Search

The Mediterranean Basin has historically been subject to alien plant invasions that threaten its unique biodiversity. This seasonally dry and densely populated region is undergoing severe climatic and socioeconomic changes, and it is unclear whether these changes will worsen or mitigate plant invasions. Predictions are often biased, as species may not be in equilibrium in the invaded environment, depending on their invasion stage and ecological characteristics. To address future predictions uncertainty, we identified invasion hotspots across multiple biased modelling scenarios and ecological characteristics of successful invaders. We selected 92 alien plant species widespread in Mediterranean Europe and compiled data on their distribution in the Mediterranean and worldwide. We combined these data with environmental and propagule pressure variables to model global and regional species niches, and map their current and future habitat suitability. We identified invasion hotspots, examined their potential future shifts, and compared the results of different modelling strategies. Finally, we generalised our findings by using linear models to determine the traits and biogeographic features of invaders most likely to benefit from global change. Currently, invasion hotspots are found near ports and coastlines throughout Mediterranean Europe. However, many species occupy only a small portion of the environmental conditions to which they are preadapted, suggesting that their invasion is still an ongoing process. Future conditions will lead to declines in many currently widespread aliens, which will tend to move to higher elevations and latitudes. Our trait models indicate that future climates will generally favour species with conservative ecological strategies that can cope with reduced water availability, such as those with short stature and low specific leaf area. Taken together, our results suggest that in future environments, these conservative aliens will move farther from the introduction areas and upslope, threatening mountain ecosystems that have been spared from invasions so far. [ABSTRACT FROM AUTHOR]

Sustainable agriculture revolves around soil organic carbon (SOC), which is essential for numerous soil functions and ecological attributes. Farmers are interested in conserving and adding additional soil organic carbon to certain fields in order to improve soil health and productivity. The relationship between soil and environment that has been discovered and standardized throughout time has enhanced the progress of digital soil-mapping techniques; therefore, a variety of machine learning techniques are used to predict soil properties. Studies are thriving at how effectively each machine learning method maps and predicts SOC, especially at high spatial resolutions. To predict SOC of soil at a 30 m resolution, four machine learning models—Random Forest, Support Vector Machine, Adaptive Boosting, and k-Nearest Neighbour were used. For model evaluation, two error metrics, namely R2 and RMSE have been used. The findings demonstrated that the calibration and validation sets’ descriptive statistics sufficiently resembled the entire set of data. The range of the calculated SOC content was 0.06 to 1.76 %. According to the findings of the study, Random Forest showed good results for both cases, i.e. evaluation using cross validation and without cross validation. Using cross validation, RF confirmed highest R2 as 0.5278 and lowest RMSE as 0.1683 for calibration dataset while without cross validation it showed R2 as 0.8612 and lowest RMSE as 0.0912 for calibration dataset. The generated soil maps will help farmers adopt precise knowledge for decisions that will increase farm productivity and provide food security through the sustainable use of nutrients and the agricultural environment. [ABSTRACT FROM AUTHOR]

Geospatial vector data with semantic annotations are a promising but complex data source for spatial prediction tasks such as land use and land cover (LULC) classification. These data describe the geometries and the types (i.e., semantics) of geo-objects, such as a Shop or an Amenity. Unlike raster data, which are commonly used for such prediction tasks, geospatial vector data are irregular and heterogenous, making it challenging for deep neural networks to learn based on them. This work tackles this problem by introducing novel encodings which quantify the geospatial vector data allowing deep neural networks to learn based on them, and to spatially predict. These encodings were evaluated in this work based on a specific use case, namely LULC classification. We therefore classified LULC based on the different encodings as input and an attention-based deep neural network (called Perceiver). Based on the accuracy assessments, the potential of these encodings is compared. Furthermore, the influence of the object semantics on the classification performance is analyzed. This is performed by pruning the ontology, describing the semantics and repeating the LULC classification. The results of this work suggest that the encoding of the geography and the semantic granularity of geospatial vector data influences the classification performance overall and on a LULC class level. Nevertheless, the proposed encodings are not restricted to LULC classification but can be applied to other spatial prediction tasks too. In general, this work highlights that geospatial vector data with semantic annotations is a rich data source unlocking new potential for spatial predictions. However, we also show that this potential depends on how much is known about the semantics, and how the geography is presented to the deep neural network. [ABSTRACT FROM AUTHOR]

Abstract Background Culicoides biting midges transmit viruses resulting in disease in ruminants and equids such as bluetongue, Schmallenberg disease and African horse sickness. In the past decades, these diseases have led to important economic losses for farmers in Europe. Vector abundance is a key factor in determining the risk of vector-borne disease spread and it is, therefore, important to predict the abundance of Culicoides species involved in the transmission of these pathogens. The objectives of this study were to model and map the monthly abundances of Culicoides in Europe. Methods We obtained entomological data from 904 farms in nine European countries (Spain, France, Germany, Switzerland, Austria, Poland, Denmark, Sweden and Norway) from 2007 to 2013. Using environmental and climatic predictors from satellite imagery and the machine learning technique Random Forests, we predicted the monthly average abundance at a 1 km2 resolution. We used independent test sets for validation and to assess model performance. Results The predictive power of the resulting models varied according to month and the Culicoides species/ensembles predicted. Model performance was lower for winter months. Performance was higher for the Obsoletus ensemble, followed by the Pulicaris ensemble, while the model for Culicoides imicola showed a poor performance. Distribution and abundance patterns corresponded well with the known distributions in Europe. The Random Forests model approach was able to distinguish differences in abundance between countries but was not able to predict vector abundance at individual farm level. Conclusions The models and maps presented here represent an initial attempt to capture large scale geographical and temporal variations in Culicoides abundance. The models are a first step towards producing abundance inputs for R0 modelling of Culicoides-borne infections at a continental scale.

Geospatial vector data with semantic annotations are a promising but complex data source for spatial prediction tasks such as land use and land cover (LULC) classification. These data describe the geometries and the types (i.e., semantics) of geo-objects, such as a Shop or an Amenity. Unlike raster data, which are commonly used for such prediction tasks, geospatial vector data are irregular and heterogenous, making it challenging for deep neural networks to learn based on them. This work tackles this problem by introducing novel encodings which quantify the geospatial vector data allowing deep neural networks to learn based on them, and to spatially predict. These encodings were evaluated in this work based on a specific use case, namely LULC classification. We therefore classified LULC based on the different encodings as input and an attention-based deep neural network (called Perceiver). Based on the accuracy assessments, the potential of these encodings is compared. Furthermore, the influence of the object semantics on the classification performance is analyzed. This is performed by pruning the ontology, describing the semantics and repeating the LULC classification. The results of this work suggest that the encoding of the geography and the semantic granularity of geospatial vector data influences the classification performance overall and on a LULC class level. Nevertheless, the proposed encodings are not restricted to LULC classification but can be applied to other spatial prediction tasks too. In general, this work highlights that geospatial vector data with semantic annotations is a rich data source unlocking new potential for spatial predictions. However, we also show that this potential depends on how much is known about the semantics, and how the geography is presented to the deep neural network.

Background: Culicoides biting midges transmit viruses resulting in disease in ruminants and equids such as bluetongue, Schmallenberg disease and African horse sickness. In the past decades, these diseases have led to important economic losses for farmers in Europe. Vector abundance is a key factor in determining the risk of vector-borne disease spread and it is, therefore, important to predict the abundance of Culicoides species involved in the transmission of these pathogens. The objectives of this study were to model and map the monthly abundances of Culicoides in Europe. Methods: We obtained entomological data from 904 farms in nine European countries (Spain, France, Germany, Switzerland, Austria, Poland, Denmark, Sweden and Norway) from 2007 to 2013. Using environmental and climatic predictors from satellite imagery and the machine learning technique Random Forests, we predicted the monthly average abundance at a 1 km2 resolution. We used independent test sets for validation and to assess model performance. Results: The predictive power of the resulting models varied according to month and the Culicoides species/ensembles predicted. Model performance was lower for winter months. Performance was higher for the Obsoletus ensemble, followed by the Pulicaris ensemble, while the model for Culicoides imicola showed a poor performance. Distribution and abundance patterns corresponded well with the known distributions in Europe. The Random Forests model approach was able to distinguish differences in abundance between countries but was not able to predict vector abundance at individual farm level. Conclusions: The models and maps presented here represent an initial attempt to capture large scale geographical and temporal variations in Culicoides abundance. The models are a first step towards producing abundance inputs for R0 modelling of Culicoides-borne infections at a continental scale. [ABSTRACT FROM AUTHOR]

For marine fishes of commercial interest, defining how individuals vary in certain attributes, through ontogeny, and across space and time, can help expose the structure of harvested populations and guide their management. The chemical markers locked within otoliths are, in principle, ideal for such applications, providing life-long, bio-environmental records for individuals and populations. However, substantial uncertainty still surrounds how intrinsic (i.e., within-individual) and extrinsic (i.e., environmental) processes shape chemical incorporation into otoliths, constraining our ability to predict marker variability when sampling is patchy. To tackle these issues, we measured otolith elemental (Li, Mg, Ca, Mn, Zn, Sr, Ba) and stable isotopic (δ13C, δ18O) concentrations from juvenile Atlantic herring (Clupea harengus) captured in Icelandic nursery sites, and built models to isolate ontogenetic (i.e., age-, growth-related) from environmental influences on otolith chemistry. Otolith Li:Ca, Mg:Ca, Mn:Ca and Sr:Ca declined with age within sites. Age slopes differed among sites for Li:Ca and Mg:Ca, and Sr:Ca was lower in larger fish within an age-class. Individual-level variation (i.e., within site, within age-class) was high for all markers, our models highlighting the importance of temperature and salinity (or proxies these represent) in explaining population-level δ13C and δ18O trends. Age- and year-specific predictions for each marker, at each site, generally accorded well with empirical observations, providing inference on island-wide heterogeneity in otolith chemistry across the juveniles' full distributional range. Such 'isoscapes', generated from mechanistically-focused models as presented here, might benefit investigations of population structure for other exploited species, particularly where sampling limitations hamper fishery-management efforts. [ABSTRACT FROM AUTHOR]

Fuzzy set theory has shown potential for reducing uncertainty as a result of data sparsity and also provides advantages for quantifying gradational changes like those of pollutant concentrations through fuzzy clustering based approaches. The ability to lower the sampling frequency and perform laboratory analyses on fewer samples, yet still produce an adequate pollutant distribution map, would reduce the initial cost of new remediation projects. To assess the ability of fuzzy modeling to make spatial predictions using fewer sample points, its predictive ability was compared with the ordinary kriging (OK) and inverse distance weighting (IDW) methods under increasingly sparse data conditions. This research used a Takagi–Sugeno (TS) fuzzy modelling approach with fuzzy c-means (FCM) clustering to make spatial predictions of the lead concentrations in soil. The performance of the TS model was very dependent on the number of outliers in the respective validation set. For modeling under sparse data conditions, the TS fuzzy modeling approach using FCM clustering and constant width Gaussian shaped membership functions did not show any advantages over IDW and OK for the type of data tested. Therefore, it was not possible to speculate on a possible reduction in sampling frequency for delineating the extent of contamination for new remediation projects.

Spatial predictive methods are increasingly being used to generate predictions across various disciplines in environmental sciences. Accuracy of the predictions is critical as they form the basis for environmental management and conservation. Therefore, improving the accuracy by selecting an appropriate method and then developing the most accurate predictive model(s) is essential. However, it is challenging to select an appropriate method and find the most accurate predictive model for a given dataset due to many aspects and multiple factors involved in the modeling process. Many previous studies considered only a portion of these aspects and factors, often leading to sub-optimal or even misleading predictive models. This study evaluates a spatial predictive modeling process, and identifies nine major components for spatial predictive modeling. Each of these nine components is then reviewed, and guidelines for selecting and applying relevant components and developing accurate predictive models are provided. Finally, reproducible examples using spm, an R package, are provided to demonstrate how to select and develop predictive models using machine learning, geostatistics, and their hybrid methods according to predictive accuracy for spatial predictive modeling; reproducible examples are also provided to generate and visualize spatial predictions in environmental sciences. [ABSTRACT FROM AUTHOR]

Abstract Land cover change has been identified as a major driver of global environmental change. In agricultural landscapes that are undergoing land cover conversions, there is need to investigate the nature, extent and pattern of these conversions. This helps to assess the resultant forest cover changes and their consequences, as well as to help inform sustainable biodiversity conservation measures. This study tests whether and in what way agricultural expansion drives land cover change in resettled landscapes of Zimbabwe now and in the future. Classified Landsat Thematic Mapper images of 2000, 2009, and Landsat 8 image of 2013 were analyzed, using the widely used Markov modelling framework, to determine land use and land cover changes between the year 2000 and 2013, as well as future land cover dynamics in the area of interest. Specifically, land cover data was used to simulate land cover transitions up to 2050, using Markov chain model. Historic land cover modelling and future spatial predictions revealed increases in cropland accompanied by decreases in plantation and forests. This study concludes that agriculture is a major driver of land cover changes, especially forest cover changes in newly resettled agricultural landscapes. Highlights • Agriculture is a major driver of land cover changes. • Historic land cover modelling and future predictions revealed increases in cropland and a decreases in plantation forests. • Land cover change has been identified as a major driver of global environmental change. • Validation of the Markov transition model showed that the model is able to predict the trend of land cover transitions. [ABSTRACT FROM AUTHOR]

Spatial predictive methods are increasingly being used to generate predictions across various disciplines in environmental sciences. Accuracy of the predictions is critical as they form the basis for environmental management and conservation. Therefore, improving the accuracy by selecting an appropriate method and then developing the most accurate predictive model(s) is essential. However, it is challenging to select an appropriate method and find the most accurate predictive model for a given dataset due to many aspects and multiple factors involved in the modeling process. Many previous studies considered only a portion of these aspects and factors, often leading to sub-optimal or even misleading predictive models. This study evaluates a spatial predictive modeling process, and identifies nine major components for spatial predictive modeling. Each of these nine components is then reviewed, and guidelines for selecting and applying relevant components and developing accurate predictive models are provided. Finally, reproducible examples using spm, an R package, are provided to demonstrate how to select and develop predictive models using machine learning, geostatistics, and their hybrid methods according to predictive accuracy for spatial predictive modeling; reproducible examples are also provided to generate and visualize spatial predictions in environmental sciences.

Seabed sediment predictions at regional and national scales in Australia are mainly based on bathymetry-related variables due to the lack of backscatter-derived data. In this study, we applied random forests (RFs), hybrid methods of RF and geostatistics, and generalized boosted regression modelling (GBM), to seabed sand content point data and acoustic multibeam data and their derived variables, to develop an accurate model to predict seabed sand content at a local scale. We also addressed relevant issues with variable selection. It was found that: (1) backscatter-related variables are more important than bathymetry-related variables for sand predictive modelling; (2) the inclusion of highly correlated predictors can improve predictive accuracy; (3) the rank orders of averaged variable importance (AVI) and accuracy contribution change with input predictors for RF and are not necessarily matched; (4) a knowledge-informed AVI method (KIAVI2) is recommended for RF; (5) the hybrid methods and their averaging can significantly improve predictive accuracy and are recommended; (6) relationships between sand and predictors are non-linear; and (7) variable selection methods for GBM need further study. Accuracy-improved predictions of sand content are generated at high resolution, which provide important baseline information for environmental management and conservation.

In this study, a 150 mJ laser-induced breakdown spectroscopy (LIBS) system was assessed to determine calcium content in infant formula (IF) samples. LIBS is a promising emission spectroscopic technique for elemental analysis, which offers advantages over conventional methods such as real-time analyses, little to no sample preparation and ease of use. The aim of this work was to evaluate the feasibility of LIBS as an at-line tool for IF manufacturing. To this end, IF mixtures with varying content of calcium were prepared over a range (approx. 1.5–7 mg/g of calcium) selected to be in conformity with the guidelines provided by the Codex Alimentarius Commission. Atomic Absorption Spectroscopy (AAS) was used as the reference method for mineral content determination in IF. Partial least squares regression (PLSR) was applied to the LIBS spectral data to develop a calibration model for calcium content quantification. A validation approach was then carried out to investigate the robustness of the model, which showed a good fit with values of R 2 cv of 0.90 for cross-validation and a R 2 p 0.85 for external validation. Low values for root mean square errors of cross-validation (RMSECV) and prediction (RMSEP) of 0.62 mg/g and 0.68 mg/g were obtained. Furthermore, this study also illustrated the possibility of LIBS to provide mineral prediction maps as a useful tool for testing sample homogeneity. [ABSTRACT FROM AUTHOR]

In this study we evaluate whether the methodology of Boosted Regression Trees (BRT) suits for accurately predicting maximum wind speeds. As predictors a broad set of parameters derived from a Digital Elevation Model (DEM) acquired within the Shuttle Radar Topography Mission (SRTM) is used. The derived parameters describe the surface by means of quantities (e.g. slope, aspect) and quality (landform classification). Furthermore land cover data from the CORINE dataset is added. The response variable is maximum wind speed, measurements are provided by a network of weather stations. The area of interest is Switzerland, a country which suits perfectly for this study because of its highly dynamic orography and various landforms. [ABSTRACT FROM AUTHOR]

A comprehensive assessment of the performance of predictive models is necessary as they have been increasingly employed to generate spatial predictions for environmental management and conservation and their accuracy is crucial to evidence-informed decision making and policy. In this study, we clarified relevant issues associated with variance explained ( VEcv ) by predictive models, established the relationships between VEcv and commonly used accuracy measures and unified these measures under VEcv that is independent of unit/scale and data variation. We quantified the relationships between these measures and data variation and found about 65% compared models and over 45% recommended models for generating spatial predictions explained no more than 50% data variance. We classified the predictive models based on VEcv , which provides a tool to directly compare the accuracy of predictive models for data with different unit/scale and variation and establishes a cross-disciplinary context and benchmark for assessing predictive models in future studies. [ABSTRACT FROM AUTHOR]

• Smartphone driving behavior data & OpenStreetMap geometric data are exploited. • Harsh braking counts are spatially analyzed in an urban road network. • GWPR, CAR, and XGBoost models (randomly and spatially cross-validated) are trained. • After adjustments, counts are predicted in another network to assess transferability. • Averaging of model results leads to more balanced and accurate predictions. Harsh driving behavior events, such as harsh braking events (HBs) are road safety surrogate measures showing promising research venues towards crash mitigation, such as safety evaluations based on high-resolution driving data from smartphone sensors. This research presents a framework for aggregation and modelling of such data to highlight safety critical locations based on geometric and network characteristics. Spatial models including Geographically Weighted Poisson Regression, Bayesian Conditional autoregressive models (CAR), and variations of EXtreme Gradient Boosting (XGBoost) are implemented. The purpose is to: (i) explore parameters affecting frequencies of harsh driving events through causal spatial models in an urban road network and (ii) assess the predictive performance of models by testing the transferable components of these models in a new urban network test area. The models are trained and evaluated in terms of accuracy and transferability for HBs predictions in separate areas of Athens, Greece. Findings indicate that geometrical characteristics affect HB frequencies per road segment: Segment length and adjusted pass count are positively correlated with HBs, while gradient and neighborhood complexity are negatively correlated with HBs. Lane number and road type have more unclear and circumstantial effects overall. Two-lane segments have statistically higher HB frequencies compared to one-lane segments, while residential type segments have statistically lower HB frequencies compared to primary road segments. Furthermore, successful spatial predictions were conducted by averaging the results of all four methods, achieving accuracy of more than 87% for HB frequencies per road segment. Finally, the implications towards proactive traffic safety management and the extension possibilities for other harsh event types are also discussed. [ABSTRACT FROM AUTHOR]

Our objective in this work was to present a hierarchical Bayesian spatiotemporal model that allowed us to make spatial predictions of air pollution levels effectively and with very few computational costs. We specified a hierarchical spatiotemporal model using the Stochastic Partial Differential Equations of the integrated nested Laplace approximations approximation. This approach allowed us to spatially predict in the territory of Catalonia (Spain) the levels of the four pollutants for which there is the most evidence of an adverse health effect. Our model allowed us to make fairly accurate spatial predictions of both long-term and short-term exposure to air pollutants with a relatively low density of monitoring stations and at a much lower computation time. The only requirements of our method are the minimum number of stations distributed throughout the territory where the predictions are to be made, and that the spatial and temporal dimensions are either independent or separable. [Display omitted] • We show a hierarchical Bayesian spatiotemporal model. • Our model provides predictions of both long-term and short-term exposure. • The computational cost is low. • The model only needs a minimum number of stations being distributed throughout the territory. • The other requirement of our model is that the spatial and temporal dimensions are either independent or separable. [ABSTRACT FROM AUTHOR]

Species within the Macadamia genus (Proteaceae) are rare and threatened narrowly distributed inhabitants of subtropical lowland rainforests of eastern Australia. Despite their strong cultural links and economic importance as a source of germplasm for the macadamia nut industry, a comprehensive assessment of factors contributing to their conservation status, or the potential impacts of climate change, is lacking. We used maximum entropy models to identify the respective niche of the following three Macadamia species with overlapping extant distributions: M. integrifolia, M. ternifolia and M. tetraphylla. We used model predictions to identify and prioritise respective areas of habitat, together with change in geographic distribution of habitats between 1990 and 2070 climates. Results reveal considerable overlap in the geographic extent of habitat among the three species; however, the extent of current occupation of habitat by any individual species is limited. Relatively high levels of clearing of ecological communities strongly associated with M. integrifolia or M. ternifolia have occurred within the extent of their respective habitats, with M. tetraphylla less affected within the Queensland extent of its range. Response to climate change varies among the three species, with a general trend of shift in respective niche to areas that currently experience relatively high precipitation and lower temperature regimes. Macadamias are native to Australia and are grown worldwide for their edible nuts. However, persistence of their wild relatives is uncertain because of its limited ability to respond to the multiple challenges it faces. Our results provide insight into the patterns and processes that are key to conservation of Macadamia and have wider applicability in conservation of its rainforest habitat. [ABSTRACT FROM AUTHOR]

This paper presents an approach based on field data to model the spatial distribution of the site productivity index (SPI) of the diverse forest types in Jalisco, Mexico and the response in SPI to site and climatic conditions. A linear regression model was constructed to test the hypothesis that site and climate variables can be used to predict the SPI of the major forest types in Jalisco. SPI varied significantly with topography (elevation, aspect and slope), soil attributes (pH, sand and silt), climate (temperature and precipitation zones) and forest type. The most important variable in the model was forest type, which accounted for 35% of the variability in SPI. Temperature and precipitation accounted for 8 to 9% of the variability in SPI while the soil attributes accounted for less than 4% of the variability observed in SPI. No significant differences were detected between the observed and predicted SPI for the individual forest types. The linear regression model was used to develop maps of the spatial variability in predicted SPI for the individual forest types in the state. The spatial site productivity models developed in this study provides a basis for understanding the complex relationship that exists between forest productivity and site and climatic conditions in the state. Findings of this study will assist resource managers in making cost-effective decisions about the management of individual forest types in the state of Jalisco, Mexico. [ABSTRACT FROM AUTHOR]

Abstract: There is an increasing need to model the distribution of species and habitats for effective conservation planning, but there is a paucity of models for the marine environment. We used presence (131) and absence (219) records of the marine angiosperm Zostera marina L. from the archipelago of SW Finland, northern Baltic Sea, to model its distribution in a 5400 km2 area. We used depth, slope, turbidity, wave exposure and distance to sandy shores as environmental predictors, and compared a presence-absence method: generalised additive model (GAM), with a presence only method: maximum entropy (Maxent). Models were validated using semi-independent data sets. Both models performed well and described the niche of Z. marina fairly consistently, although there were differences in the way the models weighted the environmental variables, and consequently the spatial predictions differed somewhat. A notable outcome from the process was that with relatively equal model performance, the area actually predicted in geographical space can vary by twofold. The area predicted as suitable for Z. marina by the ensemble was almost half of that predicted by the GAM model by itself. The ensemble of model predictions increased the model predictive capability marginally and clearly shifted the model towards a more conservative prediction, increasing specificity, but at the same time sacrificing sensitivity. The environmental predictors selected into the final models described the potential distribution of Z. marina well and showed that in the northern Baltic the species occupies a narrow niche, typically thriving in shallow and moderately exposed to exposed locations near sandy shores. We conclude that a prediction based on a combination of model results provides a more realistic estimate of the core area suitable for Z. marina and should be the modelling approach implemented in conservation planning and management. [Copyright &y& Elsevier]

Abstract: The objective of this study was to underpin the capability of diffuse reflectance Fourier transform mid-infrared spectroscopy (midDRIFTS)-based partial least squares regression (PLSR) analyses to accurately predict soil microbiological properties across six temperate grassland ecosystems differing in their land-use intensity; three sites of low land-use intensity (low LUI) and three fertilized mown meadows (high LUI). In addition, the potential of midDRIFTS-PLSR analyses for spatial studies between soils of contrasting grassland ecosystems was evaluated. 304 samples were subjected to midDRIFTS-PLSR-based predictions of soil microbial biomass, activities of beta-d-glucosidase, xylosidase and urease, as well as bacterial abundance based on 16S rRNA gene quantification. Accuracies of midDRIFTS-PLSR-based predictions across both LUI were, on basis of the residual prediction deviation (RPD), ‘acceptable’ for all soil microbiological properties, in particular soil microbial biomass (coefficient of determination (R 2) = 0.92; RPD = 3.55), urease (0.91; 3.42) and beta-d-glucosidase (0.89; 3.01). Predictions of ‘moderately successful’ accuracy were developed for 16S rRNA gene copies (0.88; 2.93) and xylosidase (0.84; 2.55). Spatial midDRIFTS-PLSR-based predictions between grassland ecosystems were only ‘acceptable’ for soil microbial biomass, while the other studied soil microbiological properties revealed only ‘moderately successful’ predictions. The potential of midDRIFTS-PLSR to predict a range of soil microbiological properties including molecular data with ‘acceptable’ accuracies across the two investigated grassland ecosystems with contrasting land-use intensities was substantiated. However, the prevailing ‘moderately successful’ prediction accuracies between grassland ecosystems were probably due to the dependence on land use-specific data sets for calibration and validation. For prospective application of cost- and time-efficient midDRIFTS-PLSR-based approaches, predictions of soil microbiological properties considering particularly, but not yet ‘acceptable’ molecular data will greatly advance the understanding on the abundance and functional dynamics of soil microbial communities across spatial and temporal scales of contrasting ecosystems. [Copyright &y& Elsevier]

Aim Data shortages mean that conservation priorities can be highly sensitive to historical patterns of exploration. Here, we investigate the potential of regionally focussed species distribution models to elucidate fine-scale patterns of richness, rarity and endemism. Location Eastern Arc Mountains, Tanzania and Kenya. Methods Generalized additive models and land cover data are used to estimate the distributions of 452 forest plant taxa (trees, lianas, shrubs and herbs). Presence records from a newly compiled database are regressed against environmental variables in a stepwise multimodel. Estimates of occurrence in forest patches are collated across target groups and analysed alongside inventory-based estimates of conservation priority. Results Predicted richness is higher than observed richness, with the biggest disparities in regions that have had the least research. North Pare and Nguu in particular are predicted to be more important than the inventory data suggest. Environmental conditions in parts of Nguru could support as many range-restricted and endemic taxa as Uluguru, although realized niches are subject to unknown colonization histories. Concentrations of rare plants are especially high in the Usambaras, a pattern mediated in models by moisture indices, whilst overall richness is better explained by temperature gradients. Tree data dominate the botanical inventory; we find that priorities based on other growth forms might favour the mountains in a different order. Main conclusions Distribution models can provide conservation planning with high-resolution estimates of richness in well-researched areas, and predictive estimates of conservation importance elsewhere. Spatial and taxonomic biases in the data are essential considerations, as is the spatial scale used for models. We caution that predictive estimates are most uncertain for the species of highest conservation concern, and advocate using models and targeted field assessments iteratively to refine our understanding of which areas should be prioritised for conservation. [ABSTRACT FROM AUTHOR]

Abstract: Habitat models are widely used in ecology, however there are relatively few studies of rare species, primarily because of a paucity of survey records and lack of robust means of assessing accuracy of modelled spatial predictions. We investigated the potential of compiled ecological data in developing habitat models for Macadamia integrifolia, a vulnerable mid-stratum tree endemic to lowland subtropical rainforests of southeast Queensland, Australia. We compared performance of two binomial models—Classification and Regression Trees (CART) and Generalised Additive Models (GAM)—with Maximum Entropy (MAXENT) models developed from (i) presence records and available absence data and (ii) developed using presence records and background data. The GAM model was the best performer across the range of evaluation measures employed, however all models were assessed as potentially useful for informing in situ conservation of M. integrifolia, A significant loss in the amount of M. integrifolia habitat has occurred (p <0.05), with only 37% of former habitat (pre-clearing) remaining in 2003. Remnant patches are significantly smaller, have larger edge-to-area ratios and are more isolated from each other compared to pre-clearing configurations (p <0.05). Whilst the network of suitable habitat patches is still largely intact, there are numerous smaller patches that are more isolated in the contemporary landscape compared with their connectedness before clearing. These results suggest that in situ conservation of M. integrifolia may be best achieved through a landscape approach that considers the relative contribution of small remnant habitat fragments to the species as a whole, as facilitating connectivity among the entire network of habitat patches. [Copyright &y& Elsevier]

• Abiotic environmental conditions predict bee species richness across ecosystems. • Prediction maps can improve the efficiency of habitat enhancements schemes. • Prediction maps can also guide threatened species mapping surveys. Wild bees are key ecosystem components making their decline a cause for concern. An effective measure to increase wild bee diversity is to enhance plant diversity. However, the effect on bee diversity of augmenting plant diversity depends on site-specific environmental conditions. We aimed to make spatial predictions of where: (a) environmental conditions maximize bee diversity, so that such areas can be prioritized for augmenting plant diversity; and (b) populations of threatened wild bee species are most likely to occur. We surveyed bee communities in traditionally managed hay meadows in SE Norway and modelled bee diversity as a function of climate, habitat area, and distance to nesting substrates. We used independent data to validate our predictions and found that plant and predicted bee species richness together explained 76% and 69% of the variation in observed solitary bee species richness in forested and agricultural ecosystems, respectively. In urban areas, the predicted bee species richness alone explained 31% of the variation in observed solitary bee species richness. Using data from online species occurrence records, we found that – compared to species of lower conservation concern – threatened solitary bee species were more typically recorded in areas with a high predicted solitary bee species richness. We show that spatial predictions of bee diversity can identify sites where augmenting plant diversity is likely to be most effective. Maps of predicted bee diversity can guide species surveys and monitoring projects and increase the chances of locating populations of threatened bees. [ABSTRACT FROM AUTHOR]

1. The concept of the ecological niche relates a set of environmental variables to the fitness of species, while habitat suitability models (HSMs) relate environmental variables to the likelihood of occurrence of the species. In spite of this relationship, the concepts are weakly linked in the literature, and there is a strong need for better integration. 2. We selectively reviewed the literature for habitat suitability studies that directly addressed four common facets of niche theory: niche characteristics, niche interactions, community-wide processes and niche evolution. 3. We found that HSMs have mostly contributed to the study of niche characteristics, but the three other themes are gaining impetus. We discuss three issues that emerge from these studies: (i) commonly used environmental variables and their link with ecological niches; (ii) the causes of false absences and false presences in species data, and associated issues; (iii) the three axes of model generalization (interpolation and extrapolation): environmental, spatial and temporal. Finally, we propose a list of 12 recommendations to strengthen the use of HSMs for wildlife management. 4. Synthesis and applications. This selective review provides conservation biologists with a list of pointers to key niche-theory concepts and a wide palette of related HSM studies. It also brings together frameworks that are often separated: theoretical and applied ecology studies; botany, zoology and parasitology; and different HSM frameworks, such as Resource Selection Functions, Species Distribution Modelling, Ecological Niche Modelling, and Gradient Analysis. We hope that integration of all these slices of knowledge will improve the quality and reliability of HSM predictions. [ABSTRACT FROM AUTHOR]

A new ecosystem-based climate envelope modeling approach was applied to assess potential climate change impacts on forest communities and tree species. Four orthogonal canonical discriminant functions were used to describe the realized climate space for British Columbia's ecosystems and to model portions of the realized niche space for tree species under current and predicted future climates. This conceptually simple model is capable of predicting species ranges at high spatial resolutions far beyond the study area, including outlying populations and southern range limits for many species. We analyzed how the realized climate space of current ecosystems changes in extent, elevation, and spatial distribution under climate change scenarios and evaluated the implications for potential tree species habitat. Tree species with their northern range limit in British Columbia gain potential habitat at a pace of at least 100 km per decade, common hardwoods appear to be generally unaffected by climate change, and some of the most important conifer species in British Columbia are expected to lose a large portion of their suitable habitat. The extent of spatial redistribution of realized climate space for ecosystems is considerable, with currently important sub-boreal and montane climate regions rapidly disappearing. Local predictions of changes to tree species frequencies were generated as a basis for systematic surveys of biological response to climate change. [ABSTRACT FROM AUTHOR]

Aim This study used data from temperate forest communities to assess: (1) five different stepwise selection methods with generalized additive models, (2) the effect of weighting absences to ensure a prevalence of 0.5, (3) the effect of limiting absences beyond the environmental envelope defined by presences, (4) four different methods for incorporating spatial autocorrelation, and (5) the effect of integrating an interaction factor defined by a regression tree on the residuals of an initial environmental model. Location State of Vaud, western Switzerland. Methods Generalized additive models (GAMs) were fitted using the grasp package (generalized regression analysis and spatial predictions, http://www.cscf.ch/grasp). Results Model selection based on cross-validation appeared to be the best compromise between model stability and performance (parsimony) among the five methods tested. Weighting absences returned models that perform better than models fitted with the original sample prevalence. This appeared to be mainly due to the impact of very low prevalence values on evaluation statistics. Removing zeroes beyond the range of presences on main environmental gradients changed the set of selected predictors, and potentially their response curve shape. Moreover, removing zeroes slightly improved model performance and stability when compared with the baseline model on the same data set. Incorporating a spatial trend predictor improved model performance and stability significantly. Even better models were obtained when including local spatial autocorrelation. A novel approach to include interactions proved to be an efficient way to account for interactions between all predictors at once. Main conclusions Models and spatial predictions of 18 forest communities were significantly improved by using either: (1) cross-validation as a model selection method, (2) weighted absences, (3) limited absences, (4) predictors accounting for spatial autocorrelation, or (5) a factor variable accounting for interactions between all predictors. The final choice of model strategy should depend on the nature of the available data and the specific study aims. Statistical evaluation is useful in searching for the best modelling practice. However, one should not neglect to consider the shapes and interpretability of response curves, as well as the resulting spatial predictions in the final assessment. [ABSTRACT FROM AUTHOR]

We examined the spatial distribution of site productivity for jack pine and black spruce across a 556,000 km2 area of Ontario, Canada. Our analysis employed data from 1140 forest ecosystem classification (FEC) plots distributed across this region. Using tree-based regression, we related soil, topographic, and climatic conditions at the plots to site index (i.e. estimated height at 50 years of age) of the target species. Both species grew better on deeper mineral soils in the southern arm of the province where wetter, warmer conditions prevail. Models were tested for accuracy and bias using withheld data. Models for both species were unbiased. The jack pine model had a square root of the mean squared prediction error (root-MSPR) of 2.55 m, while that for black spruce was 2.84 m. These error levels are comparable to other rigorously tested results reported in the literature. A map was produced for each species showing the spatial distribution of site index across the study area as predicted by the regression tree models. By necessity, coarse-scale soils data were used to produce the site index maps and this increased root-MSPR errors ∼20–30% as compared to errors obtained using actual soil data recorded in the field. These maps, despite their moderate accuracy, provide a picture of relative productivity that may be used in broad-scale planning initiatives. As well, the tree-based regression models can be used to obtain point estimates of site index at locations where the required soil data are available. [Copyright &y& Elsevier]

Identification of areas containing high biological diversity (‘hotspots’) from species presence-only data has become increasingly important in species and ecosystem management when presence/absence data is unavailable. However, as presence-only data sets lack any information on absences and as they suffer from many biases associated with the ad hoc or non-stratified sampling, they are often assumed problematic and inadequate for most statistical modeling methods. In this paper, this supposition is investigated by comparing generalized additive models (GAM) fitted with 43 native New Zealand fern species presence/absence data, obtained from a survey of 19 875 forested plots, to GAM models and ecological niche factor analysis (ENFA) models fitted with identical presence data and, in the case of GAM models, computer generated ‘pseudo’ absences. By using the same presence data for all models, absence data is isolated as the varying factor allowing different techniques for generating ‘pseudo’ absences used in the GAM models to be analyzed and compared over three principal levels of investigation. GAM models fitted with an environmentally weighted distribution of ‘pseudo’ absences and ENFA models selected environmental variables more similar to the GAM presence/absence models than did the GAM models fitted with randomly distributed ‘pseudo’ absences. Average contributions for the GAM presence/absence model showed mean annual temperature and mean annual solar radiation as the most important factors followed by lithology. Comparisons of prediction results show GAM models incorporating an environmentally weighted distribution of ‘pseudo’ absences to be more closely correlated to the GAM presence/absence models than either the GAM models fitted with randomly selected ‘pseudo’ absences or the ENFA models. ENFA models were found to be the least correlated to the GAM presence/absence models. These latter models were also shown to give the most optimistic predictions overall, however, as ENFA predicts habitat suitability rather than probability of presence this was expected. Summation of species predictions used as a measure of species potential biodiversity ‘hotspots’ also shows ENFA models to give the highest and largest distribution of potential biodiversity. Additionally, GAM models incorporating ‘pseudo’ absences were more highly correlated to the GAM presence/absence model than was ENFA. However, ENFA identified more areas of potential biodiversity ‘hotspots’ similar to the GAM presence/absence model, than either GAM model incorporating ‘pseudo’ absences. [Copyright &y& Elsevier]

Summary Aim We modelled the relationship of breeding evidence for five species of forest songbirds (ruby-crowned kinglet (Regulus calendula ) Blackburnian warbler (Dendroica fusca ), black-throated blue warbler (Dendroica caerulescens) , bay-breasted warbler (Dendrioca castanea ) and Connecticut warbler (Oporornis agilis )) and a variety of macro-climate variables to examine the importance of climate as a factor determining distribution of breeding in these species and to assess the usefulness of spatial predictions generated from these models. Location Modelling was conducted over the entire province of Ontario, Canada, an area of ≈900,000 km2 . Methods Data on the distribution of breeding in the province was derived from the Breeding Bird Atlas of Ontario. We used logistic regression to model the relationship between the probability of breeding (assessed in 10 km×10 km blocks) and estimates of a variety of climate variables at the same scale. Models were selected that had the least number of explanatory variables while at the same time having close to the best possible classification accuracy. Results The final models for these five species had from one to six explanatory variables and an overall concordance of 70.4% to 86.3% indicating a good classification accuracy. Results from subsampling 50% of the original data ten times indicate that (1) the classification accuracy of the model for data used to generate the model is not very sensitive to the specific observations used to generate the model (2) the classification accuracy of test data is close to the classification accuracy of the model data and (3) the classification accuracy of the test data is not dependent on the specific observations used to generate the model. We generated a spatial prediction of the probability of occurrence of each species for Ontario using the relationships defined by the logistic regression models and using 1 km gridded estimates of the necessary climate... [ABSTRACT FROM AUTHOR]

• We test if effects of bee habitat improvement is predictable at fine spatial scales. • We successfully predict the plant species composition, a determinant of the effect of habitat improvement. • Predicting plant species composition allows predicting effects of bee habitat management. • Predictions can improve the cost-efficiency of wild bee habitat management. Prioritizing where to implement management interventions is critical because managers have limited budgets and the effect of habitat enhancement depends on site-specific environmental conditions. Field experiments can identify the conditions where habitat enhancement is most effective, but are typically of limited extent and thus not sufficient for producing spatial predictions that can guide management efforts. We tested if we could produce spatial predictions maps – showing where management interventions to enhance bee habitat would be most successful – by combining spatial predictions of plant community composition (i.e., environmental conditions) obtained from field surveys with a field experiment, in which we quantified the effect of three types of management interventions on bee species richness. Using information from digital maps, we predicted plant species composition within power line clearings across southeast Norway. The intervention type, which involved cutting and removal of the woody vegetation, resulted in the largest increase in bee species richness, but the enhanced bee species richness was limited to clearings with forb-dominated vegetation. Importantly, the estimated effects on bee species richness did not differ between models using the predicted, versus the empirically observed, plant species composition as predictor, making it possible to produce spatial predictions of the increase in bee richness from implementing different management interventions. Synthesis and applications: Combining field surveys with data from field experiments can be used to produce high-resolution maps showing where wild bee habitat enhancement is likely to have the greatest effect. Such maps can inform decisions about where to allocate costly management interventions. [ABSTRACT FROM AUTHOR]

Seabed sediment predictions at regional and national scales in Australia are mainly based on bathymetry-related variables due to the lack of backscatter-derived data. In this study, we applied random forests (RFs), hybrid methods of RF and geostatistics, and generalized boosted regression modelling (GBM), to seabed sand content point data and acoustic multibeam data and their derived variables, to develop an accurate model to predict seabed sand content at a local scale. We also addressed relevant issues with variable selection. It was found that: (1) backscatter-related variables are more important than bathymetry-related variables for sand predictive modelling; (2) the inclusion of highly correlated predictors can improve predictive accuracy; (3) the rank orders of averaged variable importance (AVI) and accuracy contribution change with input predictors for RF and are not necessarily matched; (4) a knowledge-informed AVI method (KIAVI2) is recommended for RF; (5) the hybrid methods and their averaging can significantly improve predictive accuracy and are recommended; (6) relationships between sand and predictors are non-linear; and (7) variable selection methods for GBM need further study. Accuracy-improved predictions of sand content are generated at high resolution, which provide important baseline information for environmental management and conservation. [ABSTRACT FROM AUTHOR]