基于网络和图的时空智能——概念、方法和应用

当前时空智能(SpaceTimeAI)和地理空间智能(GeoAI)已是热门的话题,该研究领域旨在将计算机科学的最新方法(如深度学习)应用于地理空间问题。虽然深度学习方法因其对栅格数据的自然适用性而在图像处理中取得了巨大成功,但仍未广泛应用于其他空间和时空数据类型。本文提出使用网络和图作为SpaceTimeAI或GeoAI的基本结构的倡议,并将其应用于城市研究中。相比于基于网格的表达,基于网络的结构更加精确和实用。图能实现对点、线、面/多边形/网格和网络等多种空间结构的表达。本文通过时空预测、聚类和时空优化等常用时空分析方法展示基于网络和图的时空智能分析的优势,并介绍其在交通出行、警务和公共卫生等领域的应用。     

Mitigating systematic error in topographic models derived from UAV and ground‐based image networks

High resolution digital elevation models (DEMs) are increasingly produced from photographs acquired with consumer cameras, both from the ground and from unmanned aerial vehicles (UAVs). However, although such DEMs may achieve centimetric detail, they can also display systematic broad‐scale error that restricts their wider use. Such errors which, in typical UAV data are expressed as a vertical ‘doming’ of the surface, result from a combination of near‐parallel imaging directions and inaccurate correction of radial lens distortion. Using simulations of multi‐image networks with near‐parallel viewing directions, we show that enabling camera self‐calibration as part of the bundle adjustment process inherently leads to erroneous radial distortion estimates and associated DEM error. This effect is relevant whether a traditional photogrammetric or newer structure‐from‐motion (SfM) approach is used, but errors are expected to be more pronounced in SfM‐based DEMs, for which use of control and check point measurements are typically more limited. Systematic DEM error can be significantly reduced by the additional capture and inclusion of oblique images in the image network; we provide practical flight plan solutions for fixed wing or rotor‐based UAVs that, in the absence of control points, can reduce DEM error by up to two orders of magnitude. The magnitude of doming error shows a linear relationship with radial distortion and we show how characterization of this relationship allows an improved distortion estimate and, hence, existing datasets to be optimally reprocessed. Although focussed on UAV surveying, our results are also relevant to ground‐based image capture. © 2014 The Authors. Earth Surface Processes and Landforms published by John Wiley & Sons Ltd.     

Some remarks on the thermodynamics of cyclones and anticyclones

Summary A quantitative theory of steady-state cyclones and anticyclones is based on a thermodynamic cycle, in which the surface frictional flow of air towards low pressure is compensated by an outflow aloft. It is shown that such a steady-state flow can only be sustained by a distribution of heat sources and sinks, depending on the surface windfield, the frictional drag, and the height of the return flow. When the two latter quantities are specified the steady-state surface wind- and pressure-pattern can be inferred from the disposition of heat sources and sinks in the atmosphere.Given the vertical profile of temperature and humidity it is possible to estimate the net heat added to or subtracted from different levels in the atmosphere by radiation processes. If, in addition, the balance between precipitation and evaporation is known, the distribution of sources and sinks in the free atmosphere may be determined, and this procedure may be made largely routine with such aids as the radiation chart. Hence it is possible to infer the structure, intensity and size of an equilibrium surface pressure-field from the computed disposition of sources and sinks in the atmosphere. In the future it should be possible to carry through such an analysis as part of routine forecasting procedure. It is shown that the surface pressure contrast in such monsoonal systems as the Siberian Winter High is of the order called for by the thermodynamic theory. The sustaining source and sink intensities needed for cyclones and anticyclones of average size and pressure-contrast are of the order of those occurring in nature.Cyclones and anticyclones are conveniently classified according to the nature of the sustaining sources or sinks. In particular, shallow systems, cold highs and warm lows are associated with low level sinks or sources, while systems of great vertical extent, warm highs and cold lows, are associated with sinks and sources at high level.

---

Zusammenfassung Eine quantitative Theorie stationärer Zyklonen und Antizyklonen wird auf einen thermodynamischen Kreisprozeß, basiert, in dem die reibungsbedingte Luftströmung an der Erdoberfläche gegen das Tiefdruckzentrum durch ein Abfließen in der Höhe kompensiert wird. Es wird gezeigt, daß eine solche stationäre Strömung nur durch eine Verteilung von Wärmequellen und Sinkstellen aufrecht erhalten werden kann, die durch das Windfeld am Boden, die Reibungsverzögerung und die Höhe des Rückkehrstromes bedingt ist. Wenn die zwei letztgenannten Größen bestimmt sind, kann aus der Verteilung der Wärmequellen und Sinkstellen in der Atmosphäre auf das Wind- und Druckfeld des stationären Zustandes am Boden geschlossen werden.Aus dem Vertikalprofil von Temperatur und Feuchtigkeit ist es möglich, den durch Strahlungsvorgänge in verschiedenen Höhen erzeugten Wärmeüberschuß oder-verlust abzuschätzen. Wenn außerdem die Bilanz zwischen Niederschlag und Verdunstung bekannt ist, kann auch die Verteilung der Quellen und Sinkstellen in der Atmosphäre bestimmt werden und dieses Verfahren kann auf Grund von Strahlungskarten Eingang in die tägliche Praxis finden. Damit ist es möglich, aus der berechneten Verteilung von Quellen und Sinkstellen Schlüsse auf die Struktur, die Intensität und die Ausdehnung eines im Gleichgewicht befindlichen Druckfeldes zu ziehen. In Zukunft mag sich auch die Möglichkeit ergeben, eine solche Analyse als Teil des täglichen Wetterdienstes durchzuführen.Es wird gezeigt, daß die Druckunterschiede am Boden in solchen Monsunsystemen wie dem sibirischen Winterhoch von der gleichen Größenordnung sind, wie sie sich aus der thermodynamischen Theorie voraussagen lassen. Die Intensität von Quellen und Sinkstellen, die in der freien Atmosphäre auftreten, hat die richtige Größe, um die Aufrechterhaltung von Zyklonen und Antizyklonen mit den Dimensionen und Druckunterschieden zu erklären, wie sie in der Natur beobachtet werden.Zyklonen und Antizyklonen werden praktischerweise auf Grund der Beschaffenheit der Quellen oder Sinkstellen eingeteilt, die sie aufrecht erhalten. Im speziellen sind seichte Systeme, wie Kältehoch und Wärmetief, mit Quellen oder Sinkstellen im tieferen Niveau verknüpft, während Systeme mit großer Vertikalerstreckung, wie Wärmehoch und Kältetief, mit Quellen oder Sinkstellen in größerer Höhe verbunden sind.

---

Résumé On développe une théorie quantitative des dépressions et anticyclones stationnaires sur la base d'un cycle thermodynamique fermé dans lequel le courant soumis au frottement et dirigé vers le centre dépressionnaire est compensé par un écoulement divergent en altitude. On montre qu'un tel système stationnaire ne peut se maintenir que par une répartition des sources et des puits de chaleur conditionnée par le champ de vent au sol, le retard dû au frottement et l'altitude du courant de retour. Si les deux dernières grandeurs sont déterminées, on peut connaître le champ de vent et de pression de l'état stationnaire au sol à partir de la répartition des sources chaudes et des puits dans l'atmosphère.Grâce au profil vertical de température et d'humidité, il est possible d'évaluer l'excès ou le défaut de chaleur produits par les processus radiatifs à différents niveaux. Si en outre le bilan entre les précipitations et l'évaporation est connu, on peut aussi déterminer la distribution des sources et des puits dans l'atmosphère et ce procédé pourra entrer dans le travail de routine à l'aide de cartes de rayonnement. Ainsi il est possible, à partir de la distribution calculée des sources et des puits, de tirer des conclusions sur la structure, l'intensité et l'étendue d'un champ de pression en équilibre. On peut envisager dans l'avenir la possibilité d'adjoindre une telle analyse à la prévision journalière du temps.On montre que les différences de pression au sol dans des systèmes de mousson tels que celui de l'anticyclone hivernal sibérien sont de l'ordre de grandeur prévu par la théorie thermodynamique. L'intensité des sources et des puits qui apparaissent dans l'atmosphère libre a la valeur voulue pour entretenir les dépressions et les anticyclones avec les dimensions et les différences de pression que l'on observe dans la nature.On peut classer les dépressions et les anticyclones d'une manière pratique d'après la nature des sources et des puits qui les entretiennent. En particulier, des systèmes de peu d'épaisseur comme des anticyclones froids et des dépressions chaudes sont liés à des sources et des puits situés à basse altitude; par contre des systèmes de grande ampleur verticale comme des anticyclones chauds et des dépressions froides sont liés à des sources ou des puits à haute altitude.

     

Hydraulic redistribution and hydrological controls on aspen transpiration and establishment in peatlands following wildfire

In the sub‐humid Western Boreal Plains of Alberta, where evapotranspiration often exceeds precipitation, trembling aspen (Populus tremuloides Michx.) uplands often depend on adjacent peatlands for water supply through hydraulic redistribution. Wildfire is common in the Boreal Plains, so the resilience of the transfer of water from peatlands to uplands through roots immediately following wildfire may have implications for aspen succession. The objective of this research was to characterize post‐fire peatland‐upland hydraulic connectivity and assess controls on aspen transpiration (as a measure of stress and productivity) among landscape topographic positions. In May 2011, a wildfire affected 90,000 ha of north central Alberta, including the Utikuma Region Study Area (URSA). Portions of an URSA glacio‐fluval outwash lake catchment were burned, which included forests and a small peatland. Within 1 year after the fire, aspen were found to be growing in both the interior and margins of this peatland. Across recovering land units, transpiration varied along a topographic gradient of upland midslope (0.42 mm hr^?1) > upland hilltop (0.29 mm hr^?1) > margin (0.23 mm hr^?1) > peatland (0.10 mm hr^?1); similar trends were observed with leaf area and stem heights. Although volumetric water content was below field capacity, P. tremuloides were sustained through roots present, likely before fire, in peatland margins through hydraulic redistribution. Evidence for this was observed through the analysis of oxygen (δ¹⁸O) and hydrogen (δ²H) isotopes where upland xylem and peat core signatures were ?10.0‰ and ?117.8‰ and ?9.2‰ and ?114.0‰, respectively. This research highlights the potential importance of hydraulic redistribution to forest sustainability and recovery, in which the continued delivery of water may result in the encroachment of aspen into peatlands. As such, we suggest that through altering ecosystem services, peatland margins following fire may be at risk to aspen colonization during succession.     

Nearshore suspended sediment variations,central Surinam coast

Nearshore suspended sediment concentration along the muddy Surinam coast is highly variable; maximum values are many times greater than on other muddy coasts. Water samples taken at four field stations during various stages of the tide range in concentration from 15 to 3,700 mg/l near the surface and from 100 to 30,000 mg/l near the bottom. Highest overall concentrations and greatest variability in concentration occur in water over large banks of fluid mud (thixotropic gel) that extend 2–3 km offshore and 5–10 km along-shore. On both the intertidal and subaqueous portions of these mudbanks highest concentrations are found at low tide. Results provide evidence that an exchange between fluid mud and suspended sediment takes place during each tidal cycle.     

Phosphorus Speciation in Stream Bed Sediments from an Agricultural Watershed: Solid-Phase Associations and Sorption Behavior

The sorption behavior and solid-phase associations of phosphorus (P) in fine-grained sediments (<63 μm) from two upstream tributaries and one downstream main stem site of the Spoon River in west-central Illinois were characterized to better understand phosphorus bioavailability in this agriculturally dominated watershed. The P sorption affinities, as indicated by linear distribution coefficients (K _d), of all sediments were 330–5,150 L/kg, and negatively correlated with equilibrium phosphorus concentration (EPC_o) values, which ranged between 0.2 and 2.2 μM. pH values measured at the conclusion of the sorption experiments varied only slightly (7.45–8.10) but were nonetheless strongly positively correlated to EPC_o values, and negatively correlated to K _d values, suggesting the importance of pH to the observed sorption behavior. K _d values were generally lower and EPC_o values higher at the main stem site than at the upstream tributary sites, suggesting dissolved reactive P (DRP) bioavailability (specifically orthophosphate) increased downstream. The solid phase associations of P were operationally assessed with the streamlined SEDEX (sedimentary extraction) procedure, and most sediment P (≥50%) was released during the step designed to determine iron oxide–associated P. On average, 70–90% of the total sediment P pool was potentially bioavailable, as estimated by the sum of the iron oxide-, authigenic carbonate-, and organic-associated P fractions. Considerable calcium was also extracted from some sediments during the step designed to specifically remove iron oxide–associated P. It is hypothesized that the severe drought conditions that persisted between April and October, 2005 allowed authigenic carbonates (perhaps partly amorphous) to accumulate, and that these carbonates dissolved during the iron oxide extraction step. The extensive benthic algal populations also present may have aided carbonate precipitation, which under more normal hydrologic conditions would be periodically flushed downstream and replaced by fresh sediment. This suggests antecedent hydrologic conditions played a dominant role in the P sorption and solid phase associations identified.     

Alfven Wave Reflection model of field-aligned currents at Mercury

An Alfven Wave Reflection (AWR) model is proposed that provides closure for strong field-aligned currents (FACs) driven by the magnetopause reconnection in the magnetospheres of planets having no significant ionospheric and surface electrical conductance. The model is based on properties of the Alfven waves, generated at high altitudes and reflected from the low-conductivity surface of the planet. When magnetospheric convection is very slow, the incident and reflected Alfven waves propagate along approximately the same path. In this case, the net field-aligned currents will be small. However, as the convection speed increases, the reflected wave is displaced relatively to the incident wave so that the incident and reflected waves no longer compensate each other. In this case, the net field-aligned current may be large despite the lack of significant ionospheric and surface conductivity. Our estimate shows that for typical solar wind conditions at Mercury, the magnitude of Region 1-type FACs in Mercury’s magnetosphere may reach hundreds of kilo-Amperes. This AWR model of field-aligned currents may provide a solution to the long-standing problem of the closure of FACs in the Mercury’s magnetosphere.     

Uplift and submarine formation of some Melanesian porphyry copper deposits: Stable isotope evidence

Hydrogen and oxygen isotope analyses of sericites and kaolinites from four young porphyry copper deposits (Ok Tedi (1.2 Ma) and Yandera (6.5 Ma), Papua New Guinea; Koloula (1.5 Ma), Solomon Islands; and Waisoi (<5 Ma), Fiji) indicate that the fluids from which these minerals precipitated were of mixed magmatic and non-magmatic sources. The non-magmatic component of the fluid from the island arc deposits (Koloula, Waisoi) was ocean water.For Ok Tedi, the non-magmatic component was a meteoric water with an isotopic composition different from that of the present meteoric water in the region. The isotopic signature of the former meteoric water is consistent with a surface elevation of 200 m a.s.l. or less at the time of mineralization. The deposit was later exposed and supergene kaolinitization commenced at approximately 1200 m a.s.l. Uplift and erosion has continued to the present at which time the elevation of the exposed deposit is 1800 m a.s.l. This rate of uplift is consistent with that known from other geological evidence. If the rate of uplift were approximately constant during the last 1.2 Ma, the age of supergene enrichment can be dated at approximately 0.4 Ma B.P.Similarly, influx of meteoric water at Yandera occurred when the ground surface above the deposit was at an elevation of approximately 600 m a.s.l. The deposit's present elevation is 1600 m a.s.l. In this case a total uplift of approximately 2.2 km is indicated, with removal of 1.2 km of overburden by erosion.     

Integral field spectroscopy with the GEMINI multiobject spectrographs

We describe the integral field unit (IFU) which converts the Gemini Multiobject Spectrograph (GMOS) installed on the Gemini-North telescope to an integral field spectrograph,which produces spectra over a contiguous field of view of 7 × 5 arcsec with spatial sampling of 0.2 arcsecover the wavelength range 0.4-1.0 μm.GMOS is converted to this mode by the remote insertion of the IFU into thebeam in place of the masks used for the multiobject mode. A separate fieldof half the area of the main field, but otherwise identical, is alsoprovided to improve background subtraction. The IFU contains 1500lenslet-coupled fibres and was the first facility of any type for integralfield spectroscopy employed on an 8/10 m telescope.We describe the design, construction and testing of the GMOS IFU and present measurements of the throughput both in the laboratory and at the telescope. We compare these with a theoretical prediction made before construction started. All are in good agreement with each other, with the on-telescope throughput exceeding 60% (averaged over wavelength). Finallywe show an example of data obtained during commissioning to illustrate the power of the device.     

Analytic Magnetic Reconnection Solutions in Cylindrical Geometry

In this paper we present a new class of exact reconnection solutions in cylindrical geometry. We point out that in the case of planar reconnection there is a natural cylindrical analog to the Cartesian Dawson function model for the magnetic field. Although the resistive energy release scalings of these solutions mimic the Cartesian models an important new feature is the presence of curvature in the current sheet. We go on to show that these solutions can be generalized to three dimensions.