The main features of daily extreme precipitation and circulation types in southern South America (SSA) were evaluated and compared in both multiple observational datasets (rain gauges, CHIRPS, CPC ...and MSWEP) and simulations from four regional climate models (RCMs) driven by ERA-Interim during 1980–2010. The inter-comparison of extreme events, characterised in terms of their intensity, frequency and spatial coverage, varied across SSA showing large differences among observational datasets and RCMs and reflecting the current observational uncertainty when evaluating precipitation extremes at a daily scale. The spread between observational datasets was smaller than for the RCMs. Most of the RCMs successfully captured the spatial pattern of extreme precipitation across SSA, although RCA4 (REMO) usually underestimated (overestimated) precipitation intensities, particularly the maximum amounts in southeastern South America (SESA), where the extremes are remarkable. The synoptic circulation was described by a classification of circulation types (CTs) using Self-Organizing Maps (SOM). Specific CTs were found to significantly enhance the occurrence of extreme precipitation events in sectorized areas of SESA. The RCMs adequately reproduced the SOM node frequencies, although they tended to simplify the predominant CTs into a more reduced number of configurations. They appropriately represented the extreme precipitation frequencies conditioned by each CT, exhibiting some limitations in the location and intensity of the resulting precipitation systems. These sorts of evaluations contribute to a better understanding of the physical mechanisms responsible for extreme precipitation and of their future projections in a climate change scenario.
The relatively rapid shift from consuming preagricultural wild foods for thousands of years, to consuming postindustrial semi-processed and ultra-processed foods endemic of the Western world less ...than 200 years ago did not allow for evolutionary adaptation of the commensal microbial species that inhabit the human gastrointestinal (GI) tract, and this has significantly impacted gut health. The human gut microbiota, the diverse and dynamic population of microbes, has been demonstrated to have extensive and important interactions with the digestive, immune, and nervous systems. Western diet-induced dysbiosis of the gut microbiota has been shown to negatively impact human digestive physiology, to have pathogenic effects on the immune system, and, in turn, cause exaggerated neuroinflammation. Given the tremendous amount of evidence linking neuroinflammation with neural dysfunction, it is no surprise that the Western diet has been implicated in the development of many diseases and disorders of the brain, including memory impairments, neurodegenerative disorders, and depression. In this review, we discuss each of these concepts to understand how what we eat can lead to cognitive and psychiatric diseases.
Land cover monitoring using remotely sensed data requires robust classification methods which allow for the accurate mapping of complex land cover and land use categories. Random forest (RF) is a ...powerful machine learning classifier that is relatively unknown in land remote sensing and has not been evaluated thoroughly by the remote sensing community compared to more conventional pattern recognition techniques. Key advantages of RF include: their non-parametric nature; high classification accuracy; and capability to determine variable importance. However, the split rules for classification are unknown, therefore RF can be considered to be black box type classifier. RF provides an algorithm for estimating missing values; and flexibility to perform several types of data analysis, including regression, classification, survival analysis, and unsupervised learning.
In this paper, the performance of the RF classifier for land cover classification of a complex area is explored. Evaluation was based on several criteria: mapping accuracy, sensitivity to data set size and noise. Landsat-5 Thematic Mapper data captured in European spring and summer were used with auxiliary variables derived from a digital terrain model to classify 14 different land categories in the south of Spain. Results show that the RF algorithm yields accurate land cover classifications, with 92% overall accuracy and a Kappa index of 0.92. RF is robust to training data reduction and noise because significant differences in kappa values were only observed for data reduction and noise addition values greater than 50 and 20%, respectively. Additionally, variables that RF identified as most important for classifying land cover coincided with expectations. A McNemar test indicates an overall better performance of the random forest model over a single decision tree at the 0.00001 significance level.
Recognising the various sources of nitrate pollution and understanding system dynamics are fundamental to tackle groundwater quality problems. A comprehensive GIS database of twenty parameters ...regarding hydrogeological and hydrological features and driving forces were used as inputs for predictive models of nitrate pollution. Additionally, key variables extracted from remotely sensed Normalised Difference Vegetation Index time-series (NDVI) were included in database to provide indications of agroecosystem dynamics.
Many approaches can be used to evaluate feature importance related to groundwater pollution caused by nitrates. Filters, wrappers and embedded methods are used to rank feature importance according to the probability of occurrence of nitrates above a threshold value in groundwater. Machine learning algorithms (MLA) such as Classification and Regression Trees (CART), Random Forest (RF) and Support Vector Machines (SVM) are used as wrappers considering four different sequential search approaches: the sequential backward selection (SBS), the sequential forward selection (SFS), the sequential forward floating selection (SFFS) and sequential backward floating selection (SBFS). Feature importance obtained from RF and CART was used as an embedded approach.
RF with SFFS had the best performance (mmce=0.12 and AUC=0.92) and good interpretability, where three features related to groundwater polluted areas were selected: i) industries and facilities rating according to their production capacity and total nitrogen emissions to water within a 3km buffer, ii) livestock farms rating by manure production within a 5km buffer and, iii) cumulated NDVI for the post-maximum month, being used as a proxy of vegetation productivity and crop yield.
•Different Feature Selection approaches (FS) based on machine learning were evaluated.•FS allowed to isolate and identify the main drivers of nitrate pollution in groundwater.•Driving forces were more useful in predicting nitrates pollution in this case study.•A novel feature, extracted from NDVI time series, was revealed as very promising.•A Random Forest based wrapper outperformed the rest FS in predicting nitrates.
A Random Forest (RF) classifier was applied to spectral as well as mono- and multi-seasonal textural features extracted from Landsat TM imagery to increase the accuracy of land cover classification ...over a complex Mediterranean landscape, with a large number of land cover categories and low inter-class separability. Five different types of geostatistical textural measure for three different window sizes and three different lags were applied creating a total of 972 potential input variables. Madograms, rodograms and direct variograms for the univariate case and cross- and pseudo-cross variograms for the multivariate case, together with multi-seasonal spectral information, were used in a RF classifier to map the land cover types. The pseudo-cross and cross variograms were used specifically to incorporate important seasonal/temporal information. Incorporating multi-scale textural features into the RF models led to an increase in the overall index of 10.71% and, for the most accurate classification, the increase was up to 30% in some classes. The differences in the kappa coefficient for the textural classification models were evaluated statistically using a pairwise Z-test, revealing a significant increase in per-class classification accuracy compared to GLCM-based texture measures. The pseudo-cross variogram between the visible and near-infrared bands was the most important textural features for general classification, and the multi-seasonal pseudo-cross variogram had an outstanding importance for agricultural classes. Overall, the RF classifier applied to a reduced subset of input variables composed of the most informative textural features led to the highest accuracy. Highly reliable classification results were obtained when the 16 most important textural features calculated at single scales (window sizes) were selected and used in the classification. The proposed methodology significantly increased the classification accuracy achieved with a spectral maximum likelihood classifier (ML). The kappa values for the textural RF and ML classifications were equal to 0.92 and 0.83, respectively.
► We assess the increase in accuracy that can be achieved by incorporating geostatistical texture in Random Forest classifiers. ► The proposed method is based on the analysis of mono- and multi-seasonal textural features. ► Pseudo-cross and cross variograms were used to incorporate the seasonal/temporal dimension. ► Our approach outperforms GLCM-based approaches. ► Random Forest classification system was utilised to determine and select the most important textural features.
The amino or carboxy-terminal regions of certain cell wall proteins are capable of anchoring foreign proteins or peptides on the cell wall of the yeast
Saccharomyces cerevisiae
. This possibility has ...resulted in the development of a methodology known as yeast display which has powerful applications in biotechnology, pharmacy, and medicine. This work describes the results of experiments in which the agglutinin Aga2p protein is used as an anchor and several leucine-based peptides have been introduced into its N-terminal or C-terminal position. We found that the sequence of these peptides can affect plasmid stability, growth kinetics, and levels of the fusion protein displayed, and we analyzed how the incubation conditions influence these parameters. Besides, we show that the introduction of these small peptides can modify the properties of cell cover; in particular, fusing five or ten leucine residues to the Aga2p protein results in greater hydrophobicity of the cell wall and also in increased resistance to the presence of the organic solvents acetonitrile and ethanol and to high salt concentrations. The introduction of the RLRLL sequence also results in higher resistance to the exposure of yeast cells to NaCl stress.
A collection of 10 high-impact extreme precipitation events occurring in Southeastern South America during the warm season has been analyzed using statistical (ESD) and dynamical downscaling ...approaches. Regional Climate Models from the CORDEX database for the South American domain at two horizontal resolutions, 50 km and 25 km, short-term simulations at 20 km and at 4 km convective-permitting resolution and statistical downscaling techniques based on the analogue method and the generalized linear model approach were evaluated. The analysis includes observational datasets based on gridded data, station data and satellite products that allow assessing the observational uncertainty that characterizes extreme events in the region. It is found that the ability of the modelling strategies in capturing the main features of the extreme rainfall varies across the events. The higher the horizontal resolution of the models, the more intense and localized the core of the rainfall event, being the location of the exit region of the low-level jet and the low-level moisture flux convergence during the initial stages of the events the most relevant features that determine models’ ability of capturing the location and intensity of the core of the heavy rainfall. ESD models based on the generalized linear approach overestimate the spatial extension of the events and underestimate the intensity of the local maxima. Weather-like convective-permitting simulations depict an overall good performance in reproducing both the rainfall patterns and the triggering mechanisms of the extreme events as expected, given that these simulations are strongly controlled by the initial conditions.
The aim of this work is to present preliminary results of the statistical and dynamical simulations carried out within the framework of the Flagship Pilot Study in southeastern South America ...(FPS-SESA) endorsed by the Coordinated Regional Climate Downscaling Experiments (CORDEX) program. The FPS-SESA initiative seeks to promote inter-institutional collaboration and further networking with focus on extreme rainfall events. The main scientific aim is to study multi-scale processes and interactions most conducive to extreme precipitation events through both statistical and dynamical downscaling techniques, including convection-permitting simulations. To this end, a targeted experiment was designed considering the season October 2009 to March 2010, a period with a record number of extreme precipitation events within SESA. Also, three individual extreme events within that season were chosen as case studies for analyzing specific regional processes and sensitivity to resolutions. Four dynamical and four statistical downscaling models (RCM and ESD respectively) from different institutions contributed to the experiment. In this work, an analysis of the capability of the set of the FPS-SESA downscaling methods in simulating daily precipitation during the selected warm season is presented together with an integrated assessment of multiple sources of observations and available CORDEX Regional Climate Model simulations. Comparisons among all simulations reveal that there is no single model that performs best in all aspects evaluated. The ability in reproducing the different features of daily precipitation depends on the model. However, the evaluation of the sequence of precipitation events, their intensity and timing suggests that FPS-SESA simulations based on both RCM and ESD yield promising results. Most models capture the extreme events selected, although with a considerable spread in accumulated values and the location of heavy precipitation.
Machine learning algorithms (MLAs) such us artificial neural networks (ANNs), regression trees (RTs), random forest (RF) and support vector machines (SVMs) are powerful data driven methods that are ...relatively less widely used in the mapping of mineral prospectivity, and thus have not been comparatively evaluated together thoroughly in this field.
The performances of a series of MLAs, namely, artificial neural networks (ANNs), regression trees (RTs), random forest (RF) and support vector machines (SVMs) in mineral prospectivity modelling are compared based on the following criteria: i) the accuracy in the delineation of prospective areas; ii) the sensitivity to the estimation of hyper-parameters; iii) the sensitivity to the size of training data; and iv) the interpretability of model parameters. The results of applying the above algorithms to epithermal Au prospectivity mapping of the Rodalquilar district, Spain, indicate that the RF outperformed the other MLA algorithms (ANNs, RTs and SVMs). The RF algorithm showed higher stability and robustness with varying training parameters and better success rates and ROC analysis results. On the other hand, all MLA algorithms can be used when ore deposit evidences are scarce. Moreover the model parameters of RF and RT can be interpreted to gain insights into the geological controls of mineralization.
PDF
