Patch Size Research Articles

Protected area creation and its limited effect on deforestation: Insights from the Kiziba-Baluba hunting domain (DR Congo)

The study examines the spatiotemporal dynamics of landscape anthropization in the Kiziba-Baluba Hunting Domain (KBHD), near Lubumbashi in southeastern Democratic Republic of Congo, facing increasing human threats. It assesses these dynamics from 1989 to 2023 using remote sensing, Geographic Information Systems (GIS), and landscape ecology principles. The results reveal a significant decrease in forest cover, declining from 70.33 % in 1989 to 26.22 % in 2023, with an annual deforestation rate of -1.84 %. This deforestation has led to the expansion of savannas (63.93 %), agriculture (5.76 %), and built-up and bare soil (0.93 %) through patch creation and aggregation. The level of landscape disturbance has increased sixfold over 34 years, from 0.42 in 1989 to 2.81 in 2023. The reduction in the size of the largest forest patch and increased spatial isolation show rising fragmentation and dissection, often followed by the attrition of residual patches. These findings highlight the inefficiency of current conservation measures in KBHD, indicating a need for restructuring management, redefining protected area boundaries, developing a suitable management plan, implementing reforestation programs, strengthening enforcement of environmental laws, and actively involving local communities.

Determining microbial metabolic limitation under the influence of moss patch size from soil extracellular enzyme stoichiometry

Determining microbial metabolic limitation under the influence of moss patch size from soil extracellular enzyme stoichiometry

Forest disturbance regimes and trends in continental Spain (1985–2023) using dense landsat time series

Forest disturbance regimes across biomes are being altered by interactive effects of global change. Establishing baselines for assessing change requires detailed quantitative data on past disturbance events, but such data are scarce and difficult to obtain over large spatial and temporal scales. The integration of remote sensing with dense time series analysis and cloud computing platforms is enhancing the ability to monitor historical disturbances, and especially non-stand replacing events along climatic gradients. Since the integration of such tools is still scarce in Mediterranean regions, here, we combine dense Landsat time series and the Continuous Change Detection and Classification - Spectral Mixture Analysis (CCDC-SMA) method to monitor forest disturbance in continental Spain from 1985 to 2023. We adapted the CCDC-SMA method for improved disturbance detection creating new spectral libraries representative of the study region, and quantified the year, month, severity, return interval, and type of disturbance (stand replacing, non-stand replacing) at a 30 m resolution. In addition, we characterised forest disturbance regimes and trends (patch size and severity, and frequency of events) of events larger than 0.5 ha at the national scale by biome (Mediterranean and temperate) and forest type (broadleaf, needleleaf and mixed). We quantified more than 2.9 million patches of disturbed forest, covering 4.6 Mha over the region and period studied. Forest disturbances were on average larger but less severe in the Mediterranean than in the temperate biome, and significantly larger and more severe in needleleaf than in mixed and broadleaf forests. Since the late 1980s, forest disturbances have decreased in size and severity while increasing in frequency across all biomes and forest types. These results have important implications as they confirm that disturbance regimes in continental Spain are changing and should therefore be considered in forest strategic planning for policy development and implementation.

Spatial patterning and species coexistence: A case study using concentric circular vegetation patches in saline land

Spatial patterns in plant community structures within stressed ecosystems have drawn much attention in the field of ecology. However, the mechanisms underlying spatial formation and its impact on species coexistence and diversity remain controversial. In this study, we investigated concentric circular vegetation patches in coastal saline land, and analysed the spatial patterning of plant communities and associated soil physicochemical properties. Thereafter, we tested how the soil conditioned by plant communities from different locations within the vegetation patches influence the species growth and inter-specific competition. Our results show soil salinity enlarges in a centrifugal manner in horizontal direction in all patches. Soil salinity decreased and species diversity increased along with the increase of patch size. In addition, we found significant shifts in both the composition of plant communities and in soil physicochemical properties from outer to center. The results indicate that the pioneer species Suaeda salsa facilitated the subsequent species. However Suaeda salsa was inhibited and became inferior competitor in the soil conditioned by the subsequent species. We infer that the less-visible spatial patterns of soil physicochemical properties at small scales create ecological niches for specialized species, allowing them to coexist but not mix. We suggest that a trade-off between tolerance to salt stress and competitive ability under ameliorated conditions may underlie mechanisms of pattern formation in small scale. Our findings lend support to the idea that soil stress constraints community assembly and triggers spatial patterns, which, in turn, buffer the stress on plant communities and enhance species diversity.

Land use history and landscape forest cover determine tropical forest recovery

Abstract To conserve biodiversity and combat climate change it is vital to restore forest ecosystems. Natural forest regrowth is a nature‐based solution to restore forests, but it has rarely been evaluated how this is affected by the combination of previous land use intensity and surrounding forest cover, and how this varies between the two main tropical forest types; dry and wet forests. Thirty‐three plots were established on abandoned agricultural fields in a dry (13 plots) and wet (20 plots) tropical forest in Mexico and monitored 3 years for the following tree community attributes: structure, diversity, regeneration mode, potential symbioses with N fixing bacteria and mycorrhizal fungi. Previous land use intensity was described using interviews, and landscape forest cover and fragmentation within a 1000 m radius surrounding the plots were quantified using satellite images. Variable importance analyses indicated that land use intensity was more important than forest cover and fragmentation for the state of the tree community attributes after 3 years. This suggests that previous land use impacts the start of succession and leaves important legacies on the vegetation. Land use intensity, forest cover and fragmentation were equally important in determining the change in tree community attributes over time, indicating that both management practices and dispersal shape subsequently community assembly. A higher land use intensity decreases tree richness recovery, while size of the largest forest patch decreases tree density and connectivity of these patches increases tree density. The dry forest had a faster increase in tree density recovery compared to the wet forest through a high initial resprouting capacity and abiotically dispersed trees. Synthesis and applications: Both a higher land use intensity and fragmented forest landscape decreases the speed of forest recovery, indicating that human actions and landscape transformation shape the course of succession. Restoration in fragmented forest landscapes through natural regeneration is more suitable in areas with low previous land use intensity, higher forest connectivity and abiotically dispersed species. Effective restoration should therefore consider both land use history and landscape forest cover and be tailored to local socio‐ecological conditions.

Direct measurement of self-diffusiophoretic force generated by active colloids of different patch coverage using optical tweezers

Hypothesis: Synthetic micro/nanomotors are gaining extensive attention for various biomedical applications (especially in drug delivery) due to their ability to mimic the motion of biological micro/nanoscale swimmers. The feasibility of these applications relies on tight control of propulsion speed, direction, and type of motion (translation, circular, etc.) along with the exerted self-propulsive force. We propose to exploit the variation of both self-propulsion speed and force of active colloids with different patch coverages (with and without supporting layer) for engineering diffusiophoretic micro/nanomotors.Experiments: The microswimmers were designed at various patch coverages (10°, 30°, and 90°) with (Ti/Pt) and without (Pt) an adhesion layer for the catalytic patch through glancing angle metal deposition (GLAD) technique. Mean-square displacement (MSD) analysis was performed to obtain the self-propulsion parameters like speed and angular speed. Using optical tweezers (OT), the self-propulsive force was measured from the force power spectral density.Findings: The findings of our experiments suggest the non-requirement of any adhesion layer preceding the catalyst deposition since the Pt 10° colloidal batch had the maximal self-propulsion speed (4.61±0.3μm/s) and force (345±57fN) for 5% w/v H2O2 fuel concentration. Moreover, the self-propulsion speed and force decreased with increasing patch size, contrary to theoretical estimates. Also, the self-propulsive force obtained from MSD is 2 to 4 times lower in magnitude than the OT based force values. We believe that the self-propelling motion of the micromotors is possibly hindered due to interactions with the surface of the quartz cuvette during the optical microscopic analysis. Further, the MSD is limited to the self-propulsive motion in two dimensions. On the other hand, OT based force measurement involve trapping the particles in the bulk of the solution entirely avoiding the particle–substrate interactions. Hence, OT based force measurements are better than the propulsion velocity based stokes drag force estimates. We believe that this study can lay the foundation in designing efficient micro/nanomotors for translational biomedical applications.

Land Cover Mapping in a Mangrove Ecosystem Using Hybrid Selective Kernel-Based Convolutional Neural Networks and Multi-Temporal Sentinel-2 Imagery

Mangrove ecosystems provide numerous ecological services and serve as vital habitats for a wide range of flora and fauna. Thus, accurate mapping and monitoring of relevant land covers in mangrove ecosystems are crucial for effective conservation and management efforts. In this study, we proposed a novel approach for mangrove ecosystem mapping using a Hybrid Selective Kernel-based Convolutional Neural Network (HSK-CNN) framework and multi-temporal Sentinel-2 imagery. A time series of the Normalized Difference Vegetation Index (NDVI) products derived from Sentinel-2 imagery was produced to capture the temporal behavior of land cover types in the dynamic ecosystem of the study area. The proposed algorithm integrated Selective Kernel-based feature extraction techniques to facilitate the effective learning and classification of multiple land cover types within the dynamic mangrove ecosystems. The model demonstrated a high Overall Accuracy (OA) of 94% in classifying eight land cover classes, including mangrove, tidal zone, water, mudflat, urban, and vegetation. The HSK-CNN demonstrated superior performance compared to other algorithms, including random forest (OA = 85%), XGBoost (OA = 87%), Three-Dimensional (3D)-DenseNet (OA = 90%), Two-Dimensional (2D)-CNN (OA = 91%), Multi-Layer Perceptron (MLP)-Mixer (OA = 92%), and Swin Transformer (OA = 93%). Additionally, it was observed that the structure of the network, such as the types of convolutional layers and patch sizes, affected the classification accuracy using the proposed model and, thus, the optimum scenarios and values of these parameters should be determined to obtain the highest possible classification accuracy. Overall, it was observed that the produced map could offer valuable insights into the distribution of different land cover types in the mangrove ecosystem, facilitating informed decision-making for conservation and sustainable management efforts.

Patency evaluation after venous resection and reconstruction with autologous peritoneum in laparoscopic pancreatectomy

BackgroundConcomitant venous resection during pancreatectomy for pancreatic space-occupying lesions with veinous involvement is increasingly performed to achieve oncological resection. This study aims to report a single-center experience in peritoneal patch graft for vascular reconstruction during laparoscopic pancreatectomy. MethodsA retrospective analysis of all 6 patients who underwent laparoscopic pancreatectomy as well as the venous reconstruction with autologous peritoneum graft were compared with the control group, 8 patients who received Laparoscopic pancreaticoduodenectomy with venous resection and primary anastomosis. Postsurgical complications and outcomes were evaluated. We specifically assessed the reconstructed venous patency ratio by Cross-sectional Area Algorithm which was based on post-surgical CT-Venograms, thereafter the overall patency ratio was compared with the control group. ResultsAll 6 patients underwent venous resection and reconstruction using peritoneum patches of appropriate size, which includes 5 cases of lateral patch repairing and 1 case of tubular repairing. There are no significant differences noticed in post-surgical complications and outcomes between the peritoneum graft group and the control group. However, most of the peritoneum graft cases showed stenosis or complete occlusion after surgery. Partial venous stenosis or complete occlusion did not cause abnormal liver function as newly formed venous collaterals were early detected and progressively robust. ConclusionsUse of peritoneal patch as a venous substitute during the laparoscopic pancreatectomy contains satisfied operability and safety. However, the post-surgical reconstructed venous patency was an issue that need to be noticed. Overall, the peritoneum repairing technique is a considerable choice that assisted the oncological resection in certain situations.

Changes in urban green space configuration and connectivity using spatial graph-based metrics in Ardabil developing city, Iran.

Urban planning is essential for managing the diverse impacts of urban green spaces, such as public access, stormwater control, urban life quality, and landscape aesthetics, promoting sustainable urban development and urban residents' well-being by integrating green space considerations into city planning. The aim of this study is to use graph-based metrics to calculate the connectivity of UGS across the main municipal zones of Ardabil city over consecutive periods under different population growth rates. Another objective of this study is to compare the connectivity values of UGS in the four municipal zones and to evaluate changes in the connectivity indices at various distance thresholds of UGS patches. After identifying UGS in different periods, the changes in graph-based connectivity indices at various distance thresholds of UGS patches were analyzed. Additionally, the changes in connectivity indices over different periods and across various municipal zones were compared and analyzed. The findings reveal that UGS areas were larger in the past but have recently had smaller patch sizes. Connectivity between UGS nodes (dNL) decreased at various distances over the study years, showing a declining trend in different connectivity indices. UGS connectivity decreased in municipal zones 1, 2, and 3 but increased in recent years after a decline until 2012 across all four zones of Ardabil city. Zone 4 had the highest UGS connectivity due to newly developed urban areas and well-allocated UGSs. Integrating the ecological impacts of UGS connectivity in urban development and design will enhance trade-offs between conservation, public health, and social equity. New urban areas should allocate sufficient land for UGS and parks, ensuring accessibility to support health and leisure through municipal planning. The study highlights the need for sustainable urban development policies that prioritize the allocation and maintenance of UGSs.

The importance of heuweltjie patch isolation, size, and quality for arthropods in the Succulent Karoo, South Africa

AbstractA characteristic feature of Earth's drylands is the patchy nature of the vegetation, often referred to as a two‐phase mosaic landscape, comprised of a homogenous matrix containing distinctive vegetated patches. The latter are considered vital for ecosystem functioning as they provide refuge to biota from unsuitable conditions. Ground‐living (epigeic) and foliage arthropods contribute to dryland biodiversity patterns and processes, but little is known of how their richness, abundance, and composition varies between patches and the matrix in these systems. Throughout the Succulent Karoo, South Africa, such patches (earthen mounds referred to as heuweltjies) are hotspots for both floral and faunal diversity. We investigate how epigeic and foliage‐dwelling arthropod species richness, abundance, and community composition respond to heuweltjie patch characteristics, particularly isolation (distance to nearest neighbor, average distance to nearest neighbor, patch density), size (area), and quality (average plant height, dead plant cover, leaf litter cover, termite frass) during a severe drought. Patch isolation and quality were significantly correlated with arthropod richness, abundance, and community composition. More specifically, average proximity of sampled heuweltjies to other neighboring heuweltjies, termite frass, and vegetation structure (height, leaf litter and dead plant cover) were key determinants for epigeic and foliage‐dwelling arthropods' species richness, abundance, and community composition. The uniqueness of these mounds as landscape features increases niche and microhabitat availability for arthropods. This emphasizes their importance as promoters of landscape heterogeneity and highlight heuweltjies as key to the spatial distribution of arthropod assemblages in the Succulent Karoo.

Pansharpening with multi-CAE: impact of patch size and overlapping pixels on spectral and spatial distortion

Pansharpening with multi-CAE: impact of patch size and overlapping pixels on spectral and spatial distortion

Scale effect of landscape characteristics on undergrowth vegetation variance with different ecological traits

Herbaceous understory vegetation plays a significant role in temperate forested ecosystems, with its diversity strongly dependent on both local and landscape conditions. However, the contribution of these factors across different spatial scales (e.g., scale effect) remains inadequately understood. The objective of this study was to determine the most important landscape and local characteristics influencing the species richness of the herb layer. We surveyed and sampled 53 sites in a mountainous forest within the Baihuashan National Nature Reserve of Beijing. Using generalized linear models, we investigated the simultaneous effect of local (e.g. tree density, average tree height, and cover of tree, shrub, and ground vegetation) and landscape (e.g. Area-weighted Mean Patch Size, Area-weighted Mean Euclidean Nearest Neighbor Distance, Area-weighted Mean Fractal Dimension Index, forest area, and shrubbery area) variables on the richness of species in the understory herb layer. Variation partitioning was used to examine the individual contribution of local and landscape variables. Our results identified 366 plant species from 95 families across 53 sampling sites. Significant differences in species richness were found among life forms, seed dispersal types, and habitat preferences (P < 0.001). Parsimonious models indicated that combining local and landscape factors at a 2 km scale explained the most variation in phanerophytes (R2 = 0.658). Locally, average tree height significantly influenced the species richness of biotic dispersal, phanerophytes, and geophytes. At the landscape level, variables such as the Area-weighted Mean Fractal Dimension Index and farmland area significantly impacted plant community traits across different scales. Understanding these effects is crucial for developing effective sustainable forest ecosystem management strategies and for establishing robust ecological conservation policies.

Bird Habitat Suitability Analysis of The New City Centre Gede Bage Bandung Using Landscape Metrics

Abstract One focus of landscape ecology studies is the study of landscape patterns. Understanding the landscape patterns can help to find suitable habitat for the development or restoration of wildlife habitats, which can partly be expressed quantitatively with measurable or computable indicators by Geographic Information System, i.e. landscape metrics. This paper aims to analyze suitable habitat for bird habitat in The New City Centre in Gede Bage Bandung using statistical relationships between the occurrence of birds and the landscape pattern. Our study focus on habitat for Javan Pond Heron (A. speciosa) and Cattle Egret (B. ibis), two bird species which has been lived since 1970 in Ranca Bayawak Village inside The New City Centre Gede Bage Bandung. This study using landscape metrics software FRAGSTAT 4.2, parameter characterising landscape pattern were determined for the UTM 4 km x 4 km cells covering the whole of The New City Centre Gede Bage Bandung. The characteristic bird habitat define based on three types of landscape pattern : built-up areas, green spaces and water bodies. The study results based on statistical calculations found green spaces landscape metrics especially paddy field, forest and green open spaces are suitable for bird habitat, the type of patch that has the second largest patch size from total area. From statistical calculations also known built-up areas landscape metrics has the highest probability of fragmentation and the highest edge effect as a result of The New City Centre Gede Bage Bandung development.

Assessing four decades of fire behavior dynamics in the Cerrado biome (1985 to 2022)

BackgroundFire significantly transforms ecology and landscapes worldwide, impacting carbon cycling, species interactions, and ecosystem functions. In the Brazilian Cerrado, a fire-dependent savanna, the interaction between fire, society, and the environment is evident. Given that wildfires significantly contribute to greenhouse gas emissions, our study aimed to analyze four decades of burned area data to understand changes in fire dynamics, using Collection 2 of annual MapBiomas Fire maps (1985 to 2022). Our study examined spatiotemporal patterns, fire recurrence, fire distribution across land uses, temporal changes in fire scar size, burned area variations across ecoregions, and their correlation with farming areas.ResultsFrom 1985 to 2022, fire impacted 40% (792,204 km2) of the Cerrado biome, with 63% burning more than once. Natural vegetation was the most affected, primarily due to human-driven ignition during the dry season. A noticeable trend of later peaks in fire activity, concentrated towards the end of the dry season, along with an increase in patch size over time, characterized a clear shift in the Cerrado fire regime. Recently, the MATOPIBA region and the northern biome exhibited significant fire clusters, with burned areas rising alongside farming expansion. The ecoregion-based analysis identified fire hotspots, with the "Bananal" ecoregion, the largest wetland area in the biome, exhibiting increased fire recurrence and larger patch size over time.ConclusionsOur four-decade analysis of fire dynamics in the Cerrado revealed human-induced changes in the fire regime, originally shifting from July to September to a new fire season from August to October. This shift poses several environmental threats given their overlap with the driest months of the year. This study improved our understanding of changes in fire patterns and their impacts on each ecoregion and land use. Wetlands experienced the highest relative burned area, highlighting their ecological importance and increased vulnerability. In the southern Cerrado, where farming is established and natural vegetation more fragmented, fire events tend to decrease; while in the north, with recent farming expansion, fire susceptibility rises. Conservation-oriented strategies, like the Brazilian Integrated Fire Management (MIF), are crucial for mitigating impacts while enhancing the Cerrado’s resilience to climate change.

Controlled Porosity of Selective Laser Melting-Produced Thermal Pipes: Experimental Analysis and Machine Learning Approach for Pore Recognition on Pipes Surfaces.

This study investigates the methods for controlling porosity in thermal pipes manufactured using selective laser melting (SLM) technology. Experiments conducted include water permeability tests and surface roughness measurements, which are complemented by SEM image ML-based analysis for pore recognition. The results elucidate the impact of SLM printing parameters on water permeability. Specifically, an increase in hatch and point distances leads to a linear rise in permeability, while higher laser power diminishes permeability. Using machine learning (ML) techniques, precise pore identification on SEM images depicting surface microstructures of the samples is achieved. The average percentage of the surface area containing detected pores for microstructure samples printed with laser parameters (laser power (W) _ hatch distance (µm) _ point distance (µm)) 175_ 80_80 was found to be 5.2%, while for 225_120_120, it was 4.2%, and for 275_160_160, it was 3.8%. Pore recognition was conducted using the Haar feature-based method, and the optimal patch size was determined to be 36 pixels on monochrome images of microstructures with a magnification of 33×, which were acquired using a Leica S9 D microscope.

Fire refugia patch dynamics differ between prescribed fires and wildfires in longleaf pine savanna

Increasing wildfire activity has spurred an increasing push for the application of prescribed fire to reduce wildfire risk while simultaneously acting as a surrogate for fire’s historical role as a fundamental ecosystem process. However, prescribed fires are often ignited using uniform ignitions to maintain operational control and there are concerns that they may not be able to replicate the landscape heterogeneity, particularly patterns in unburned patches, generated by historical fires. Fire refugia, unburned areas within fire perimeters, play an integral role in determining post-fire recovery and community structure. We assess patterns in fire refugia across 443 large (>200 ha) wildfire and prescribed fire perimeters using remotely sensed fire severity data in longleaf pine (Pinus palustris) savannas of the Florida Panhandle. Contrary to concerns, large prescribed fires had a significantly greater proportion of unburned area than wildfires, driven by larger refugia patch sizes. Drier conditions promoted smaller and more numerous fire refugia patches. Our study demonstrates differences in wildfires versus prescribed fire outcomes on landscape structure, with implications for future longleaf pine savanna management.

Semantic Web Techniques for Extracting and Analyzing of Cropland Abandonment in Hilly Areas

Due to rising rural labor costs, farmland abandonment is common in China's hilly areas. Timely, accurate extraction of its spatiotemporal changes is crucial for sustainable farmland use. This study presents a novel approach for extracting abandoned cropland using NDVI time series from Sentinel-2 satellite images, exploiting the difference in vegetation phenology between abandoned and cultivated croplands. Dynamic Time Warping (DTW) algorithm was used to determine the similarity between NDVI time-series curves of different cropland types. The similarity metrics were used to find the optimal NDVI threshold for abandoned cropland through F1-score evaluation. The method's overall accuracy is 92.4%, higher than comparison methods at 84.60% and 75.6%. The approach captures year-round vegetation changes, expands time dimension data, and improves accuracy. Spatiotemporal analysis revealed decreased patch size, increased shape complexity, and more dispersed distribution of abandoned cropland over time. Major factors were proximity to settlements, roads, water bodies.

Fusing Multispectral and LiDAR Data for CNN-Based Semantic Segmentation in Semi-Arid Mediterranean Environments: Land Cover Classification and Analysis

Spectral confusion among land cover classes is quite common, let alone in a complex and heterogenous system like the semi-arid Mediterranean environment; thus, employing new developments in remote sensing, such as multispectral imagery (MSI) captured by unmanned aerial vehicles (UAVs) and airborne light detection and ranging (LiDAR) techniques, with deep learning (DL) algorithms for land cover classification can help to address this problem. Therefore, we propose an image-based land cover classification methodology based on fusing multispectral and airborne LiDAR data by adopting CNN-based semantic segmentation in a semi-arid Mediterranean area of northeastern Aegean, Greece. The methodology consists of three stages: (i) data pre-processing, (ii) semantic segmentation, and (iii) accuracy assessment. The multispectral bands were stacked with the calculated Normalized Difference Vegetation Index (NDVI) and the LiDAR-based attributes height, intensity, and number of returns converted into two-dimensional (2D) images. Then, a hyper-parameter analysis was performed to investigate the impact on the classification accuracy and training time of the U-Net architecture by varying the input tile size and the patch size for prediction, including the learning rate and algorithm optimizer. Finally, comparative experiments were conducted by altering the input data type to test our hypothesis, and the CNN model performance was analyzed by using accuracy assessment metrics and visually comparing the segmentation maps. The findings of this investigation showed that fusing multispectral and LiDAR data improves the classification accuracy of the U-Net, as it yielded the highest overall accuracy of 79.34% and a kappa coefficient of 0.6966, compared to using multispectral (OA: 76.03%; K: 0.6538) or LiDAR (OA: 37.79%; K: 0.0840) data separately. Although some confusion still exists among the seven land cover classes observed, the U-Net delivered a detailed and quite accurate segmentation map.

Projecting the long‐term effects of large‐scale human influence on the spatial and functional persistence of extant longleaf pine ecosystems in the Florida Flatwoods Pyrome

AbstractDecades of human activities and fire suppression have adversely affected longleaf pine (Pinus palustris) ecosystems, which are home to high levels of diversity and endemism. These iconic ecosystems also now face challenges from urbanization and climate change, which will alter conservation outcomes over the remainder of the 21st century. To explore how long‐term, large‐scale human influences could affect the spatial and functional persistence of extant longleaf pine ecosystems in the Florida Flatwoods Pyrome, we extracted a set of 2400 longleaf pine patches ≥40 ha in size from the Florida Longleaf Pine Ecosystem Geodatabase. Projections from the FUTURES urban growth model and the Florida 2070 project indicate that development will lead to losses of existing longleaf pine habitat, reductions in longleaf pine patch size, and patches that are predominantly located in close proximity to developed areas. Finer‐scale patterns of longleaf pine loss in three focal landscapes highlighted differences in land protection, ecological setting, and development pressure and the value of using of multiple urbanization iterations. The occurrence of suitable conditions to conduct prescribed fires, a crucial tool for maintaining, improving, and restoring longleaf pine ecosystems, is projected to decrease seasonally throughout the study area. As a result, the functional persistence of ecosystems is at risk due to climate changes that increase barriers to the safe and reliable application of intentional fire. The long‐term viability of this critical ecosystem will warrant the evaluation of adaptive strategies that explicitly account for the individual and compounding effects of urban development and changing fire management conditions when considering options for ecosystem protection, management, and restoration.

Improving lung nodule segmentation in thoracic CT scans through the ensemble of 3D U-Net models.

The current study explores the application of 3D U-Net architectures combined with Inception and ResNet modules for precise lung nodule detection through deep learning-based segmentation technique. This investigation is motivated by the objective of developing a Computer-Aided Diagnosis (CAD) system for effective diagnosis and prognostication of lung nodules in clinical settings. The proposed method trained four different 3D U-Net models on the retrospective dataset obtained from AIIMS Delhi. To augment the training dataset, affine transformations and intensity transforms were utilized. Preprocessing steps included CT scan voxel resampling, intensity normalization, and lung parenchyma segmentation. Model optimization utilized a hybrid loss function that combined Dice Loss and Focal Loss. The model performance of all four 3D U-Nets was evaluated patient-wise using dice coefficient and Jaccard coefficient, then averaged to obtain the average volumetric dice coefficient (DSCavg) and average Jaccard coefficient (IoUavg) on a test dataset comprising 53 CT scans. Additionally, an ensemble approach (Model-V) was utilized featuring 3D U-Net (Model-I), ResNet (Model-II), and Inception (Model-III) 3D U-Net architectures, combined with two distinct patch sizes for further investigation. The ensemble of models obtained the highest DSCavg of 0.84 ± 0.05 and IoUavg of 0.74 ± 0.06 on the test dataset, compared against individual models. It mitigated false positives, overestimations, and underestimations observed in individual U-Net models. Moreover, the ensemble of models reduced average false positives per scan in the test dataset (1.57 nodules/scan) compared to individual models (2.69-3.39 nodules/scan). The suggested ensemble approach presents a strong and effective strategy for automatically detecting and delineating lung nodules, potentially aiding CAD systems in clinical settings. This approach could assist radiologists in laborious and meticulous lung nodule detection tasks in CT scans, improving lung cancer diagnosis and treatment planning.