Class Map Research Articles

Remote estimates of soil organic carbon using multi-temporal synthetic images and the probability hybrid model

Soil organic carbon (SOC) plays a key role in soil function, ecosystem services, and the global carbon cycle. Digital SOC mapping is essential for agricultural production management. Digital SOC mapping based on multi-source remote sensing data has been integrated well into prediction models and methodological approaches on different mapping scales. However, the mixed use of synthetic images and the application of the hybrid model considering the soil classification probability are rare. Here, we propose a probability hybrid model to estimate and map SOC content and distribution. Multi-temporal synthetic images were used to build the probability hybrid model. One hundred forty topsoil samples were collected in Suihua, a city in a typical black soil region in China. Cloud-free Sentinel-2 images were acquired from the bare soil periods between 2018 and 2021. Minimum, maximum, mean, and median synthetic images were calculated using single images from the same period. The random forest and support vector machine models were used for discriminating soil class and calculating soil classification probability, and then random forest regression model was applied to SOC mapping. Soil class mapping and classification probability were performed for Phaeozems, Chernozems, and Cambisols of the World Reference Base for Soil Resources (WRB) based on soil texture, climatic factors, and the normalized differential vegetation index (NDVI). Based on soil class mapping results, global models using single temporal images and multi-temporal images, the hybrid model using multi-temporal images for the three soil classes as well as a probability hybrid model using multi-temporal images were built and compared their SOC predictions. Our study showed that (1) Phaeozems, Chernozems and Cambisols could be classified accurately, and the overall validation accuracy of random forest model was 91.67%. (2) The correlations of SOC with bands and band indices improved using multi-temporal images, and the use of mixed synthetic images was better than the use of only one synthetic image. (3) The hybrid model performed far better than the global models, and the probability hybrid model led to the highest prediction accuracy, a validation R2 of 0.77 and an RMSE of 2.30 g kg−1. (4) The probability hybrid model was more accurate than the original hybrid model for digital SOC mapping, and the SOC distribution in boundary regions was smoother and more continuous. Our results suggest that the probability hybrid model has a large potential for SOC prediction and mapping.

Towards a New Political Economy of Turkish Capitalism: Three Worlds

ABSTRACT Political economists investigating Turkey's turbulent path in recent years predominantly work from within three different characterizations of Turkish capitalism: authoritarian neoliberalism, crony capitalism, and state capitalism. This article critically reviews these competing visions and identifies directions for future research. I argue that, fundamental differences aside, these approaches together illustrate the indispensability of a political economy perspective for comprehending Turkey’s current predicament, in particular its authoritarian turn and ongoing systemic crisis. Yet meeting the potential of this research program also requires resisting rigid macro conceptualizations and aiming instead for empirically rich analyses of nuts-and-bolts phenomena such as changes in the class map, sectoral regimes, and challenges of development, with a view to identifying feasible strategies of renewal post-AKP.

Homological methods in certain Picard group computations

Let G be a connected complex semisimple Lie group, Γ be a cocompact, irreducible and torsionless lattice in G and K be a maximal compact subgroup of G. Assume Γ acts by left multiplication and K acts by right multiplication on G. Let MΓ=Γ﹨G, X=G/K and XΓ=Γ﹨X. In this article we prove that for any n≥0, the composition Hn(XΓ,C)→Hn(MΓ,C)→Hn(MΓ,OMΓ) is an isomorphism. As an application when G is simply connected, we compute the Picard group of MΓ for the cases rank(G) =1,2. More precisely we show that if rank(G) =1, Pic(MΓ)=(Cr/Zr)⊕A and if rank(G) =2, then Pic(MΓ)≅A via the first Chern class map, where A is the torsion subgroup of H2(MΓ,Z) and r is the rank of Γ/[Γ,Γ].

Digital mapping of soil texture classes for efficient land management in the Piedmont plain of Iran

AbstractAccurate prediction of digital soil maps allows for the evaluation of larger areas with respect to the design of efficient land management plans at the regional scale. Nowadays, there is an increasing demand for high spatial resolution‐gridded soil data for crop planning and management because it saves time and costs. One of the most essential soil physical properties affecting water holding capacity, nutrient availability and crop growth is soil texture. It exhibits a high spatial variability, but accurate maps for larger scales are lacking. The aim of this research was to produce gridded maps of soil texture fractions (clay, silt and sand) using regression‐based approaches and to establish soil texture classes using classification‐based techniques for the semi‐arid Piedmont plain of Iran. To this end, a digital elevation model and derived topographic indices, vegetation and soil‐based indices generated from 4‐year timeseries of remote sensing products of Landsat 8 OLI were used as covariates. The decision tree (linear) and its improved version of random forest (nonlinear) algorithms were used for both the regression and classification analysis. For both algorithms, the topography‐based indices and remotely sensed products were the most important predictors for the soil particle fractions. For the estimation of the different textural classes with multiple algorithms, we recorded a moderate overall accuracy rate of 54% and a Kappa coefficient of 17% for the validation datasets. It was observed that the nonlinear classification method of the random forest was more effective, and this was also the case for the regression modelling. In general, the random forest algorithm produced a more useful gridded map to help to design regional management plans based on soil properties.

Method of designing the system of classes of the subject area model on the basis of the architecture of integrated objects

The article presents a method of designing the system of software classes, which includes the stages of conceptual, logical and physical design of classes with the implementation of a software system in the architecture of integrated objects. The input data for design is the model of the subject area, which is obtained as the result of object-oriented analysis, which is represented by conceptual, mathematical and semantic models. The conceptual model defines the entities and attributes of entities and is considered as the basis for designing the properties of classes. Mathematical model, as a dynamic model of entity behavior, is formally or informally defined by mathematical formulas, algorithms, diagrams or descriptions of the implementation, which is the basis for designing class methods. The semantic model defines the concepts of subject area in relation to the elements of its model and includes identifiers, meaningful names and descriptions used for the design, programming and implementation of system functionality in terms of subject area. The proposed design method is based on the following principles: normalization of classes (reduction of data to atomic values, reduction of class methods arguments and removal of redundant class methods), replacement of inheritance by logical composition and physical integration, decapsulation (separation of physical functional model from logical data model), virtualization of classes (logical model of classes are considered as a set of relational data model and physical functional model of class methods), interrelation of models (end-to-end mapping of conceptual model on logical and physical model, and class implementation). The essence of the method is to perform successive stages of design: normalization of conceptual model, logical design of classes (transformation of mathematical model into normalized functional model, design and analysis of class characteristics), physical design of classes in architecture of integrated objects (creation of classes, mapping of logical classes on physical classes, implementation of classes in programming language). Actually, the developed design method leads to the creation of a model focused on solving problems, when instances of entities are not created entirely and only the properties and methods that are needed to calculate the desired results for given input values are used. An example of application of the method for a certain subject area with demonstration of design results is considered, and also the approach on the basis of query language of formulation of problems in terms of subject area with generation of source code in programming language is shown. The application of the method allows to increase the efficiency of creating software systems due to the non-redundancy of the system of classes and parameters of class methods, reuse of code and selection of the minimum required set of properties and class methods when solving specific problems.

A general model for creating robust choropleth maps

Choropleth maps visualize areal geographical data by grouping data into a few map classes and assigning different colors, shades, or patterns. Recent studies show that data uncertainty, commonly observed in real-life applications, should also be accounted for when determining the best classification scheme. Due to data uncertainty, a few studies note that map units might be placed in a wrong class, and the concept of map robustness has been introduced to minimize such misplacement. Recently, an algorithm has been developed to integrate robustness into the design of the optimal map classification scheme. However, the existing algorithm has two limitations: first, it is only suitable for certain robustness metrics. Second, when identifying the optimal class breaks, the existing algorithm requires predefined candidate class break values, which might lead to sub-optimal solutions. This paper resolves these issues by proposing a new model, namely, the Continuous Robust Map Classification Problem (CRMCP), and the associated solution approach. The CRMCP allows mapmakers to customize robustness metrics according to their data and applications. In addition, a particle swarm optimization algorithm is developed to solve the CRMCP. The model and algorithm are tested using American Community Survey data. Test results suggest that the new approach can find better solutions than the existing algorithm. The study improves the usability of choropleth maps when uncertain geographical attributes are involved and allows spatial analysts and decision-makers to incorporate robustness into the mapmaking process more flexibly.

Management of Counseling and Skill-building Services for Junior High School Students

Guidance and counseling services in schools help students develop their potential and overcome personal and peer-related issues. Studies have shown a general lack of learning motivation in students, which further lowers their interest in learning. Thus, the current study explored the development of learning interest in students and motivating them in the process. A personality potential test was conducted to see the students’ abilities, followed by continued mapping of specialization classes based on their potential. Because this study was conducted at an Islamic boarding school that has introduced new specialization courses, which include specialized classes in Arabic, English, Tahfidzul Qur’an, and Entrepreneur, researchers realized that students’ abilities and their development vary according to their skills. The method used in this research was documenting, reviewing all articles, analyzing, determining the focus of the topic of scientific work, collecting data from several pieces of literature relevant to the subject, and then drawing conclusions. In total, 25 articles from national and international accredited journals published between 2014 and 2021 were obtained from Google Scholar, Science Direct, Sinta, and lens.org, however, only 18 articles relevant to the topic were used. The results indicated that counseling and skill-building services produce students who are graceful in character and excel in work.
 Keywords: management counseling guidance services, skills building

A Multiscale Attention-Guided UNet++ with Edge Constraint for Building Extraction from High Spatial Resolution Imagery

Building extraction from high spatial resolution imagery (HSRI) plays an important role in the remotely sensed imagery application fields. However, automatically extracting buildings from HSRI is still a challenging task due to such factors as large size variations of buildings, background complexity, variations in appearance, etc. Especially, it is difficult to extract both crowded small buildings and large buildings with accurate boundaries. To address these challenges, this paper presents an end-to-end encoder–decoder model to automatically extract buildings from HSRI. The designed network, called AEUNet++, is based on UNet++, attention mechanism and multi-task learning. Specifically, the AEUNet++ introduces the UNet++ as the backbone to extract multiscale features. Then, the attention block is used to effectively fuse different-layer feature maps instead of direct concatenation in the output of traditional UNet++, which can assign adaptive weights to different-layer feature maps as their relative importance to enhance the sensitivity of the mode and suppress the background influence of irrelevant features. To further improve the boundary accuracy of the extracted buildings, the boundary geometric information of buildings is integrated into the proposed model by a multi-task loss using a proposed distance class map during training of the network, which simultaneously learns the extraction of buildings and boundaries and only outputs extracted buildings while testing. Two different data sets are utilized for evaluating the performance of AEUNet++. The experimental results indicate that AEUNet++ produces greater accuracy than U-Net and the original UNet++ architectures and, hence, provides an effective method for building extraction from HSRI.

SEMANTIC SEGMENTATION USING A UNET ARCHITECTURE ON SENTINEL-2 DATA

Abstract. This paper presents the development of a methodological framework, based on deep learning, for the efficient mapping of main land cover classes (built-up, vegetation, barren land, water body) on different urban and suburban landscapes. In particular, the proposed framework integrates the superpixel segmentation (an essential procedure) with deep learning. A combination of spectral bands and indices is introduced to produce optimal results, ensuring adequate discrimination between built-up and barren land classes. A UNET architecture is implemented, which can learn the characteristics of main land cover classes from the input data that can be deployed from a Colab notebook without excessive computational needs. The resulted classifications depict promising accuracy values (above 90%).

Farm-scale digital soil mapping of soil classes in South Africa

This study involved the evaluation of farm-scale digital soil classification in the Sandspruit catchment of the Western Cape Province, South Africa. The study aimed to evaluate a digital soil mapping (DSM) method, from feature selection, spatial predictions and sample design. The results showed that feature selection with the least absolute shrinkage and selection operator (LASSO) technique is a robust method as it had a high relative efficiency and achieved the highest accuracy for three out of the four soil classes predicted. This implies that covariate selection is the most notable aspect in DSM at the farm-scale. The top-performing predictive models achieved satisfactory results for soil associations (kappa = 0.64, accuracy = 74%), presence of a bleached topsoil (kappa = 0.64, accuracy = 74%) and soil depth (kappa = 0.48, accuracy = 74%), whereas only moderate results were achieved for soil texture (kappa = 0.43, accuracy = 66%). Lastly, the expert sampling locations had a higher average probability of occurrence (geographic and feature space distribution coverage) yet achieved similar performance to conditioned Latin hypercube sampling (cLHS).

Global coastal geomorphology – integrating earth observation and geospatial data

Coastal geomorphic classification identifying the type of depositional environment, is an important indicator for coastal state and vulnerability, e.g., beach, bedrock and wetland (tidal flat and estuarine/delta/back-barrier depression). Accurate and current maps of these geomorphology classes from local (100 m-1 km) to global (105 km) scales are important for a range of objectives in the coastal zone. Previous methods to create coastal geomorphic maps relied heavily on manual efforts and expert opinions. However, these approaches are labour intensive and impractical for consistent application at a global scale, while still including local details. Existing global coastal geomorphology datasets are a compilation of different local-scale datasets, with varying spatial, categorical resolutions and accuracies. The recent availability of “big” and “open” data as well as increased processing capabilities resulted in the use of satellite images and digital elevation models to identify sandy beach and cliff separately. In our work, these raster data were processed on Google Earth Engine, combined with shape descriptors of coastline vectors and then fitted into machine learning models to classify the intertidal coastal geomorphology into bedrock, beach or wetland classes on the global scale (from 56° S to 60° N, excluding coral reef/fringing reef and fjord dominated coasts). A unique attribute of this new method was the incorporation of shape descriptors extracted from shoreline vector data, which substantially improved the classification accuracy. Using existing data from European Union (EU), the United States of America (US) and Australia for training and testing, the model achieved averaged 85% accuracy for testing datasets in these continents and 84% accuracy for 10,000 independent global validation samples beyond these continents. Successful classification was achieved for coasts dominated by a single geomorphology class, as well as more complex environments, such as headland bay and barrier island systems, where multiple classes appear next to each other. The model had relatively less confidence in coasts with sandy cliffs, tall canopy cover and engineered structures. Complete statistics from this study covering 56° S to 60° N showed 36.8% (142,845 km) wetland, 26.7% (103,762 km) beach and 36.5% (141,579 km) bedrock. Beach and bedrock geomorphic classes were commonly detected at mid-latitudes (30o to 60o N/S), while wetland dominated tropical latitudes (0o to 30o N/S). The output dataset can be used for different coastal management purposes. The accuracy and categorical resolution of the classification can be further improved with the development of Earth Observation Big Data in the future.

Comparison of sampling designs for calibrating digital soil maps at multiple depths

Digital soil mapping (DSM) aims to produce detailed maps of soil properties or soil classes to improve agricultural management and soil quality assessment. Optimized sampling design can reduce the substantial costs and efforts associated with sampling, profile description, and laboratory analysis. The purpose of this study was to compare common sampling designs for DSM, including grid sampling (GS), grid random sampling (GRS), stratified random sampling (StRS), and conditioned Latin hypercube sampling (cLHS). In an agricultural field (11 ha) in Quebec, Canada, a total of unique 118 locations were selected using each of the four sampling designs (45 locations each), and additional 30 sample locations were selected as an independent testing dataset (evaluation dataset). Soil visible near-infrared (Vis-NIR) spectra were collected in situ at the 148 locations (1 m depth), and soil cores were collected from a subset of 32 locations and subdivided at 10-cm depth intervals, totaling 251 samples. The Cubist model was used to elucidate the relationship between Vis-NIR spectra and soil properties (soil organic matter (SOM) and clay), which was then used to predict the soil properties at all 148 sample locations. Digital maps of soil properties at multiple depths for the entire field (148 sample locations) were prepared using a quantile random forest model to obtain complete model maps (CM-maps). Soil properties were also mapped using the samples from each of the 45 locations for each sampling design to obtain sampling design maps (SD-maps). The SD-maps were evaluated using the independent testing dataset (30 sample locations), and the spatial distribution and model uncertainty of each SD-map were compared with those of the corresponding CM-map. The spatial and feature space coverage were compared across the four sampling designs. The results showed that GS resulted in the most even spatial coverage, cLHS resulted in the best coverage of the feature space, and GS and cLHS resulted in similar prediction accuracies and spatial distributions of soil properties. The SOM content was underestimated using GRS, with large errors at 0–50 cm depth, due to some values not being captured by this sampling design, whereas larger errors for the deeper soil layers were produced using StRS. Predictions of SOM and clay contents had higher accuracy for topsoil (0–30 cm) than for deep subsoil (60–100 cm). It was concluded that the soil sampling designs with either good spatial coverage or feature space coverage can provide good accuracy in 3D DSM, but their performances may be different for different soil properties.

Intelligent fault diagnosis and visual interpretability of rotating machinery based on residual neural network

Accuracy of machinery fault diagnosis and interpretability of diagnosis methods are fundamental to safe operation of machinery and help to improve the universality of the model. Mechanical vibration signals can reflect the operating state of the machine. Therefore, to improve the accuracy of fault diagnosis, this paper constructs a 6-layer residual neural network (ResNet06), which embeds two residual blocks to fully extract features of the mechanical vibration signals. Then, we use the gradient-based class activation map (Grad-CAM) and eigenvector-based class activation map (Eigen-CAM) to interpret the ResNet06 visually and to verify the ResNet06 correctness. Experimental results indicate that the fault diagnosis accuracy of our proposed model can reach almost 100%, and it can be seen that the model can accurately capture the fault points by the visualization of the model.

Digital mapping of soil erodibility factor in northwestern Iran using machine learning models.

Understanding the spatial distribution of soil erodibility factor (K-factor) at the district scale is essential for managing water erosion risk. In this research, we performed to predict the low and high classes of K-factor in the northwest of Iran. Based on this, soil sampling was performed at 64 points using the grid sampling method with 1km spacing. To calculate the K-factor, the distribution of particle size and organic carbon (OC) were determined. In addition, 21 terrain attributes were calculated by Digital Elevation Model (DEM) to add value to the soil data. Then, K-factor was modeled using Random Forest (RF) and Artificial Neural Network (ANN) models. In the next step, a non-linear Multiple Logistic Regression (NMLR) was used to obtain low and high classes of K-factor. The results showed that the performance of RF is superior to ANN with a high coefficient of determination [R2 = 0.85] and good accuracy [RMSE = 0.003 (Mg ha h/ha MJ mm)]. Therefore, the RF was employed for predicting the K-factor spatial distribution. Finally, using the NMLR model, the study area was divided into low and high classes of K-factor with good correlation [R2 Cox and Snell = 0.78, R2 Nagelkerke = 0.65]. The areas of these two classes were 60.4% for low class and 39.6% for the high class of K-factor. Based on these results, it was concluded that the resultant map of low and high classes of K-factor could be used by farmers and managers for managing soil water erosion risks in the study area.

Metagenomics Insights Into the Microbial Diversity and Microbiome Network Analysis on the Heterogeneity of Influent to Effluent Water.

Sanitizing the water sources of local communities is important to control the spread of microbial resistance genes, especially those for water-borne illnesses. The activities of antibiotic resistance gene (ARG)-host pathogens pose a threat to public health, and it has been estimated that the infection will lead up to 10 million deaths globally by the year 2050. Hence, in this study, we aim to analyze the efficiency of our municipal wastewater treatment plant (WWTP) process in producing pathogen-free water by investigating the microbial composition between influent and effluent water sites. Shotgun metagenomics sequencing using the Illumina platform was performed on the influent and effluent samples of six different WWTP sites located in Johore, Malaysia. After raw data pre-processing, the non-redundant contigs library was then aligned against BLASTP for taxonomy profiling and the Comprehensive Antibiotic Resistance Database for ARG annotation. Interestingly, the alpha-diversity result reported that effluent site samples showed higher abundance and diverse heterogeneity compared to the influent site. The principal component analysis (PCA) and non-metric multidimensional scaling (NMDS) plots also suggested that effluent sites showed high variation in the genetic material due to loosely clustered sample plots, as compared to the tightly clustered influent samples. This study has successfully identified the top three abundant phyla in influent—Proteobacteria, Firmicutes, and Bacteroidetes—and effluent—Proteobacteria, Actinobacteria, and Bacteroidetes—water. Despite the overlap within the top three abundant phyla in influent and effluent sites (Proteobacteria and Bacteroidetes), the ARG composition heat map and drug class phenotype plot bar exhibits a general trend of a downward shift, showing the efficiency of WWTP in reducing opportunistic pathogens. Overall, it was demonstrated that our municipal WWTP efficiently eliminated pathogenic microbes from the influent water before its total discharge to the environment, though not with the total elimination of microorganisms. This metagenomics study allowed for an examination of our water source and showed the potential interaction of species and ARGs residing in the influent and effluent environment. Both microbial profile structure and co-occurrence network analysis provide integrated understanding regarding the diversity of microorganisms and interactions for future advanced water sanitation treatments.

중도장애 유아를 위한 증강현실 기반 수업에 대한 인식과 실행 방안

The purpose of this study is to examine the necessity and implementation strategy of augmented reality(AR)-based classes for young children with moderate disabilities. Regarding the research method, a conceptual diagram analysis was performed to confirm the perception of AR-based classes and the empirical and theoretical knowledge of experts. First, in-depth interviews were conducted with 22 experts in the special education, educational engineering, and engineering fields. Afterwards, a multidimensional scale analysis, hierarchical cluster analysis, and validity evaluation were performed on the developed statements to analyze the conceptual map of AR-based classes for young children with moderate disabilities. As a result of the study, first, experts positively recognized the need for AR-based classes for young children with severe disabilities. Second, we confirmed that AR-based classes require the development of teaching strategies according to the systematic instructional design process. Third, we suggested methods such as action induction, context maintenance, and sensory information to improve the usability of AR contents for young children with moderate disabilities. Based on the above study results, we discussed the necessity and implementation plan for AR-based classes for young children with moderate disabilities, and the characteristics and usability improvement principles of AR contents. Subsequent research requires the development of various empirical studies and instructional models for the AR-based teaching strategy developed in this study. In addition, it is necessary to develop immersive content based on the curriculum and centered on teachers, students with disabilities, and developers.

I. S. MELEKHOV’S SCALE OF NATURAL FIRE DANGER OF FOREST ECOSYSTEMS: OVERVIEW OF MODERN RUSSIAN METHODOLOGICAL APPROACHES

The article reviews various methodological approaches to assessing natural fire danger (NFD) and the creation, updating, and application of NFD maps, which modern domestic researchers develop. Herein we introduce and analyse the currently accepted natural fire danger assessment scale by I. S. Melekhov. As pointed out by modern researchers, the methodological drawbacks of the scale are listed. The paper reviews the development of a new methodological approach to the compilation of regional scales for assessing the natural fire danger of forests, considering the links between forest growth conditions and seasonal and climatic conditions in the regions of the Russian Federation. The methodology of mapping natural fire danger based on vegetation fuel maps proposed by a scientific group of Sukachev Forest Institute of SВ RAS was studied. The works of the Mytishchi branch of Bauman Moscow State Technical University were reviewed to investigate the possibility of applying mathematical modelling methods for long-term forecasting of changes in NFD under different scenarios of forest management. The method of annual mapping of NFD classes, as proposed in CEPF RAS, is described. An example of NFD maps to assess the probability of forest fires in the Institute for Complex Analysis of the FEB RAS Regional Problems is considered. Future research areas are identified, namely, a cartographic representation of the created regional scales of NFD and the results of mathematical modelling of long-term changes in NFD.

Mapping Canopy Cover in African Dry Forests from the Combined Use of Sentinel-1 and Sentinel-2 Data: Application to Tanzania for the Year 2018

High-resolution Earth observation data is routinely used to monitor tropical forests. However, the seasonality and openness of the canopy of dry tropical forests remains a challenge for optical sensors. In this study, we demonstrate the potential of combining Sentinel-1 (S1) SAR and Sentinel-2 (S2) optical sensors in order to map the tree cover in East Africa. The overall methodology consists of: (i) the generation of S1 and S2 layers, (ii) the collection of an expert-based training/validation dataset and (iii) the classification of the satellite data. Three different classification workflows, together with different approaches to incorporating the spatial information to train the classifiers, are explored. Two types of maps were derived from these mapping approaches over Tanzania: (i) binary tree cover–no tree cover (TC/NTC) maps, and (ii) maps of the canopy cover classes. The overall accuracy of the maps is >95% for the TC/NTC maps and >85% for the forest types maps. Considering the neighboring pixels for training the classification improved the mapping of the areas that are covered by 1–10% tree cover. The study relied on open data and publicly available tools and can be integrated into national monitoring systems.

Assessment of Canopy Health with Drone-Based Orthoimagery in a Southern Appalachian Red Spruce Forest

Consumer-grade drone-produced digital orthoimagery is a valuable tool for conservation management and enables the low-cost monitoring of remote ecosystems. This study demonstrates the applicability of RGB orthoimagery for the assessment of forest health at the scale of individual trees in a 46-hectare plot of rare southern Appalachian red spruce forest on Whitetop Mountain, Virginia. We used photogrammetric Structure from Motion software Pix4Dmapper with drone-collected imagery to generate a mosaic for point cloud reconstruction and orthoimagery of the plot. Using 3-band RBG digital orthoimagery, we visually classified 9402 red spruce individuals, finding 8700 healthy (92.5%), 251 declining/dying (2.6%), and 451 dead (4.8%). We mapped individual spruce trees in each class and produced kernel density maps of health classes (live, dead, and dying). Our approach provided a nearly gap-free assessment of the red spruce canopy in our study site, versus a much more time-intensive field survey. Our maps provided useful information on stand mortality patterns and canopy gaps that could be used by managers to identify optimal locations for selective thinning to facilitate understory sapling regeneration. This approach, dependent mainly on an off-the-shelf drone system and visual interpretation of orthoimagery, could be applied by land managers to measure forest health in other spruce, or possibly spruce-fir, communities in the Appalachians. Our study highlights the usefulness of drone-produced orthoimagery for conservation monitoring, presenting a valid and accessible protocol for the monitoring and assessment of forest health in remote spruce, and possibly other conifer, populations. Adoption of drone-based monitoring may be especially useful in light of climate change and the possible displacement of southern Appalachian red spruce (and spruce-fir) ecosystems by the upslope migration of deciduous trees.

Can a Hierarchical Classification of Sentinel-2 Data Improve Land Cover Mapping?

Monitoring of land cover plays an important role in effective environmental management, assessment of natural resources, environmental protection, urban planning and sustainable development. Increasing demand for accurate and repeatable information on land cover and land cover changes causes rapid development of the advanced, machine learning algorithms dedicated to land cover mapping using satellite images. Free and open access to Sentinel-2 data, characterized with high spatial and temporal resolution, increased the potential to map and to monitor land surface with high accuracy and frequency. Despite a considerable number of approaches towards land cover classification based on satellite data, there is still a challenge to clearly separate complex land cover classes, for example grasslands, arable land and wetlands. The aim of this study is to examine, whether a hierarchal classification of Sentinel-2 data can improve the accuracy of land cover mapping and delineation of complex land cover classes. The study is conducted in the Lodz Province, in central Poland. The pixel-based land cover classification is carried out using the machine learning Random Forest (RF) algorithm, based on a time series of Sentinel-2 imagery acquired in 2020. The following nine land cover classes are mapped: sealed surfaces, woodland broadleaved, woodland coniferous, shrubs, permanent herbaceous (grassy cover), periodically herbaceous (i.e., arable land), mosses, non-vegetated (bare soil) and water bodies. The land cover classification is conducted following two approaches: (1) flat, where all land cover classes are classified together, and (2) hierarchical, where the stratification is applied to first separate the most stable land cover classes and then classifying the most problematic once. The national databases served as the source of the reference sampling plots for the classification process. The process of selection and verification of the reference sampling plots is performed automatically. To assess the stability of the classification models the classification processes are performed iteratively. The results of this study confirmed that the hierarchical approach gave more accurate results compared to the commonly used flat approach. The median of the overall accuracy (OA) of the hierarchical classification was higher by 3–9 percentage points compared to the flat one. Of interest, the OA of the hierarchical classification reached 0.93–0.99, whereas the flat approach reached 0.90. Individual classes are also better classified in the hierarchical approach.