Analysis, modelling and classification of geospatial data using machine learning

The research considers the problem of spatial data classification using machine learning algorithms: probabilistic neural networks (PNN) and support vector machines (SVM). As a benchmark model simple k-nearest neighbor algorithm is considered. PNN is a neural network reformulation of well known nonparametric principles of probability density modeling using kernel density estimator and Bayesian optimal or maximum a posteriori decision rules. PNN is well suited to problems where not only predictions but also quantification of accuracy and integration of prior information are necessary. An important property of PNN is that they can be easily used in decision support systems dealing with problems of automatic classification. Support vector machine is an implementation of the principles of statistical learning theory for the classification tasks. Recently they were successfully applied for different environmental topics: classification of soil types and hydro-geological units, optimization of monitoring networks, susceptibility mapping of natural hazards. In the present paper both simulated and real data case studies (low and high dimensional) are considered. The main attention is paid to the detection and learning of spatial patterns by the algorithms applied.

Decomposition and nutrient release of leguminous plants in coffee agroforestry systems

Leguminous plants used as green manure are an important nutrient source for coffee plantations, especially for soils with low nutrient levels. Field experiments were conducted in the Zona da Mata of Minas Gerais State, Brazil to evaluate the decomposition and nutrient release rates of four leguminous species used as green manures (Arachis pintoi, Calopogonium mucunoides, Stizolobium aterrimum and Stylosanthes guianensis) in a coffee agroforestry system under two different climate conditions. The initial N contents in plant residues varied from 25.7 to 37.0 g kg-1 and P from 2.4 to 3.0 g kg-1. The lignin/N, lignin/polyphenol and (lignin+polyphenol)/N ratios were low in all residues studied. Mass loss rates were highest in the first 15 days, when 25 % of the residues were decomposed. From 15 to 30 days, the decomposition rate decreased on both farms. On the farm in Pedra Dourada (PD), the decomposition constant k increased in the order C. mucunoides < S. aterrimum < S. guianensis < A. pintoi. On the farm in Araponga (ARA), there was no difference in the decomposition rate among leguminous plants. The N release rates varied from 0.0036 to 0.0096 d-1. Around 32 % of the total N content in the plant material was released in the first 15 days. In ARA, the N concentration in the S. aterrimum residues was always significantly higher than in the other residues. At the end of 360 days, the N released was 78 % in ARA and 89 % in PD of the initial content. Phosphorus was the most rapidly released nutrient (k values from 0.0165 to 0.0394 d-1). Residue decomposition and nutrient release did not correlate with initial residue chemistry and biochemistry, but differences in climatic conditions between the two study sites modified the decomposition rate constants.

Neural regulation of pancreatic islet cell mass and function.

Intracellular glucose signalling pathways control the secretion of glucagon and insulin by pancreatic islet α- and β-cells, respectively. However, glucose also indirectly controls the secretion of these hormones through regulation of the autonomic nervous system that richly innervates this endocrine organ. Both parasympathetic and sympathetic nervous systems also impact endocrine pancreas postnatal development and plasticity in adult animals. Defects in these autonomic regulations impair β-cell mass expansion during the weaning period and β-cell mass adaptation in adult life. Both branches of the autonomic nervous system also regulate glucagon secretion. In type 2 diabetes, impaired glucose-dependent autonomic activity causes the loss of cephalic and first phases of insulin secretion, and impaired suppression of glucagon secretion in the postabsorptive phase; in diabetic patients treated with insulin, it causes a progressive failure of hypoglycaemia to trigger the secretion of glucagon and other counterregulatory hormones. Therefore, identification of the glucose-sensing cells that control the autonomic innervation of the endocrine pancreatic and insulin and glucagon secretion is an important goal of research. This is required for a better understanding of the physiological control of glucose homeostasis and its deregulation in diabetes. This review will discuss recent advances in this field of investigation.

Carbon stock and its compartments in a subtropical oxisol under long-term tillage and crop rotation systems

Soil organic matter (SOM) plays a crucial role in soil quality and can act as an atmospheric C-CO2 sink under conservationist management systems. This study aimed to evaluate the long-term effects (19 years) of tillage (CT-conventional tillage and NT-no tillage) and crop rotations (R0-monoculture system, R1-winter crop rotation, and R2- intensive crop rotation) on total, particulate and mineral-associated organic carbon (C) stocks of an originally degraded Red Oxisol in Cruz Alta, RS, Southern Brazil. The climate is humid subtropical Cfa 2a (Köppen classification), the mean annual precipitation 1,774 mm and mean annual temperature 19.2 ºC. The plots were divided into four segments, of which each was sampled in the layers 0-0.05, 0.05-0.10, 0.10-0.20, and 0.20-0.30 m. Sampling was performed manually by opening small trenches. The SOM pools were determined by physical fractionation. Soil C stocks had a linear relationship with annual crop C inputs, regardless of the tillage systems. Thus, soil disturbance had a minor effect on SOM turnover. In the 0-0.30 m layer, soil C sequestration ranged from 0 to 0.51 Mg ha-1 yr-1, using the CT R0 treatment as base-line; crop rotation systems had more influence on soil stock C than tillage systems. The mean C sequestration rate of the cropping systems was 0.13 Mg ha-1 yr-1 higher in NT than CT. This result was associated to the higher C input by crops due to the improvement in soil quality under long-term no-tillage. The particulate C fraction was a sensitive indicator of soil management quality, while mineral-associated organic C was the main pool of atmospheric C fixed in this clayey Oxisol. The C retention in this stable SOM fraction accounts for 81 and 89 % of total C sequestration in the treatments NT R1 and NT R2, respectively, in relation to the same cropping systems under CT. The highest C management index was observed in NT R2, confirming the capacity of this soil management practice to improve the soil C stock qualitatively in relation to CT R0. The results highlighted the diversification of crop rotation with cover crops as a crucial strategy for atmospheric C-CO2 sequestration and SOM quality improvement in highly weathered subtropical Oxisols.

Long-term C-CO2 emissions and carbon crop residue mineralization in an oxisol under different tillage and crop rotation systems

Soil C-CO2 emissions are sensitive indicators of management system impacts on soil organic matter (SOM). The main soil C-CO2 sources at the soil-plant interface are the decomposition of crop residues, SOM turnover, and respiration of roots and soil biota. The objectives of this study were to evaluate the impacts of tillage and cropping systems on long-term soil C-CO2 emissions and their relationship with carbon (C) mineralization of crop residues. A long-term experiment was conducted in a Red Oxisol in Cruz Alta, RS, Brazil, with subtropical climate Cfa (Köppen classification), mean annual precipitation of 1,774 mm and mean annual temperature of 19.2 ºC. Treatments consisted of two tillage systems: (a) conventional tillage (CT) and (b) no tillage (NT) in combination with three cropping systems: (a) R0- monoculture system (soybean/wheat), (b) R1- winter crop rotation (soybean/wheat/soybean/black oat), and (c) R2- intensive crop rotation (soybean/ black oat/soybean/black oat + common vetch/maize/oilseed radish/wheat). The soil C-CO2 efflux was measured every 14 days for two years (48 measurements), by trapping the CO2 in an alkaline solution. The soil gravimetric moisture in the 0-0.05 m layer was determined concomitantly with the C-CO2 efflux measurements. The crop residue C mineralization was evaluated with the mesh-bag method, with sampling 14, 28, 56, 84, 112, and 140 days after the beginning of the evaluation period for C measurements. Four C conservation indexes were used to assess the relation between C-CO2 efflux and soil C stock and its compartments. The crop residue C mineralization fit an exponential model in time. For black oat, wheat and maize residues, C mineralization was higher in CT than NT, while for soybean it was similar. Soil moisture was higher in NT than CT, mainly in the second year of evaluation. There was no difference in tillage systems for annual average C-CO2 emissions, but in some individual evaluations, differences between tillage systems were noticed for C-CO2 evolution. Soil C-CO2 effluxes followed a bi-modal pattern, with peaks in October/November and February/March. The highest emission was recorded in the summer and the lowest in the winter. The C-CO2 effluxes were weakly correlated to air temperature and not correlated to soil moisture. Based on the soil C conservation indexes investigated, NT associated to intensive crop rotation was more C conserving than CT with monoculture.

Chemical and structural characterization of soil humic substances under agroforestry and conventional systems

Studies have proven that the agroforestry systems in the semi-arid region of the State of Ceará, Brazil, induce an increase in soil organic C levels. Notwithstanding, there is no information if this increase also results in qualitative changes in different pools of soil organic matter. The objective of this study was to verify the possible chemical and structural alterations in fulvic and humic acids of a Luvisol in areas adopting agroforestry, traditional intensive cultivation and native forest in a long-term experiment conducted in the semi-arid region of Ceará State, Brazil. The study was conducted in an experimental area of the National Goat Research Center (Embrapa) in Sobral, CE. The following treatments were evaluated: agrosilvopasture (AGP), silvopasture (SILV), intensive cultivation under fallow (ICF), and areas with native forest (NF). Soil fulvic and humic acids fractions were extracted from the 0-6 and 6-12 cm layers and characterized by elemental composition, thermogravimetry and infrared spectroscopy analyses. The elemental composition analysis of humic acids confirmed the data found for fulvic acids, showing reduction in the C, H and N levels, followed by an increase in O contents in the AGP and ICF treatments over SILV and NF. In all treatments, except to SILV in the 0-6 cm layer, the percentage of mass loss was highest (300-600 °C) for humic acids in the thermally most stable region. Despite the similarity between infrared spectra, soil fulvic acids in the SILV treatment extracted from 6-12 cm depth decrease the absorption bands at 1708 and 1408 cm-1 followed by an increase in the absorption band at 1608 cm-1 attributed to aromatic C=C groups. This behavior suggests an increase in the aromatic character of the structure. The AGP and ICF treatments, which increase the soil tilling, favored the maintenance of humic substances with a more aromatic character in the soil than SILV and NF. The less aromatic humic substances in the SILV treatment resulted in an increase of exchange sites of soil organic matter, indicating improved nutrient cycling and maintenance of productivity in the system.

Mult-class differentiation of cannabis seedlings in a forensic context

This article presents an experimental study about the classification ability of several classifiers for multi-classclassification of cannabis seedlings. As the cultivation of drug type cannabis is forbidden in Switzerland lawenforcement authorities regularly ask forensic laboratories to determinate the chemotype of a seized cannabisplant and then to conclude if the plantation is legal or not. This classification is mainly performed when theplant is mature as required by the EU official protocol and then the classification of cannabis seedlings is a timeconsuming and costly procedure. A previous study made by the authors has investigated this problematic [1]and showed that it is possible to differentiate between drug type (illegal) and fibre type (legal) cannabis at anearly stage of growth using gas chromatography interfaced with mass spectrometry (GC-MS) based on therelative proportions of eight major leaf compounds. The aims of the present work are on one hand to continueformer work and to optimize the methodology for the discrimination of drug- and fibre type cannabisdeveloped in the previous study and on the other hand to investigate the possibility to predict illegal cannabisvarieties. Seven classifiers for differentiating between cannabis seedlings are evaluated in this paper, namelyLinear Discriminant Analysis (LDA), Partial Least Squares Discriminant Analysis (PLS-DA), Nearest NeighbourClassification (NNC), Learning Vector Quantization (LVQ), Radial Basis Function Support Vector Machines(RBF SVMs), Random Forest (RF) and Artificial Neural Networks (ANN). The performance of each method wasassessed using the same analytical dataset that consists of 861 samples split into drug- and fibre type cannabiswith drug type cannabis being made up of 12 varieties (i.e. 12 classes). The results show that linear classifiersare not able to manage the distribution of classes in which some overlap areas exist for both classificationproblems. Unlike linear classifiers, NNC and RBF SVMs best differentiate cannabis samples both for 2-class and12-class classifications with average classification results up to 99% and 98%, respectively. Furthermore, RBFSVMs correctly classified into drug type cannabis the independent validation set, which consists of cannabisplants coming from police seizures. In forensic case work this study shows that the discrimination betweencannabis samples at an early stage of growth is possible with fairly high classification performance fordiscriminating between cannabis chemotypes or between drug type cannabis varieties.

Soil water erosion under different cultivation systems and different fertilization rates and forms over 10 years

The action of rain and surface runoff together are the active agents of water erosion, and further influences are the soil type, terrain, soil cover, soil management, and conservation practices. Soil water erosion is low in the no-tillage management system, being influenced by the amount and form of lime and fertilizer application to the soil, among other factors. The aim was to evaluate the effect of the form of liming, the quantity and management of fertilizer application on the soil and water losses by erosion under natural rainfall. The study was carried out between 2003 and 2013 on a Humic Dystrupept soil, with the following treatments: T1 - cultivation with liming and corrective fertilizer incorporated into the soil in the first year, and with 100 % annual maintenance fertilization of P and K; T2 - surface liming and corrective fertilization distributed over five years, and with 75 % annual maintenance fertilization of P and K; T3 - surface liming and corrective fertilization distributed over three years, and with 50 % annual maintenance fertilization of P and K; T4 - surface liming and corrective fertilization distributed over two years, and with 25 % annual maintenance fertilization of P and K; T5 - fallow soil, without liming or fertilization. In the rotation the crops black oat (Avena strigosa ), soybean (Glycine max ), common vetch (Vicia sativa ), maize (Zea mays ), fodder radish (Raphanus sativus ), and black beans (Phaseolus vulgaris ). The split application of lime and mineral fertilizer to the soil surface in a no-tillage system over three and five years, results in better control of soil losses than when split in two years. The increase in the amount of fertilizer applied to the soil surface under no-tillage cultivation increases phytomass production and reduces soil loss by water erosion. Water losses in treatments under no-tillage cultivation were low in all crop cycles, with a similar behavior as soil losses.

Microfabricated dual sprayer for on-line mass tagging of phosphopeptides.

Phosphopeptides tagging reactions by dinuclear zinc(II) complexes (1,3-bis[bis(2-pyridylmethyl)amino]-propan-2-olato dizinc(II)3+, called tag) were performed with a dual-channel microsprayer in electrospray ionization mass spectrometry. The reaction is first studied ex situ and analyzed with a commercial electrospray source. In situ reactions (i.e., inside the Taylor cone) were achieved with a dual-channel microsprayer both with the tag synthesized chemically before the experiments and with the tag electrogenerated by in situ oxidation of a zinc electrode, also used to apply the electrospray current. The device consists of a polyimide microchip with two microchannels (20 microm x 50 microm x 1 cm) etched on each side of the structure and connecting only at the tip of the microchip. We demonstrate here that mixing two solutions with different physicochemical properties inside the Taylor cone can be used to selectively tag target molecules.

Characterization and classification of matrix effects in biological samples analyses.

An exhaustive classification of matrix effects occurring when a sample preparation is performed prior to liquid-chromatography coupled to mass spectrometry (LC-MS) analyses was proposed. A total of eight different situations were identified allowing the recognition of the matrix effect typology via the calculation of four recovery values. A set of 198 compounds was used to evaluate matrix effects after solid phase extraction (SPE) from plasma or urine samples prior to LC-ESI-MS analysis. Matrix effect identification was achieved for all compounds and classified through an organization chart. Only 17% of the tested compounds did not present significant matrix effects.

A Small World Perspective on Urban Systems

The theory of small-world networks as initiated by Watts and Strogatz (1998) has drawn new insights in spatial analysis as well as systems theory. The theoryâeuro?s concepts and methods are particularly relevant to geography, where spatial interaction is mainstream and where interactions can be described and studied using large numbers of exchanges or similarity matrices. Networks are organized through direct links or by indirect paths, inducing topological proximities that simultaneously involve spatial, social, cultural or organizational dimensions. Network synergies build over similarities and are fed by complementarities between or inside cities, with the two effects potentially amplifying each other according to the âeurooepreferential attachmentâeuro hypothesis that has been explored in a number of different scientific fields (Barabási, Albert 1999; Barabási A-L 2002; Newman M, Watts D, Barabàsi A-L). In fact, according to Barabási and Albert (1999), the high level of hierarchy observed in âeurooescale-free networksâeuro results from âeurooepreferential attachmentâeuro, which characterizes the development of networks: new connections appear preferentially close to nodes that already have the largest number of connections because in this way, the improvement in the network accessibility of the new connection will likely be greater. However, at the same time, network regions gathering dense and numerous weak links (Granovetter, 1985) or network entities acting as bridges between several components (Burt 2005) offer a higher capacity for urban communities to benefit from opportunities and create future synergies. Several methodologies have been suggested to identify such denser and more coherent regions (also called communities or clusters) in terms of links (Watts, Strogatz 1998; Watts 1999; Barabási, Albert 1999; Barabási 2002; Auber 2003; Newman 2006). These communities not only possess a high level of dependency among their member entities but also show a low level of âeurooevulnerabilityâeuro, allowing for numerous redundancies (Burt 2000; Burt 2005). The SPANGEO project 2005âeuro"2008 (SPAtial Networks in GEOgraphy), gathering a team of geographers and computer scientists, has included empirical studies to survey concepts and measures developed in other related fields, such as physics, sociology and communication science. The relevancy and potential interpretation of weighted or non-weighted measures on edges and nodes were examined and analyzed at different scales (intra-urban, inter-urban or both). New classification and clustering schemes based on the relative local density of subgraphs were developed. The present article describes how these notions and methods contribute on a conceptual level, in terms of measures, delineations, explanatory analyses and visualization of geographical phenomena.

A global marine sedimentary response to the end-Permian mass extinction: Examples from southern Turkey and the western United States

The end-Permian mass extinction greatly diminished marine diversity and brought about a whole-scale restructuring of marine ecosystems; these ecosystem changes also profoundly affected the sedimentary record. Data presented here, attained through facies analyses of strata deposited during the immediate aftermath of the end-Permian mass extinction (southern Turkey) and at the close of the Early Triassic (southwestern United States), in combination with a literature review, show that sedimentary systems were profoundly affected by: (1) a reduction in biotic diversity and abundance and (2) long-term environmental fluctuations that resulted from the end-Permian crisis. Lower Triassic strata display widespread microbialite and carbonate seafloor fan development and contain indicators of suppressed infaunal bioturbation such as flat-pebble conglomerates and wrinkle structures (facies considered unusual in post-Cambrian subtidal deposits). Our observations suggest that depositional systems, too, respond to biotic crises, and that certain facies may act as barometers of ecologic and environmental change independent of fossil assemblage analyses. Close investigation of facies changes during other critical times in Earth history may serve as an important tool in interpreting the ecology of metazoans and their environment.

Applications of MALDI-TOF mass spectrometry in clinical diagnostic microbiology.

Until recently, microbial identification in clinical diagnostic laboratories has mainly relied on conventional phenotypic and gene sequencing identification techniques. The development of matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) devices has revolutionized the routine identification of microorganisms in clinical microbiology laboratories by introducing an easy, rapid, high throughput, low-cost, and efficient identification technique. This technology has been adapted to the constraint of clinical diagnostic laboratories and has the potential to replace and/or complement conventional identification techniques for both bacterial and fungal strains. Using standardized procedures, the resolution of MALDI-TOF MS allows accurate identification at the species level of most Gram-positive and Gram-negative bacterial strains with the exception of a few difficult strains that require more attention and further development of the method. Similarly, the routine identification by MALDI-TOF MS of yeast isolates is reliable and much quicker than conventional techniques. Recent studies have shown that MALDI-TOF MS has also the potential to accurately identify filamentous fungi and dermatophytes, providing that specific standardized procedures are established for these microorganisms. Moreover, MALDI-TOF MS has been used successfully for microbial typing and identification at the subspecies level, demonstrating that this technology is a potential efficient tool for epidemiological studies and for taxonomical classification.

Potato plant growth and macronutrient uptake as affected by soil tillage and irrigation systems

The objective of this study was to evaluate potato plant growth and macronutrient uptake, as affected by soil tillage methods, in sprinkle and drip irrigated experiments. Eight treatments were set: T1, no tillage, except for furrowing before planting; T2, one subsoiling (SS); T3, twice rotary hoeing (RH); T4, one disc plowing (DP) + twice disc harrow leveling (DL); T5, 1DP + 2DL + 1RH; T6, 1DP + 2DL + 2RH; T7, 1SS + T6; T8, one moldboard plowing (MP) + 2DL. Treatments were arranged in a randomized block design with four replications. In both irrigation systems, plants presented higher emergence velocity index (EVI), when the soil was not tillaged, and the EVI was inversely related to the maximum tuber dry mass production. In both experiments, a functional direct relationship was found between the leaf area index and maximum tuber dry mass yield. The growth of plant organs (tuber, leaf, stem and root) and the macronutrient (N, P, K, Ca and Mg) contents in potato plant responded positively to a deeper soil revolving caused by plowing, especially with moldboard plow.