Search

This study investigates the impact of untreated wastewater on a section of the Euphrates River in Al-Diwaniyah, Iraq. The unregulated discharge of sewage and industrial effluents has caused severe pollution in this area. The research measures electrical conductivity, turbidity, total dissolved solids, dissolved oxygen, total hardness, biological oxygen demand and heavy metals (Ni, Cr, Pb, Co, Cd, Cu, Fe) using Inductively Coupled Plasma Atomic Emission Spectrometry. Approximately 40 water sampling stations were selected for analysis and the Heavy Metal Pollution Index (HPI) was used to assess water quality. The study employed Inverse Distance Weighting (IDW) tool, to predict HPI values and their spatial distribution. The accuracy of prediction maps was evaluated using regression analysis by comparing observed values with predicted values from the maps. The study emphasizes the excessive pollution levels in the Euphrates River, specifically exceeding Iraqi standard thresholds for Ni, Fe, and Cd concentrations. [ABSTRACT FROM AUTHOR]

To manage soil and agriculture sustainably, it is crucial to keep updated on changes in the pools of soil organic carbon (SOC). Using tools from digital soil mapping (DSM), this case study simulates the dynamics of total and active organic carbon. The aim of present study was to assess the performance of various machine learning techniques for SOC fractions. The SRTM DEM and its derivatives were used to predict total organic carbon (TOC) and its fractions for a 4000 ha study area in Central State Farm, Sardargragh, Rajasthan. For prediction, Extreme Gradient Boosting (XG Boost), Cubist, and Random Forest (RF) models were used. For modeling purposes, the SOC fractions like very labile C (VLC), labile C (LC), less labile C (LLC) was analyzed. The range of the WBC content was 0.97 to 6.85 g kg-1. Both the calibration and validation sets of VLC showed outstanding results from the RF model (R2c = 0.948, RMSEc = 0.174 g kg-1and R2v = 0.204, RMSEv = 0.213 g kg-1). The XG Boost model, on the other hand, had the next lowest accuracy and explained just around 10.06–36.8% of the variation of C fractions. The cubist model performed the worst in both the calibration and validation sets. Clay is the most significant predictor of the TOC, WBC, LLC, NLC and PC. The predicted SOC varied from 3.06 to 9.41 g kg-1, 0.34 to 1.42 g kg-1, 0.73 to 2.45 g kg-1, 0.28 to 3.30 g kg-1 and 1.05 to 4.12 g kg-1 in TOC, VLC, LC, LLC and NLC, respectively. RF proved to be the most accurate and least uncertain model when it came to predicting regional SOC. Based on the findings of our simulation; it appears that SOC may be approximated rather accurately and with ease. In general, stakeholders, decision-makers, and applicants in agricultural management approaches toward precision agriculture may find their high-resolution maps helpful. [ABSTRACT FROM AUTHOR]

Many soil functions and processes are controlled by the soil particle size distribution. Accordingly, nationwide geoinformation on this soil property is required to enable climate-smart and resilient land management. This study presents a new deep learning approach to simultaneously model the contents of the three particle sizes of sand, silt, and clay and their variations with depth throughout the landscape. The approach allows for the consideration of the natural soil horizon boundaries and the inclusion of the surrounding landscape context of each soil profile to investigate the soil–landscape relation. Applied to the agricultural soil landscape of Germany, the approach generated a three-dimensional continuous data product with a resolution of 100 m in geographic space and a depth resolution of 1 cm. The approach relies on a patch-wise multi-target convolutional neural network (CNN) model. Genetic algorithm optimization was applied for CNN parameter tuning. Overall, the effectiveness of the CNN algorithm in generating multidimensional, multivariate, national-scale soil data products was demonstrated. The predictive performance resulted in a median root mean square error of 17.8 mass-% for the sand, 14.4 mass-% for the silt, and 9.3 mass-% for the clay content in the top ten centimeters. This increased to 20.9, 16.5, and 11.8 mass-% at a 40 cm depth. The generated data product is the first of its kind. However, even though the potential of this deep learning approach to understand and model the complex soil–landscape relation is virtually limitless, its limitations are data driven concerning the approximation of the soil-forming factors and the available soil profile data. [ABSTRACT FROM AUTHOR]

Soil temperature and soil moisture are key drivers of various soil ecological processes, which implies a significant importance of datasets including their variations in space, depth and time (4D). Current gridded products typically have a low resolution, either spatially or temporally. Here, we aim at modelling and explaining high-resolution soil temperature and soil moisture patterns in 4D for a 400 km2 study area in a heterogeneous landscape. Our target resolution of 10 m in space, 10 cm in depth, and 1 h in time allows capturing small-scale variations as well as short-term dynamics. We used multi-depth soil temperature and soil moisture measurements from 212 locations and linked them to 45 potential predictors, representing meteorological conditions, soil parameters, vegetation coverage, and landscape relief. We trained random forest models that were able to predict soil temperature with a mean absolute error of 0.93 °C and soil moisture with a mean absolute error of 4.64 % volumetric water content. Continuous model predictions enabled a comprehensive analysis of 4D patterns and confirmed that the selected resolution is essential to capture soil temperature and soil moisture variations at the landscape scale. In addition to a strongly pronounced annual cycle and recognisable influences on the diurnal cycle, there were significant differences between the land uses and patterns due to the relief, which also varied along the depth gradient. By applying interpretable machine learning techniques, we investigated the detailed influence of all drivers and discussed overlapping effects that led to the prediction patterns.

The Random Forest (RF) and K nearest neighbors (KNN) machine learning (ML) algorithms were evaluated for their ability to predict ophiolite occurrences, in the East Vardar Zone (EVZ) of central North Macedonia. A predictive map of the investigated area was created using three data sources: geophysical data (digital elevation model, gravity and geomagnetic), multispectral optical satellite images (Landsat 7 ETM + and their derivatives), and geological data (distance to fault map and ophiolite outcrops map). The research included a comparison and discussion on the statistical and geological findings derived from different training dataset class ratios in relation to a testing dataset characterized by significant class imbalance. The results suggest that the precise selection of a suitable class balance for the training dataset is a critical factor in achieving accurate ophiolite prediction with RF and KNN algorithms. The analysis of feature importance revealed that the Bouguer gravity anomaly map, total intensity of the Earth's magnetic field reduced to the pole map, distance to fault map, band ratio BR3 map obtained from multispectral satellite images, and digital elevation model are the most significant features for predicting ophiolites within the EVZ. KNN showed poorer results compared to RF in terms of both the evaluation metrics and visual analysis of prediction maps. The methods applied in this research can be applied for predictive mapping of complex geo-tectonic units covered by dense vegetation, and may indicate the presence of these units even if they were not previously mapped, particularly when geophysical data are used as features. [ABSTRACT FROM AUTHOR]

Soil salinity is a significant challenge in numerous regions across the globe, including Egypt. The potential consequences encompass negative impacts on crop yield, human well-being, and eco-logical systems. The utilization of remote sensing and machine learning techniques is increasingly becoming recognized as cost-effective methodologies for the cartographic representation of soil salinity. The present work employed Landsat 8 satellite imaging data and sophisticated machine learning techniques to delineate and assess soil salinity levels in Sharkia Governorate, Egypt. In this work, several machine learning techniques were employed to forecast the salinity values of Total Dissolved Solids (TDS) in the designated geographical region. These algorithms encompassed support vector machines (SVM), regression trees, Gaussian linear regression, and tree-based ensemble in addition to linear regression analysis. A variety of instances were generated to develop an optimal model that accurately characterizes the salinity TDS values within the study area. This was achieved by utilizing the band values extracted from the Landsat 8 satellite imagery. The approach that demonstrated the highest performance was observed when employing the Blue, Red, and shortwave infrared bands in conjunction with the SVM-Quadratic SVM model. This particular configuration yielded an R2 value of 0.86 and an RMSE value of 175.98. The findings of this work demonstrate the feasibility of precisely mapping soil salinity through the utilization of Landsat 8 satellite imaging data and machine learning techniques. The provided data can be utilized to identify regions characterized by elevated levels of soil salinity, as well as for the formulation of effective approaches aimed at addressing this issue. [ABSTRACT FROM AUTHOR]

ABSTRACT We present the first Brazilian record of Chiroxiphia napensis, documented by an adult male specimen collected on the right bank of the upper Japurá River. We also compiled verifiable records of all Amazonian taxa in the C. pareola complex to update their range maps. New records extended considerably the ranges of all taxa. By assuming river-delimited distributions to infer areas from point records, we generated a predictive distributional map that suggests testable hypotheses about occurrence in unstudied areas and that pinpoints topics for future research. We interpret C. napensis as endemic to the Amazon-Japurá interfluve, its range extending narrowly beyond this area in the foothills of the Andes. We suggest that the unusual distribution pattern of C. regina could be the result of river avulsion. We also detected a region of possible contact between C. regina and C. pareola, and a large area in northern Amazonia from which the complex appears to be absent.

The determination of soil properties, in addition to requiring great human effort, also involves a number of technical activities of high financial cost. Seeking to show that it is possible to reduce these high financial costs, this study presents a methodology that combines remote sensing techniques and mathematical modeling for estimating the concentration of soil organic matter (SOM). Technological advances have provided great improvements in equipment for capturing terrestrial information, allowing a greater availability of quality data for spatial analysis, which has brought about better estimates in the models. Thus, this study evaluated the capacity of the Sentinel-1 Synthetic Aperture Radar (SAR) satellite to determine the concentration of SOM in areas with different types of agricultural use in a hydrographic basin localted in the souutheastern region of the state of São Paulo, Brazil. The partial least squares regression method was used to build models for estimating the SOM, taking into account the SAR backscattering values and soil samples obtained in situ, which were collected at a depth of 0–10 cm. The results obtained indicated that the accuracy of the model adjusted based on backscattering values in the vertical/vertical (VV), vertical/horizontal (VH) polarizations, product of VHxVV and soil moisture presented a coefficient of determination of 0.502 with the SOM for independent data, at p <0.0001. These results indicate that the model derived from the SAR data for the mapping of SOM presented potential for prediction, but only within acceptable limits for predicting SOM in areas with similar characteristics of land use. [ABSTRACT FROM AUTHOR]

Many soil functions and processes are controlled by the soil particle size distribution. Accordingly, nationwide geoinformation on this soil property is required to enable climate-smart and resilient land management. This study presents a new deep learning approach to simultaneously model the contents of the three particle sizes of sand, silt, and clay and their variations with depth throughout the landscape. The approach allows for the consideration of the natural soil horizon boundaries and the inclusion of the surrounding landscape context of each soil profile to investigate the soil–landscape relation. Applied to the agricultural soil landscape of Germany, the approach generated a three-dimensional continuous data product with a resolution of 100 m in geographic space and a depth resolution of 1 cm. The approach relies on a patch-wise multi-target convolutional neural network (CNN) model. Genetic algorithm optimization was applied for CNN parameter tuning. Overall, the effectiveness of the CNN algorithm in generating multidimensional, multivariate, national-scale soil data products was demonstrated. The predictive performance resulted in a median root mean square error of 17.8 mass-% for the sand, 14.4 mass-% for the silt, and 9.3 mass-% for the clay content in the top ten centimeters. This increased to 20.9, 16.5, and 11.8 mass-% at a 40 cm depth. The generated data product is the first of its kind. However, even though the potential of this deep learning approach to understand and model the complex soil–landscape relation is virtually limitless, its limitations are data driven concerning the approximation of the soil-forming factors and the available soil profile data.

Rapid climate change and invasive species introductions threaten ecological communities across the globe. Freshwaters are particularly vulnerable and impacted, especially when these stresses coincide. We document the migration of an invasive piscine predator, the Sacramento pikeminnow (Ptychocheilus grandis), within its introduced range, the South Fork Eel River, California, USA. Snorkel surveys and temperature monitoring in 2015–2019 showed that pikeminnow migrate upstream during spring and early summer, with earlier migration in warmer years. We developed a statistical temperature model to forecast the timing and extent of upstream migration by pikeminnow under varying combinations of discharge and air temperature. Modeled river temperature increased with air temperature and downstream and decreased with discharge. In years with low discharge and high air temperature, we predict pikeminnow will move upstream earlier, increasing spatial and temporal overlap in their summer range with native fishes. Managing conditions that reduce pikeminnow co-occurrence with native fishes (i.e., decreasing river temperature) could increase amount and duration of predator-free habitat for native fishes. We predict invasive pikeminnow will have larger impacts on invaded riverine communities with global warming and increasing drought severity. Knowledge of life history and phenology, for pikeminnow and other organisms, can guide effective management as conditions change and help to limit adverse impacts of introduced organisms on native species. [ABSTRACT FROM AUTHOR]

Soil Organic Carbon (SOC) is a crucial indicator of ecosystem health and soil quality. Machine learning (ML) models that predict soil quality based on environmental parameters are becoming more prevalent. However, studies have yet to examine how well each ML technique performs when predicting and mapping SOC, particularly at high spatial resolutions. Model predictors include topographic variables generated from SRTM DEM; vegetation and soil indices derived from Landsat satellite images predict SOC for the Lakhimpur district of the upper Brahmaputra Valley of Assam, India. Four ML models, Random Forest (RF), Cubist, Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM), were utilized to predict SOC for the top layer of soil (0–15 cm) at a 30 m resolution. The results showed that the descriptive statistics of the calibration and validation sets were close enough to the total set data and calibration dataset, representing the complete samples. The measured SOC content varied from 0.10 to 1.85%. The RF model's performance was optimal in the calibration and validation sets (R2c = 0.966, RMSEc = 0.159%, R2v = 0.418, RMSEv = 0.377%). The SVM model, on the other hand, had the next-lowest accuracy, explaining 47% of the variation (R2c = 0.471, RMSEc = 0.293, R2v = 0.081, RMSEv = 0.452), while the Cubist model fared the poorest in both the calibration and validation sets. The most-critical variable in the RF model for predicting SOC was elevation, followed by MAT and MRVBF. The essential variables for the Cubist model were slope, TRI, MAT, and Band4. AP and LS were the most-essential factors in the XGBoost and SVM models. The predicted OC ranged from 0.44 to 1.35%, 0.031 to 1.61%, 0.035 to 1.71%, and 0.47 to 1.36% in the RF, Cubist, XGBoost, and SVM models, respectively. Compared with different ML models, RF was optimal (high accuracy and low uncertainty) for predicting SOC in the investigated region. According to the present modeling results, SOC may be determined simply and accurately. In general, the high-resolution maps might be helpful for decision-makers, stakeholders, and applicants in sericultural management practices towards precision sericulture. [ABSTRACT FROM AUTHOR]

Recent years have witnessed an explosion of interest in predictive policing, with a clear opposition emerging between supporters and critics of its implementation. While critical accounts conventionally centre on opacities and operational asymmetries of the algorithmic construct (biased training, feedback loop, etc.), I argue that a different critique is first needed. Focussing on place-based techniques, I maintain that contemporary predictive mapping basically perpetuates the political and epistemic dictates which have historically framed the conceptualisation of crime in relation to space. Through a review of sources spanning from the Cartographic School to current predictive policing literature, I identify two main conceptual axes which operationalise this heritage: first, an explanatory framework of crime that has never detached from the socio-economic deficit archetype; and secondly, an ontologisation of crime alternative to biologicist positivism, nonetheless integral to the etiologic paradigm. Therefore, without first disputing these ideological bottlenecks, no initiative towards a transparent use of predictive policing is plausible, neither does a sharp distinction between place-based and person-based predictions seem tenable. En los últimos años se ha asistido a una explosión de interés por la vigilancia policial predictiva y al surgimiento de una clara oposición entre partidarios y detractores de su implementación. Mientras que los relatos críticos se centran convencionalmente en opacidades y asimetrías operacionales del constructo algorítmico (entrenamiento sesgado, bucle de retroalimentación, etc.), lo que aquí se plantea es la necesidad preliminar de otro tipo de crítica. Con el foco puesto en las técnicas de predicción geoespacial, se sostiene que los métodos de mapeo predictivo tienden a perpetuar los dictados políticos y epistemológicos que históricamente han enmarcado la conceptualización del delito en su relación con el espacio. Mediante una revisión de fuentes que abarcan desde la Escuela Cartográfica a la literatura contemporánea sobre vigilancia predictiva, se detectan dos grandes ejes conceptuales que operativizan dicho legado: primero, un marco explicativo del delito que nunca supo emanciparse del arquetipo del déficit socioeconómico, y segundo, una ontologización del delito alternativa a la del positivismo biologicista pero aun así parte integral del paradigma etiológico. Así pues, sin antes cuestionar dichas constricciones ideológicas, ninguna iniciativa encaminada a un uso transparente de la policía predictiva es plausible, ni una distinción real entre mapeo predictivo y predicción individualizada parece defendible. Available from: https://doi.org/10.35295/osls.iisl/0000-0000-0000-1360

Coral reefs are experiencing unprecedented levels of stress from global warming, ocean acidification, fishing, and water pollution. In the Caribbean and Western Atlantic, multiple stressors have resulted in widespread losses of the dominant reef‐building Acroporid corals, two of which are listed as threatened species under the 1973 U.S. Endangered Species Act. In response, active coral reef restoration through the outplanting of live corals has become a widespread intervention technique. To increase restoration success, active coral reef restoration requires significant investment and careful planning, and selection of suitable sites for coral outplanting is an essential early step with considerable influence on restoration outcomes. We applied a maximum entropy model to predict and map habitat suitability for the reef‐building coral species, Acropora palmata, around the island of St. Croix in the U.S. Virgin Islands. Based mostly on bathymetry and benthic habitat type, the highest performing model predicted approximately 21.75 km2 of the highest probability of suitable habitat, of which over half occurred within existing marine protected areas (MPAs). Outplanted coral at 60% of sites coincided with predicted maximum habitat suitability index values greater than 0.75 and 35% with values greater than 0.90. The model reveals that all three statutory MPAs with shallow water coral reefs have a considerable area (13.24 km2) of predicted high suitability seabed with potential for active A. palmata restoration efforts. The predictive spatial modeling approach provides a cost‐effective tool to inform future coral restoration design and to evaluate the habitat suitability of coral outplanting sites. [ABSTRACT FROM AUTHOR]

This study was performed to examine the separate and simultaneous influence of predictive models’ choice alongside sample ratios selection in soil organic matter (SOM). The research was carried out in northern Morocco, characterized by relatively cold weather and diverse geological conditions. The dataset herein used accounted for 1591 soil samples, which were randomly split into the following ratios: 10% (∼150 sample ratio), 20% (∼250 sample ratio), 35% (∼450 sample ratio), 50% (∼600 sample ratio) and 95% (∼1200 sample ratio). Models herein involved were ordinary kriging (OK), regression kriging (RK), multiple linear regression (MLR), random forest (RF), quantile regression forest (QRF), Gaussian process regression (GPR) and an ensemble model. The findings in the study showed that the accuracy of SOM prediction is sensitive to both predictive models and sample ratios. OK combined with 95% sample ratio performed equally to RF in conjunction with all the sample ratios, as the latter did not show much sensitivity to sample ratios. ANOVA results revealed that RF with a ∼10% sample ratio could also be optimum for predicting SOM in the study area. In conclusion, the findings herein reported could be instrumental for producing cost-effective detailed and accurate spatial estimation of SOM in other sites. Furthermore, they could serve as a baseline study for future research in the region or elsewhere. Therefore, we recommend conducting series of simulation of all possible combinations between various predictive models and sample ratios as a preliminary step in soil organic matter prediction.

Soil Organic Carbon (SOC) is a crucial indicator of ecosystem health and soil quality. Machine learning (ML) models that predict soil quality based on environmental parameters are becoming more prevalent. However, studies have yet to examine how well each ML technique performs when predicting and mapping SOC, particularly at high spatial resolutions. Model predictors include topographic variables generated from SRTM DEM; vegetation and soil indices derived from Landsat satellite images predict SOC for the Lakhimpur district of the upper Brahmaputra Valley of Assam, India. Four ML models, Random Forest (RF), Cubist, Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM), were utilized to predict SOC for the top layer of soil (0–15 cm) at a 30 m resolution. The results showed that the descriptive statistics of the calibration and validation sets were close enough to the total set data and calibration dataset, representing the complete samples. The measured SOC content varied from 0.10 to 1.85%. The RF model’s performance was optimal in the calibration and validation sets (R2c = 0.966, RMSEc = 0.159%, R2v = 0.418, RMSEv = 0.377%). The SVM model, on the other hand, had the next-lowest accuracy, explaining 47% of the variation (R2c = 0.471, RMSEc = 0.293, R2v = 0.081, RMSEv = 0.452), while the Cubist model fared the poorest in both the calibration and validation sets. The most-critical variable in the RF model for predicting SOC was elevation, followed by MAT and MRVBF. The essential variables for the Cubist model were slope, TRI, MAT, and Band4. AP and LS were the most-essential factors in the XGBoost and SVM models. The predicted OC ranged from 0.44 to 1.35%, 0.031 to 1.61%, 0.035 to 1.71%, and 0.47 to 1.36% in the RF, Cubist, XGBoost, and SVM models, respectively. Compared with different ML models, RF was optimal (high accuracy and low uncertainty) for predicting SOC in the investigated region. According to the present modeling results, SOC may be determined simply and accurately. In general, the high-resolution maps might be helpful for decision-makers, stakeholders, and applicants in sericultural management practices towards precision sericulture.

The spatial distribution of soil organic carbon (SOC) serves as critical geographic information for assessing ecosystem services, climate change mitigation, and optimal agriculture management. Digital mapping of SOC is challenging due to the complex relationships between the soil and its environment. Except for the well-known terrain and climate environmental covariates, vegetation that interacts with soils influences SOC significantly over long periods. Although several remote-sensing-based vegetation indices have been widely adopted in digital soil mapping, variables indicating long term vegetation growth have been less used. Vegetation phenology, an indicator of vegetation growth characteristics, can be used as a potential time series environmental covariate for SOC prediction. A CNN-LSTM model was developed for SOC prediction with inputs of static and dynamic environmental variables in Xuancheng City, China. The spatially contextual features in static variables (e.g., topographic variables) were extracted by the convolutional neural network (CNN), while the temporal features in dynamic variables (e.g., vegetation phenology over a long period of time) were extracted by a long short-term memory (LSTM) network. The ten-year phenological variables derived from moderate-resolution imaging spectroradiometer (MODIS) observations were adopted as predictors with historical temporal changes in vegetation in addition to the commonly used static variables. The random forest (RF) model was used as a reference model for comparison. Our results indicate that adding phenological variables can produce a more accurate map, as tested by the five-fold cross-validation, and demonstrate that CNN-LSTM is a potentially effective model for predicting SOC at a regional spatial scale with long-term historical vegetation phenology information as an extra input. We highlight the great potential of hybrid deep learning models, which can simultaneously extract spatial and temporal features from different types of environmental variables, for future applications in digital soil mapping. [ABSTRACT FROM AUTHOR]

Many mammals cross large rivers to access seasonal habitats, yet river crossing behavior is poorly acknowledged or understood. Crossing large rivers is inherently risky, with vessel traffic and flooding events increasing the risk of drowning. We investigated river crossing behavior by threatened bison (Bison bison) in the boreal forest, using data from 11 GPS-collared animals to identify when and where they cross a major river. We used kernel densities of GPS locations to identify crossing hotspots, and developed resource selection function (RSF) models that used habitat characteristics to explain crossing behavior based on a priori hypotheses. We then predicted high-probability crossing areas along a 400-km stretch of river using the best-supported RSF model. Bison regularly crossed the river (x̄ = 8.6 crossings/100 days), particularly during summer months. Model selection results suggested that bison may have chosen river crossing areas that maximized access to forage. Bison preferred to cross the river near islands and areas with high seismic line densities, both of which were likely preferred summer foraging habitats. In addition, islands may have been used as temporary stopovers while crossing the river, or for relief from biting insects. Bison crossed the river where the channel was relatively narrow (x̄ = 400 ± 213 m [SD]) and chose days when river discharge was low, likely to reduce the risk of drowning, exhaustion, or hypothermia. While based on a small sample of collared bison, predictions about high-probability crossing areas may be used to inform mitigation measures aimed at reducing bison drowning caused by vessel traffic, which is expected to increase as resource development expands in the region. Our approach may be informative for identifying river crossing hotspots for other mammals that cross major rivers. [ABSTRACT FROM AUTHOR]

One consequence of global change causing widespread concern is the possibility of ecosystem conversions from one type to another. A classic example of this is vegetation type conversion (VTC) from native woody shrublands to invasive annual grasslands in the biodiversity hotspot of Southern California. Although the significance of this problem is well recognized, understanding where, how much, and why this change is occurring remains elusive owing to differences in results from studies conducted using different methods, spatial extents, and scales. Disagreement has arisen particularly over the relative importance of short‐interval fires in driving these changes. Chronosequence approaches that use space for time to estimate changes have produced different results than studies of changes at a site over time. Here we calculated the percentage woody and herbaceous cover across Southern California using air photos from ~1950 to 2019. We assessed the extent of woody cover change and the relative importance of fire history, topography, soil moisture, and distance to human infrastructure in explaining change across a hierarchy of spatial extents and regions. We found substantial net decline in woody cover and expansion of herbaceous vegetation across all regions, but the most dramatic changes occurred in the northern interior and southern coastal areas. Variables related to frequent, short‐interval fire were consistently top ranked as the explanation for shrub to grassland type conversion, but low soil moisture and topographic complexity were also strong correlates. Despite the consistent importance of fire, there was substantial geographical variation in the relative importance of drivers, and these differences resulted in different mapped predictions of VTC. This geographical variation is important to recognize for management decision‐making and, in addition to differences in methodological design, may also partly explain differences in previous study results. The overwhelming importance of short‐interval fire has management implications. It suggests that actions should be directed away from imposing fires to preventing fires. Prevention can be controlled through management actions that limit ignitions, fire spread, and the damage sustained in areas that do burn. This study also demonstrates significant potential for changing fire regimes to drive large‐scale, abrupt ecological change. [ABSTRACT FROM AUTHOR]

Baseline data on the distribution and extent of biogenic habitat-forming species at a high spatial resolution are essential to inform habitat management strategies, preserve ecosystem integrity, and achieve effective conservation objectives in the nearshore. Model-based approaches to map suitable habitat for these species are a key tool to address this need, filling in gaps where observations are otherwise unavailable and remote sensing methods are limited by turbid waters or cannot be applied at scale. We developed a high resolution (35 m) ensemble species distribution model to predict the distribution of eelgrass (Zostera marina) along the Atlantic coast of Nova Scotia, Canada where the observational coverage of eelgrass occurrence is sparse and nearshore waters are optically complex. Our ensemble model was derived as a performance-weighted average prediction of 7 different modeling methods fit to 6 physical predictors (substrate type, depth, wave exposure, slope, and two bathymetric position indices) and evaluated with a 5-fold spatially-blocked cross-validation procedure. The ensemble model showed moderate predictive performance (Area Under the Receiver-Operating Characteristic Curve (AUC) = 0.803 ± 0.061, True Skill Statistic (TSS) = 0.531 ± 0.100; mean ± SD), high sensitivity (92.0 ± 4.5), and offered some improvement over individual models. Substrate type, depth, and relative wave exposure were the most influential predictors associated with eelgrass occurrence, where the highest probabilities were associated with sandy and sandy-mud sediments, depths ranging 0 m – 4 m, and low to intermediate wave exposure. Within our study region, we predicted a total extent of suitable eelgrass habitat of 38,130 ha. We found suitable habitat was particularly extensive within the long narrow inlets and extensive shallow flats of the South Shore, Eastern Shore, and Bras d’Or Lakes. We also identified substantial overlap of eelgrass habitat with previously identified Ecologically and Biologically Significant Areas that guide regional conservation planning while also highlighting areas of greater prediction uncertainty arising from disagreement among modeling methods. By offering improved sensitivity and insights into the fine-scale regional distribution of a habitat-forming species with associated uncertainties, our ensemble-based modeling approach provides improved support to numerous nearshore applications including conservation planning and restoration, marine spatial and emergency response planning, environmental impact assessments, and fish habitat protection.

Recovery of bryophyte diversity following silvicultural treatments depends upon the reestablishment of favorable microhabitats and microclimatic conditions. Without sources of propagules (reproductive structures) within the managed landscape, however, even optimal habitat conditions would not be sufficient to ensure bryophyte diversity. To identify sources of propagules and ensure their protection, we used indices that were derived from a Digital Elevation Model (DEMs) and an airborne point cloud (LiDAR; Light Detection and Ranging) as explanatory variables to predict bryophyte biodiversity. Bryophytes were collected in the intensively managed Black Brook District of New Brunswick, Canada, in eight mature managed and unmanaged forest types (n = 38). Our results show a strong bryophyte community gradient between wetter stands (Cedar, riparian zone and Spruce-Fir) and drier stands (Tolerant Harwood and Plantation) forming two distinctive groups. Indices explaining bryophyte composition and richness were related to moisture (closest distance to a stream), canopy (canopy relief ratio, canopy closure and density) and microtopography (Topographic Position Index). Models obtained from these indices explained 75% of bryophyte composition and predicted composition with a certainty of 71% The predominance of the closest distance to a stream in our model reinforces the great importance of buffer along the hydrological network. Buffers represent a substantial propagule source for the landscape and notably increase its ecological connectivity. Although wetter sites had greater richness, the completely different composition find at drier sites suggest that biodiversity management efforts to maintain bryophytes should not be restricted to wetter stands. Our model demonstrates the potential of airborne LiDAR-derived indices as surrogates for field data in estimating and mapping bryophyte compositions to understand the variation in diversity across the managed landscape. This model can be used as a dynamic tool to target areas that represent the overall bryophyte diversity of the managed landscape to ensure protection of propagule sources and favors reestablishment.

This study used a novel approach combining biological, environmental, and ecosystem function data of the Logachev cold-water coral carbonate mound province to predictively map coral framework (bio)mass. A more accurate representation and quantification of cold-water coral reef ecosystem functions such as Carbon and Nitrogen stock and turnover were given by accounting for the spatial heterogeneity. Our results indicate that 45% is covered by dead and only 3% by live coral framework. The remaining 51%, is covered by fine sediments. It is estimated that 75,034–93,534 tons (T) of live coral framework is present in the area, of which ∼10% (7,747–9,316 T) consists of Cinorg and ∼1% (411–1,061 T) of Corg. A much larger amount of 3,485,828–4,357,435 T (60:1 dead:live ratio) dead coral framework contained ∼11% (418,299–522,892 T) Cinorg and

Soil aggregate stability (SAS) is a critical parameter of soil quality and its mapping can help determine erosion hotspots. Despite this importance, SAS is less documented in available literature due to limited number of analyzes besides being a time consuming. For this reason, many researchers have turned to alternative methods that often use readily available variables such as soil parameters or remote sensing indices to estimate this variable. In that framework, the aim of the present study focused on the investigation of the feasibile use of adapted Leo Breiman’s random forest algorithm (RF) to mapping different mean weight diameter (MWD) tests as an index of SAS (mechanical breakdown (MWDmb), slow wetting (MWDsw), fast wetting (MWDfw) and the mean of the three tests (MWDmean)). The model was built with 77 samples distributed in the three watersheds of the study area located at Settat Ben-Ahmed, in Morocco and with the use of several environmental variables such as soil parameters (organic matter and clay), remote sensing indices (band 2, band 3, band 4, band 5, normalized difference vegetation index (NDVI) and transformed normalized difference vegetation index (TNDVI)), topography (elevation, slope, curvature plane and the topographic wetness index (TWI)) along with additional categorical variables as geological maps, land use and soil classes. The results showed a good level of accuracy for the training phase (75% of samples) for the different tests (R2 > 0.92, RMSE and MAE < 0.15) and were satisfactory for the testing phase (25% of samples, R2 > 0.65, RMSE and MAE < 0.31). Also, organic matter, topography and geology were the most important parameters in the spatial prediction of SAS. Finally, the maps build during this study could be of great use to identify areas of less stable soils in the perspective for taking the necessary measures to improve their quality.

In deeply weathered terrains, with high geological diversity and scarce outcrops, the lateritic regolith features support mineral research. Predictive regolith mapping uses spatial modeling and airborne gamma-ray spectrometric data to locate weathered residual products, such as the ferruginous lateritic duricrusts found in tropical regions. Although the regolith materials exhibit qualitative similarities in their gamma-ray spectrometric responses in RGB (KThU) images, the Boolean and fuzzy approaches highlighted differences that allows for mapping the regolith weathering residual products on a wide geodiversity protoliths. Two inference systems were used, one identified the ferruginous lateritic duricrusts and oxisols derived from the felsic/pelitc and mafic rocks types (TP test procedures), and the other identified solely those derived exclusively from mafic rocks (TPM test procedures). Cutoff values for the conjunction eTh/K ∩ eU/K ∩ eTh ∩ K (TP-4) using Boolean logic, provided the most accurate regolith map (accuracy = 95%) among the various tests carried out and indicate that in 28.6% of the total study area, there are lateritic duricrust and oxisols relative to rocks, saprolites, mottled horizons and other soils type. The eTh cutoff value > 4.4 ppm was used to exclude mafic rocks and associated soils, and the K value < 1.3% to exclude rocks with eTh/K and eU/K values similar to the lateritic duricrusts and oxisols. For mapping the lateritic duricrusts derived exclusively from the mafic rocks the conjunction of eTh ∩ eU/K ∩ eTh/K ∩ K ∩ eU (TPM-3) fuzzified by the large membership function was the most efficient. The eU variable with values < 1.21 ppm helps improve the effectiveness of this procedure. The generated predictive map indicated that in about ∼12% (657.9 km2) of the total study area, there are lateritic duricrusts and oxisols derived from mafic rocks. Therefore, a successful method of mapping vast areas from complicated lateritic regolith mosaic derived from a large geodiversity is to apply specific plugins for data fuzzification, supported by observations from field outcrops. • Boolean and fuzzy techniques in airborne gamma-ray spectrometric data. • Regolith with extensive lateritic residual cover. • The gamma spectrometric data reflects the chemical composition of the regolith. • Prediction of lateritic duricrusts derived from mafic rocks. • Prediction of regolith derived from metasedimentary, felsic metaplutonic and metavolcanosedimentary rocks. [ABSTRACT FROM AUTHOR]

Aim: This study aimed to predict the alpha and beta plant diversity of an arid mountain based on environmental variables derived from remotely sensed and ground truth data. Location: Gebel Elba, Egypt. Methods: Based on 133 vegetation plots of 100 m2, we calculated alpha (Shannon index) and beta [the first ordination axis of nonmetric multidimensional scaling (NMDS1)] plant diversity. Generalized additive models (GAMs) were used to map alpha and beta diversity based on various environmental variables derived from a digital elevation model, the SoilGrids dataset, and very high resolution PlanetScope satellite imagery. The predictive models for alpha and beta diversity were mapped within the northern slopes of Gebel Elba. An ANOVA post hoc test was used to compare Shannon index and NMDS1 values among plant communities. Results: The selected models revealed the importance of altitude, landforms, solar insolation, catchment area, and modified soil‐adjusted vegetation index for Shannon diversity and NMDS1. The GAMs explained 54.9% of Shannon diversity and 80.6% of NMDS1. The predicted diversity maps showed that the mountainous area was more diverse and substantially different from the open desert. The post‐hoc test revealed a clear separation of mountain and desert vegetation. Conclusions: Employing remotely sensed variables combined with ground truth data offers great opportunities for exploring spatial patterns of biodiversity. By mapping alpha and beta diversity, it was possible to determine the spatial distribution of plant diversity in Gebel Elba; the results highlighted the importance of the wadi systems and higher slopes of this mountain area. We expect our findings can be generalized to similar arid mountains in the region.In this paper, we have created the first model‐based plant diversity maps for the arid mountain Gebel Elba. We found that topographical parameters and plant productivity played important roles in explaining predictive models of alpha and beta diversity on the northern slopes of this mountain. We expect that our findings can be generalized to similar arid mountains in the region. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we introduce two methods to forecast apartment burglaries that are based on repeat and near repeat victimization. While the first approach, the “heuristic method” generates buffer areas around each new apartment burglary, the second approach concentrates on forecasting near repeat chain links. These near repeat chain links are events that follow a near repeat pair of an originating and (near) repeat event that is close in space and in time. We name this approach the “near repeat chain method”. This research analyzes apartment burglaries from November 2013 to November 2016 in Vienna, Austria. The overall research goal is to investigate whether the near repeat chain method shows better prediction efficiencies (using a capture rate and the prediction accuracy index) while producing fewer prediction areas. Results show that the near repeat chain method proves to be the more efficient compared to the heuristic method for all bandwidth combinations analyzed in this research.

This study presents a novel approach resulting in the first cold-water coral reef biomass maps, used to assess associated ecosystem functions, such as carbon (C) stock and turnover. We focussed on two dominant ecosystem engineers at the Mingulay Reef Complex, the coral Lophelia pertusa (rubble, live and dead framework) and the sponge Spongosorites coralliophaga. Firstly, from combining biological (high-definition video, collected specimens), environmental (extracted from multibeam bathymetry) and ecosystem function (oxygen consumption rate values) data, we calculated biomass, C stock and turnover which can feed into assessments of C budgets. Secondly, using those values, we employed random forest modelling to predictively map whole-reef live coral and sponge biomass. The whole-reef mean biomass of S. coralliophaga was estimated to be 304 T (range 168–440 T biomass), containing 10 T C (range 5–18 T C) stock. The mean skeletal mass of the coral colonies (live and dead framework) was estimated to be 3874 T (range 507–9352 T skeletal mass), containing a mean of 209 T of biomass (range 26–515 T biomass) and a mean of 465 T C (range 60–1122 T C) stock. These estimates were used to calculate the C turnover rates, using respiration data available in the literature. These calculations revealed that the epi- and microbial fauna associated with coral rubble were the largest contributor towards C turnover in the area with a mean of 163 T C year−1 (range 149–176 T C year−1). The live and dead framework of L. pertusa were estimated to overturn a mean of 32 T C year−1 (range 4–93 T C year−1) and 44 T C year−1 (range 6–139 T C year−1), respectively. Our calculations showed that the Mingulay Reef overturned three to seven (with a mean of four) times more C than a soft-sediment area at a similar depth. As proof of concept, the supply of C needed from surface water primary productivity to the reef was inferred. Since 65–124 T C year−1 is supplied by natural deposition and our study suggested that a mean of 241 T C year−1 (range 160–400 T C year−1), was turned over by the reef, a mean of 117–176 T C year−1 (range 36–335 T C year−1) of the reef would therefore be supplied by tidal downwelling and/or deep-water advection. Our results indicate that monitoring and/or managing surface primary productivity would be a key consideration for any efforts towards the conservation of cold-water coral reef ecosystems. [ABSTRACT FROM AUTHOR]

Questions: Is it possible to map floristic gradients in heterogeneous boreal vegetation by using remote‐sensing data? Does a continuous vegetation map enable the creation of a spatially continuous map of seasonal permafrost soil thaw depth? Location: Bonanza Creek LTER, Fairbanks, Alaska, USA. Methods: Vegetation records are subjected to an ordination to extract the predominant floristic gradient. The ordination scores are then extrapolated using Sentinel 2 imagery and a digital elevation model (DEM). As the relation between vegetation pattern and seasonal thaw depth was confirmed in this study, the spatial distribution of ordination scores is then used to predict seasonal thaw depth over the same area. Results: The first dimension of the ordination space separates species corresponding to moist and cold soil conditions from species associated with well‐drained soils. This floristic gradient was successfully mapped within the sampled plant communities. The extrapolated thaw depths follow the typical distribution along a topographical and geomorphological gradient for this region. Besides vegetation information also DEM derivatives show high contributions to the thaw depth modeling. Conclusion: We demonstrate that floristic gradient mapping in boreal vegetation is possible. The accuracy of the thaw depth prediction model is comparable to that in previous analyses but uses a more parsimonious set of predictors, underlining the efficacy of this approach. [ABSTRACT FROM AUTHOR]

The tropical dry forests of NW Peru are heavily shaped by the El Niño Southern Oscillation (ENSO), where especially El Niño brings rain to arid to semi‐arid areas. However, the resulting effects on biodiversity patterns remain largely unknown as well as the effect of environmental variables on the floristic composition under varying rainfall patterns. Therefore, we studied the spatio‐temporal effects of different ENSO episodes on floristic biodiversity along a climatic gradient ranging from the coastal desert to the Andean foothills. We sampled 50 vegetation plots in four years representing different ENSO episodes. To highlight the spatio‐temporal changes in floristic composition and beta diversity across ENSO episodes, we predicted ordination scores with a Generalized Additive Model. We applied variation partitioning to test if topographic or edaphic variables gained in importance during more humid ENSO episodes. Additionally, we executed an irrigation–fertilization experiment to quantify the beneficial effects of the water–nutrient interaction under different simulated ENSO rainfall scenarios. Plant species richness increased under humid conditions during the humid La Niña (2012) and the moderate El Niño (2016), and slightly decreased under the very humid conditions during the coastal El Niño (2017). The spatial prediction revealed that specific vegetation formations became more pronounced with increasing water input, but that a large water surplus led to the disruption of the strict order along the climatic gradient. Edaphic and topographic variables gained in importance with increased water availability (2012 and 2016), however, this effect was not further amplified under very wet conditions (2017). The experiment showed that plant cover under Super Niño conditions was three times higher when fertilized. Overall, our spatial predictions concede detailed insights into spatio‐temporal ecosystem dynamics in response to varying rainfall caused by different ENSO episodes while the results of the experiment can support farmers regarding a sustainable agrarian management. [ABSTRACT FROM AUTHOR]

Monitoring tillage practices is important for explaining soil quality and yield trends, and their impact on environmental quality. However, a common problem in sustainable residue management is scarcity of accurate residue maps. Because predictive insights on soil quality dynamics across a spatial domain are vital, this entry explicates on a new remote sensing-based technique for assessing surface residue cover. Here, an empirical model for mapping surface residue cover was created by integrating line-transect % residue cover field measurements with information gleaned from ground spectroradiometers and Advanced Wide-Field Sensor (AWiFS) satellite imagery. This map was validated using non-photosynthetic vegetation (NPV) fractional component extracted by spectral mixture analysis (SMA). SMA extracts fractional components of sensed signals in imagery, which within agricultural fields are NPV, green vegetation, bare soil, and shade. A stepwise linear regression between residue estimates by line transect and map generated using satellite imagery had R2 = 87%. Upon map categorization according to surface residue for a single AWiFS imagery encompassing an area of 836,868 ha, but focused on corn (Zea mays) fields within South Dakota, revealed that <4% of these corn fields had >15% surface residue cover left in the field by November 2009. Findings such as these may guide policy on soil quality, which is directly correlated with residue management. In the future, the spatial distribution of surface residues remaining after harvest in field planted with other crops and other seasons will be mapped. Besides, the efficacy of integrating hyperspectral sensor data to enhance accuracy will be investigated. [ABSTRACT FROM AUTHOR]

Predictive mapping of environment is an important means for environment assessment and management. The selection of predictor variables (or environmental covariates) is the first and key step in predictive mapping. A number of machine learning and statistical models have been developed to select what and how many environmental covariates in a wide range of predictive mapping. Nevertheless, those models require a large amount of field data for model training and calibration, which can be problematic in applying to the areas with no or very limited field data available. To overcome the shortcoming, this paper proposes the most similar case method for selecting environmental covariates for predictive mapping. First, we describe the basic idea and the development procedures of the most similar case method; second, as an experimental test, we employ the proposed method to select the topographic covariates for inputting to the predictive soil mapping; third, we evaluate the effectiveness of the proposed method in the designed experiment using the leave-one-out cross-validation method. In total, 191 evaluation cases are included in the experimental case base and the test results show that 58.7% of the topographic covariates originally used in each evaluation case are correctly selected by the proposed method, which suggests that the proposed most-similar-case method perform reasonably well even with a relatively limited size of the case base. The future work should include the selection of other types of environmental covariates (e.g., climate, organism, etc.) and the development of an automatic method to extract the existing application cases from literature. [ABSTRACT FROM AUTHOR]

Rajasthan generates 1055 million litres per day as wastewater, out of which 27 million litres is treated and nearly 1028 million litres untreated wastewater is discharged in various water resources. The present study is based on the impact of Kota stone or limestone mining on water resources. Among those villages and census towns, experiencing mining activity, a total of 26 surface water and groundwater samples were tested and analysed. Mining waste often creates eutrophication, toxification, temporary hardness and sometimes permanent hardness. The mining belt was 17.54 km² in the year 2000 which further increased to 24.25 km² in the year 2018. The parameters analysed were pH, EC, TDS, alkalinity, total hardness, calcium and magnesium hardness, DO, COD, chloride, sodium and potassium. The predictive mapping for the mining belt was executed in Arc GIS software using Inverse Distance Weightage (IDW) method. The mean of pH was 9.13, TDS 457.12 mg/L, total hardness 593.52 mg/L, calcium hardness is 205.54 mg/L, magnesium hardness 387.53 mg/L, COD 442.2 mg/L, Na+ 139.9 mg/L, K+ 19.40 mg/L, Cl- 318.29, DO 3.04mg/L and alkalinity 14.02 mg/L. [ABSTRACT FROM AUTHOR]

A new approach to modeling spatiotemporal dynamics in forest-steppe ecosystems amid climate change is proposed. This approach involves a morphometric analysis of the digital elevation model (SRTM). The resultant forecast is based on an analysis of the geological–geomorphological framework and takes into account the functional significance of specific relief forms and types and their impacts on the static and dynamic properties of ecosystems. Plain areas are most resistant to climatic changes, while areas with lacustrine relief are most vulnerable both under humidization and aridization conditions. Predictive mapping based on GIS and remote sensing (RS) data confirms the hypothesis of the focal–discrete nature of spatial changes observed in the forest-steppe zone of Western Siberia amid different climatic trends. [ABSTRACT FROM AUTHOR]

Understanding how animals use the space in which they are distributed is important for guiding management decisions in conservation, especially where human disturbance can be spatially managed. Here we applied distribution modelling to examine common dolphin (Delphinus sp.) habitat use in the Hauraki Gulf (36°S, 175°E), New Zealand. Given the known importance of the area for foraging and nursing, we assessed which variables affect Delphinus occurrence based on generalised additive models (GAMs), and modelled probability of encounter. Behavioural information was included to assess habitat use by feeding and nursing groups and determine whether persistent hotspots for such activities could be identified and meaningfully used as a spatial management tool. Using data collected from dedicated boat surveys during 2010–2012, depth and sea surface temperature (SST) were frequently identified as important variables. Overall, seasonal predictive occurrence maps for the larger population resembled predictive maps of feeding groups more than nursery groups, suggesting prey availability has important implications for the distribution of Delphinus in this region. In this case, static spatial exclusions would not be the best management option as the core areas of use identified for these activities were large and shifted temporally. It appears that at the scale examined, most of the Hauraki Gulf is important for feeding and nursing rather than specific smaller regions being used for these functions. In cases where static management is not the optimal tool, as suggested here for a highly mobile species, a dynamic approach requires more than a boundary line on a map. [ABSTRACT FROM AUTHOR]

Multi-element geochemical surveys of rocks, soils, stream/lake/floodplain sediments and regolith are typically carried out at continental, regional and local scales. The chemistry of these materials is defined by their primary mineral assemblages and their subsequent modification by comminution and weathering. Modern geochemical datasets represent a multi-dimensional geochemical space that can be studied using multivariate statistical methods from which patterns reflecting geochemical/geological processes are described (process discovery). These patterns form the basis from which probabilistic predictive maps are created (process validation). Processing geochemical survey data requires a systematic approach to effectively interpret the multi-dimensional data in a meaningful way. Problems that are typically associated with geochemical data include closure, missing values, censoring, merging, levelling different datasets and adequate spatial sample design. Recent developments in advanced multivariate analytics, geospatial analysis and mapping provide an effective framework to analyse and interpret geochemical datasets. Geochemical and geological processes can often be recognized through the use of data discovery procedures such as the application of principal component analysis. Classification and predictive procedures can be used to confirm lithological variability, alteration and mineralization. Geochemical survey data of lake/till sediments from Canada and of floodplain sediments from Australia show that predictive maps of bedrock and regolith processes can be generated. Upscaling a multivariate statistics-based prospectivity analysis for arc-related Cu–Au mineralization from a regional survey in the southern Thomson Orogen in Australia to the continental scale, reveals a number of regions with a similar (or stronger) multivariate response and hence potentially similar (or higher) mineral potential throughout Australia. Thematic collection: This article is part of the Exploration 17 collection available at: https://www.lyellcollection.org/cc/exploration-17 [ABSTRACT FROM AUTHOR]

Predictive mapping of seabed sediments based on multibeam bathymetric (BM), and backscatter (BS) data is effective for mapping the spatial distribution of the substrate. A robust modeling technique, the random forest decision tree (RFDT), was used to predict the seabed sediments in an area of the Joseph Bonaparte Gulf, Northern Australia, using the multibeam data and seabed sediment samples collected simultaneously. The results showed that: (1) Using multibeam bathymetry data in addition to multibeam backscatter data improves the prediction performance of the RFDT. In comparison to only multibeam backscatter data, the prediction performance achieved a ~10% improvement in sediment properties; it achieved a ~44.45% improvement of overall accuracy in sediment types, and a ~0.55 improvement in Kappa. (2) The underlying relationships between sediment properties and multibeam data show that there is an opposite non-linear correlation between sediment property-BS and sediment property-BM. For example, there is an obvious negative relationship between %mud-BS at incidence angles of 13° and 21°, but the relationship between %mud-BM is positive. As such, the RFDT is a useful and well-performing method in predicting the relationship between sediment properties and multibeam data and in predicting the distribution of sediment properties and types. However, the sediment prediction method in deep-water areas with high gravel content needs to be further evaluated.

Volunteered geographic information (VGI) contains valuable field observations that represent the spatial distribution of geographic phenomena. As such, it has the potential to provide regularly updated low-cost field samples for predictively mapping the spatial variations of geographic phenomena. The predictive mapping of geographic phenomena often requires representative samples for high mapping accuracy, but samples consisting of VGI observations are often not representative as they concentrate on specific geographic areas (i.e. spatial bias) due to the opportunistic nature of voluntary observation efforts. In this article, we propose a representativeness-directed approach to mitigate spatial bias in VGI for predictive mapping. The proposed approach defines and quantifies sample representativeness by comparing the probability distributions of sample locations and the mapping area in the environmental covariate space. Spatial bias is mitigated by weighting the sample locations to maximize their representativeness. The approach is evaluated using species habit suitability mapping as a case study. The results show that the accuracy of predictive mapping using weighted sample locations is higher than using unweighted sample locations. A positive relationship between sample representativeness and mapping accuracy is also observed, suggesting that sample representativeness is a valid indicator of predictive mapping accuracy. This approach mitigates spatial bias in VGI to improve predictive mapping accuracy. [ABSTRACT FROM AUTHOR]

Highlights • Models the geographical distribution of two mosquito genera, vectors of human pathogens, in rural Kenya using GIS techniques. • Seeks environmental and anthropogenic factors affecting mosquito distributions. • Provides predictive risk maps for Culex and Stegomyia distributions. • Shows the applicability of the predictive models produced by The Biomod2 ensemble platform in R. • Future work focuses on studying viruses of sampled mosquitoes using GIS applications. Abstract Mosquitoes are vectors for numerous pathogens, which are collectively responsible for millions of human deaths each year. As such, it is vital to be able to accurately predict their distributions, particularly in areas where species composition is unknown. Species distribution modeling was used to determine the relationship between environmental, anthropogenic and distance factors on the occurrence of two mosquito genera, Culex Linnaeus and Stegomyia Theobald (syn. Aedes), in the Taita Hills, southeastern Kenya. This study aims to test whether any of the statistical prediction models produced by the Biomod2 package in R can reliably estimate the distributions of mosquitoes in these genera in the Taita Hills; and to examine which factors best explain their presence. Mosquito collections were acquired from 122 locations between January–March 2016 along transects throughout the Taita Hills. Environmental-, anthropogenic- and distance-based geospatial data were acquired from the Taita Hills geo-database, satellite- and aerial imagery and processed in GIS software. The Biomod2 package in R, intended for ensemble forecasting of species distributions, was used to generate predictive models. Slope, human population density, normalized difference vegetation index, distance to roads and elevation best estimated Culex distributions by a generalized additive model with an area under the curve (AUC) value of 0.791. Mean radiation, human population density, normalized difference vegetation index, distance to roads and mean temperature resulted in the highest AUC (0.708) value in a random forest model for Stegomyia distributions. We conclude that in the process towards more detailed species-level maps, with our study results, general assumptions can be made about the distribution areas of Culex and Stegomyia mosquitoes in the Taita Hills and the factors which influence their distribution. [ABSTRACT FROM AUTHOR]

Environmental demands for greener, more sustainable urban areas are increasing worldwide and are particularly magnified in megacities such as Mexico City, the largest city in North America. Strategic land use planning initiatives aimed at integrated and sustainable urban green infrastructure across broader city landscapes require an understanding of urban environments, their urban forest, and their spatial distribution. However, this approach has not been widely used in megacities. Here, we use the urban biotope approach to address the knowledge gap and help inform land-use planning to support greater urban forest sustainability in Mexico City. Two levels of biotope classes, broader-level (3 classes defined by canopy cover) and finer-level (7 classes describing urban forest characteristics), were modeled and mapped across the city. Statistical modeling was done using a random forest algorithm, a set of uncorrelated predictor variables, and field observations to predict biotope presence. Models of the broader-level biotopes yielded better prediction results (AUC > 0.90) than the finer-level biotopes (AUC values from 0.61 to 0.88). Classification errors ranged from 6.5 to 100 %, indicating different levels of predictive confidence. Of all mapping units of the broader-level biotopes 87 % were predicted while of the finer-level biotopes only 12.7 %. As the finer-level predictions often overlap, more field data and better predictors are required to accurately capture biotope classes. The modeled biotopes can inform urban planning across spatial scales, from the neighborhood and borough level to the city-wide scale, and across the residential-commercial land use, which requires more greening efforts. Results suggest that more field and spatial data, and a smaller grain size, can improve biotope models and maps to better represent the variation in urban forest characteristics across Mexico City. Biotope maps are a practical tool to guide spatial planning, identify priority areas for canopy protection and tree planting, and set tree diversification goals. • Predictive modeling is used to develop biotope maps for an urban landscape. • Biotope maps can guide multi-scale urban forest spatial planning in Mexico City. • Biotopes models reflect specific urban forest characteristics but need improvement. • Machine learning supports cost-efficient decision-making for urban forest planning. • Discussion on the potential and limitations of predictive modeling for biotope mapping. [ABSTRACT FROM AUTHOR]

Soil organic carbon (SOC) is an important indicator of soil quality and directly determines soil fertility. Hence, understanding its spatial distribution and controlling factors is necessary for efficient and sustainable soil nutrient management. In this study, machine learning algorithms including artificial neural network (ANN), support vector machine (SVM), cubist regression, random forests (RF), and multiple linear regression (MLR) were chosen for advancing the prediction of SOC. A total of sixty (n = 60) soil samples were collected within the research area at 30 cm soil depth and measured for SOC content using the Walkley–Black method. From these samples, 80% were used for model training and 21 auxiliary data were included as predictors. The predictors include effective cation exchange capacity (ECEC), base saturation (BS), calcium to magnesium ratio (Ca_Mg), potassium to magnesium ratio (K_Mg), potassium to calcium ratio (K_Ca), elevation, plan curvature, total catchment area, channel network base level, topographic wetness index, clay index, iron index, normalized difference build-up index (NDBI), ratio vegetation index (RVI), soil adjusted vegetation index (SAVI), normalized difference vegetation index (NDVI), normalized difference moisture index (NDMI) and land surface temperature (LST). Mean absolute error (MAE), root-mean-square error (RMSE) and R2 were used to determine the model performance. The result showed the mean SOC to be 1.62% with a coefficient of variation (CV) of 47%. The best performing model was RF (R2 = 0.68) followed by the cubist model (R2 = 0.51), SVM (R2 = 0.36), ANN (R2 = 0.36) and MLR (R2 = 0.17). The soil nutrient indicators, topographic wetness index and total catchment area were considered an indicator for spatial prediction of SOC in flat homogenous topography. Future studies should include other auxiliary predictors (e.g., soil physical and chemical properties, and lithological data) as well as cover a broader range of soil types to improve model performance.

Soil salinization, one of the most severe global land degradation problems, leads to the loss of arable land and declines in crop yields. Monitoring the distribution of salinized soil and degree of salinization is critical for management, remediation, and utilization of salinized soil; however, there is a lack of thorough assessment of various data sources including remote sensing and landscape characteristics for estimating soil salinity in arid and semi-arid areas. The overall goal of this study was to develop a framework for estimating soil salinity in diverse landscapes by fusing information from satellite images, landscape characteristics, and appropriate machine learning models. To explore the spatial distribution of soil salinity in southern Xinjiang, China, as a case study, we obtained 151 soil samples in a field campaign, which were analyzed in laboratory for soil electrical conductivity. A total of 35 indices including remote sensing classifiers (11), terrain attributes (3), vegetation spectral indices (8), and salinity spectral indices (13) were calculated or derived and correlated with soil salinity. Nine were used to model and estimate soil salinity using four predictive modelling approaches: partial least squares regression (PLSR), convolutional neural network (CNN), support vector machine (SVM) learning, and random forest (RF). Testing datasets were divided into vegetation-covered and bare soil samples and were used for accuracy assessment. The RF model was the best regression model in this study, with R2 = 0.75, and was most effective in revealing the spatial characteristics of salt distribution. Importance analysis and path modeling of independent variables indicated that environmental factors and soil salinity indices including digital elevation model (DEM), B10, and green atmospherically resistant vegetation index (GARI) showed the strongest contribution in soil salinity estimation. This showed a great promise in the measurement and monitoring of soil salinity in arid and semi-arid areas from the integration of remote sensing, landscape characteristics, and using machine learning model.

Abstract: We have determined the informational value of the Normalized Difference Vegetation Index (NDVI) for predictive mapping of the content of available lithium in the arable horizon of soils of different slope positions: the first (280-310 m) and the second (240-280 m) altitudinal levels. The NDVI is not informative for the diagnostics or mapping of the content of available lithium in soils of small drainage valleys, the width of which is smaller than the resolution of the satellite image (30 m). In the regression model, the NDVI explains 28% of the variation in the content of available lithium in soils. Based on this model, a predictive map of the content of available lithium in soils has been compiled. Data on the spatial distribution pattern of the NDVI calculated based on a Landsat 8 satellite image (resolution of 30 m) were used as an indicator and the cartographical basis for digital mapping. The accuracy of the prediction of the content of available lithium in soils is good (MAPE is 16.9%). It has been revealed that the NDVI values and the content of available lithium in soils of the first altitudinal level are higher than in the second. The differences between NDVI in the drainage valley and on the first altitudinal level are not insignificant. [ABSTRACT FROM AUTHOR]

In this paper, we introduce two methods to forecast apartment burglaries that are based on repeat and near repeat victimization. While the first approach, the “heuristic method” generates buffer areas around each new apartment burglary, the second approach concentrates on forecasting near repeat chain links. These near repeat chain links are events that follow a near repeat pair of an originating and (near) repeat event that is close in space and in time. We name this approach the “near repeat chain method”. This research analyzes apartment burglaries from November 2013 to November 2016 in Vienna, Austria. The overall research goal is to investigate whether the near repeat chain method shows better prediction efficiencies (using a capture rate and the prediction accuracy index) while producing fewer prediction areas. Results show that the near repeat chain method proves to be the more efficient compared to the heuristic method for all bandwidth combinations analyzed in this research. [ABSTRACT FROM AUTHOR]