic of these monospecific woodlands, the local distribution of these woodlands was associated with the soil features that they favoured. [ABSTRACT FROM AUTHOR]

Light absorbing particles (LAPs) present high absorbance and contribute to reducing the snow albedo when deposited on snow surfaces. This deposition can be caused by aerosols transported from natural or anthropogenic, either distant or nearby sources. In this study, snow was artificially contaminated with soil samples collected in the Central Andes (near El Yeso dam) to simulate the most common nearby source of Mineral Dust (MD) deposition onto snow surface. Andean soil samples previously conditioned were characterized through Single Particle Optical Sizing (SPOS), X-ray diffraction (XRD) analysis, and Scanning Electron Microscope (SEM) for the determination of optical properties. Spectral snow albedo was measured in situ with a spectroradiometric system. To evaluate the heterogeneity of the particle distribution over the snow surface, aerial photographs were taken with a drone to apply a visual color segmentation of the surface and to determine the equivalent MD concentration. Experimental snow albedo was compared with theoretical values obtained with the OptiPar radiative transfer model. Inputs for the model were: the MD refractive index (calculated from the mineralogical composition and morphology of MD) and particle size, cloudiness, snow density, surface roughness, snow grain size, and LAPs concentration (obtained from the snow samples collected during the experiments and analyzed in the laboratory). Small black carbon concentrations were found in natural snow and considered in the simulations. Spectral albedo measurements showed high albedo reductions in the UV and VIS range (300–800 nm), being less significant in the NIR range (800–1700 nm). A nonlinear behavior was observed in broadband albedo when increasing MD concentration. For lower values of MD concentration (lower than 1500 mg ⋅ kg−1), a significant albedo reduction rate of 0.1 units per 1000 mg ⋅ kg−1 was found, while at higher concentrations (¿ 3500 mg ⋅ kg−1), such reduction tends to the minimum. Simulated values with OptiPar are in agreement with measured albedo, but some differences are observed, probably due to the refractive index considered, the snow surface roughness, and the non-uniform MD concentration in snow. • Andean MD particles decrease snow spectral albedo mainly in UV/VIS wavelength range. • XRD and SEM classify Andean soil, matching local geology. • OptiPar radiative transfer model albedo simulations agree with experimental data. • A nonlinear behavior was observed broadband albedo decrease. [ABSTRACT FROM AUTHOR]

Accurate digital elevation models represent the basic tool for a large spectrum of applications, including geological, architectonic, archaeological, and urbanistic studies. However, aggressive urbanization may significantly alter the morphology of areas of interest. Such is the case of the active Campi Flegrei caldera in southern Italy, where all the buildings, facilities, infrastructures, quarries, and landfills altered the original volcanic morphology. To avoid these limitations, we analyzed a set of vintage aerial photographs acquired in 1943 by Italy's Military Geographic Institute (IGM). We reconstructed the study area's topography before the deep anthropic modification from the 1960s onwards by applying Structure-from-Motion photogrammetric processing. On the reconstructed orthomosaic and a historical-Digital Surface Model (hDMS), we conducted geomorphic analyses along 18 longitudinal stream profiles outside, across the border, and inside the caldera, underlining that the rectilinear sections with preferred orientations and increased incision values suggest a strong structural underpinning on valley incision and stream paths. The analysis of the spatial distribution of sinuosity, SL index, Chi (χ) value and knickpoint maps suggests the presence of tectonic lineaments that influence the stream network from the pre-caldera (>40 ka) to recent times, producing articulated caldera and craters geometry as well as affecting the localization of volcanic vents. The comparison between the 1943 hDSM and 2013 LiDAR (Light Detection and Ranging) DSM allowed us to spotlight the areas that show the most significant differences due to anthropic intervention that obliterated critical features, thus supporting our motivation to use this base. Overall, the results indicate that long-lived volcano-tectonic and tectonic structures control the orientation of drainage patterns and their re-arrangements during volcano-tectonic deformation phases. Finally, a comprehensive structural map, based on the merged 1943 hDSM and the bathymetric DEM, is presented together with a conceptual evolutionary model of the stream network across the caldera border. Under the proper acquisition conditions, vintage aerial photographs can provide a useful tool for morphological analysis and can be applied to several topics in the geosciences. • The original topography of Campi Flegrei is restored using historical aerial photographs. • Several geomorphic indices highlight the activity of volcano-tectonic structures. • Rectilinear and deeply incised valleys occur in different sectors. • NE–SW and NW–SE oriented tectonic features are overprinted by E–W oriented features. • Tectonic lineaments control the occurrence of volcanism through time. [ABSTRACT FROM AUTHOR]

• A novel rotating object detector is proposed. • The detector relies on the anchor refinement and two-stage sample selective strategy. • Performance of the detector is competitive against most state-of-the-art methods. • The new drone-based dataset promotes the research in rotating object detection. Rotating object detection has recently attracted increasing attention in aerial photographs, remote sensing images, etc. In this paper, we propose a rotating object detector TS4Net, which contains anchor refinement module (ARM) and two-stage sample selective strategy (TS4). The ARM can convert the preset horizontal anchors into high-quality rotated anchors through two-stage refinement, which also adopts a rotated Intersection-over-Union(IoU) prediction branch to improve localization accuracy in the second stage. TS4 module utilizes different constrained sample selective strategies to allocate positive and negative samples, which is adaptive to the regression task in different stages. Benefiting from the ARM and TS4, the TS4Net can achieve superior performance for rotating object detection solely with one preset horizontal anchor. Considering that most rotating object detection datasets mainly focus on the field of remote sensing and are shot in high-altitude scenes. We present a low-altitude drone-based dataset, named UAV-ROD, aiming to promote the research and development in rotating object detection and UAV applications. The UAV-ROD dataset can be used in rotating object detection, vehicle orientation recognition, and object counting tasks. Extensive experiments on the UAV-ROD dataset and four datasets demonstrate that our method achieves competitive performance against most state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The Chaochou Fault in southern Taiwan has long been recognized as an active fault, but its exact location is still uncertain. In this study, we focused on the middle part of the Chaochou Fault, where a flight of fluvial terraces was deformed and preserved. High-resolution aerial photos were first used to observe and map these terraces and their vicinity, followed by comprehensive field investigations including RTK-GPS surveys for high-resolution topographic profiles of the deformed terraces and OSL sample collections for obtaining the ages of the terraces. A series of active faults, consisting of Fault A ("FA"), Fault B ("FB"), and Fault Bb ("FBb"), are collectively named the Chaochou Active Fault Zone. The fault zone is recognized and characterized by range-facing scarps, frontal flexural scarps, and discontinuous slopes deformed by FA, FB, and FBb, respectively. The repeated activities of FA were recorded by the incremental fault scarp heights from young to old terraces. Considering the highest fault scarp height of 65.9 ± 6 m observed in T1 and its OSL age of 35.3 ± 4.3 ka, the long-term fault slip rate of FA is about 2 mm/yr. • Detailed air photo mapping and fieldwork reveal the Chaochou Active Fault Zone. • A subsidiary fault of the Chaochou Fault is first identified in this study. • High-resolution topo profiles confirm repeated activities of the back thrust. • The Chaochou Fault is undoubtedly active and maybe even moved in the Holocene. • The long-term fault slip rate of FA ranges from 0.5 to 2 mm/yr. [ABSTRACT FROM AUTHOR]

Derailments significantly harm railroads in terms of severity and fatality rates. Manually monitoring railway tracks is a tedious and often insufficient task to prevent derailment accidents. This necessitates the development of an automated monitoring system to oversee the condition of railway tracks. In the foreseeable future, Unmanned Aerial Vehicles (UAVs), artificial intelligence, and computer vision will play a crucial role in periodically assessing the railway environment to ensure passenger safety. This paper proposes a lightweight and efficient railroad semantic segmentation network (Lite-RSNet), which segments real-time aerial railroad images into rails, gauges, and backgrounds. The segmented regions then compute the distance between rails and obstacles, identifying potential derailment hazards in railroad applications. Lite-RSNet incorporates a multi-level feature fusion framework that includes a residual layer, a multi-head attention layer, a Concurrent Spatial and Channel Squeeze Excitation (CSSE) block, and a Convolutional Block Attention Module (CBAM). The designed approach enhances onboard processing efficiency by streamlining the model to use fewer parameters and reduce storage demands. Performance assessment of the model utilized the Rail Segmentation Dataset (RSD), which consists of high-resolution, low-altitude aerial photographs of Indian railroads. The Lite-RSNet model outperforms current leading methods, maintaining a lean structure with only 6 million parameters and achieving notable metrics of 0.971 for the Dice Coefficient (DC) and 0.947 for the Jaccard Index (JI) on the test dataset. Furthermore, distance estimation between an obstacle and rail has been implemented using image processing techniques as an additional work of this proposal to know about the caution of an obstacle. [ABSTRACT FROM AUTHOR]

• The socio-cultural response to the morphological transformations of water regimes in peri -urban areas is investigated. • Land cover changes at the catchment level of a large river in Hong Kong from 1994 to 2022 are analyzed. • Mixed-methods research combining remote sensing (analysis of aerial and satellite photographs) and in-depth semi-structured interviews. • Integrating environmental mitigation strategies in channelization can moderate the negative impacts of engineering-driven fluvial morphological transformations, retaining place attachment and land care among local inhabitants. Rural-to-urban transformation in Southeast Asia must be discussed as a vicious cycle of hydrologic alteration, environmental degradation, and destructive land use changes that collectively jeopardize millions of lives. Particularly, top-down flood control schemes have been linked to significant degradation of the socioeconomic and biophysical features of rural landscapes and increasingly faster agricultural decline and environmental degradation. In recent years, growing awareness has focused on ecosystem fragility and the need for a water-sensitive approach in planning. Water management and stream renaturation are expanding and rapidly developing disciplines that have produced radical advancements in ecological, geomorphological, and hydrologic areas. Despite such progress, however, these disciplines have failed to offer major advancements in socially related fields. A paucity of scientific research exists on the association between hydrologic alteration and the transformation of the landscape—which is a socially constructed phenomenon. As a result, the socio-cultural response to the morphological transformations of water regimes in peri -urban and desakota areas remains under-investigated. The current study questions whether a specific morphological transformation of water streams, namely channelization cum environmental mitigation, can catalyze socio-cultural reactions and thus shape peculiar land use patterns. To examine that impact, we conducted a study on the Sheung Yue River in the New Territories of Hong Kong. Longitudinal data on the morphological change of the landscape at the catchment level from 1994 to 2022 were interpreted from aerial photographs and high-resolution satellite imagery, controlling for external factors. Alongside that data, 20 semi-structured interviews with local farmers provided in-depth knowledge of societal responses to the landscape mutation. Eleven interviews with knowledgeable informants also enhanced the understanding of river management and agricultural conservation. The overall findings revealed that integrating environmental mitigation strategies in channelization can moderate the negative impacts of the engineering-driven fluvial morphological transformation. Based on these findings, the study stresses the significance of socio-environmental responsive river management and integrated urban planning strategies for policymaking in Southeast Asia. [ABSTRACT FROM AUTHOR]

This study investigated how structural changes of a historical traditionally used forest landscape in Central Europe have affected traits of vegetation and wild bee communities. We compared the extent of gaps in the forest using aerial photographs between 1945 and 2020. And we found historic vegetation and bee surveys from 1975 and 1990, which we repeated in 2010 and 2020. We characterised the vegetation of the closed forest, the forest gaps and the small-scale meadows adjacent to the forest as well as the wild bee community by traits and investigated trait changes with Kruskal-Wallis tests. By NMDS we characterised sample plots and transect walks of timepoints with traits and species. Area of forest gaps decreased by 88% from 1945 to 2020 and by 74% from 1975 to 2020. In the traits of vegetation, Ellenberg Indicator Values (EIV) for moisture and soil nutrients significantly increased in closed forest, forest gaps and meadows adjacent to forest. The EIVs for light and temperature, and the number of red list species decreased. The number of wild bee species that specialise in visiting flowers with long tubes and hylophilic species declined. The number of eremophilic species, species with longer phenologies, and recently Red list species increased. In the NMDS figure of forest gaps and of adjacent meadows, different species and traits characterised more recent and older sample plots of vegetation and confirm the results of the Kruskal Wallis tests. Wild bee communities of 2010 and 2020 were characterised by frequent species, while those of 1975 and 1990 were not characterised by any species. The traits characterised the wild bee communities of the older and younger transect walks in accordance with the results of the Kruskal Wallis tests. In addition, oligolectic bees characterised the older transect walks. The loss of forest gaps represents a direct loss of habitat for many plant and wild bee species. In addition, the qualitative changes in vegetation indicate a deterioration in habitat quality for plants of open forests and wild bees. Among wild bees, specialists were negatively affected by the changes described. These developments were reflected in the initial downward trend in Red List bee species. The increase in bee species with longer phenology, in eremophilic species, the decrease in hylophilic species, and the recent increase in red-listed species can indicate climate change. Therefore, this study underlines the importance of the preservation and development of ecologically valuable forest gaps, particularly in historical traditionally used forests with a typical species composition. [Display omitted] • loss of 88% in forest gaps from 1945 to 2020. • increase in EIV for nitrogen and moisture of vegetation from 1975 to 2020. • decrease in EIV for light, for temperature and Red List species of vegetation. • decline in specialised and hylophilic wild bee species from 1975 to 2020. • increase in eremophilic wild bee species and species with a longer phenology. [ABSTRACT FROM AUTHOR]

The association between villages in the city (ViCs) and crime is an important topic but has not been rigorously investigated in crime geography. This study developed three new variables in explaining burglary at the community level: the existence of ViCs, the area of ViCs and the ViC-based segregation index. We extracted all ViCs in the study area from satellite images using machine learning. The relationships between these variables and burglary are tested using two negative binomial models. The existence of ViCs in a community is significantly associated with burglary. Results showed that the ViC communities have more burglary than non-ViC communities. Further, the size of ViCs in a community is the most powerful variable in explaining burglary in ViC communities. Communities with larger ViCs have more burglary. This ViC area variable is critical, given that the number of households, a traditional measure of burglary target, is not effective in a city with ViCs. The co-variates pertaining to disadvantaged population, mixture of locals and non-locals, informal controls, crime generators and accessibility shed light on why ViCs contribute to crime. Finally, ViC-based segregation is not significantly related to burglary. This finding is an important addition to the literature on segregation and crime, a challenging research problem that requires future attention. The high concentration of crime in ViCs calls for targeted intervention strategies by law enforcement agencies. • Extracted villages in the city (ViCs) from air photos using machine learning • Developed a theoretical framework and a ViC-based segregation index • Revealed that ViC communities have more burglaries than non-ViC communities • Discovered that the area of ViCs best explains burglaries in a ViC community • Found that segregation is not associated with burglaries. [ABSTRACT FROM AUTHOR]

Landslides, the most significant geohazards in Iran, adversely affect the region's socioeconomic conditions and the environment. Landslide susceptibility mapping is crucial for proactive risk management, sustainable development, and protecting human lives, infrastructure, and the environment. It enables decision-makers to make informed choices, implement appropriate mitigation measures, and plan for potential landslide events, leading to safer and more resilient communities. Using a recommended subdivision methodology, this research has developed a landslide susceptibility map for a significant Gilan, Iran region. The production of the map involved utilizing an artificial neural network (ANN) model. This model incorporated sixteen causal components from several characteristics, including topographic and geomorphologic features, geological factors, land use patterns, hydrological aspects, and hydrogeological properties. Three hundred seventy instances were identified using multiple verified sources and analyzing aerial photographs to create the landslide inventory map. Examining and verifying the weights assigned to the causal elements were conducted per accepted mathematical standards, incorporating sensitivity analysis, earlier research, and empirical data on landslides. The area under the receiver operating characteristic curve (AUROC) was used to compare the various models. The study's findings indicated that the best swarm size value for COA-MLP 1 1 Cuckoo Optimization Algorithm (COA). is equal to 450, and the accuracy indices for this algorithm were 0.998 and 0.995 in training and testing datasets, respectively. Similarly, the AUC 2 2 Area under the curve (AUC). for the HS-MLP, 3 3 Harmony Search Algorithm (HS). SFS-MLP, 4 4 Stochastic Fractal Search (SFS). and TLBO-MLP 5 5 Teaching- Learning-Based Optiization (TLBO). were 0.997, 0.999, and 0.999 in training and 0.995, 0.996, and 0.995 in the testing dataset, respectively. Therefore, the use of optimization algorithms leads to an increase in the performance and accuracy of the neural network, and the high accuracy of the SFS-MLP model demonstrates the existence of a dependable criterion for delineating the susceptibility zones about forthcoming landslide occurrences. The model is a cost-effective and potentially indispensable tool for urban planners in growing cities and municipalities. Its effectiveness is demonstrated by comparisons with previous susceptibility analyses conducted in the region. [ABSTRACT FROM AUTHOR]

Source–sink systems in arid and semiarid regions contain low plant cover areas (inter-patches) functioning as sources of runoff, sediments, and nutrients that are retained by vegetated patches, which act as sinks. The spatial configuration of these hydrological units largely determines the connectivity network of runoff sources. At the same time, their dynamics are essential for assessing ecohydrological functionality, which mining can modify through the alteration of the system's ability to retain water and sediments, causing increased runoff and soil erosion. The objectives of this work were to: (1) identify the spatio-temporal changes in ground cover (i.e., vegetated patches and inter-patches) in two areas with mining activity: Cerro de San Pedro (1970–2015) and Villa de la Paz (1983–2014); and (2) evaluate, from an ecohydrological perspective, the changes in hydrological and sediment connectivity as the result of changes in ground cover in the two periods within each study area. Ground cover maps were obtained from aerial photographs and high-resolution WorldView 2 satellite images calibrated with 2 × 2 m plots measured in the field. For each study area, a transition matrix was used to assess spatio-temporal changes of patches and inter-patches, while hydrological and sediment connectivity were evaluated with the index of connectivity (IC). The results showed that the area with vegetated patches has decreased in Cerro de San Pedro and Villa de la Paz, leading to increased hydrological and sediment connectivity related to mining infrastructure and urbanization. The results can be used to prioritize areas for restoring ecohydrological functioning by reducing the connectivity of runoff sources. [Display omitted] • Mining may modify vegetation cover and alter hydrological and sediment connectivity. • Vegetation, topography, and mine infrastructure were mapped to compute connectivity. • The index of connectivity weighting factor was derived from rainfall simulation. • Hydrological and sediment connectivity increased with changes in vegetation cover. • Vegetation function and connectivity can support the rehabilitation of mining areas. [ABSTRACT FROM AUTHOR]

Landslides are hillslope processes controlled by natural changing topographic conditions. Landslides are also influenced by human activities. Yet, understanding the space-time occurrence of landslides and their interactions with these typically long-term natural and short-term human-induced controls remains a key challenge in many regions, especially in tropical environments where data scarcity is commonplace. Here we decipher these dynamic processes in the Ruzizi Gorge located in the Kivu Rift (Central Africa), that is an exceptional geomorphological landmark whose origin is associated with the rerouting of >7000 km2 of drainage area from Lake Kivu during the Holocene. This bedrock river has also seen its landscape disturbed over the past decades by the development of the city of Bukavu (DR Congo). In this study, we combine detailed field observations, historical aerial photographs, archive analysis and satellite imagery to compile a multi-temporal inventory of 385 landslides and constrain their dynamics. We show that extremely high incision rates during the early stage of the formation of the gorge explain the space-time clustering of thousand-year-old large (up to ∼2 km2) landslides, independently from the lithological context. These landslides are currently non-active and poorly eroded. Their deposit areas partly cover the riverbed with boulders, armouring the channel and inhibiting further incision. The landslides that occurred over the last 60 years are shallower slope failures of smaller size and higher mobility. They tend to disappear rather quickly from the landscape, sometimes within a few years. Their distribution is primarily controlled by threshold slopes, lithology, and the past large landslides, the influence of the land use being less pronounced. Overall, the sediment mobilization rates associated with these high frequency landslides significantly outpace the extreme landslide erosion pulse associated with the gorge formation. Our results provide insight on interactions between channel-hillslope coupling and feedbacks among landslide processes and river gorge formation in a unique environment. • Extremely high bedrock incision rates at the early stage of the gorge formation. • Incision dominates over lithology as the main control on a pulse of large landslides. • Large landslides are old, non-active, and inhibit further incision. • Recent, smaller landslides are controlled by threshold slopes and lithology, less by land use. • Sediment mobilization rates of recent landslides outpace that of the large landslide pulse. [ABSTRACT FROM AUTHOR]

Our understanding of tundra fire effects in Northern Alaska is limited because fires have been relatively rare. We sampled a 70+ year-old burn visible in a 1948 aerial photograph for vegetation composition and structure, soil attributes, terrain rugosity, and thermokarst pit density. Between 1948 and 2017 the burn initially became wetter as ice wedges melted but then drained and dried as the troughs became hydrologically connected. The reference tundra has become wetter over the last few decades and appears to be lagging through a similar sequence. The burn averaged 2.5 °C warmer than the reference tundra at 30 cm depth. Thinning of organic soil following fire appears to dramatically accelerate the background degradation of ground-ice features in response to climate change and promotes a plant community that is distinct in terms of taxa and structure, dominated by tall willows and other competitive, rather than cold-tolerant, species. The cover of sedges and mosses is low while that of willows and grass is high relative to the reference tundra. The changes in plant community composition and structure, increasing ground temperature, and thermokarst lead us to expect the observed biophysical changes to the tundra will persist centuries into the future. • An old tundra burn was found in a 1948 aerial photograph in northern Alaska. • Fire resulted in melted ground-ice, thermokarst, and conspicuously tall willows. • The adjacent unburned tundra also shows signs of thermokarst related to climate change. • There was a shift from cold-tolerant to competitive plants within the burn. • Ground temperature continues to warm relative to the unburned tundra. [ABSTRACT FROM AUTHOR]

Crowd counting on aerial images using the embedded system is a challenging task, due to high-definition images, low computing power, and limited memory. To tackle this task, we propose an efficient deep learning model named Flounder-Net structured like a flounder. In the Flounder-Net, a novel interleaved group convolution is proposed to eliminate the redundancy of network, and a rapid shrink of feature maps is employed to tackle the high-resolution problem. Since we would like to investigate the case of online aerial surveillance, we use the embedded system of a drone to run our algorithm. We also use the vision system of this drone to collect a set of high-definition aerial photographs as a benchmark. Extensive experiments on existing datasets and our aerial dataset show that Flounder-Net achieves FCN-level accuracy with three types of photograph devices: handheld cameras, surveillance cameras, and drone-based cameras. Additionally, Flounder-Net has 17 × fewer parameters and 20 × faster speed than FCN and allows an input image with arbitrary sizes. [ABSTRACT FROM AUTHOR]

• Water hyacinth was present in the uThongathi Estuary by the late 1970s. • Numerous Category 1b prohibited aquatic plant species now infest the estuary. • Increased nutrient availability and freshwater flow modifications are key drivers. • Expansion of the invasive aquatic plants reduced the open water habitat by 20–40%. • Restoring estuary function is required to prevent continued infestation. Invasive alien aquatic plants (IAAPs) are rapidly spreading in South African estuaries because of anthropogenic nutrient loading and persistent freshwater conditions. This study investigated changes in IAAP cover in relation to development pressures in the temporarily closed, subtropical uThongathi Estuary. Changes over time (1937-2018) were assessed using historical aerial photographs and rectified satellite imagery. The IAAP cover was assessed quarterly in relation to river flow, physico-chemical and nutrient monitoring data (2015-2018). The estuary floodplain has been transformed by sugarcane farming, urban development, uncultivated fields and sand mining operations, occupying 27% of the estuarine functional zone. In the past, agricultural return flow and urban runoff facilitated the sporadic growth of IAAPs in the estuary. Since 1983, discharges from the Tongaat Central Wastewater Treatment Works (WWTW) increased daily nutrient loads (current contribution: 1.8 kg d−1 N and 9.63 kg d−1 P), resulting in persistent IAAP growth in the estuary. Discharge from the WWTW represents 15% of the estuary volume and the increased inflow has changed the estuary functionality to a predominantly open, low salinity state. Aerial photograph analysis showed that IAAPs have established along the entire length of the estuary (∼7 km) and are abundant during all mouth states. They are flushed out of the estuary during high flow events (≥ 5 m3s−1) but steadily re-establish (within 3 months) after floods. The likely source of IAAPs are heavily infested upstream dams. A nutrient management reduction strategy integrated with manual/biological control methods and increased salinity (> 5) are needed to restore estuary function and prevent infestations. [ABSTRACT FROM AUTHOR]

Historical aerial photographs provide salient information on the historical state of the landscape. The exploitation of these archives is often limited by accessibility and the time-consuming process of digitizing the analogue copies at a high resolution and processing them with a proper photogrammetric workflow. Furthermore, these data are characterised by limited spectral information since it occurs very often in a single band. Our work presents a first application of deep learning for the extraction of land cover from historical aerial panchromatic photographs of the African cities of Goma, Bukavu and Bujumbura. We evaluate the suitability of deep learning for land cover generation from a challenging dataset of photographs from the 1940s and 1950s that covers large geographical extents and is characterised by radiometric variations between dates and locations. A fully convolutional approach is investigated by considering two network architectures with different strategies of exploiting contextual information: one used atrous convolutional layers without downsampling, whereas the second network has both downsampling and learned upsampling convolutional layers (U-NET). The networks are trained to detect three main classes namely, buildings, high vegetation and a mixed class of bare land and low vegetation class. High overall accuracies of >90% in Goma-Gisenyi and Bukavu, and >85% in Bujumbura are obtained. This work provides a novel methodology that outperforms a baseline standard machine learning classifier for the exploitation of the vast archives of historical aerial photographs that can aid long-term environmental baseline studies. Future work will entail developing domain adaptation strategies in order to make the trained network robust for different image mosaics. [ABSTRACT FROM AUTHOR]

• LiDAR and WorldView-3 data were used for the automatic generation of stand map. • LiDAR/WorldView-3 data set provides a better image segmentation quality. • The combination of LiDAR and multispectral image improves classification accuracy. • The stand map derived is not sufficient for tactical forest management plans. Although stand delineation approach based on aerial photographs and field survey produces high accuracy maps, it is labour-intensive and time consuming. Furthermore, conventional forest stand maps may have some uncertainties that can hardly be verified due to the experiments and skills of photo-interpreters. Therefore, researchers have been seeking more objective and cost-effective methods for forest mapping. LiDAR (Light Detection and Ranging) data have a high potential to automatically delineate forest stands. Unlike optical sensors, LiDAR height data provides information about both the horizontal and vertical structural characteristics of forest stands. However, it deprives of spectral data that may be successfully used in separating tree species. In this study, we investigate the potential of LiDAR – WorldView-3 data synergy for the automatic generation of a detailed forest stand map which can be used for a tactical forest management plan. Firstly, image segmentation was applied to LiDAR data alone and LiDAR/WorldView-3 data set in order to obtain the most suitable image objects representing forest stands. Visual inspection of the segmentation results showed that image objects based on the LiDAR/WorldView-3 data set were more compatible with the reference forest stand boundaries. After the segmentation process, the LiDAR and LiDAR/WorldView-3 data sets were independently classified using object-based classification method. We tested two levels of classification. The first was a detailed classification with 14 classes considering reference stand types. The second was the rough classification with 9 classes where some stand types were combined. The mean, standard deviation and texture features of LiDAR metrics and spectral information were used in the classification. The accuracy assessment results of LiDAR data showed that the Overall Accuracy (OA) was calculated as 0.31 and 0.43, and the Kappa Index (KIA) was calculated as 0.26 and 0.32 for the detailed and rough classifications, respectively. For the LiDAR/WorldView-3 data set, the OA values were calculated as 0.50 and 0.61, and the KIA were calculated as 0.46 and 0.55 for the detailed and rough classifications, respectively. These results showed that the forest stand map derived from the LiDAR/WorldView-3 data synergy is more compatible with the reference forest stand map. In conclusion, it can be said that the forest stand maps produced in this study may provide strategic forest planning needs, but it is not sufficient for tactical forest management plans. [ABSTRACT FROM AUTHOR]

The Otofuke River is a tributary of the Tokachi River, Hokkaido Japan, consisting of a channelized reach with a steep slope and a relatively large designed width. In August 2016, the river witnessed four major flooding events with a record-breaking discharge within a time span of 15 days. The Otofuke River originally had a straight channelized reach. However, because of a series of flooding events, the river promptly followed a sinusoidal path, damaging the riverbank embankments. The straight channelized reach of Otofuke River faced active erosion of both banks, leading to the formation of a meandering channel planform. This also led to levee breaches at the seven locations. In order to understand the mechanism of such a quick shift in river channel planform, we collected various field survey data and conducted computational analysis. We investigated the underlying processes by capturing the complex interactions between the rapid channel migration, unsteady discharge, and sediment transport. Field surveys including aerial photographs and images showed rapid formation of a meandering planform in the Otofuke River as a result of the typhoon on 31 August 2016. According to the maximum water level traces, bank erosion developed after the peak discharge. Numerical investigation revealed that alternate sandbars continuously developed and propagated downstream until the moment of peak discharge, and a significant accumulation of sediment subsequently started on the bars at the mid-channel during the flood falling stage. Such morphological processes led to strong flow deflections transversely toward the embankment that resulted in lateral channel shifting and levee breaches at a location where vegetation along the embankments had already disappeared. [ABSTRACT FROM AUTHOR]

A riparian vegetation dynamics model was proposed to consider the mutual influence of hydro-morphology and vegetation dynamics in a sand-bed river. The simulation model consists of four sub-models: river flow and sediment transport, recruitment, growth and expansion, and destruction of vegetation. The numerical simulation was applied to the downstream part of a sand-bed river for predicting the vegetation dynamics over a long period. The model performance is validated by comparison with the field investigation results obtained from analysis of aerial photographs. The simulation results showed that the growth of vegetation starts from the shoreline and the downstream part of sandbars, and the riparian vegetation increases the stability and the relative elevation of the bars. From the good agreements of the temporal tendencies of vegetation recruitment, expansion, and destruction with those obtained from the aerial photographs analysis, the vegetation dynamics model proposed in this study was verified to reproduce the long-term trend of riparian vegetation dynamics fairly well. Finally, the model was applied to different sandbar modes, i.e., alternating bars and multiple bars. The simulation results indicated that the expansion and destruction rates of vegetation show similar tendencies for both bar mode cases, and they were mainly determined by the magnitude of the annual maximum flood. However, the amount of vegetation coverage rate in the alternating bar case was higher than that in the multiple bar case. [ABSTRACT FROM AUTHOR]

The Sefidrud, a river flowing into south west Caspian Sea, is prone to avulsion. Previous studies of the river delta support the idea that several major avulsions have occurred during the past few thousand years. During the most recent major avulsion, a 23 km shift occurred along the coast. However, in literature the timing of the event is subject to uncertainty and therefore the causes of the avulsion remain unclear. Lying in the most populated area in Iran explains the importance to investigate the Sefidrud river avulsion. In this study, an 11 m-long sequence was obtained from an area where the river splits off into an old and a new distributary, and where a large pre-avulsion wetland had developed. Palaeo-environmental studies including sedimentology, macro-remain, and palynology were performed to detect changes in environment. Five radiocarbon dates were obtained to determine the avulsion age. In parallel an analysis of historical maps around the timing of the supposed avulsion and for more recent times provided a complementary way of refining the avulsion age. The results from this novel multidisciplinary study show that the last major avulsion occurred at a date after the early1800s and before 1929, during a period of rapid water level fall. Hence this is several centuries more recent than so far suggested. Studying aerial photographs, this research confirms the occurrence of minor avulsion between 1955 and 2014 where the avulsion point is around 2 Km from the coast line, and during a period of sea level fall. The results of this study also suggest that beyond the immediate effects on society and agriculture, avulsion increases the rate of coastal erosion near the abandoned distributary, as sediments are temporarily sequestered on the floodplain and causes intensive erosion. Avulsion has also major impact on coastal lagoons and inland wetlands by river diverting into the lagoons and wetlands. [ABSTRACT FROM AUTHOR]

Since the 1950s, Moenkopi Wash, in Arizona, United States, has been transformed from a relatively wide river with little riparian vegetation, to a narrow, heavily vegetated river that is less than half of its former width. We analyzed a ∼95-years-long instantaneous-discharge record, an extensive sediment-transport record, oblique and aerial photographs, historical channel surveys, and historical stage-discharge rating relations to determine the primary mechanisms responsible for this transformation. Frequent, large floods dominated the early part of the discharge record between 1926 and 1940. A dramatic ∼30 % decrease in annual mean discharge, and ∼27 % decrease in the mean of the partial-duration flood series occurred in 1941. This decline did not result in widespread channel change but provided an opportunity for vegetation to establish along the channel margins. Widespread channel narrowing began after a second decline in mean and peak discharge in 1956 at which time the channel banks became heavily vegetated. Between 1952 and 2019, the river channel narrowed by 57–59 %. Approximately 2–4 m of bed aggradation occurred at most study sites. As the channel narrowed and became more vegetated, large floods maintained the narrower channel width but did not cause rewidening. Suspended-sediment transport data illustrate the sediment trapping effects of vegetation that lead to channel narrowing; suspended-sand concentrations decline, thereby indicating sediment deposition, as vegetation becomes progressively inundated during floods. Furthermore, dense channel-margin and floodplain vegetation provide increased roughness, resulting in flood-peak attenuation and alteration of hydrograph shape. Vegetation expansion causes positive feedbacks whereby sediment deposition during floods is exacerbated by the roughness and sediment-trapping effects of vegetation leading to further narrowing. These positive feedbacks have reduced sediment delivery to the Little Colorado River downstream. Channel widening is not likely to occur unless there are very large floods that exceed the erosional threshold of the channel-margin vegetation, or unless large-scale vegetation removal efforts are undertaken. • Vegetation plays a stronger role than hydrology alone in causing channel change. • Channel change was caused by biogeomorphic feedbacks associated with vegetation. • Channel narrowing began as vegetation trapped sediment during a low flow period. • Biogeomorphic feedbacks caused flood attenuation and reduced flood-peak discharge. • Biogeomorphic feedbacks reduced sediment delivery to downstream rivers. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging can be conceptualized as a three-dimensional dataset of spectral information related to a particular landscape. Generally speaking, these are aerial photographs captured by Earth observation satellites. A useful analogy for a hyperspectral image is one of a cube formed with the image acquired along the X and Y axes and a third dimension of spectral bands of varying wavelengths. Given the wealth of data contained within these images, they have been employed in both civilian and military applications such as terrain recognition, urban development supervision, recognition of rare minerals, and various other objectives. The increased utilization of these images has garnered the interest of researchers striving to create solutions that may enable faster processing of the images via parallel processing. In this context, FPGA technology is an option capable of facilitating the implementation of such a system for observation satellites. This research is situated within this framework and aims to develop an FPGA-synthesized hardware accelerator to facilitate real-time hyperspectral image categorization. By taking this approach, hardware-specific solutions can be implemented for embedded applications that process hyperspectral images and can also be integrated with further image processing steps. The proposed accelerator was constructed based on an advanced algorithmic model, resulting in outcomes consistent with those generated by the software-based solution. The experimental results demonstrate that the engineered accelerator can attain a pixel classification time equal to or less than the pixel acquisition time, thus conforming to the real-time processing criteria concerning classification time. Further, the manufactured accelerator exhibits scalability that can classify distinct datasets with varying classes concurrently while maintaining a uniform logic resource utilization. [ABSTRACT FROM AUTHOR]

This contribution presents a novel U-Net convolutional neural network (CNN)-based workflow for automated mapping of bedrock fracture traces from 0.55 cm spatial resolution aerial photographs, acquired by unmanned aerial vehicles (UAV), over the Wiborg Rapakivi granite outcrops in the islands off the coast of the Loviisa Region in Southern Finland. The workflow comprised training a U-Net CNN using a small subset of photographs with manually traced fractures for optimizing the network parameters using the root mean squared propagation optimizer and sigmoidal focal loss function for semantic segmentation of input images and pixel-wise identification of fracture traces. The ridge detection algorithm was then applied to the U-Net prediction results, followed by vectorization of the fracture-traces pixels as vector polylines representing the traces of fractures. Both intensity values of the pixels and topological connectivity were used in the process of vectorization. Quantitatively the results were assessed using various accuracy assessment metrics. Qualitative evaluations of the results were implemented by comparisons of orientations and length-frequency distributions of automatically- and manually-mapped traces. The results show that the model has the class-balanced accuracy score 0.945, predicting 88.79% of the fracture-trace pixels. Bedrock outcrops with well-exposed surfaces present high true positive rates (>99%). The demonstration (test) site has the class-balanced accuracy score of 0.873 and 75% true positive rate. Additionally, the fracture trace networks replicate the orientation distributions of the manually digitized traces. The length distributions of automatic traces, however, differ with varying intensity from the manual trace length distributions. However, this study demonstrates that the workflow can be successfully applied to UAV-acquired images for fast and efficient automated mapping of bedrock-fracture traces during the initial phases of structural characterization of a region. • Semantic segmentation of UAV-acquired bedrock images with deep neural networks. • Bedrock fracture traces mapped automatically from UAV images using U-Net CNN. • Automated 2D structural characterization of Rapakivi granite islands in Finland. [ABSTRACT FROM AUTHOR]

A large-scale landslide occurred in 2015 in Senninvalley, Mt. Hakusan, Central Japan, releasing large amounts of sediment. To mitigate continuous erosion in the landslide area and its downstream adverse effects, reforestation activities were implemented from 2016 to 2021. This study assessed the status of vegetation recovery and erosion condition after 7 years using aerial photographs, light detection and ranging data, satellite imagery, and published documents. The results show that the greenery (herbaceous and woody plants) is mainly recovering on the upper slope of the landslide area where reforestation works were implemented, while the lower slope, where no reforestation activities have been applied, still lacks greenery. The lower slope suffered nearly 1–4 m of continued erosion from 2016 to 2022, while upper slope has no significant erosion. Within this 7-year period, the vegetation coverage on the upper slope of the landslide, where artificial restoration was applied, increased significantly from 10 to 43%. [Display omitted] • Large-scale landslides can result in denuded slopes that may experience prolonged erosion over an extended period. • Restoration of vegetation on landslide slopes can be achieved by carefully selecting native plant species. • Air photos and remote sensing data can assist in assessing erosion status and vegetation recovery in remote landslides. [ABSTRACT FROM AUTHOR]

Two-dimensional (2D) Sn-based organic-inorganic hybrid perovskites are useful materials for white light emitters due to their super high fluorescence efficiency and higher stability when compared to their three-dimensional (3D) counterparts. Herein, 2D Ruddlesden-Popper (RP) perovskites (BA) 2 Sn(Br 1-x I x) 4 (BA = n-butylammonium, x: 0–1) were prepared and regulated using halogen doping. As iodine content increased, the band gap of (BA) 2 Sn(Br 1-x I x) 4 was finely regulated from 2.35 eV to 2.0 eV, thereby tuning the photoluminescence from yellow to red with relatively high PLQY exceeding 40 %. The use of small amounts of Pb2+ to stabilize Sn2+ through the addition of PbO in the precursor solution resulted in high air- and photo stable (BA) 2 Pb 0.1 Sn 0.9 (Br–I) 4 phosphors with bright yellow to red light. The mixture with (BA) 2 PbBr 4 (blue light phosphor) followed by excitement by UV LED chips (285 nm) resulted in cold and warm white Sn–Br–I perovskite LEDs. The CIE coordinates and correlated color temperature (CCT) indicated Br 0.4 I 0.6 LED as an ideal warm-white emitter with suitable CIE (0.352, 0.330) and CCT (4376.3 K). The prepared white LEDs showed high photo and air stability without significant luminescence decay (<5 % relative) under continuous UV light excitation over 200 min and exposure to air for 3 months. [Display omitted] • The PL color of the prepared (BA) 2 Sn(Br 1-x I x) 4 was tuned by I dopant. • Sn2+ was effectively stabilized by small amount of Pb2+. • High air- and photo stable (BA) 2 Pb 0.1 Sn 0.9 (Br 0.4 I 0.6) 4 phosphor was obtained. • Warm white Sn–Br–I perovskite LEDs with long-duration stability were prepared. [ABSTRACT FROM AUTHOR]

[Display omitted] • We estimated SOC and TN transport rates due to landslides during 64 years. • Mean SOC and TN transport rates were 5.8–9.3 MgC km−2 yr−1 and 0.5–0.9 MgN km−2 yr−1. • SOC and TN transports in young forests (<20 yr) accounted for 87% of the total. • Forest harvesting increased landslide frequency, and SOC and TN transport rates. • Our results are useful for optimizing forestry practice to sequestrate more carbon. Landslides considerably affect soil organic carbon (SOC) and total nitrogen (TN) transports. However, the quantitative assessment of these effects remain limited. Furthermore, it is unresolved how forestry practices, especially clearcutting and subsequent planting, affect the SOC and TN transports by facilitating landslides. Therefore, we quantified the SOC and TN transports attributable to landslides from 1948 to 2012 in a steep mountainous area in Japan and examined the impacts of forest clearcutting on SOC and TN transports through landslides. We used 12 aerial photographs and two LiDAR DEMs to identify the landslide areas and to estimate their scar depths. In addition, we sampled soil at 17 points to examine the relationship between SOC and TN and soil depth. Subsequently, we estimated the temporal changes in SOC and TN transports due to landslides and examined the relationships between forest age and rates of SOC or TN transport due to landslides. During the 64 years, the average annual rate of landslide was 3.7 × 10−2 ha km−2 yr−1. Large temporal variability existed in rates of landslide, and the maximum rate of landslide was around 11.0 × 10−2 ha km−2 yr−1, while almost no landslide occurred in some periods. The average annual SOC and TN transport rates were 5.8–9.3 MgC km−2 yr−1 and 0.5–0.9 MgN km−2 yr−1, respectively. Many landslides occurred at places where trees were 15 years old or less in plated forests. The amount of SOC and TN transports in landslide areas with < 20 forest age accounted for 87% of the total amounts moved owing to landslides. That is, our results clearly showed that forest harvesting increased landslide frequency and SOC and TN transport rates. We expect our results to be useful for predicting carbon stock in forest ecosystems and optimizing forestry practices in steep mountainous regions. [ABSTRACT FROM AUTHOR]

• Soil pipe collapse is more related to early winter rainfall events. • Area loss is related to absence of vegetation's defensive shield effect. • Soil pipe orientation is related to stream bank orientation and tension cracks. • Massive area loss from sinkholes is related to catastrophic rainfall events. Subsurface soil erosion causes irreversible damage to agricultural fields by generating elongated pipes and elliptical sinkholes that can further develop into rills and gullies. However, despite their severe consequences, there is still limited available quantitative information on spatiotemporal dynamics of soil pipes and sinkholes, especially in semiarid regions; evidence of subsurface erosion rates is based on a small number of time points during a sampling period that is usually limited to a few successive years. Here, temporal changes in sinkhole area, and their level of circularity and orientation were quantified over a decade (2010–2020), in loess-derived soils in the semiarid Negev desert, Israel. Data were measured using high-resolution aerial photographs (1 pixel = 15 cm length), and were related to high-temporal-resolution rainfall data. The results revealed that: (1) soil pipe collapse is related to early winter rainfall events more than to mid or end-of-winter events, possibly due to the development of annual herbaceous vegetation that shields the soil and increases its resistance to erosion; (2) sinkhole and soil pipe morphological development and orientation are related to the orientation of stream banks and the tension cracks developed along those banks; and (3) changes in sinkhole area are related to catastrophic rainfall events and not to gradually accumulated annual rainfall. [ABSTRACT FROM AUTHOR]

Human interventions on sandy barriers disturb natural barrier dynamics, to the extent of having become key forces in modifying geomorphological evolution. This work identifies natural and human-induced drivers and analyses their importance to the multi-decadal evolution of the Ria Formosa barrier island system, in South Portugal. Aerial photographs from the last six decades and historical maps are used to assess changes in cross-shore rates, morphological characteristics (barrier and dune widths, inlet morphology and migration) and barrier areas, through systematic methods that can be easily transferred to other barrier systems. Interventions, and especially hard engineering ones (jetties, inlet stabilisations), affected barrier evolution trends. Shore-perpendicular works increased shoreline progradation updrift and initiated coastal retreat downdrift, with strongest erosive impacts along the edges of the system. Inlet stabilisations changed tidal inlet hydrodynamics and initiated ebb-shoal attachment to the barriers on either side of a non-migrating inlet that experienced loss of tidal prism. This shoal attachment was the main factor for the increase in total barrier area of Ria Formosa during the 60 years of analysis. Barrier growth after 2005 was slower, which could indicate that the system is reaching morphodynamic stability. Unlabelled Image • Human interventions interacting with natural processes changed barrier morphology. • Inlet stabilisations played a key role in barrier evolution. • Barrier growth was promoted by inlet tidal prism reduction and shoal attachment. • Disruption of longshore drift induced coastal retreat and barrier rollover. [ABSTRACT FROM AUTHOR]

• A complex of factors at different spatial scales affects beech growth response. • The variability in responses to an identical event implies uncertainty in reconstructing disturbance histories. • A detailed examination of all important parameters of the growth reaction is suggested. Despite improvements of dendrochronological techniques in many forest ecosystems, studies describing the growth responses of trees following disturbance events including comprehensive data on factors and processes behind tree-growth release are rare, especially for European temperate forests, limiting the interpretation and generalization of dendroecological studies. Here we examine the effect of key factors on the growth reaction of Fagus sylvatica L. to disturbances in 8 old-growth forests in the Czech Republic, independently dated using three different data sources – tree censuses, dendrochronology and historical aerial photographs. Using 117 increment cores, we studied four parameters of radial tree-growth surrounding canopy gaps: magnitude, duration, abruptness and response shift, in relation to factors of different spatial scales, i.e. tree-, gap- and stand-level variables. Our results indicate that the beech growth response to gap formation strongly differs across a complex of factors at diverse spatial levels, frequently occurring in noteworthy interactions. A positive trend for the gap size * radiation interaction in the analysis of magnitude was found, suggesting that the effect of gap area differs among topographical features. As for duration and abruptness, the interaction of distance and mortality mode was identified, indicating a longer and more gradual reaction along with increasing distance of the cored tree from the gap centre, varying by the type of the gap maker's death. In addition, an altitudinal gradient was found to predict the magnitude of growth changes and the time interval between the disturbance and the reaction of nearby trees. Surprisingly, the diameter of the surviving trees was not a good predictor of radial growth, with a significant positive relationship to just abruptness. An effect of locality was not observed, facilitating the broad applicability of our findings across a wide range of beech-dominated temperate forests. Our results emphasize the general limits of dendrochronological data in disturbance spatial pattern studies, as indicated by the considerable variability in the responses of neighbouring trees to identical disturbance event. A detailed examination of the growth reaction, considering not only the response magnitude but also other important parameters, plays a key role in the precision of detecting growth releases as well as in reducing the sample size required for summary disturbance history evaluations. In focusing on this issue we believe we contribute to future dendroecological research by demonstrating the need for more precise disturbance history evaluations. [ABSTRACT FROM AUTHOR]

• A method for row detection from photographs taken from cereal fields was developed. • The method was tested with drone photographs from a rye field. • 2D Fourier transform is used to find the angle between the rows and the picture axis. • High speed and high precision were achieved. Automatic locating of weeds from fields is an active research topic in precision agriculture. A reliable and practical plant identification technique would enable the reduction of herbicide amounts and lowering of production costs, along with reducing the damage to the ecosystem. When the seeds have been sown row-wise, most weeds may be located between the sowing rows. The present work describes a clustering-based method for recognition of plantlet rows from a set of aerial photographs, taken by a drone flying at approximately ten meters. The algorithm includes three phases: segmentation of green objects in the view, feature extraction, and clustering of plants into individual rows. Segmentation separates the plants from the background. The main feature to be extracted is the center of gravity of each plant segment. A tentative clustering is obtained piecewise by applying the 2D Fourier transform to image blocks to get information about the direction and the distance between the rows. The precise sowing line position is finally derived by principal component analysis. The method was able to find the rows from a set of photographs of size 1452 × 969 pixels approximately in 0.11 s, with the accuracy of 94 per cent. [ABSTRACT FROM AUTHOR]

Glaciers of the Svalbard archipelago react very rapidly to climate change in the polar environment. Combining the study of aerial photographs, digital elevation models and in situ measurements, it is possible to measure glacier geometric changes. Apart from the standard remote sensing data evaluation we focus on in situ measurements by ground penetrating radar. This study calculates length, area, volume and elevation changes of three glaciers in Dickson Land, central Spitsbergen, during the period 1990–2015. Ground penetrating radar surveys indicate ice thickness, which was used for calculating the volume and the bed topography of the glaciers. The mean ice thickness ranges from 21.2 to 52.4 m. Between 1990 and 2015, the glacierized area decreased from 5.37 ± 1.02 km2 to 4.45 ± 0.25 km2 (- 17.5%) and the volume reduced from 309.93 ± 4.75 mil m3 to 215. 44 ± 1.27 mil m3 (- 30.5%). The mean surface elevation decreased by 23.0 ± 8.4 m (Bertilbreen), 22.3 ± 10.2 (Ferdinandbreen) and 10.7 ± 7.8 m (Elsabreen). The average surface elevation (0.4 ± 0.3 to 1.0 ± 0.4 m a−1) and volumetric (- 0.22 ± 0.02 to - 3.61 ± 0.12 mil m3 a−1) changes correspond to values reported from this and other regions of Svalbard. [ABSTRACT FROM AUTHOR]

Aerial photographs depict objects from an overhead position, which gives them several unusual visual characteristics that are challenging for viewers to perceive and memorize. However, even for untrained viewers, aerial photographs are still meaningful and rich with contextual information. Such visual stimulus properties are considered appropriate and important when testing for expertise effects in visual recognition memory. The current experiment investigated memory recognition in expert image analysts and untrained viewers using two types of aerial photographs. The experts were better than untrained viewers at recognizing both vertical aerial photographs, which is the domain of their expertise, and oblique aerial photographs. Thus, one notable finding is that the superior memory performance of experts is not limited to a domain of expertise but extends to a broader category of large-scale landscape scenes. Furthermore, the experts' recognition accuracy remained relatively stable throughout the experimental conditions, illustrating the ability to use semantic information over strictly visual information in memory processes. [ABSTRACT FROM AUTHOR]

Abstract Sinkholes are the most typical surface feature of karst landscapes. Their correct mapping is of great importance for a proper assessment of the possible occurrence of new events, and the likely effects they could have on the anthropogenic environment. In this work we compare and discuss different ways to map sinkholes, including heuristic approach, automatic mapping through different resolution DTMs, and digital aerial photo interpretation, aimed at highlighting advantages and drawbacks of the multiple approaches. For our purpose we use a study area in the Murge of Apulia, SE Italy, where many sinkholes of several typologies are present to characterize the landscape. To evaluate the reliability of the sinkhole mapping, we set up a conceptual methodology that integrates a semi-automatic method with the heuristic approach, with a special focus on the importance of incorporating digital stereoscopic interpretation. In detail, we used as input two DTMs at different resolution, aimed to test the robustness of the automatic method in different circumstances. Comparison of the data produced by the different approaches illustrated how the number, density and boundaries of the sinkholes may vary as a function of the input data. The automated mapping provides reliable results in terms of number, dimensions, and morphometric attributes of sinkholes, and appears to be particularly useful when large areas must be analyzed. Within the whole process, we stress the role of the expert geomorphologist who must determine the accuracy of the product obtained with the automated mapping by correcting/eliminating features not attributable to karst processes. The presented approach highlights the potential of digital stereoscopy in substantially decreasing the errors encountered with traditional analog mapping, simultaneously reducing the working time significantly. Highlights • Automated mapping sinkholes using different resolution DTMs gives different results. • The quality of the inventory can be biased by uncertainties linked to multiple factors. • Combining multiple mapping techniques reduce bias obtaining more reliable maps. [ABSTRACT FROM AUTHOR]

Abstract Technological progress in remote sensing has enabled digital representation of terrain through new techniques (e.g. digital photogrammetry) and instruments (e.g. 3D laser scanners). However, the use of old aerial images remains important in geosciences to reconstruct past landforms and detect long-term topographic changes. Administrations have recently expressed growing interest in sharing photogrammetric datasets on public repositories, providing opportunities to exploit these resources and detect natural and anthropogenic topographic changes. The SfM-MVS photogrammetric technique was applied to scanned historical black and white aerial photos of the Serra de Fontcalent (Alicante, Spain), as well as to recent high-quality digital aerial photos. Ground control points (GCPs) extracted from a LiDAR-derived three-dimensional point cloud were used to georeference the results with non-linear deformations. Two point clouds obtained with SfM-MVS were compared with the LiDAR-derived reference point cloud. Based on the result, the quality of the models was analysed through the comparison of the stages on stable areas, i.e., lands where no variations were detected, and active areas, with quarries, new infrastructures, fillings, excavations or new buildings. This study also indicates that errors are higher for old aerial photos (up to 5 m on average) than recent digital photos (up to 0.5 m). The application of SfM-MVS to open access data generated 3D models that enhance the geomorphological analysis, compared to stereophotogrammetry, and effectively detected activities in quarries and building of landfills. Highlights • Existing techniques enable the use of historical and recent aerial photos to reconstruct DEMS • An area of interest is reconstructed using historical printed and digital aerial photos, obtained from readily available resources, through Structure-from-Motion • An airbone LiDAR derived point cloud was used as a benchmark and for the extraction of ground control points • Three different stages of the landform were analysed and compared • The quality of the reconstruction and the capability of monitoring changes were analysed [ABSTRACT FROM AUTHOR]

Abstract Agricultural activity evolution during the second half 20th century has profoundly changed agricultural landscapes in western Europe. Numerous scientific studies have shown that the consequences on erosive runoff processes are important. This study aims show one of the important consequences of these transformations that is the structural connectivity between cultivated plots and the river. The study is based on the analysis of aerial images taken at four different dates (1947, 1967, 1988 and 2014) and on the use of the LASCAR (LAndscape StruCture And Runoff) multi-agent model. It shows how agricultural landscapes have evolved over the past 70 years and how this has altered the connectivity of cultivated plots to the river. In addition to the development of spatial analysis indices, the results of this study offer key understandings relevant to all spatial scales (global to local). The study was carried out on a small agricultural watershed in the northwest of France (Lingèvres, 17.6 km2), whose recent landscape evolution is part of the overall dynamics of northwest French agricultural landscapes transformation. Highlights • Multi-agent simulation is a relevant approach for assessing hydrological connectivity. • Percentage of cultivated areas that are connected to river has more than tripled in 70 years. • Landscape entities organisation in space plays a key role in this hydrological connectivity. [ABSTRACT FROM AUTHOR]

Abstract Aerial image classification is of great significance in the remote sensing community, and many researches have been conducted over the past few years. Among these studies, most of them focus on categorizing an image into one semantic label, while in the real world, an aerial image is often associated with multiple labels, e.g., multiple object-level labels in our case. Besides, a comprehensive picture of present objects in a given high-resolution aerial image can provide a more in-depth understanding of the studied region. For these reasons, aerial image multi-label classification has been attracting increasing attention. However, one common limitation shared by existing methods in the community is that the co-occurrence relationship of various classes, so-called class dependency, is underexplored and leads to an inconsiderate decision. In this paper, we propose a novel end-to-end network, namely class-wise attention-based convolutional and bidirectional LSTM network (CA-Conv-BiLSTM), for this task. The proposed network consists of three indispensable components: (1) a feature extraction module, (2) a class attention learning layer, and (3) a bidirectional LSTM-based sub-network. Particularly, the feature extraction module is designed for extracting fine-grained semantic feature maps, while the class attention learning layer aims at capturing discriminative class-specific features. As the most important part, the bidirectional LSTM-based sub-network models the underlying class dependency in both directions and produce structured multiple object labels. Experimental results on UCM multi-label dataset and DFC15 multi-label dataset validate the effectiveness of our model quantitatively and qualitatively. [ABSTRACT FROM AUTHOR]

Abstract At the latitude of Reno, Nevada, USA, ~15–25% of the Pacific/North American plate boundary strain is accommodated east of the Sierra Nevada through a combination of normal oblique (dextral) slip along the eastern Sierra Nevada range front and dextral slip along faults in the northern Walker Lane. The North Valleys region sits within the northern Walker Lane directly east of the Sierra range front and is characterized by a series of small north-striking basins bound by Quaternary active faults that accommodate dextral oblique extension. Geodetic studies of the North Valleys estimate that the faults collectively accommodate 0.9–1.7 mm/y of extension and <0.3 mm/y of dextral slip; however, few Quaternary geologic/paleoseismic studies have been conducted. In this paper, geologic observations on the tectonic geomorphic expression and relative activity of faults within and bounding the North Valleys are described, including the Last Chance, Petersen Mountain, Fred's Mountain, and Spanish Springs Valley fault zones, and several un-named faults in Lemmon and Hungry valleys. Geologic observations from the interpretation of lidar-derived hillshade images, aerial photographs, satellite imagery, and field reconnaissance indicate that the faults all exhibit evidence of late Quaternary deformation, including offset alluvial fan deposits and abrupt range-front escarpments. The results indicate that the faults of the North Valleys are capable seismic sources characterized by extension rates on the order of hundredths to tenths of millimeters per year. The faults act to transfer slip from the Sierra Nevada range front to the northeastern Walker Lane by accommodating dextral oblique extension consistent with pull-apart basin kinematics within a simple shear right-lateral system. Highlights • Evidence for Quaternary deformation is documented for faults in the North Valleys area in the northern Walker Lane. • The cumulative geologic extension rate across the North Valleys system of faults approaches rates of strain accumulation. • Kinematic links between the Sierra Nevada and the northern Walker Lane is consistent with right-lateral simple shear. • Short, low slip faults may play an important role in partitioning slip in trans-tensional tectonic settings. [ABSTRACT FROM AUTHOR]

Abstract The assessment of multidecadal scale change in a polythermal glacial landsystem in the high Arctic is facilitated by a quantitative approach that utilises time series of aerial photographs, satellite images, digital elevation models, and field geomorphological mapping. The resulting spatiotemporal analysis illustrates a transition from glacial to proglacial/paraglacial conditions indicating that (1) the areal coverage of ice between the maximum LIA extent and 2013 decreased from 29.35 to 16.07 km2, which is a reduction in the glacierized area in the catchment from 62% to 34%; (2) the ice volume loss in the proglacial area amounted to 214.9 (±3%) million m3, which was attributed mostly to ablation of the glacier snout but to a lesser extent the degradation of ice-cored landforms; (3) the transition from areas formerly covered by glacier ice to ice-cored moraines, glacifluvial deposits, and other landforms was the most intense in the period 1990–2013; (4) two end member scenarios (polythermal glacial landsystem domains) evolve during glacier recession, each one dictated by the volume of debris in englacial and supraglacial positions, and include (a) subglacial surfaces (limited englacial and supraglacial debris) related to temperate basal ice and (b) ice-cored lateral moraines and moraine-mound complexes (significant supraglacial debris accumulations) related to marginal cold-based ice. An additional assemblage of geometric ridge networks (discrete or linear englacial and supraglacial debris concentrations) relates to crevasse and hydrofracture infill branching out from an esker complex and is indicative of either surging or later rapid release of pressurised meltwater from temperate to cold-based parts of the former glacier snout. Graphical abstract Unlabelled Image Highlights • Multidecadal changes in the high Arctic polythermal landsystem have been quantified. • The glacierized area in the catchment decreased from 62% to 34% (LIA-2013). • Two end member scenarios (related to former thermal regime) evolve during recession. • Azonal domains include geometric ridge networks and glacifluvial outwash plains. • We proposed four stages model of paraglacial development of Hørbyebreen foreland. [ABSTRACT FROM AUTHOR]

Understanding the islets morphodynamics of polynesian atolls requires a study of their evolution over several decades. These low-lying islands are exposed to highly variable ocean/weather conditions, especially high energy events. This study is based on a diachronic analysis of aerial photographs, in the field measurements and the U/Th dating of megablocks in three atolls in the western part of the Tuamotu archipelago. It presents an investigation of these dynamics from the 19th century until now. This period makes it possible to record the movements of sediments accumulated during decades of calm weather but also to observe the impact of moderate cyclones such as that of 1983. This article also highlights that exceptional cyclones such as those of 1822 and 1906 can dislocate the coral reef and move up the famous Rangiroa megablock whose volume has been significantly refined by photogrammetry at 369 m3. In Rangiroa and Anaa, we have identified three blocks weighting >500 tons that were put in place by these cyclones. • Unconsolidated ridges are the result of decades of calm weather-marine conditions. • Usual cyclonic submersions are strong islet reshaping drivers. • Transfer of sediments induced by cyclones help to raise and enlarge the islets. • Usual cyclones have limited effects on megablocks of more than ten cubic meters. • Exceptional cyclones (Hs > 15 m) can produce megablocks of >300 m3. [ABSTRACT FROM AUTHOR]

Lava flow hazards are usually thought to end when the erupting vent becomes inactive, but this is not always the case. At Kīlauea in August 2014, a spiny ʻaʻā flow erupted from the levee of a crusted perched lava lake that had been inactive for a month, and the surface of the lava lake subsided as the flow advanced downslope over the following few days. Topography constructed from oblique aerial photographs using structure-from-motion (SfM) software shows that the volume of the flow (∼68,000 m3) closely matches the volume of subsidence of the crusted lava lake (∼64,000 m3). The similarity of these volumes, along with the textural characteristics of the lava, shows that the lava that fed the flow had been stored beneath the surface of the perched lava lake, and that the flow was not generated by reactivation of the vent. This extends the duration of the local lava flow hazard presented by perched lava lakes and similar flow field structures that store lava, such as rootless shields. The flow probably occurred because the density of the lava beneath the crusted surface of the perched lava lake increased through loss of gas bubbles until it was able to penetrate the less-dense levee, which was composed of relatively vesicular overflows. The flow is thus equivalent to the lava seeps described previously at Kīlauea and elsewhere. We present a simple physical model for the pressure change at the base of a densifying body of lava, which we apply to this case study, and which could be applied to similar scenarios elsewhere. • Spiny lava flow broke through levee of perched lake inactive for about one month. • Surface of inactive perched lava lake subsided at same time. • Topographic modeling shows lava flow volume matches lava lake subsidence volume. • Breakout caused by densification of stored lava rather than by vent reactivation. • Extends duration of hazards posed by flow field structures that store lava. [ABSTRACT FROM AUTHOR]

Understanding land surface change in and sediment export out of proglacial landscapes is critical for understanding geohazard and flood risks over engineering timescales and characterizing landscape evolution over geomorphic timescales. We used automated Structure from Motion software to process historical aerial photographs and, with modern lidar data, generated a high-resolution DEM time series with coverage over 10 glacierized watersheds on Mount Baker, Washington, USA for the time period between 1947 and 2015. We measured basin-wide sediment yields and sediment redistribution on hillslopes and in stream channels. Slopes within most measured erosion sites are above theoretical and observed debris-flow thresholds. We observed significant erosion of hillslopes and limited deposition on hillslopes and in stream channels. Sediment delivery ratios during time periods with net erosion averaged 0.73. We determined, consistent with previous field observations, that debris flows originating from moraines are a primary erosion mechanism in proglacial zones on Mount Baker. Time series measurements indicate that temporal variability in erosion rates is associated with climate oscillations, with higher erosion rates during cooler-wetter periods. Basin-wide sediment yield is positively correlated with lithology (r2 = 0.54), hillslope angle (r2 = 0.52), drainage area (r2 = 0.82), and negatively correlated with stream channel slope (r2 = 0.67). Topographic differences between high and low yielding basins indicate that spatial variability in erosion on Mount Baker is sensitive to Pleistocene and Holocene glacial and volcanic activity. Specific sediment yields in six basins averaged 4600 ton/km2/yr, consistent with global measurements in glacierized catchments. Specific sediment yield decreased with increasing basin area, with total loads in the downstream main stem Nooksack River estimated between 480 and 820 ton/km2/yr. Proglacial sediment yields account for between 18 and 32 % of total sediment load in the main stem Nooksack River and exceed contributions by bluff and terrace erosion, which account for between 8 and 13 % of total load. Our findings indicate that erosion in glacierized basins is sensitive to decadal climate oscillations and that high proglacial sediment yields provide an important contribution to river systems downstream, particularly in catchments where upland topography and lithology is favorable. [Display omitted] • Erosion rates approximately follow Pacific Decadal Oscillation (PDO) climate cycles. • Denudation rates in 8 proglacial basins range from 0.2 to 4 mm/yr. • Mean specific sediment yield across 8 proglacial basins is 4600 ton/km2/yr. • Drainage area, hillslope angle, and surface lithology predict sediment yield. • Proglacial headwater basins supply 18–32 % of sediment in the downstream river system. [ABSTRACT FROM AUTHOR]

Fishponds play a key role in current pondscapes in many developed countries. They provide multiple ecosystem services affected by trade-offs mediated by different types of fishpond management. Many fishponds were impacted by the shift towards more intensive aquaculture and widespread eutrophication in the 20th century. These changes can undermine various ecosystem functions including the maintenance of aquatic and riparian biodiversity in the littoral areas, but their extent is not well documented. Here we use aerial photographs to investigate the changes of littoral areas of 46 protected and 20 unprotected fishponds in the Czech Republic between the years 1950 and 2019. Protected fishponds had initially larger littoral areas than the unprotected ones, and the difference persisted over time. However, littoral areas decreased markedly in 38 protected (mean area decrease by 22.5%) and all unprotected (mean area decrease by 16%) fishpond categories, especially during the second half of the 20th century. Within protected fishponds, the trend was unaffected by the reserve establishment year, fishpond area and conservation target. Our results suggests that legal protection did not prevent the initial catastrophic loss of littoral areas and facilitated at most minor recovery in the last two decades, with negative implications for long-term maintenance of aquatic diversity. We attribute the losses of fishpond littoral areas to high fish stocks and eutrophication resulting from additional feeding, pond manuring, and ongoing nutrient inputs from the pond catchments. We conclude that littoral areas of fishponds urgently need effective protection. This would require a paradigm shift towards less intensive fish stock management, more frequent summer drainage, and effective reduction of all nutrient inputs to increase the water quality. Such measures could help recover the littoral areas and the associated biota. [Display omitted] • We evaluated long-term changes in the littoral areas of fishponds in Czechia. • The areas decreased markedly in both protected and unprotected ponds since 1950. • Reserve duration, fishpond area and conservation target did not affect the trends. • Changes in fishpond management are recommended for recovery of littoral areas. [ABSTRACT FROM AUTHOR]

Highlights • First attempt to tell why aerial photos are less susceptible to criticisms than ground ones. • Explain differences between aerial and ground photos in relation to land use. • Examples of mis-using aerial photos provided. Abstract Although there is growing interest in the subjectivity of ground photos, similar criticism of aerial photos as a media product widely used in land use, environmental planning and management is seemingly absent. Inspired by pioneering work of Dorrian and Pousin (2013) and informed by the idea of Farman (2010) that users of aerial images can re-contextualise and subvert "master representation", this paper attempts to offer an explanation for this contrast and argues that it is harder to wage a subjectivist battle against aerial photography than against ground photos and discusses the possible better use of government possessed aerial photo data in a digital form. Four reasons for this based on disinterested observation, data neutrality, psychology and focus are offered. Two thought experiments and examples are used to help explanation. [ABSTRACT FROM AUTHOR]

Abstract This paper examines trends in woody vegetation and tree species composition in the Sudan zone of West Africa, using the Kano region of northern Nigeria as a case study. The study compares data on tree density, fuelwood production and tree species composition from fieldwork conducted in 1981 and 2016, as well as on several dates of aerial and satellite images since the 1960s. Recent satellite-based reports of greening in arid West Africa as a response to recovery from droughts in the 1970s and 1980s, are examined to explain the observed trends. Tree densities in the goods and services hinterland of Kano city have at least doubled since the drought period, and no decline, rather a slight increase was observed during the drought decades. This contradicts reports of woody vegetation trends from the more arid and less densely populated Sahel zone, which generally observed decline during the drought years and current recovery to pre-drought levels. The remarkable increase in tree numbers in Kano region is accompanied by increasing fuelwood production as suggested by greater concentration by farmers on tree species highly valued for fuel, at the expense of other traditional species. The main driver of such trends is thought to be rapid population growth in the context of a remaining dependence on wood as fuel by both urban and rural populations in Nigeria. Climate is thought to play only a minor role in explaining the trends. These observations confirm trends in fuelwood production observed in Kano region more than three decades previously, and indicate a somewhat Boserupian response to Malthusian-type pressures on available resources. Nevertheless, a return to rainfall levels of the drought decades combined with climate change predictions of increasing temperatures in dryland Africa, may have serious consequences for rural households if energy sources are not diversified. [ABSTRACT FROM AUTHOR]

Abstract Soil loss due to sheet or rill soil erosion is a critical problem in watersheds of Taiwan. However, an order-of-magnitude discrepancy of soil loss in the literature raises many questions. In this study, we conducted a new analysis using the most recent available data and the Universal Soil Loss Equation (USLE) to compute the amounts of sheet and rill erosion of the Shihmen reservoir watershed in northern Taiwan. Using four different Digital Elevation Models (DEMs), we identified relatively high soil erosion sites and found them to be located at similar locations despite of the difference in DEM. We also determined that the average soil erosion in the Shihmen reservoir watershed is comparable to other watersheds in Asia, but higher than those of the European Union. Furthermore, soil erosion is not uniformly distributed throughout the study area. It is found that the distribution of soil erosion is highly skewed to the right (right-tailed), which means that the majority of the distribution is concentrated to the left side (many cells with low soil erosion). Based on our model, approximately 2% of the areas account for 30% of the soil erosion. In other words, a small proportion of the areas contribute to a large proportion of the total soil loss. Moreover, the DEM created from airborne LiDAR yields the highest amount of soil erosion, the two DEMs created from satellite images yield the lowest amounts of soil erosion, and the DEM created from aerial photographs yields an in-between soil erosion amount. Their vertical resolutions range from high to low. It appears that the amount of soil erosion is influenced by the vertical accuracy of DEMs. In addition to the comparison of DEMs, we demonstrated rudimentary steps to visualize areas of high soil erosion risk using freely available tool for long-term monitoring. [ABSTRACT FROM AUTHOR]

Abstract Understanding land use changes provides data suitable for decision making processes in biodiversity conservation and landscape planning. We examined temporal change in landscape and vegetation formations on Elba Island (Tyrrhenian Sea, Italy) using digital aerial photographs (1954, 1978, 2000) and deriving land-use maps from photo interpretation. The detected variations of landscape composition, patch size and isolation revealed that the decline of agriculture is the main anthropogenic driver causing significant changes in the landscape, leading to a complete shift from an island under cultivation to a largely naturally vegetated island. The patchiness and the fragmentation have decreased so that natural vegetation with localized intensive cultivation and settlements, especially along the coast, remained the dominant elements of the mosaic. These changes in landscape composition and pattern are expected to have dramatic functional consequences and a landscape management plan avoiding land-use polarization through tourism is needed. Highlights • Significant changes have occurred in the landscape over the last 50 years. • The decline of agriculture is the main anthropogenic driver. • Natural vegetation increase and urbanisation are the main opposite modifying processes. • Natural vegetation with localized cultivation are dominant elements of the mosaic. • A landscape management plan avoiding land-use polarization through tourism is needed. [ABSTRACT FROM AUTHOR]

Abstract Remote sensing data analysis to infer land cover and the subsequent modelling of land use change are subject to uncertainties, which may have an impact on the accuracy of future land-use predictions. Part of these uncertainties come from the visual interpretation of remote sensing data as during this phase a specific knowledge and expertise are crucial. Visual interpretation includes the meaning of the image content but also goes beyond what can be seen on the image to recognize spatial and landscape patterns. The quality of recognition depends on the expertise in image interpretation and visual perception. Based on this statement, it is necessary to visualize also the information about uncertainty. This paper describes different perspectives of uncertainty visualization on the example of visually interpreted aerial photographs in a mining area. The study is focused on visualization techniques for uncertainty awareness analysis of land cover data. This approach should contribute to better understanding, assessment and potential spreading of visual information of uncertainty and help to appraise the data critically. The goal of this paper is to raise uncertainty awareness among practitioners who deal with land cover data and stress the visualizations techniques as a more reliable means to assess the quality, and hence the uncertainty, of these data. Highlights • Subjectivity in photo visual interpretation of aerial images plays key role. • Uncertainty of land cover data can be expressed by visual variables. • Method of transparent grid and transparent circles were ranked as best within all methods and metrics evaluated. [ABSTRACT FROM AUTHOR]

The gain-loss approach for greenhouse gas inventories requires estimates of areas of human activity and estimates of emissions per unit area for each activity. Stratified sampling and estimation have emerged as a popular and useful statistical approach for estimation of activity areas. With this approach, a map depicting classes of activity is used to stratify the area of interest. For each map class used as a stratum, map units are randomly selected and assessed with respect to an attribute such as forest/non-forest or forest land cover change. Ground observations are generally accepted as the most accurate source of information for these assessments but may be cost-prohibitive to acquire for remote and inaccessible forest regions. In lieu of ground observations, visual interpretations of remotely sensed data such as aerial imagery or satellite imagery are often used with the caveat that the interpretations must be of greater quality than the map data. An unresolved issue pertains to the effects of interpreter error on the bias and precision of the stratified estimators of activity areas. For a 7500-km 2 study area in north central Minnesota in the United States of America, combinations of forest inventory plot observations, visual interpretations of aerial imagery, and two forest/non-forest maps were used to assess the effects of interpreter error on stratified estimators of proportion forest and corresponding standard errors. The primary objectives related to estimating the bias and precision of the stratified estimators in the presence of interpreter errors, identifying factors and the levels of those factors that affect bias and precision, and facilitating planning to circumvent and/or mitigate the effects of bias. The primary results were that interpreter error induces bias into the stratified estimators of both land cover class proportion and its standard error. Bias increased with greater inequality in stratum weights, smaller map and interpreter accuracies, fewer interpreters and greater correlations among interpreters. Failure to account for interpreter error produced stratified standard errors that under-estimated actual standard errors by factors as great as 2.3. Greater number of interpreters mitigated the effects of interpreter error on proportion forest estimates, and a hybrid variance estimator accounted for the effects on standard errors. [ABSTRACT FROM AUTHOR]