Simplifying the interpretation of continuous time models for spatio-temporal networks

Autoregressive and moving average models for temporally dynamic networks treat time as a series of discrete steps which assumes even intervals between data measurements and can introduce bias if this assumption is not met. Using real and simulated data from the London Underground network, this paper illustrates the use of continuous time multilevel models to capture temporal trajectories of edge properties without the need for simultaneous measurements, along with two methods for producing interpretable summaries of model results. These including extracting ‘features’ of temporal patterns (e.g. maxima, time of maxima) which have utility in understanding the network properties of each connection and summarising whole-network properties as a continuous function of time which allows estimation of network properties at any time without temporal aggregation of non-simultaneous measurements. Results for temporal pattern features in the response variable were captured with reasonable accuracy. Variation in the temporal pattern features for the exposure variable was underestimated by the models. The models showed some lack of precision. Both model summaries provided clear ‘real-world’ interpretations and could be applied to data from a range of spatio-temporal network structures (e.g. rivers, social networks). These models should be tested more extensively in a range of scenarios, with potential improvements such as random effects in the exposure variable dimension.

     

Mineral-chemical studies of metamorphosed hydrothermal alteration in the Kristineberg volcanogenic massive sulfide district,Sweden

Mineralium Deposita - The massive sulfide deposits of the Kristineberg area, Sweden, occur within a 2- to 3-km-thick succession of felsic volcaniclastic rocks belonging to the Skellefte Group. The...     

Thermodynamic properties of hematite — ilmenite — geikielite solid solutions

A solution model is developed for rhombohedral oxide solid solutions having compositions within the ternary system ilmenite [(Fe ₂₊ ^s Ti ₄₊ ^1–s ) ^A (Fe ₂₊ ^1–s Ti ₄₊ ^s ) ^B O₃]-geikielite [(Mg ₂₊ ^t Ti ₄₊ ^1–t ) ^A (Mg ₂₊ ^1–t Ti ₄₊ ^t ) ^B O₃]-hematite [(Fe³⁺) ^A (Fe³⁺) ^B O₃]. The model incorporates an expression for the configurational entropy of solution, which accounts for varying degrees of structural long-range order (0s, t1) and utilizes simple regular solution theory to characterize the excess Gibbs free energy of mixing within the five-dimensional composition-ordering space. The 13 model parameters are calibrated from available data on: (1) the degree of long-range order and the composition-temperature dependence of the transition along the ilmenite-hematite binary join; (2) the compositions of coexisting olivine and rhombohedral oxide solid solutions close to the Mg–Fe²⁺ join; (3) the shape of the miscibility gap along the ilmenite-hematite join; (4) the compositions of coexisting spinel and rhombohedral oxide solid solutions along the Fe²⁺–Fe³⁺ join. In the course of calibration, estimates are obtained for the reference state enthalpy of formation of ulvöspinel and stoichiometric hematite (–1488.5 and –822.0 kJ/mol at 298 K and 1 bar, respectively). The model involves no excess entropies of mixing nor does it incorporate ternary interaction parameters. The formulation fits the available data and represents an internally consistent energetic model when used in conjuction with the standard state thermodynamic data set of Berman (1988) and the solution theory for orthopyroxenes, olivines and Fe–Mg titanomagnetite-aluminate-chromate spinels developed by Sack and Ghiorso (1989, 1990a, b). Calculated activity-composition relations for the end-members of the series, demonstrate the substantial degree of nonideality associated with interactions between the ordered and disordered structures and the dominant influence of the miscibility gap across much of the ternary system. The predicted shape of the miscibility gap, and the orientation of tie-lines relating the compositions of coexisting phases, display the effects of coupling between the excess enthalpy of solution and the degree of long-range order. One limb of the miscibility gap follows the composititiontemperature surface corresponding to the ternary second-order transition.     

Pollutant transport in a shallow unconfined aquifer in Perry,Ohio

The groundwater system in northern Perry Township in Lake County, Ohio, is a shallow, unconfined aquifer consisting of periglacial lake beach deposits and less permeable lacustrine plain deposits. Groundwater flow is generally toward Lake Erie from south to north and is controlled by the top of the Ashtabula Till, but strong, local variations are caused by northward flowing streams During the study period, water levels in most wells exhibited a seasonal fluctuation of less than 0 3 m from their mean values. The areal distributions of chloride and nitrate concentrations indicate that road salt runoff easily infiltrates the aquifer and that nitrate may be sourced from fertilizer application. Ground-water flow and solute transport models indicate that in excess of 27 years are required to obtain chemical steady-state under hydrologic steady-state conditions. The simulations also demonstrate that nitrate loading must occur in more than one cultivated field in order to obtain the observed wide-spread nitrate distribution.     

Horseshoe crabs (Limulus polyphemus) in an urban estuary (Jamaica Bay, New York) and the potential for ecological restoration

We assessed the suitability of intertidal habitats for spawning by horseshoe crabs (Limulus polyphemus) at 12 proposed restoration sites identified by the United States Army Corps of Engineers along the shore of Jamaica Bay, a highly developed estuary in New York City. Based on beach geomorphology, we chose to quantify horseshoe crab activity at five of the sites during the May–July 2000 breeding season. Horseshoe crabs spawned intensively on small patches of suitable sand within larger areas of eroding shoreline with bulkheads and rubble fill. Small areas of sand behind grounded barges at Brant Point and Dubos Point had densities of over 100,000 eggs m⁻², which was equal to or greater than the egg densities on longer, more natural appearing beaches at Spring Creek and Dead Horse Bay, or at a sand spit at Bayswater State Park. There were no significant differences in the percentage of Jamaica Bay horseshoe crab eggs that completed development when cultured using water from Jamaica Bay or lower Delaware Bay, a less polluted location. Only 1% of the embryos from Jamaica Bay exhibited developmental anomalies, a frequency comparable to a previously studied population from Delaware Bay. We suggest that the distribution and abundance of horseshoe crabs at our study areas in Jamaica Bay is presently limited by the availability of suitable shoreline for breeding, rather than by water quality. Restoration efforts that increase the amount of sandy beach in this urban estuary have a good likelihood of benefiting horseshoe crabs and providing additional value to migrating shorebirds that use horseshoe crab eggs as food.     

剪断翼龙的翅膀：对于翼展估计和多样性的评论

Giant azhdarchids are perhaps the most spectacular pterosaurs known and, although they lack the elaborate headcrests or striking dentition that adorn other charismatic pterosaurs, the sheer enormity of their proportions has drawn a wealth of popular and scientific interest. Their remains, however, are so fragmentary that only limited interpretations of their anatomy and taxonomy can be made with certainty. Here, we report on recent investigations into giant azhdarchid taxonomy and size estimates that suggest both may be over-inflated.     

Stable carbon isotope ratio variations of organic matter in Orca Basin sediments

Orca Basin is a highly reducing basin on the slope of the Gulf of Mexico. Stable carbon isotope ratios and total organic carbon percentages were determined for two cores within the basin and one control core outside the basin. The results show that the organic carbon content of the basin cores is consistently 2–3 times greater than that of the control core. The Pleistocene-Holocene boundary, indicated by a break in the δ¹³C depth profile, occurs at a greater sediment depth in the basin cores than in the control core. A small sampling interval has made it possible to detect an unexplained fine structure in the δ¹³C profile not previously observed.     

Macro and microstructures as indicators of the development of syntectonic granitoids and host rocks in the Camboriú region,Santa Catarina,Brazil

In the northeastern Dom Feliciano Belt, Santa Catarina/Brazil, Paleoproterozoic rocks (mainly the Camboriú Complex) and Neoproterozoic granitoids – with the older Itapema Granite and the younger Corre-mar, Rio Pequeno and Serra dos Macacos granites – experienced a deformation history from magmatic to greenschist facies temperatures, under different rheological conditions. The concordance of flat amphibolite facies structures of the Camboriú Complex and magmatic and subsolidus structures in the Itapema Granite indicate the late-tectonic character of the latter. Based on tectonic features, the Corre-mar Granite is interpreted as older than the Rio Pequeno Granite and as related to transcurrent tectonics of the Southern Brazilian Shear Belt.In all granites, microstructures point to widespread magmatic alignment, followed by weak subsolidus and, locally, amphibolite to greenschist facies deformation. Magmatic foliations are progressively weaker in the younger granites. Synmagmatic shear zones in the Rio Pequeno Granite are possibly concentrated at the intrusive contact. The weak solid-state deformation at late-magmatic conditions argues for magmatism within a low-strain zone, which is compatible with the location of the area relative to the Major Gercino and Itajaí shear zones. The amphibolite to greenschist facies deformation structures are attributed to continuous deformation within the same low-strain zone during decreasing temperatures.     

Spawning origin of small,late-hatched Atlantic herring (Clupea harengus) larvae in a Maine estuary

Atlantic herring (Clupea harengus) larvae have been collected for resource monitoring purposes in the Sheepscot River in mid-coastal Maine during October–February, for the past 20 years. During this period, the larval population in the river has typically peaked in October-early November and has been composed of larvae derived from August–September spawning in eastern Maine and New Brunswick waters and from September-early October spawning along the central Maine coast. Larvae from eastern coastal spawning areas are transported to the river by the prevailing westerly coastal current. The appearance of small (≤15 mm SL) larvae in the river during December and January 1985–1989 suggested an additional time and area of origin. Aging procedures based on enumeration of daily otolith increments showed the majority of these small larvae were spawned from mid October to mid November when spawning usually occurs in western Maine coastal waters and in the vicinity of Jeffreys Ledge. Comparison of back-calculated hatching dates for small larvae collected in the river with wind direction and velocity data from mid October through November suggested that larvae were transported eastward against a weakened Gulf of Maine coastal current to the Sheepscot River by complex wind-driven surface currents that occur off the western Maine coast in the fall. *** DIRECT SUPPORT *** A01BY059 00003