This study aims to introduce contextual Neural Gas (CNG), a variant of the Neural Gas algorithm, which explicitly accounts for spatial dependencies within spatial data. The main idea of the CNG is to map spatially close observations to neurons, which are close with respect to their rank distance. Thus, spatial dependency is incorporated independently from the attribute values of the data. To discuss and compare the performance of the CNG and GeoSOM, this study draws from a series of experiments, which are based on two artificial and one real-world dataset. The experimental results of the artificial datasets show that the CNG produces more homogenous clusters, a better ratio of positional accuracy, and a lower quantization error than the GeoSOM. The results of the real-world dataset illustrate that the resulting patterns of the CNG are theoretically more sound and coherent than that of the GeoSOM, which emphasizes its applicability for geographic analysis tasks. 相似文献
The majority of cities are rapidly growing. This makes the monitoring and modeling of urban change’s spatial patterns critical to urban planners, decision makers, and environment protection activists. Although a wide range of methods exists for modeling and simulating urban growth, machine learning (ML) techniques have received less attention despite their potential for producing highly accurate predictions of future urban extents. The aim of this study is to investigate two ML techniques, namely radial basis function network (RBFN) and multi-layer perceptron (MLP) networks, for modeling urban change. By predicting urban change for 2010, the models’ performance is evaluated by comparing results with a reference map and by using a set of pertinent statistical measures, such as average spatial distance deviation and figure of merit. The application of these techniques employs the case study area of Mumbai, India. The results show that both models, which were tested using the same explanatory variables, produced promising results in terms of predicting the size and extent of future urban areas. Although a close match between RBFN and MLP is observed, RBFN demonstrates higher spatial accuracy of prediction. Accordingly, RBFN was utilized to simulate urban change for 2020 and 2030. Overall, the study provides evidence that RBFN is a robust and efficient ML technique and can therefore be recommended for land use change modeling. 相似文献
We investigate the production of gamma-rays in the inverse Compton (IC) scattering process by leptons accelerated inside relativistic
blobs in jets of active galactic nuclei. Leptons are injected homogeneously inside the spherical blob and initiate IC e± pair cascade in the synchrotron radiation (produced by the same population of leptons, SSC model), provided that the optical
depth for gamma-rays is larger than unity. It is shown that for likely parameters internal absorption of gamma-rays has to
be important. We suggest that new type of blazars might be discovered by the future simultaneous X-ray and γ-ray observations, showing peak emissions in the hard X-rays, and in the GeV γ-rays. Moreover, the considered scenario might be also responsible for the orphan X-ray flares recently reported from BL Lac
type active galaxies.
相似文献
Previous biogeographical research has investigated the mortality of woody plants, emphasizing tree size, shade tolerance, and competition individually as critical factors. Few researchers, though, have empirically evaluated the simultaneous and relative importance of these three factors in a single study. Based on a thirty-five-year study within the University of Kentucky Arboretum, we report that none of the factors exerted an overriding influence on the tree survivorship alone. Rather, they were tightly intertwined in a complex way. Our study indicates that the three factors have contributed perhaps equally or at least simultaneously to the overall demographic processes in the woodland. 相似文献
This paper reports an investigation on the performance of ethylene-vinyl acetate (EVA) foam when used as a cushion layer for rigid barriers used to resist debris flow. Large-scale pendulum impact tests were conducted to study the effects of layer thickness on cushion performance under six successive impacts. Results show that for the first impact at 70 kJ, the peak contact force with the EVA foam thickness of 500 mm is about twice larger compared to that of 1000 mm. Results also reveal that the cushion mechanism of elastic collapse of cell walls in the EVA foam results in peak contact forces and maximum transmitted loads that are up to 30 and 50% lower compared to gabions for the first impact at 70 kJ, respectively. The elastic behavior of EVA foam provides consistent cushioning efficiency. Furthermore, EVA foam is found to be susceptible to degradation by ultraviolet light so that a suitable coating layer is required for outdoor use. Polyurea was identified as a suitable coating material and a small-scale coating trial was performed to confirm this. Findings presented in this paper will have direct implications on the future design of cushion layers for rigid barriers used to intercept debris flows. 相似文献
Nitrogen (N) is one of the primary nutrients required to build biomass and is therefore in high demand in aquatic ecosystems. Estuaries, however, are frequently inundated with high concentrations of anthropogenic nitrogen, which can lead to substantially degraded water quality. Understanding drivers of biogeochemical N cycling rates and the microbial communities responsible for these processes is critical for understanding how estuaries are responding to human development. Estuaries are notoriously complex ecosystems: not only do individual estuaries by definition encompass gradients of salinity and other changing environmental conditions, but differences in physical parameters (e.g., bathymetry, hydrodynamics, tidal flushing) lead to a tremendous amount of variability in estuarine processes between ecosystems, as well. Here, we review the current knowledge of N cycling processes in estuaries carried out by bacteria and archaea, including both biogeochemical rate measurements and molecular characterizations of N cycling microbial communities. Particular attention is focused on identifying key environmental factors associated with distinct biogeochemical or microbial regimes across numerous estuaries. Additionally, we describe novel metabolisms or organisms that have recently been discovered but have not yet been fully explored in estuaries to date. While the majority of research has been conducted in the benthos, we also describe data from estuarine water columns. Understanding both the common patterns and the differences between estuaries has important implications for how these critical ecosystems respond to changing environmental conditions. 相似文献
Conventional hydrogeological practice is to formulate a conceptual model, which is often the basis of a numerical model. The numerical model is then used to test groundwater management strategies. A workflow is proposed, employing the numerically enhanced conceptual model (NECoM) of the Mean Sea Level Aquifer (MSLA) on the island of Malta. The Malta MSLA is overexploited and under threat of salinization. Data (heads, chloride concentrations, electrical conductivity logs, tidal tests and qualitative analyses) were assimilated into a fast-running numerical model. Simultaneously, strategies for optimal acquisition of further data were examined through the modelling process. The model was delivered through the Energy and Water Agency, with suggestions for flexible model deployment. These workflows will, hopefully, spawn model improvements through further revision of the base concepts. The model allows the agency to make predictions, which have uncertainties that are quantified and reduced through data assimilation as new data become available. Contemplated management plans can therefore be properly assessed before implementation. The proposed NECoM approach can be generalized since it bases model usage on the premise that modelling should make maximum use of existing data by assimilating its information content, thereby highlighting the uncertainties of decision-critical predictions that remain because of data insufficiency. Thus, the presently disjointed process of modelling on the one hand, and data acquisition on the other, can be better aligned. Conceptual and numerical model development become parallel, rather than sequential, activities. Together, they enable predictions of future system behaviour for which bias is reduced and uncertainties quantified.
Hydrogeology Journal - Geological and hydrogeological conceptualizations of the five main aquifers of Malta were performed by means of characterization of the groundwater bodies’ geometries... 相似文献