Search

Spatial prediction is essential for obtaining the spatial distribution of geographic variables and selecting appropriate covariates for this process can be challenging, especially for non-expert users. For easing the burden of selecting the appropriate covariates, two case-based reasoning strategies, namely the most-similar-case and covariate-classification strategies, have been proposed for automated covariate selection. The former may suggest nonessential covariates due to its case-level reasoning way. And the latter with covariate-level reasoning may overlook related covariates and recommend fewer covariates than the case-level reasoning. In this study, we propose a new strategy of integrating case-level and covariate-level reasoning to effectively leverage the strengths of both previous strategies while also addressing their limitations. The proposed strategy is validated through a case study of automatically selecting covariates for digital soil mapping under reasoning with a case base containing 189 cases. The leave-one-out evaluation demonstrated that our proposed strategy outperformed the previous two strategies. [ABSTRACT FROM AUTHOR]

Climate change is already reducing carbon sequestration in Central European forests dramatically through extensive droughts and bark beetle outbreaks. Further warming may threaten the enormous carbon reservoirs in the boreal forests in northern Europe unless disturbance risks can be reduced by adaptive forest management. The European spruce bark beetle (Ips typographus) is a major natural disturbance agent in spruce‐dominated forests and can overwhelm the defences of healthy trees through pheromone‐coordinated mass‐attacks.We used an extensive dataset of bark beetle trap counts to quantify how climatic and management‐related factors influence bark beetle population sizes in boreal forests. Trap data were collected during a period without outbreaks and can thus identify mechanisms that drive populations towards outbreak thresholds.The most significant predictors of bark beetle population size were the volume of mature spruce, the extent of newly exposed clearcut edges, temperature and soil moisture. For clearcut edge, temperature and soil moisture, a 3‐year time lag produced the best model fit. We demonstrate how a model incorporating the most significant predictors, with a time lag, can be a useful management tool by allowing spatial prediction of future beetle population sizes.Synthesis and Applications: Some of the population drivers identified here, i,e., spruce volume and clearcut edges, can be targeted by adaptive management measures to reduce the risk of future bark beetle outbreaks. Implementing such measures may help preserve future carbon sequestration of European boreal forests. [ABSTRACT FROM AUTHOR]

Spatial prediction software is a valuable tool for predicting the spatial distribution of geographic variables and have a wide applicability with numerous non-expert users. However, as the input data to spatial prediction tasks gets more various in sources and larger in volume in the big data era, existing software are faced with challenges when processing such data on personal computers, which is one of the main platforms for running the software. This study uses the design and implementation of a similarity-based spatial prediction desktop software, the intelligent Soil Land Inference Model (iSoLIM), as an example to enable existing software to easily execute spatial prediction tasks with multi-source and large-volume data. The implemented similarity-based method was originally designed for digital soil mapping and has been applied to many other geographic variables. In the presented software, an expandable knowledge representation schema is adopted to intelligently manage multi-source knowledge data and enable knowledge integration; block division-based parallelization is adopted for intelligent computation arrangement on large-volume data. Experiments attest that, compared to existing software, iSoLIM can achieve higher accuracy from multi-source knowledge data by knowledge integration; and shows higher efficiency when processing data in large volume. [ABSTRACT FROM AUTHOR]

Soil particle size fractions (PSFs) are important properties for understanding the physical and chemical processes in soil systems. Knowledge about the distribution of soil PSFs is critical for sustainable soil management. Although log-ratio transformations have been widely applied to soil PSFs prediction, the statistical distribution of original data and the transformed data given by log-ratio transformations is different, resulting in biased estimates of soil PSFs. Therefore, multivariate random forest (MRF) was utilized for the simultaneous prediction of soil PSFs, as it is able to capture dependencies and internal relations among the three components. Specifically, 243 soil samples collected across the Loess Plateau were used. Meanwhile, Landsat data, terrain attributes, and climatic variables were employed as environmental variables for spatial prediction of soil PSFs. The results depicted that MRF gave satisfactory soil PSF prediction performance, where the R2 values were 0.62, 0.53, and 0.73 for sand, silt, and clay, respectively. Among the environmental variables, nighttime land surface temperature (LST_N) presented the highest importance in predicting soil PSFs in the Loess Plateau, China. Maps of soil PSFs and texture were generated at a 30 m resolution, which can be utilized as alternative data for soil erosion management and ecosystem conservation. [ABSTRACT FROM AUTHOR]

This study aims to understand the sources and pollution of soil heavy metals. Taking Taicang City, Jiangsu Province as an example, the soil surface samples were collected to conduct the heavy metal content analysis. A novel receptor (UNMIX) model was selected to analyze the source and contribution of soil heavy metals. Geoaccumulation index and multivariate geostatistical techniques were used to quantitatively evaluate for the spatial prediction of heavy metal pollution. The results showed that: (1) The contents of As, Cd, Cr, Cu, Hg, Ni, and Zn were relatively high, exceeding the background values of soil heavy metals in the study area. The enrichment degree of Cd, Cu, Hg, Pb, and Zn was relatively high, while there were the normal enrichment coefficients of As, Cr, and Ni similar to the natural basic level. Except for Cr, the average content of the rest heavy metal elements exceeded the background values of soil heavy metals. The coefficients of variation for Cu and Zn were 0.60 and 0.64, respectively, indicating a high level of variation and a high skewness state. (2) The distribution trend of As and Pb gradually increased from west to east, with the highest content in Chengxiang and Ludu towns. The soil was mainly developed from basalt and metamorphic rocks, containing a certain amount of Ni element. There was the similar pattern of Cd, Cr, Cu, and Zn spatial distribution, where the high value areas were located in Shuangfeng Town and Chengxiang Town in the southwest. Overall, the high and low value areas of the eight heavy metal elements were distributed alternately with the islandlike spatial feature, indicating the negative effects of human activities on the soil environment. (3) Soli heavy metals were comprehensively affected by transportation-industrial, natural-agricultural, industrial-natural, and agricultural-industrial sources, with the contribution rates of 35.18%, 27.32%, 20.26%, and 17.24%, respectively. Among them, Hg, Pb, Ni, Cu, and Zn were mainly from transportation-industrial sources, Cr, Zn, and Cu were from natural-agricultural sources, As and Ni were from industrial-natural sources, and Cd was from agricultural-industrial sources. (4) The ground accumulation index of Hg was ranked first among all elements with a mean of 1.12. A high degree of artificial accumulation of Hg was greatly influenced by non-natural factors. The elements with a ground accumulation index greater than 0 were Ni and Cd, with values of 0.68 and 0.42, respectively. The pollution levels of As, Cu, and Zn were in the middle. The soil condition was less affected by Cr and Pb. (5) The pollution areas of Ni, Cu, and Hg were the largest, reaching 107.42, 75.56, and 55.02 km² respectively. The area of Cd pollution was 342 km² (accounting for 8.6% of the total area) that distributed in industrial, tailings surrounding, and urban suburbs. The potential pollution space was distributed in patches, which should be given more attention. At the same time, it was also necessary to strictly prevent the expansion of Cr and Pb pollution space. This soil investigation can provide the scientific basis for the soil environmental management and heavy metal pollution remediation [ABSTRACT FROM AUTHOR]

Global climate change is a serious threat to food and energy security. Crop growth modelling is an important tool for simulating crop food production and assisting in decision making. Planting date is one of the important model parameters. Larger-scale spatial distribution with high accuracy for planting dates is essential for the widespread application of crop growth models. In this study, a planting date prediction method based on environmental similarity was developed in accordance with the third law of geography. Spring maize planting date observations from 124 agricultural meteorological experiment stations in China over the years 1992–2010 were used as the data source. Samples spanning from 1992 to 2009 were allocated as training data, while samples from 2010 constituted the independent validation set. The results indicated that the root mean square error (RMSE) for spring maize planting date based on environmental similarity was 10 days, which is better than that of multiple regression analysis (RMSE = 13 days) in 2010. Additionally, when applied at varying scales, the accuracy of national-scale prediction was better than that of regional-scale prediction in areas with large differences in planting dates. Consequently, the method based on environmental similarity can effectively and accurately estimate planting date parameters at multiple scales and provide reasonable parameter support for large-scale crop growth modelling. [ABSTRACT FROM AUTHOR]

Predicting landslide hazards benefits geological disaster prevention and control. A novel cellular automaton (CA) integrating spatial case-based reasoning (SCBR), namely SCBR-CA, is proposed in this paper to predict landslide hazards at a local scale. The proposed model not only extracts spatial scene features for computations but also achieves dynamic prediction, which means that only one input is needed to obtain continuous predictions. Experiments were performed in Lushan, Sichuan, China. After using a convolutional neural network (CNN) to obtain the initial static landslide hazard zoning results, the landslide hazard zoning results for 2016–2025 were predicted with the SCBR-CA model. For comparison, a CA combined with a CNN (CNN-CA), was introduced. The area under the curve (AUC) of the receiver operating characteristic curve and Moran's I index were used to assess the performance of the model. The experimental results showed that SCBR-CA yields slightly better AUC and Moran's I index values than CNN-CA, and the dynamically predicted landslide hazard zoning results are equivalent or superior to those of static zoning, which indicates that the SCBR-CA model effectively predict local landslide hazards. [ABSTRACT FROM AUTHOR]

The spatial distribution of rockhead profile has a significant impact on the design and construction of geotechnical engineering structures. Limited by the economic and technical conditions, the borehole data of specific sites are often sparse, which brings great challenges to the accurate prediction of rockhead profile. In this study, a Gaussian process regression (GPR) method is used to predict the rockhead profile. Borehole data from a construction site in Hong Kong are used to evaluate the ability of the GPR method to predict rockhead profile. Besides, the influences of the amount of borehole data on the accuracy of the GPR model and the spatial prediction results of rockhead are investigated. The results indicate that the GPR model based on the square exponential and the Matérn covariance function can obtain more accurate prediction results, and the GPR model with limited borehole data can provide a reasonable prediction interval for rockhead depth. With the increase of borehole data, the generalization ability and prediction accuracy of the GPR model gradually improves. In engineering practice, the prediction accuracy and uncertainty degree of the GPR model can be used to judge whether it is necessary to continue to increase borehole data. [ABSTRACT FROM AUTHOR]

Spatial prediction methods are an important means of predicting the spatial variation of groundwater level change. Existing methods extract spatial or statistical relationships from samples to represent the study area for inference and require a representative sample set that is usually in large quantity and is distributed across geographic or covariate space. However, samples for groundwater are usually sparsely and unevenly distributed. In this paper, an approach based on the Third Law of Geography is proposed to make predictions by comparing the similarity between each individual sample and unmeasured site. The approach requires no specific number or distribution of samples and provides individual uncertainty measures at each location. Experiments in three different watersheds across the U.S. show that the proposed methods outperform machine learning methods when available samples do not well represent the area. The provided uncertainty measures are indicative of prediction accuracy by location. The results of this study also show that the spatial prediction based on the Third Law of Geography can also be successfully applied to dynamic variables such as groundwater level change. [ABSTRACT FROM AUTHOR]

Spatial prediction of soil ammonia (NH3) plays an important role in monitoring climate warming and soil ecological health. However, traditional machine learning (ML) models do not consider optimal parameter selection and spatial autocorrelation. Here, we present an integration method (tree-structured Parzen estimator–machine learning–ordinary kriging (TPE–ML–OK)) to predict spatial variability of soil NH3 from Sentinel-2 remote sensing image and air quality data. In TPE–ML–OK, we designed the TPE search algorithm, which encourages gradient boosting decision tree (GBDT), random forest (RF), and extreme gradient boosting (XGB) models to pay more attention to the optimal hyperparameters' high-possibility range, and then the residual ordinary kriging model is used to further improve the prediction accuracy of soil NH3 flux. We found a weak linear correlation between soil NH3 flux and environmental variables using scatter matrix correlation analysis. The optimal hyperparameters from the TPE search algorithm existed in the densest iteration region, and the TPE–XGB–OK method exhibited the highest predicted accuracy (R2 = 85.97%) for soil NH3 flux in comparison with other models. The spatial mapping results based on TPE–ML–OK methods showed that the high fluxes of soil NH3 were concentrated in the central and northeast areas, which may be influenced by rivers or soil water. The analysis result of the SHapley additive explanation (SHAP) algorithm found that the variables with the highest contribution to soil NH3 were O3, SO2, PM10, CO, and NDWI. The above results demonstrate the powerful linear–nonlinear interpretation ability between soil NH3 and environmental variables using the integration method, which can reduce the impact on agricultural nitrogen deposition and regional air quality. [ABSTRACT FROM AUTHOR]

Urban area data are strategically important for public safety, urban management, and planning. Previous research has attempted to estimate the values of unsampled regular areas, while minimal attention has been paid to the values of irregular areas. To address this problem, this study proposes a hierarchical geospatial graph neural network model based on the spatial hierarchical constraints of areas. The model first characterizes spatial relationships between irregular areas at different spatial scales. Then, it aggregates information from neighboring areas with graph neural networks, and finally, it imputes missing values in fine-grained areas under hierarchical relationship constraints. To investigate the performance of the proposed model, we constructed a new dataset consisting of the urban statistical values of irregular areas in New York City. Experiments on the dataset show that the proposed model outperforms state-of-the-art baselines and exhibits robustness. The model is adaptable to numerous geographic applications, including traffic management, public safety, and public resource allocation. [ABSTRACT FROM AUTHOR]

Symbolic data analysis deals with complex data with symbolic objects, such as lists, histograms, and intervals. Spatial analysis for symbolic data is relatively underexplored. To fill the gap, this paper proposes a statistical framework for spatial interval-valued data (SIVD) analysis. We provide geostatistical methods for spatial prediction, predictive performance measure for prediction assessment, and visualization for mapping SIVD. The proposed methods are illustrated with both simulated and real examples. [ABSTRACT FROM AUTHOR]

This article presents an overview of methodologies for spatial prediction of functional data, focusing on both stationary and non-stationary conditions. A significant aspect of the functional random fields analysis is evaluating stationarity to characterize the stability of statistical properties across the spatial domain. The article explores methodologies from the literature, providing insights into the challenges and advancements in functional geostatistics. This work is relevant from theoretical and practical perspectives, offering an integrated view of methodologies tailored to the specific stationarity conditions of the functional processes under study. The practical implications of our work span across fields like environmental monitoring, geosciences, and biomedical research. This overview encourages advancements in functional geostatistics, paving the way for the development of innovative techniques for analyzing and predicting spatially correlated functional data. It lays the groundwork for future research, enhancing our understanding of spatial statistics and its applications. [ABSTRACT FROM AUTHOR]

Gully erosion is a worldwide threat with numerous environmental, social, and economic impacts. The purpose of this research is to evaluate the performance and robustness of six machine learning ensemble models based on the decision tree principle: Random Forest (RF), C5.0, XGBoost, treebag, Gradient Boosting Machines (GBMs) and Adaboost, in order to map and predict gully erosion-prone areas in a semi-arid mountain context. The first step was to prepare the inventory data, which consisted of 217 gully points. This database was then randomly subdivided into five percentages of Train/Test (50/50, 60/40, 70/30, 80/20, and 90/10) to assess the stability and robustness of the models. Furthermore, 17 geo-environmental variables were used as potential controlling factors, and several metrics were examined to evaluate the performance of the six models. The results revealed that all of the models used performed well in terms of predicting vulnerability to gully erosion. The C5.0 and RF models had the best prediction performance (AUC = 90.8 and AUC = 90.1, respectively). However, according to the random subdivisions of the database, these models exhibit small but noticeable instability, with high performance for the 80/20% and 70/30% subdivisions. This demonstrates the significance of database refining and the need to test various splitting data in order to ensure efficient and reliable output results. [ABSTRACT FROM AUTHOR]

The National Forestry Commission of Mexico continuously monitors forest structure within the country's continental territory by the implementation of the National Forest and Soils Inventory (INFyS). Due to the challenges involved in collecting data exclusively from field surveys, there are spatial information gaps for important forest attributes. This can produce bias or increase uncertainty when generating estimates required to support forest management decisions. Our objective is to predict the spatial distribution of tree height and tree density in all Mexican forests. We performed wall‐to‐wall spatial predictions of both attributes in 1‐km grids, using ensemble machine learning across each forest type in Mexico. Predictor variables include remote sensing imagery and other geospatial data (e.g., mean precipitation, surface temperature, canopy cover). Training data is from the 2009 to 2014 cycle (n > 26,000 sampling plots). Spatial cross validation suggested that the model had a better performance when predicting tree height r2 =.35 [.12,.51] (mean [min, max]) than for tree density r2 =.23 [.05,.42]. The best predictive performance when mapping tree height was for broadleaf and coniferous‐broadleaf forests (model explained ~50% of variance). The best predictive performance when mapping tree density was for tropical forest (model explained ~40% of variance). Although most forests had relatively low uncertainty for tree height predictions, e.g., values <60%, arid and semiarid ecosystems had high uncertainty, e.g., values >80%. Uncertainty values for tree density predictions were >80% in most forests. The applied open science approach we present is easily replicable and scalable, thus it is helpful to assist in the decision‐making and future of the National Forest and Soils Inventory. This work highlights the need for analytical tools that help us exploit the full potential of the Mexican forest inventory datasets. [ABSTRACT FROM AUTHOR]

Spatial point process models are theoretically useful for mapping discrete events, such as plant or animal presence, across space; however, the computational complexity of fitting these models is often a barrier to their practical use. The log-Gaussian Cox process (LGCP) is a point process driven by a latent Gaussian field, and recent advances have made it possible to fit Bayesian LGCP models using approximate methods that facilitate rapid computation. These advances include the integrated nested Laplace approximation (INLA) with a stochastic partial differential equations (SPDE) approach to sparsely approximate the Gaussian field and an extension using pseudodata with a Poisson response. To help link the theoretical results to statistical practice, we provide an overview of INLA for point process data and then illustrate their implementation using freely available data. The analyzed datasets include both a completely observed spatial field and an incomplete data situation. Our well-commented R code is shared in the online supplement. Our intent is to make these methods accessible to the practitioner of spatial statistics without requiring deep knowledge of point process theory. [ABSTRACT FROM AUTHOR]

The accurate mapping of soil organic carbon (SOC) distribution is important for carbon sequestration and land management strategies, contributing to mitigating climate change and ensuring agricultural productivity. The Heihe River Basin in China is an important region that has immense potential for SOC storage. Phenological variables are effective indicators of vegetation growth, and hence are closely related to SOC. However, few studies have incorporated phenological variables in SOC prediction, especially in alpine areas such as the Heihe River Basin. This study used random forest (RF) and extreme gradient boosting (XGBoost) to study the effects of phenological variables (e.g., Greenup, Dormancy, etc.) obtained from MODIS (i.e., Moderate Resolution Imaging Spectroradiometer) product (MCD12Q2) on SOC content prediction in the middle and upper reaches of Heihe River Basin. The current study also identified the dominating variables in SOC prediction and compared model performance using a cross validation procedure. The results indicate that: (1) when phenological variables were considered, the R 2 (coefficient of determination) of RF and XGBoost were 0.68 and 0.56, respectively, and RF consistently outperforms XGBoost in various cross validation experiments; (2) the environmental variables MAT, MAP, DEM and NDVI play the most important roles in SOC prediction; (3) the phenological variables can account for 32–39% of the spatial variability of SOC in both the RF and XGBoost models, and hence were the most important factor among the five categories of predictive variables. This study proved that the introduction of phenological variables can significantly improve the performance of SOC prediction. They should be used as indispensable variables for accurately modeling SOC in related studies. [ABSTRACT FROM AUTHOR]

Understanding geographical characteristics of distribution patterns and spatial association is essential for spatial statistical inference such as factor exploration and spatial prediction. The geographical similarity principle was recently developed to explain the association between geographical variables. It describes the comprehensive degree of approximation of a geographical structure instead of explicit relationships between variables. However, there are still challenges for geographical similarity-based methods. For instance, all samples are used for prediction, leading to increased calculation burden and reduced prediction accuracy due to the noise and unrelated data in large spatial data sets. This study develops a geographically optimal similarity (GOS) model for accurate and reliable spatial prediction based on the geographical similarity principle. In GOS, the geographical configurations are first characterized, and similarities between unknown and known observation locations are assessed. Next, an optimal threshold is determined to select a small number of observations with optimal similarities for the prediction at each unknown location. Finally, a reliable uncertainty assessment approach is developed to assess and map uncertainties of GOS predictions. A new R package "geosimilarity" is developed to conduct GOS models. In this study, GOS is implemented in predicting spatial distributions of trace elements in a mining region in Australia. Results show that GOS can use a small number of observations to derive more accurate and reliable spatial predictions than linear regression and basic configuration similarity models. In addition, pattern characteristics of predictions can be improved by GOS by eliminating the phenomenon wherein predictions are clustered near mean values and contain striped textures. Therefore, GOS demonstrates greater potential for implementing the geographical similarity principle in spatial predictions by bringing information from samples with relatively high similarities at any location across space for more accurate and effective predictions in broader fields and practice. [ABSTRACT FROM AUTHOR]

Aim: Global‐scale maps of the environment are an important source of information for researchers and decision makers. Often, these maps are created by training machine learning algorithms on field‐sampled reference data using remote sensing information as predictors. Since field samples are often sparse and clustered in geographic space, model prediction requires a transfer of the trained model to regions where no reference data are available. However, recent studies question the feasibility of predictions far beyond the location of training data. Innovation: We propose a novel workflow for spatial predictive mapping that leverages recent developments in this field and combines them in innovative ways with the aim of improved model transferability and performance assessment. We demonstrate, evaluate and discuss the workflow with data from recently published global environmental maps. Main conclusions: Reducing predictors to those relevant for spatial prediction leads to an increase of model transferability and map accuracy without a decrease of prediction quality in areas with high sampling density. Still, reliable gap‐free global predictions were not possible, highlighting that global maps and their evaluation are hampered by limited availability of reference data. [ABSTRACT FROM AUTHOR]

Spatial prediction(SP) based on machine learning(ML) has been applied to soil water quality, air quality, marine environment, etc. However, there are still deficiencies in dealing with the problem of small samples. Normally, ML requires large amounts of training samples to prevent underfitting. And the data augmentation(DA) methods of mixup and synthetic minority over-sampling technique(SMOTE) ignore the similarity of geographic information. Therefore, this paper proposes a modified upsampling method and combines it with the random forest spatial interpolation(RFSI) to deal with the small sample problem in geographical space. The modified upsampling is mainly reflected in the following two aspects. Firstly, in the process of selecting the nearest points, it is to select points with similar geographic information in some aspects of the category after classification. Secondly, the selected difference is the difference of each category. In order to verify the effectiveness of the proposed method, we use daily precipitation data for January 2018 in Chongqing. The experimental results show that the combination of the modified upsampling method and RFSI effectively improves the accuracy of SP. [ABSTRACT FROM AUTHOR]

The spatial random effects model is popular in analyzing spatially referenced data sets. The model includes spatially observed covariates and unobserved spatial random effects. If spatial confounding between covariates and random effects is ignored, parameter estimation and spatial prediction would be inaccurate. In this research, we focus on discussing the estimation of regression coefficients and the selection of covariates for spatial regression when existing unmeasured confounders. First, we introduce an adjusted estimation method for regression coefficients and the consequent spatial predictor in the presence of spatial confounding. From a prediction point of view, we then propose a generalized conditional Akaike information criterion to select a subset of covariates, resulting in satisfactory variable selection and spatial prediction. Statistical inferences of the proposed methodology are justified theoretically and numerically. [ABSTRACT FROM AUTHOR]

Sampling design can significantly reduce the uncertainty in geospatial predictions. In this paper, we developed an adaptive uncertainty-guided stepwise sampling (AUGSS) method to select sampling locations to supplement existing legacy sample points whose representation should be improved. The proposed method selects supplemental samples in a stepwise manner as guided by an objective function with two weighted sub-objectives. One reduces the area with high prediction uncertainty, and the other minimizes the overall prediction uncertainty for the entire area. The method takes an adaptive approach to adjust weights for the two sub-objectives and to tune an uncertainty threshold controlling whether a location can be reliably predicted during the sampling procedure. A case study on soil property prediction shows that AUGSS outperforms the stratified random sampling (SRS) and the non-adaptive uncertainty guided sampling method (UGSS) in terms of RMSE and Lin's concordance correlation coefficient with different sample sizes. This study shows that the AUGSS method offers a potential for effectively adding supplemental samples to existing samples which are insufficient for spatial prediction. The adaptive strategy guided by predicted uncertainty provides an efficient support to improve the spatial pattern of samples, which plays a key role in the result accuracy of geospatial predictive mapping. [ABSTRACT FROM AUTHOR]

The understory is an essential ecological and structural component of forest ecosystems. The lack of efficient, accurate, and objective methods for evaluating and quantifying the spatial spread of understory characteristics over large areas is a challenge for forest planning and management, with specific regard to biodiversity and habitat governance. In this study, we used terrestrial and airborne laser scanning (TLS and ALS) data to characterize understory in a European beech and black pine forest in Italy. First, we linked understory structural features derived from traditional field measurements with TLS metrics, then, we related such metrics to the ones derived from ALS. Results indicate that (i) the upper understory density (5–10 m above ground) is significantly associated with two ALS metrics, specifically the mean height of points belonging to the lower third of the ALS point cloud within the voxel (HM1/3) and the corresponding standard deviation (SD1/3), while (ii) for the lower understory layer (2–5 m above ground), the most related metric is HM1/3 alone. As an example application, we have produced a map of forest understory for each layer, extending over the entire study region covered by ALS data, based on the developed spatial prediction models. With this study, we also demonstrated the power of hand-held mobile-TLS as a fast and high-resolution tool for measuring forest structural attributes and obtaining relevant ecological data. [ABSTRACT FROM AUTHOR]

Evaluation of spatial variability and mapping of soil properties is critical for sustainable agricultural production in arid lands. The main objectives of the present study were to spatialize soil organic carbon (SOC), soil particle size distribution(clay, sand, and silt contents), and calcium carbonate equivalent (CCE) by integrating multisource environmental covariates, including digital elevation model (DEM) and remote sensing data by machine learning (Cubist, Cu and random forest, RF) in an arid region of Iran. Additionally, Synthetic Soil Images (SySI) were achieved from multi-temporal images of bare soil pixels based on Landsats 4, 5, 7, 8, and a DEM. Three hundred topsoil samples (0-30cm depth) were collected based on the conditioned Latin hypercube sampling (cLHS) approach in Afzar district, Fars province, southern Iran. The models were calibrated and validated by the 10-fold cross-validation approach, and the performance was evaluated using root mean square error (RMSE), the ratio of the performance to interquartile distance (RPIQ), and coefficient of determination (R²). Also, the prediction accuracy was assessed by the relative RMSE (rRMSE). The performance of the best models based on the RPIQ index showed that the model for predicting clay (1.89) had a good prediction, for sand (1.64), SOC (1.55), and CCE (1.59) had a fair prediction, while the model for silt (1.13) performed poorly. We found that the Cu and RF models had the highest and lowest prediction accuracies for CCE (rRMSE = 14.31%) and SOC (rRMSE = 43.93%), respectively. We discovered that a combination of high-quality RS data (SySI) and variables derived from DEM were reasonably able to predict soil properties. We revealed a strong promise to enhance the accuracy of digital soil mapping, especially in regions with limited soil data. Moreover, the application of RS data can reduce the soil sampling cost and, accordingly, soil mapping. [ABSTRACT FROM AUTHOR]

Conventional methods of spatial prediction, such as Kriging, require assumptions such as stationarity and isotropy, which are not easy to evaluate, and often do not hold for spatial data. For these methods, the spatial dependency structure between data should be accurately modeled, which requires expert knowledge in spatial statistics. On the other hand, spatial prediction using artificial neural network (ANN) has attracted considerable interest due to ANN's ability in learning from data without the need for complex and specialized assumptions. However, ANN models require suitable input variables for better and efficient spatial prediction. This paper aims to improve the accuracy of ANNs spatial prediction using neighboring information. Given the general principle that "closer spatial data are more dependent", we tried to somehow enter data dependency into the network by using the neighboring observations. We proposed a hybrid model of ANN and inverse distance weighting, based on nearby observations. We also proposed an ANN-based model for spatial prediction based on weighted values of nearby observations. The accuracy of the models was compared through a simulation study. The results showed that using neighboring information to train ANN, dramatically increases the prediction accuracy. [ABSTRACT FROM AUTHOR]

Selection of store sites is a common but challenging task in business practices. Picking the most desirable location for a future store is crucial for attracting customers and becoming profitable. The classic multi-criteria decision-making framework for store site selection oversimplifies the local characteristics that are both high dimensional and unstructured. Recent advances in deep learning enable more powerful data-driven approaches for site selection, many of which, however, overlook the interaction between different locations on the map. To better incorporate the spatial interaction patterns in understanding neighborhood characteristics and their impact on store placement, we propose to learn a graph convolutional network (GCN) for highly effective site selection tasks. Furthermore, we present a novel dataset that encompasses land use information as well as public transport networks in Singapore as a case study to benchmark site selection algorithms. It allows us to construct a geospatial GCN based on the public transport system to predict the attractiveness of different store sites within neighborhoods. We show that the proposed GCN model outperforms the competing methods that are learning from local geographical characteristics only. The proposed case study corroborates the geospatial interactions and offers new insights for solving various geographic and transport problems using graph neural networks. [ABSTRACT FROM AUTHOR]

There is limited data regarding the nesting ecology of boreal ducks and their response to industrial development, despite this region being an important North American breeding area. We investigated how landcover and oil and gas development affect third‐order nest‐site selection of boreal ducks. We located duck nests in Alberta's western boreal forest between 2016 and 2018. We used multiscale analysis to identify how scale affects the selection of a resource using generalized linear mixed‐effects models and determined what scale‐optimized combination of landscape features were most important in describing where ducks nest. We located 136 nests of six species of upland nesting ducks between 2016 and 2018. The magnitude, direction, and best spatial scale varied by resource. For landcover, ducks selected nest‐sites associated with mineral wetlands (300 m) and open water (300 m). Ducks avoided greater densities of seismic lines (300 m) and pipelines (2500 m) but selected nest‐sites associated with borrow pits (300 m) and roads (1000 m). We used our models to predict important duck nesting habitat in the boreal forest, which can support conservation and management decisions. We recommend conservation actions target the conservation of mineral wetlands and associated habitats within this working landscape. Further research is necessary to understand the adaptive consequences of nest‐site selection and how industrial development influences important nest predators. [ABSTRACT FROM AUTHOR]

Groundwater-based irrigation has dramatically expanded over the past decades. It has important implications for terrestrial water, energy fluxes, and food production, as well as local to regional climates. However, irrigation water use is hard to monitor at large scales due to various constraints, including the high cost of metering equipment installation and maintenance, privacy issues, and the presence of illegal or unregistered wells. This study estimates irrigation water amounts using machine learning to integrate in situ pumping records, remote sensing products, and climate data in the Kansas High Plains. We use a random forest regression to estimate the annual irrigation water amount at a reprojected spatial resolution of 6 km based on various data, including remotely sensed vegetation indices and evapotranspiration (ET), land cover, near-surface meteorological forcing, and a satellite-derived irrigation map. In addition, we assess the value of ECOSTRESS ET products for irrigation water use estimation and compare with the baseline results by using MODIS ET. The random forest regression model can capture the temporal and spatial variability of irrigation amounts with a satisfactory accuracy ( R 2 = 0.82). It performs reasonably well when it is calibrated on the western portion of the study area and tested on the eastern portion that receives more rain than the western one, suggesting its potential transferability to other regions. ECSOTRESS ET and MODIS ET yield a similar irrigation estimation accuracy. [ABSTRACT FROM AUTHOR]

This study proposes a method to predict buildings' energy efficiency based on available descriptive information and without a physical visit, by merging diverse datasets and employing advanced classification techniques. By integrating geographical, structural, legal, and socio-economic data with Energy Performance Certificate (EPC) observations, our approach yields a rich learning set. Through variable selection methods like forward selection with KNN and simultaneous perturbation stochastic approximation for fuzzy KNN, we refine model variables. Comparing fuzzy and hard classification using KNN, Kriging or Random Forest approaches, we find fuzzy classification more adept at capturing nuanced energy inefficiency indicators. Our study highlights the importance of mass energy efficiency prediction for sustainable renovation efforts. [ABSTRACT FROM AUTHOR]

Forest fires threaten global ecosystems, socio-economic structures, and public safety. Accurately assessing forest fire susceptibility is critical for effective environmental management. Supervised learning methods dominate this assessment, relying on a substantial dataset of forest fire occurrences for model training. However, obtaining precise forest fire location data remains challenging. To address this issue, semi-supervised learning emerges as a viable solution, leveraging both a limited set of collected samples and unlabeled data containing environmental factors for training. Our study employed the transductive support vector machine (TSVM), a key semi-supervised learning method, to assess forest fire susceptibility in scenarios with limited samples. We conducted a comparative analysis, evaluating its performance against widely used supervised learning methods. The assessment area for forest fire susceptibility lies in Dayu County, Jiangxi Province, China, renowned for its vast forest cover and frequent fire incidents. We analyzed and generated maps depicting forest fire susceptibility, evaluating prediction accuracies for both supervised and semi-supervised learning methods across various small sample scenarios (e.g., 4, 8, 12, 16, 20, 24, 28, and 32 samples). Our findings indicate that TSVM exhibits superior prediction accuracy compared to supervised learning with limited samples, yielding more plausible forest fire susceptibility maps. For instance, at sample sizes of 4, 16, and 28, TSVM achieves prediction accuracies of approximately 0.8037, 0.9257, and 0.9583, respectively. In contrast, random forests, the top performers in supervised learning, demonstrate accuracies of approximately 0.7424, 0.8916, and 0.9431, respectively, for the same small sample sizes. Additionally, we discussed three key aspects: TSVM parameter configuration, the impact of unlabeled sample size, and performance within typical sample sizes. Our findings support semi-supervised learning as a promising approach compared to supervised learning for forest fire susceptibility assessment and mapping, particularly in scenarios with small sample sizes. • We used the transductive support vector machine (TSVM) for forest fire susceptibility assessment with limited samples. • The prediction accuracy of TSVM outperforms common supervised learning methods in small sample scenarios. • TSVM as a semi-supervised learning exploits unlabeled data to improve prediction accuracy. [ABSTRACT FROM AUTHOR]

Dust pollution poses significant risks to human health, air quality, and food safety, necessitating the identification of dust occurrence and the development of dust susceptibility maps (DSMs) to mitigate its effects. This research aims to detect dust occurrence using satellite images and prepare a DSM for Bushehr province, Iran, by enhancing the attentive interpretable tabular learning (TabNet) model through three swarm-based metaheuristic algorithms: particle swarm optimization (PSO), grey wolf optimizer (GWO), and hunger games search (HGS). A spatial database incorporating dust occurrence areas was created using Moderate Resolution Imaging Spectroradiometer (MODIS) images from 2002 to 2022, including 15 influential criteria related to climate, soil, topography, and land cover. Four models were employed for modeling and DSM generation: TabNet, TabNet-PSO, TabNet-GWO, and TabNet-HGS. Evaluation of the modeling results using performance metrics indicated that the TabNet-HGS model outperformed the other models in both training (mean absolute error (MAE) = 0.055, root-mean-square error (RMSE) = 0.1, coefficient of determination (R2) = 0.959), and testing (MAE = 0.063, RMSE = 0.114, R2 = 0.947) data. Following TabNet-HGS, the TabNet-PSO, TabNet-GWO, and TabNet models demonstrated progressively lower accuracy. The validation of the DSM was performed by assessing receiver operating characteristic (ROC) curves, revealing that the TabNet-HGS, TabNet-PSO, TabNet-GWO, and TabNet models exhibited the highest modeling accuracy, with corresponding area under the curve (AUC) values of 0.994, 0.986, 0.98, and 0.832, respectively. These results highlight the enhanced accuracy of dust susceptibility modeling achieved by integrating swarm-based metaheuristic algorithms with the TabNet model. The dust susceptibility map provides valuable insights into the sources, pathways, and impacts of dust particles on the environment and human health in the study area. [Display omitted] • Satellite imagery and comprehensive data yield precise dust susceptibility maps. • Swarm-based metaheuristics enhance TabNet accuracy for accurate dust modeling. • TabNet-HGS model outperforms others, improving dust susceptibility mapping. [ABSTRACT FROM AUTHOR]

Classical kriging models use Euclidean distance when modeling spatial autocorrelation. However for regions with irregular boundaries and holes, such as estuaries and coastlines, a measure of within-domain distance may capture a system's proximity dependencies more accurately. Standard kriging techniques are not guaranteed to yield a valid covariance structure when defined in terms of non-Euclidean distances. In this paper, we develop a new kriging model for irregularly shaped domains. Our model uses an approximation to a diffusion process to define a valid covariance structure that reflects the domain topology. A covariance matrix is defined through the use of random walks on a lattice, process convolutions, and the kriging equations. A simulation study demonstrates that for commonly encountered topologies, our diffusion kriging estimator is superior to a kriging estimator based on shortest within-domain distance. We also illustrate our method using water quality data from Puget Sound and Lake Peipsi to map chlorophyll concentration. [ABSTRACT FROM AUTHOR]

Spatial prediction tasks are challenging when observed samples are sparse and prediction samples are abundant. Gaussian processes (GPs) are commonly used in spatial prediction tasks and have the advantage of measuring the uncertainty of the interpolation result. However, as the sample size increases, GPs suffer from significant overhead. Standard neural networks (NNs) provide a powerful and scalable solution for modeling spatial data, but they often overfit small sample data. Based on conditional neural processes (CNPs), which combine the advantages of GPs and NNs, we propose a new framework called Spatial Multi-Attention Conditional Neural Processes (SMACNPs) for spatial small sample prediction tasks. SMACNPs are a modular model that can predict targets by employing different attention mechanisms to extract relevant information from different forms of sample data. The task representation is inferred by measuring the spatial correlation contained in different sample points and the relationship contained in attribute variables, respectively. The distribution of the target variable is predicted by GPs parameterized by NNs. SMACNPs allow us to obtain accurate predictions of the target value while quantifying the prediction uncertainty. Experiments on spatial prediction tasks on simulated and real-world datasets demonstrate that this framework flexibly incorporates spatial context and correlation into the model, achieving state-of-the-art results in spatial small sample prediction tasks in terms of both predictive performance and reliability. For example, on the California housing dataset, our method reduces MAE by 8% and MSE by 7% compared to the second-best method. In addition, a spatiotemporal prediction task to forecast traffic speed further confirms the effectiveness and generality of our method. [ABSTRACT FROM AUTHOR]

• A hypergraph-based hybrid graph convolutional network for intra-city human activity prediction. • Pairwise and high-order spatial relationships are modeled simultaneously via adaptive learning. • Geospatial knowledge can be interpreted from the perspective of geographic principles. • Empirical experiments on a large-scale mobile positioning data. Human activity intensity prediction, i.e., estimating the dynamic population distribution, is crucial to many location-based applications, particularly intelligent transportation systems and urban planning. Geographical artificial intelligence (GeoAI) is an emerging field that has contributed many new modeling approaches for spatiotemporal phenomena, especially spatiotemporal graph neural networks (ST-GNNs). However, there are two challenges to the prediction of human activity intensity. On the one hand, accurate predictions require modeling not only pairwise spatial relationships, e.g., distance and spatial flow but also complex high-order relationships, such as clusters and groups. On the other hand, it is not appropriate to incorporate prior knowledge on the spatial structure of data with subjective assumptions using graphs with predefined topologies or weights. To mitigate such challenges, a novel deep learning-based model was proposed, namely, the Hy pergraph-based Hy brid G raph C onvolutional N etwork (HyGCN). In our study, high-order relationships are modeled by hypergraph convolutions, and pairwise relationships are captured in graph convolutions. A spatial fusion layer is added to fuse the pairwise and high-order relationships. Then, adaptive graph and hypergraph generation methods are proposed to interpret edge weights and understand the learned geographic relationships in accordance with principles. After systematic experiments on large-scale mobility data, our results demonstrate that HyGCN outperforms state-of-the-art models. The learned geospatial knowledge can be disentangled and explained from the perspectives of geographic principles. This study offers new insights for not only revisiting geographic laws in GeoAI but also utilizing spatiotemporal knowledge in deep learning models. [ABSTRACT FROM AUTHOR]

Mapping of landslides, conducted in 2021 by the Geological Survey of Denmark and Greenland (GEUS), revealed 3202 landslides in Denmark, indicating that they might pose a bigger problem than previously acknowledged. Moreover, the changing climate is assumed to have an impact on landslide occurrences in the future. The aim of this study is to conduct the first landslide susceptibility mapping (LSM) in Denmark, reducing the geographical bias existing in LSM studies, and to identify areas prone to landslides in the future following representative concentration pathway RCP8.5, based on a set of explanatory variables in an area of interest located around Vejle Fjord, Jutland, Denmark. A subset from the landslide inventory provided by GEUS is used as ground truth data. Three well-established machine learning (ML) algorithms—Random Forest, Support Vector Machine, and Logistic Regression—were trained to classify the data samples as landslide or non-landslide, treating the ML task as a binary classification and expressing the results in the form of a probability in order to produce susceptibility maps. The classification results were validated through the test data and through an external data set for an area located outside of the region of interest. While the high predictive performance varied slightly among the three models on the test data, the LR and SVM demonstrated inferior accuracy outside of the study area. The results show that the RF model has robustness and potential for applicability in landslide susceptibility mapping in low-lying landscapes of Denmark in the present. The conducted mapping can become a step forward towards planning for mitigative and protective measures in landslide-prone areas in Denmark, providing policy-makers with necessary decision support. However, the map of the future climate change scenario shows the reduction of the susceptible areas, raising the question of the choice of the climate models and variables in the analysis. [ABSTRACT FROM AUTHOR]

Spatial confounding, that is, collinearity between fixed effects and random effects in a spatial generalized linear mixed model, can adversely affect estimates of the fixed effects. Restricted spatial regression methods have been proposed as a remedy for spatial confounding. Such methods replace inference for the fixed effects of the original model with inference for those effects under a model in which the random effects are restricted to a subspace orthogonal to the column space of the fixed effects model matrix; thus, they "deconfound" the two types of effects. We prove, however, that frequentist inference for the fixed effects of a deconfounded linear model is generally inferior to that for the fixed effects of the original spatial linear model; in fact, it is even inferior to inference for the corresponding nonspatial model. We show further that deconfounding also leads to inferior predictive inferences, though its impact on prediction appears to be relatively small in practice. Based on these results, we argue that deconfounding a spatial linear model is bad statistical practice and should be avoided. [ABSTRACT FROM AUTHOR]

Obtaining the diversity samples of invasive alien species (species presence and absence samples) is vital for species distribution models. However, because of the enhanced focus on collecting presence samples, most datasets regarding invasive species lack explicit absence samples. Thus, the generation of effective pseudo-absence samples of invasive species is a critical issue for building species distribution models. This paper proposes a pseudo-absence sampling approach based on outlier detection in the geographical characteristic space. First, principal component analysis is used to model the linear correlation of the original variables, and a statistical index is built to determine the weight of the principal components. Next, in the geographical characteristic space built based on the principal components and their corresponding weights, the local outlier factor is obtained to identify the pseudo-absence samples. The dataset regarding the invasive species Erigeron annuus in the Yangtze River Economic Belt is used to illustrate the general process of the proposed approach. The prediction results from logistical regression with the proposed approach are better than these with the spatial random sampling, surface range envelope, and one-class support vector machine models. These findings validate the effectiveness of the proposed sampling approach. [ABSTRACT FROM AUTHOR]

Small data samples are still a critical challenge for spatial predictions. Land use regression (LUR) is a widely used model for spatial predictions with observations at a limited number of locations. Studies have demonstrated that LUR models can overcome the limitation exhibited by other spatial prediction models which usually require greater spatial densities of observations. However, the prediction accuracy and robustness of LUR models still need to be improved due to the linear regression within the LUR model. To improve LUR models, this study develops a land use quantile regression (LUQR) model for more accurate spatial predictions for small data samples. The LUQR is an integration of the LUR and quantile regression, which both have advantages in predictions with a small data set of samples. In this study, the LUQR model is applied in predicting spatial distributions of annual mean PM 2.5 concentrations across the Greater Sydney Region, New South Wales, Australia, with observations at 19 valid monitoring stations in 2020. Cross validation shows that the goodness-of-fit can be improved by 25.6–32.1% by LUQR models when compared with LUR, and prediction root mean squared error (RMSE) and mean absolute error (MAE) can be reduced by 10.6–13.4% and 19.4–24.7% by LUQR models, respectively. This study also indicates that LUQR is a more robust model for the spatial prediction with small data samples than LUR. Thus, LUQR has great potentials to be widely applied in spatial issues with a limited number of observations. [ABSTRACT FROM AUTHOR]

Digital soil mapping (DSM) approaches provide soil information by utilising the relationship between soil properties and environmental variables. Calibration of DSM models requires measurements that may often have substantial measurement errors which propagate to the DSM outputs and need to be accounted for. This study applied a geostatistical‐based DSM approach that incorporates measurement error variances in the covariance structure of the spatial model, weights measurements in accordance with their measurement accuracies and assesses the effects of measurement errors on the accuracies of DSM outputs. The method was applied in the Western Cameroon, where soil samples from 480 locations were collected and analysed for pH, clay and soil organic carbon (SOC) using conventional and mid‐infrared spectroscopy methods. Variogram parameters and regression coefficients were estimated using residual maximum likelihood under two scenarios: with and without taking measurement errors into account. Performance of the spatial models in the two scenarios was compared using validation metrics obtained with three types of cross‐validation. Acknowledging measurement errors impacted the regression coefficients and influenced the variogram parameters by reducing the nugget and sill variance for the three soil properties. Validation metrics including mean error, root mean square error and model efficiency coefficient were quite similar in both scenarios, but the prediction uncertainties were more realistically quantified by the models that account for measurement errors, as indicated by accuracy plots. There were relatively small absolute differences in predicted values of soil properties of up to 0.1 for pH, 1.6% for clay and 2 g/kg for SOC between the two scenarios. We emphasised the need of incorporating measurement errors in DSM approaches to improve uncertainty quantification, particularly when applying spectroscopy for estimating soil properties. Further development of the approach is the extension to non‐linear machine learning regression methods. Highlights: Errors in soil measurements are usually not accounted for and may affect DSM results.Measurement error variances were incorporated in the geostatistical models of three soil properties.Quantifying measurement errors in DSM allows to weigh measurements in accordance with their accuracy.Accounting for measurement errors in DSM better assesses prediction accuracy. [ABSTRACT FROM AUTHOR]

Background: Human echinococcosis is affected by natural environmental factors, and its prevalence shows a distinct geographical distribution. Western China has the highest endemicity of human echinococcosis worldwide, but the spatial pattern and environmental determinants of echinococcosis are still unclear. Methods: Hot/cold spot analysis was used to investigate the spatial distribution of human echinococcosis prevalence. Geodetector was used to identify key natural factors, and a structured additive regression model was used to analyse the relationship between natural factors and human echinococcosis prevalence and spatially predict echinococcosis epidemics. Results: Hot spots for human echinococcosis prevalence include western and southeastern parts of Tibet Autonomous Region (henceforth Tibet) and the border areas between Tibet and the provinces of Qinghai and Sichuan. Spatial effects are crucial when modelling epidemics, and relative humidity, altitude and grassland area ratio were found to have the most evident effects on echinococcosis epidemics. The relationship between these three factors and echinococcosis prevalence was non-linear, and echinococcosis risk was higher in areas with high relative humidity, high altitude, and a high ratio of grassland to other land use types. The prevalence that was predicted from the investigated environmental factors was generally higher than the actual prevalence, and more epidemic hot spots were predicted for the Qinghai-Tibet Plateau, Inner Mongolia Autonomous Region, and the provinces of Yunnan and Sichuan than the rest of western China. These results indicate that prevention and control measures may effectively reduce echinococcosis prevalence. Conclusions: We suggest that the prevention and control of human echinococcosis should be prioritized in the hot spots identified here, through the rational allocation of limited medical resources to where they are most needed. Furthermore, the spatial epidemiological modelling methods used in this study can be employed in future studies on echinococcosis and similar diseases. [ABSTRACT FROM AUTHOR]

Background: Human echinococcosis is affected by natural environmental factors, and its prevalence shows a distinct geographical distribution. Western China has the highest endemicity of human echinococcosis worldwide, but the spatial pattern and environmental determinants of echinococcosis are still unclear. Methods: Hot/cold spot analysis was used to investigate the spatial distribution of human echinococcosis prevalence. Geodetector was used to identify key natural factors, and a structured additive regression model was used to analyse the relationship between natural factors and human echinococcosis prevalence and spatially predict echinococcosis epidemics. Results: Hot spots for human echinococcosis prevalence include western and southeastern parts of Tibet Autonomous Region (henceforth Tibet) and the border areas between Tibet and the provinces of Qinghai and Sichuan. Spatial effects are crucial when modelling epidemics, and relative humidity, altitude and grassland area ratio were found to have the most evident effects on echinococcosis epidemics. The relationship between these three factors and echinococcosis prevalence was non-linear, and echinococcosis risk was higher in areas with high relative humidity, high altitude, and a high ratio of grassland to other land use types. The prevalence that was predicted from the investigated environmental factors was generally higher than the actual prevalence, and more epidemic hot spots were predicted for the Qinghai-Tibet Plateau, Inner Mongolia Autonomous Region, and the provinces of Yunnan and Sichuan than the rest of western China. These results indicate that prevention and control measures may effectively reduce echinococcosis prevalence. Conclusions: We suggest that the prevention and control of human echinococcosis should be prioritized in the hot spots identified here, through the rational allocation of limited medical resources to where they are most needed. Furthermore, the spatial epidemiological modelling methods used in this study can be employed in future studies on echinococcosis and similar diseases. [ABSTRACT FROM AUTHOR]

The prediction of a spatial variable is of particular importance when analyzing spatial data. The main objective of this study is to evaluate and compare the performance of several prediction-based methods in spatial prediction through a simulation study. The studied methods include ordinary Kriging (OK), along with several neural network methods including Multi-Layer Perceptron network (MLP), Ensemble Neural Networks (ENN), and Radial Basis Function (RBF) network. We simulated several spatial datasets with three different scenarios due to changes in data stationarity and isotropy. The performance of methods was evaluated using the Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Concordance Correlation Coefficient (CCC) indexes. Although the results of the simulation study revealed that the performance of the neural network in spatial prediction is weaker than the Kriging method, but it can still be a good competitor for Kriging. [ABSTRACT FROM AUTHOR]

• We spatially predicted soil chemical attributes in Antarctica with Random Forest. • We evaluated the performance through accuracy and uncertainty issues. • Bases Sum, pH, and Na values were higher in Antarctic Peninsula and TOC, H + Al, and P were higher in Maritime Antarctica. • Climate, relief, and vegetation influenced the attributes distribution at different scales. • We created novel soil chemical maps for important regions of the Antarctic continent. Digital Soil Mapping includes quantitative estimations of soil attributes in areas with little or no soil information, being important in remote landscapes such as the Antarctic ice-free areas. Our objective was to predict the spatial distribution of major soil chemical attributes in two important Antarctic Peninsula regions; as well as to identify the main environmental drivers of these attributes. For this, we compiled one of the largest soil databases of Antarctica and applied the machine learning algorithm Random Forest to predict seven soil chemical attributes. We also used covariates selection and partial dependence analysis to better understand the relationships of the attributes with the environmental covariates. Bases sum was the attribute which presented the highest prediction performance, whereas Na and P predictions presented the lowest. Both accuracy and uncertainty indicators showed the difficulties of Random Forest in handling natural outliers. The soil attributes distribution was related to the mean annual temperature and annual precipitation, multispectral bands and indexes related to the vegetation response, rookeries distance expressing the birds' activity, and topographic attributes. The attributes showed a climatic gradient, with higher values of bases sum, pH, and Na in Northern Antarctic Peninsula, and higher total organic carbon, H + Al and P in Maritime Antarctic. This shows the climate covariates influence the soil variability in a macroscale, whereas terrain predictors control the soil at local scales. This study showed the Digital Soil Mapping potential to surpass the limitations of conventional mapping and indicated the feasibility in obtaining interpretable predictions which can be directly associated with the soil forming factors. Finally, the data generated can be used as reference to monitor the impacts of the climate changes on the soils of Antarctica. [ABSTRACT FROM AUTHOR]

Spatial prediction of any geographic phenomenon can be an intractable problem. Predicting sparse and uncertain spatial events related to many influencing factors necessitates the integration of multiple data sources. We present an innovative approach that combines data in a Discrete Global Grid System (DGGS) and uses machine learning for analysis. A DGGS provides a structured input for multiple types of spatial data, consistent over multiple scales. This data framework facilitates the training of an Artificial Neural Network (ANN) to map and predict a phenomenon. Spatial lag regression models (SLRM) are used to evaluate and rank the outputs of the ANN. In our case study, we predict hate crimes in the USA. Hate crimes get attention from mass media and the scientific community, but data on such events is sparse. We trained the ANN with data ingested in the DGGS based on a 50% sample of hate crimes as identified by the Southern Poverty Law Center (SPLC). Our spatial prediction is up to 78% accurate and verified at the state level against the independent FBI hate crime statistics with a fit of 80%. The derived risk maps are a guide to action for policy makers and law enforcement. [ABSTRACT FROM AUTHOR]

SUMMARY: This paper addresses the issue of monitoring spatial environmental phenomena of interest utilizing information collected by a network of mobile, wireless, and noisy sensors that can take discrete measurements as they navigate through the environment. It is proposed to employ Gaussian Markov random field (GMRF) represented on an irregular discrete lattice by using the stochastic partial differential equations method to model the physical spatial field. It then derives a GMRF-based approach to effectively predict the field at unmeasured locations, given available observations, in both centralized and distributed manners. Furthermore, a novel but efficient optimality criterion is then proposed to design centralized and distributed adaptive sampling strategies for the mobile robotic sensors to find the most informative sampling paths in taking future measurements. By taking advantage of conditional independence property in the GMRF, the adaptive sampling optimization problem is proven to be resolved in a deterministic time. The effectiveness of the proposed approach is compared and demonstrated using pre-published data sets with appealing results. [ABSTRACT FROM AUTHOR]

Given a spatial raster framework with explanatory feature layers, a spatial contextual layer (e.g., a potential field), as well as a set of training samples with class labels, the spatial prediction problem aims to learn a model that can predict a class layer. The problem is important in societal applications such as flood extent mapping for disaster response and national water forecasting, but is challenging due to the noise, obstacles, and heterogeneity in feature maps, implicit spatial dependency between locations based on the contextual layer (e.g., gradient directions on a potential field), and the large number of sample locations. Existing work often assumes undirected spatial dependency, or directed dependency with a total order, and thus cannot reflect complex directed dependency with a partial order. In contrast, we recently proposed geographical hidden Markov tree, a probabilistic graphical model that generalizes the common hidden Markov model from a one-dimensional sequence to a two-dimensional map. Partial order class dependency is incorporated in the hidden class layer with a reverse tree structure. We also investigated computational algorithms for reverse tree construction, model parameter learning and class inference. This paper extends our recent model with overlaying class nodes between observation nodes and underlying hidden class nodes. The additional overlaying class layer makes the model more robust to large scale feature obstacles. We also proposed corresponding learning and inference methods. Extensive evaluations on real world datasets show that our models outperform multiple baselines in flood mapping applications, our algorithms are scalable on large data sizes, and the proposed extension enhances classification performance. [ABSTRACT FROM AUTHOR]

Aim: Species distribution models (SDMs) are valuable for rare species conservation and are commonly used to extrapolate predictions of habitat suitability geographically to regions where species occurrence is unknown (i.e., transferability). Spatially structured cross‐validation can be used to infer transferability, yet, few studies have evaluated how delineation of cross‐validation folds affects model complexity and predictions. We developed SDMs using multiple cross‐validation approaches to understand the implications for predicting habitat suitability for northern Idaho ground squirrels, a rare, federally threatened species that has been extensively surveyed in regions where known populations occur, resulting in >8000 presence locations. Location: Idaho, USA. Methods: We delineated cross‐validation folds by mimicking the manner in which predictions would be geographically extrapolated or by using existing dispersal barriers. We varied the distance between, number, and directionality of folds. We conducted a grid search on statistical regularization parameters to optimize model complexity, covering a range of values exceeding that typically implemented. For each cross‐validation approach, we selected optimal regularization and model complexity based on out‐of‐sample predictive ability. Results: Delineation of cross‐validation folds substantially affected resulting model complexity and extrapolated predictions. All cross‐validation approaches resulted in models with apparently high out‐of‐sample predictive ability, yet optimal model complexity varied substantially among the approaches. Regularization demonstrated a noisy relationship between model complexity and prediction, where local optima in predictive performance were common at small values. Main conclusion: Subtle modelling decisions can have large consequences for predictions of habitat suitability and transferability of SDMs. When transferability is the goal, cross‐validation approaches should be considered carefully and mimic the manner in which spatial extrapolation will occur, else overly complex models with inflated assessments of predictive accuracy may result. Further, spatially structured cross‐validation may not guard against over‐parameterization, and assessing a broader range of regularization parameters may be necessary to optimize model complexity for transferability. [ABSTRACT FROM AUTHOR]

Spatial data (e.g., phytopathogenic data) do not always meet assumptions such as stationarity, isotropy and Gaussian distribution, thereby requiring complex spatial methods and models. Some deterministic assumption-free methods such as the inverse distance weighting can also be applied to predict spatial data, but their output is limited for graphical solutions (mapping). We adapted a computer-based prediction method called Circular Variable Radius Moving Window (CVRMW) that is based on two others: moving window kriging (MWK) and inverse squared-distance weighting (ISDW). The algorithm is developed to meet an objective function that minimizes the index of variation of the spatial observations inside the moving window. A code in R language is presented and thoroughly described. The outputs include the range of the spatial dependence as the radius calculated at every target location and the standard error of the predicted values, mapped to provide a useful tool for spatial exploratory analysis. The method does not make any assumptions about the spatial process, and it is an alternative for dealing with spatial heterogeneity. [ABSTRACT FROM AUTHOR]

Flash floods are one of the worst natural disasters, causing massive economic losses and many deaths. Creating a flood susceptibility map (FSM) that pinpoints the areas most at risk of flooding is a crucial non-structural solution for managing floods. This study aimed to assess the efficacy of combinations of the random forest (RF) model with three biology-inspired metaheuristic algorithms, namely invasive weed optimization (IWO), slime mould algorithm (SMA), and satin bowerbird optimization (SBO), for flood susceptibility mapping in Estahban town, Iran. Initially, synthetic-aperture radar (SAR) (Sentinel-1) and optical (Landsat-8) satellite images were integrated to monitor the flooded areas during the July 2022 monsoon in the study area. A dataset of 509 flood occurrence points was created to identify flood-prone areas using remote sensing techniques, considering the monitored flood areas. The dataset also included twelve flood-related criteria: topography, land cover, and climate. The holdout method was employed for modeling, with a ratio of 70:30 used for the train/test split. Data pre-processing techniques were conducted to improve model performance, including determining criteria importance and addressing multicollinearity issues using certainty factor (CF), multicollinearity, and information gain ratio (IGR) methods. Then FSM was prepared using RF, RF-IWO, RF-SBO, and RF-SMA models. The findings of this research revealed that the RF-IWO model was the best predictive model of flood susceptibility modeling, with root-mean-square-error (RMSE) (0.211 and 0.0.27), mean-absolute-error (MAE) (0.103 and 0.15), and coefficient-of-determination (R2) (0.821 and 0.707) in the training and testing phases, respectively. Receiver operating characteristic (ROC) curve analysis of FSM revealed that the most accurate models were the RF-IWO (area under the curve (AUC) = 0.983), RF-SBO (AUC = 0.979), RF-SMA (AUC = 0.963), and RF (AUC = 0.959), respectively. Integrating biology-inspired computing algorithms with machine learning algorithms presents a novel approach to enhancing the accuracy of FSMs. [Display omitted] • Detected monsoon floods and identified vulnerable areas. • Integrated optical and radar satellite images to improve flood monitoring accuracy. • The biology-inspired metaheuristic algorithms increased flood prediction accuracy. • Strong spatial association of flood probability (0.959–0.983) among models. [ABSTRACT FROM AUTHOR]

Landslides pose a significant threat to human life and infrastructure, underscoring the ongoing need for accurate landslide susceptibility mapping (LSM) to effectively assess risks. This study introduces an innovative approach that leverages multi-objective evolutionary fuzzy algorithms for landslide modeling in Khalkhal town, Iran. Two algorithms, namely the non-dominated sorting genetic algorithm II (NSGA-II) and the evolutionary non-dominated radial slots-based algorithm (ENORA), were employed to optimize Gaussian fuzzy rules. By utilizing 15 landslide conditioning factors (aspect, altitude, distance from the fault, soil, slope, lithology, rainfall, distance from the road, the normalized difference vegetation index (NDVI), land cover, plan curvature, profile curvature, topographic wetness index (TWI), stream power index (SPI), and distance from the river) and historical landslide events (153 landslide locations), we randomly partitioned the input data into training (70%) and validation (30%) sets. The training set determined the weight of conditioning factor classes using the frequency ratio (FR) approach. These weights were then used as inputs for the NSGA-II and ENORA algorithms to generate an LSM. The NSGA-II algorithm achieved a root-mean-square error (RMSE) of 0.25 during training and 0.43 during validation. Similarly, the ENORA algorithm demonstrated an RMSE of 0.28 in training and 0.48 in validation. The findings revealed that the LSM created by the NSGA-II algorithm exhibited superior predictive capabilities (area under the receiver operating characteristic curve (AUC) = 0.867) compared to the ENORA algorithm (AUC = 0.844). Additionally, a particle swarm optimization (PSO) algorithm was employed to determine the importance of conditioning factors, identifying lithology, land cover, and altitude as the most influential factors. [ABSTRACT FROM AUTHOR]