Systematic Temporal Variations Research Articles

Probabilistic Forecasting of Bus Travel Time with a Bayesian Gaussian Mixture Model

Accurate forecasting of bus travel time and its uncertainty is critical to service quality and operation of transit systems: it can help passengers make informed decisions on departure time, route choice, and even transport mode choice, and it also support transit operators on tasks such as crew/vehicle scheduling and timetabling. However, most existing approaches in bus travel time forecasting are based on deterministic models that provide only point estimation. To this end, we develop in this paper a Bayesian probabilistic model for forecasting bus travel time and estimated time of arrival (ETA). To characterize the strong dependencies/interactions between consecutive buses, we concatenate the link travel time vectors and the headway vector from a pair of two adjacent buses as a new augmented variable and model it with a mixture of constrained multivariate Gaussian distributions. This approach can naturally capture the interactions between adjacent buses (e.g., correlated speed and smooth variation of headway), handle missing values in data, and depict the multimodality in bus travel time distributions. Next, we assume different periods in a day share the same set of Gaussian components, and we use time-varying mixing coefficients to characterize the systematic temporal variations in bus operation. For model inference, we develop an efficient Markov chain Monte Carlo (MCMC) algorithm to obtain the posterior distributions of model parameters and make probabilistic forecasting. We test the proposed model using the data from two bus lines in Guangzhou, China. Results show that our approach significantly outperforms baseline models that overlook bus-to-bus interactions, in terms of both predictive means and distributions. Besides forecasting, the parameters of the proposed model contain rich information for understanding/improving the bus service, for example, analyzing link travel time and headway correlation using covariance matrices and understanding time-varying patterns of bus fleet operation from the mixing coefficients. Funding: This research is supported in part by the Fonds de Recherche du Quebec-Societe et Culture (FRQSC) under the NSFC-FRQSC Research Program on Smart Cities and Big Data, the Canadian Statistical Sciences Institute (CANSSI) Collaborative Research Teams grants, and the Natural Sciences and Engineering Research Council (NSERC) of Canada. X. Chen acknowledges funding support from the China Scholarship Council (CSC). Supplemental Material: The e-companion is available at https://doi.org/10.1287/trsc.2022.0214 .

Spatial and temporal variation in δ13C values of methane emitted from a hemiboreal mire: methanogenesis, methanotrophy, and hysteresis

Abstract. The reasons for spatial and temporal variation in methane emission from mire ecosystems are not fully understood. Stable isotope signatures of the emitted methane can offer clues to the causes of these variations. We measured the methane emission (FCH4) and 13C signature (δ13C) of emitted methane by automated chambers at a hemiboreal mire for two growing seasons. In addition, we used ambient methane mixing ratios and δ13C to calculate a mire-scale 13C signature using a nocturnal boundary-layer accumulation approach. Microbial methanogenic and methanotrophic communities were determined by a captured metagenomics analysis. The chamber measurements showed large and systematic spatial variations in δ13C-CH4 of up to 15 ‰ but smaller and less systematic temporal variation. According to the spatial δ13C–FCH4 relations, methanotrophy was unlikely to be the dominating cause for the spatial variation. Instead, these were an indication of the substrate availability of methanogenesis being a major factor in explaining the spatial variation. Genetic analysis indicated that methanogenic communities at all sample locations were able to utilize both hydrogenotrophic and acetoclastic pathways and could thus adapt to changes in the available substrate. The temporal variation in FCH4 and δ13C over the growing seasons showed hysteresis-like behavior at high-emission locations, indicative of time-lagged responses to temperature and substrate availability. The upscaled chamber measurements and nocturnal boundary-layer accumulation measurements showed similar average δ13C values of −81.3 ‰ and −79.3 ‰, respectively, indicative of hydrogenotrophic methanogenesis at the mire. The close correspondence of the δ13C values obtained by the two methods lends confidence to the obtained mire-scale isotopic signature. This and other recently published data on δ13C values of CH4 emitted from northern mires are considerably lower than the values used in atmospheric inversion studies on methane sources, suggesting a need for revision of the model input.

Proterozoic supercontinent break-up as a driver for oxygenation events and subsequent carbon isotope excursions.

Oxygen and carbon are 2 elements critical for life on Earth. Earth's most dramatic oxygenation events and carbon isotope excursions (CIE) occurred during the Proterozoic, including the Paleoproterozoic Great Oxidation Event and the associated Lomagundi CIE, the Neoproterozoic Oxygenation event, and the Shuram negative CIE during the late Neoproterozoic. A specific pattern of a long-lived positive CIE followed by a negative CIE is observed in association with oxygenation events during the Paleo- and Neo-proterozoic. We present results from a carbon cycle model designed to couple the surface and interior cycling of carbon that reproduce this pattern. The model assumes organic carbon resides in the mantle longer than carbonate, leading to systematic temporal variations in the δ13C of volcanic CO2 emissions. When the model is perturbed by periods of enhanced continental weathering, increased amounts of carbonate and organic carbon are buried. Increased deposition of organic carbon allows O2 accumulation, while positive CIEs are driven by rapid release of subducted carbonate-derived CO2 at arcs. The subsequent negative CIEs are driven by the delayed release of organic C-derived CO2 at ocean islands. Our model reproduces the sequences observed in the Paleo- and Neo-proterozoic, that is oxygenation accompanied by a positive CIE followed by a negative CIE. Periods of enhanced weathering correspond temporally to supercontinent break-up, suggesting an important connection between global tectonics and the evolution of oxygen and carbon on Earth.

Source models of long-period seismic events at Galeras volcano, Colombia

SUMMARY Long-period (LP) seismic events have occurred repeatedly at Galeras volcano, Colombia, during the transition from effusive dome formation to explosive Vulcanian eruptions. Since 1989, two types of LP events have been observed there: one characterized by long-lasting, decaying harmonic oscillations (NLP events) and the other by non-harmonic oscillatory features (BLP events). NLP events are attributed to resonances of a dusty gas-filled crack in the magma plugging the eruptive conduit. Sixteen episodes of NLP events occurred at Galeras during 1992–2010, each characterized by systematic temporal variations in the frequencies and quality factors of NLP events. Our and previous estimates of crack model parameters during three of those NLP episodes indicate that the similar temporal variations in crack geometry and fluid properties can be explained by an increase in the ash content within the crack and a decrease in crack volume. We found that NLP events, associated with low SO2 fluxes, are anticorrelated with BLP events, which are accompanied by high SO2 emissions. From our observations and analytical results, we inferred that BLP events are generated by resonances of open cracks in the uppermost magma plug, corresponding to tuffisite veins, that efficiently transfer volcanic gases. After sufficient degassing and densification, the magma plug effectively seals the conduit. The growing overpressure in the deeper magma is then released through a shear fracture along the conduit margin. The intrusion of deeper, vesiculated magma into the shear fracture depressurizes and fragments the magma, producing a dusty gas and triggering the crack resonances that generate NLP events. Our results thus indicate that the evolution of the properties of the magma plug controls the occurrences of BLP and NLP events at Galeras. Although NLP events do not always precede explosive eruptions, they indicate that an important overpressure is building in the shallow conduit.

Associations between maternal circulating 25-hydroxyvitamin D concentration and birth outcomes-Mode of delivery and episiotomy rate: A prospective cohort study.

AimThe objective of this study was to evaluate associations between maternal concentrations of 25‐hydroxyvitamin D (25(OHD)) and birth outcomes: mode of delivery and episiotomy rate.Design and MethodsOne hundred and seventeen pregnant women were enrolled in an observational, longitudinal, prospective cohort study. Multivariable linear regression analyses were performed to assess relationships between maternal 25(OHD) concentrations and mode of delivery. To account for systematic temporal variation in 25(OHD), a cosinor model to the data was fitted.ResultsNo significant statistical associations were found between adjusted maternal 25(OHD) concentrations and risk of eutocic vaginal delivery, instrumented delivery (OR 1.05 [95% CI: 0.97–1.13]), primary Caesarean section (OR 0.99 [95% CI: 0.88–1.11]) or Caesarean section for any other causes (OR 1.04 [95% CI: 0.95–1.14]). High 25(OHD) levels tended to show a protective effect on performance of episiotomy, without reaching statistical significance (OR 0.36 [95% CI: 0.09, 1.37]).

Neotethyan Subduction Ignited the Iran Arc and Backarc Differently

AbstractMost arcs show systematic temporal and spatial variations in magmatism with clear shifts in igneous rock compositions between those of the magmatic front (MF) and those in the backarc (BA). It is unclear if similar magmatic polarity is seen for extensional continental arcs. Herein, we use geochemical and isotopic characteristics coupled with zircon U‐Pb geochronology to identify the different magmatic style of the Iran convergent margin, an extensional system that evolved over 100 Myr. Our new and compiled U‐Pb ages indicate that major magmatic episodes for the NE Iran BA occurred at 110–80, 75–50, 50–35, 35–20, and 15–10 Ma. In contrast to NE Iran BA magmatic episodes, compiled data from MF display two main magmatic episodes at 95–75 and 55–5 Ma, indicating more continuous magmatism for the MF than for the BA. We show that Paleogene Iran serves as a useful example of a continental arc under extension. Our data also suggest that there is not a clear relationship between the subduction velocity of Neotethyan Ocean beneath Iran and magmatic activity in Iran. Our results imply that the isotopic compositions of Iran BA igneous rocks do not directly correspond to the changes in tectonic processes or geodynamics, but other parameters such as the composition of lithosphere and melt source(s) should be considered. In addition, changes in subduction zone dynamics and contractional versus extensional tectonic regimes influenced the composition of MF and BA magmatic rocks. These controls diminished the geochemical and isotopic variations between the magmatic front and backarc.

Rhenium in Molybdenite: a Database Approach to Identifying Geochemical Controls on the Distribution of a Critical Element

Molybdenite is the world’s principal source of rhenium (Re), a critical element in multiple high-tech applications. However, the Re contents in molybdenite vary by orders of magnitude on scales ranging from single grains to whole deposits. In order to better understand the systematics of this variation and what geochemical factors control molybdenite Re concentration, and hence overall Re resources, we examine global patterns in molybdenite Re contents through a compilation of > 3000 measurements of Re in molybdenite from > 700 mainly ore-bearing moderate- to high-temperature hydrothermal systems of different types. Our results are similar to but expand on those of earlier studies. Rhenium concentration in molybdenite has a lognormal distribution and varies systematically with type of geologic system, intrusive lithology, and Mo grade. The lowest-Re molybdenite occurs in greisens (geometric mean 1 ppm ± a multiplicative standard deviation of 9), quartz vein-hosted W-Sn deposits (2 ± 5 ppm), unmineralized granites and granodiorites (12 ± 8 ppm), intrusion-related deposits (24 ± 8 ppm), and porphyry W-Sn deposits (16 ± 11 ppm). Rhenium is most enriched in molybdenites from volcanic sublimates (23,800 ± 5 ppm), with skarn Fe and Au (560 ± 5 ppm and 540 ± 3 ppm respectively) and porphyry Cu and Cu-Au deposits next (470 ± 4 and 430 ± 7 ppm respectively). Among porphyries, skarns, and quartz vein-hosted deposits, Re is most highly concentrated in molybdenites from Cu and Au systems and its concentration decreases systematically through Cu-Mo, Mo, Sn, and W deposits. In nearly all cases, molybdenites from systems associated with intermediate igneous rocks contain more Re than molybdenites from systems of the same type with more felsic rock associations. The disparity between Re contents of molybdenite in felsic and intermediate systems is largest for porphyries, quartz vein-hosted, and skarn deposits and is near zero for subeconomic or barren granite and granodiorite Mo systems; felsic intrusion-related deposits have slightly higher molybdenite Re than their equivalents associated with intermediate intrusions. In most systems, molybdenite Re content does not correlate with metal grade, but may have an inverse correlation with Au grade in intrusion-related deposits (based on a small number of data points) and does exhibit a strong inverse correlation with deposit Mo grade. Dilution of Re through larger amounts (higher deposit grades) of molybdenite explains about 40% of this correlation, but the relative enrichment of Re in molybdenite from low-Mo deposits must also reflect some selective enrichment of Re/Mo in porphyry Cu systems compared to porphyry Mo systems. We found no evidence for secular increase or other systematic temporal variation in molybdenite Re content. The data regarding the use of molybdenite Re content as a proxy for mantle influence are ambiguous. Nearly all observed empirical correlations can be traced back to differences in redox state and sulfide concentration, the two geochemical factors identified here and by previous experimental work as the controlling influences on Re mobility under hydrothermal conditions. Hydrothermal systems with reducing conditions (W- and Sn-rich) tend to have low molybdenite Re even though compiled whole-rock data indicate that their source rocks have as much or more Re as those of more oxidized systems (e.g., Cu-rich). Vapor-phase exsolution, crustal assimilation, and mixing with external fluids may all enrich molybdenite Re concentrations in individual deposits and deposit types, but their extent and importance in overall hydrothermal concentration of Re is uncertain. Thus, it appears that the available molybdenite Re resource in an ore deposit largely depends on how the deposit’s redox and sulfidation conditions have varied over time and space during the timespan of hydrothermal activity. Oxidized, high-sulfide conditions tend to concentrate Re in molybdenite, whereas reducing conditions tend to leave Re dispersed at low concentrations in the bulk rock.

Unfold: an integrated toolbox for overlap correction, non-linear modeling, and regression-based EEG analysis.

Electrophysiological research with event-related brain potentials (ERPs) is increasingly moving from simple, strictly orthogonal stimulation paradigms towards more complex, quasi-experimental designs and naturalistic situations that involve fast, multisensory stimulation and complex motor behavior. As a result, electrophysiological responses from subsequent events often overlap with each other. In addition, the recorded neural activity is typically modulated by numerous covariates, which influence the measured responses in a linear or non-linear fashion. Examples of paradigms where systematic temporal overlap variations and low-level confounds between conditions cannot be avoided include combined electroencephalogram (EEG)/eye-tracking experiments during natural vision, fast multisensory stimulation experiments, and mobile brain/body imaging studies. However, even “traditional,” highly controlled ERP datasets often contain a hidden mix of overlapping activity (e.g., from stimulus onsets, involuntary microsaccades, or button presses) and it is helpful or even necessary to disentangle these components for a correct interpretation of the results. In this paper, we introduce unfold, a powerful, yet easy-to-use MATLAB toolbox for regression-based EEG analyses that combines existing concepts of massive univariate modeling (“regression-ERPs”), linear deconvolution modeling, and non-linear modeling with the generalized additive model into one coherent and flexible analysis framework. The toolbox is modular, compatible with EEGLAB and can handle even large datasets efficiently. It also includes advanced options for regularization and the use of temporal basis functions (e.g., Fourier sets). We illustrate the advantages of this approach for simulated data as well as data from a standard face recognition experiment. In addition to traditional and non-conventional EEG/ERP designs, unfold can also be applied to other overlapping physiological signals, such as pupillary or electrodermal responses. It is available as open-source software at http://www.unfoldtoolbox.org.

Solar Wind Dynamic Pressure Upstream From Saturn: Estimation From Magnetosheath Properties and Comparison With SKR

AbstractAn analytical method is developed by which measurements made by the Cassini spacecraft in Saturn's magnetosheath can be used to infer the upstream solar wind parameters, specifically the solar wind speed (Vsw) and the dynamic pressure (Pd). The method is validated by comparing the results with other estimates of these parameters, including the mSWiM MHD model and magnetopause and bow shock models applied to observed boundary crossings. The comparisons suggest that the new inferred Vsw are on average ~40 km/s lower than the mSWiM values, and the dynamic pressure values are slightly lower as well. We find few of the lower Pd values predicted by mSWiM, probably because Cassini would have been inside the expanded magnetosphere under such conditions. Systematic temporal variations such as interplanetary shocks do seem to be captured well, with arrival times within several days of the MHD prediction. Compared to dynamic pressures estimated from boundary crossings with well‐known magnetopause and bow shock models, the magnetosheath‐inferred dynamic pressure tends to be somewhat lower, but within the uncertainties of the analytical derivation. Comparison of the inferred dynamic pressure with observed Saturn's kilometric radiation (SKR) activity reveals several episodes of very good temporal tracking between dynamic pressure and SKR intensity, with relatively short time delays (4–5 hr), suggesting rather direct driving. Such good tracking intervals occur almost exclusively on the dawnside of the magnetosphere, where the dominant SKR source is visible. When the tracking is good, the SKR fluxes vary roughly as the square of the dynamic pressure.

Temporal Variations of Near‐Surface Anisotropy Induced by Hydraulic Fracturing at a Shale Play Site in Southwest China

AbstractKnowledge of the geometric properties of fractures and cracks in a petroleum reservoir is important to reservoir exploitation. When aligned and partially connected, fractures and cracks can act as conduits for fluid flow and thus can significantly increase the permeability of the reservoir. The aligned fractures and cracks, on the other hand, are an effective means to generate seismic anisotropy. In this study, we utilize the seismic data recorded by a vertical array installed in a shallow borehole at a shale play site in southwest China. By applying seismic interferometry to the ambient noise data recorded by 12 three‐component geophones, we extractPandSwaves propagating vertically along the borehole. TheSwaves show up to 20% velocity variations with respect to their polarization directions. Such largeSwave anisotropy can be explained by the horizontal transverse isotropic model and is likely caused by natural fractures that are widely present in the area and align approximately in the NE‐SW direction. During the 13‐day period of hydraulic fracking treatment, we also observe large and systematic temporal variations inSwave velocity, degree ofSwave polarization anisotropy, and fast polarization direction. By comparing our observations with normal strain changes calculated with a half‐space elastic model, we speculate that strain changes induced by hydraulic injection and fracturing are likely to be responsible for the observed temporal variations in seismic anisotropy. As such, seismic interferometry with shallow borehole acquisition might provide an alternative means to monitor hydraulic fracturing and wastewater injection in the future.

Let's talk about stress, baby! Infant-feeding practices and stress in the ancient Atacama desert, Northern Chile.

The transition to an agricultural economy is often presumed to involve an increase in female fertility related to changes in weaning practice. In particular, the availability of staple crops as complementary foods is hypothesized to allow earlier weaning in agricultural populations. In this study, our primary aim is to explore whether this model fits the agricultural transition in the Atacama Desert using incremental isotopic analysis. A secondary aim of this study is to identify isotopic patterns relating to weaning, and assess how these may be differentiated from those relating to early life stress. We use incremental isotopic analysis of dentine to examine changes in δ15 N and δ13 C values from infancy and childhood in sites of the Arica region (n = 30). We compare individuals from pre-agricultural and agricultural phases to establish isotopic patterns and relate these patterns to maternal diet, weaning trajectory and physiological stress. We find that there is no evidence for systematic temporal or geographic variation in incremental isotopic results. Instead, results from all time periods are highly variable, with weaning completed between 1.5 and 3.5 years. Characteristics of the incremental profiles indicate that both in utero and postnatal stress were a common part of the infant experience in the Atacama. In the Atacama Desert it appears that the arrival of agricultural crops did not result in uniform shifts in weaning behavior. Instead, infant and child diet seems to have been dictated by the broad-spectrum diets of the mothers, perhaps as a way of mitigating the stresses of the harsh desert environment.

A 65 k.y. time series from sediment-hosted glasses reveals rapid transitions in ocean ridge magmas

Studies of ocean ridge magmatism have been hampered by the difficulty in constructing time-series data over more than a few thousand years. Sediment rapidly covers newly formed ocean crust, and older rocks, even when recovered from fault scarps, cannot be dated accurately. Ridge eruptions, however, disperse pyroclastic glass over distances as far as 5 km, and these glasses have been shown to persist for thousands of years in on-ridge sediment push cores. Here we present data on such glasses from a piston core that impacted basement in much older (600 ka) sediment. The age of deposition was determined using established stratigraphic methods to date the host sediment, yielding an average sample resolution of a few thousand years and a continuous 65 k.y. time series. The new time-series data show systematic temporal variations in magma compositions related to a change to the dynamics of crustal storage, which led to greater extents of pre-eruptive differentiation. Shortly thereafter was a small but discernable shift toward more enriched primary melt compositions. These events coincide with the onset of enhanced crustal production, previously identified using seismic data and interpreted to reflect the capture of a hotspot by the ridge. These results show the long-term preservation of pyroclastic glasses and suggest that the construction of high-resolution volcanic stratigraphy over a million years or more may be possible at ocean ridges, using multiple piston cores that impact basement. Sediment-hosted glasses have the potential to transform ocean ridges from the volcanic setting with the worst time-series data to that with the best.

Comparison of four methods for spatial interpolation of estimated atmospheric nitrogen deposition in South China.

Spatial interpolation methods have been applied in many environmental research studies. However, it is still a controversial issue to select an appropriate interpolation method for the conversion of discrete sampling sites into continuous maps. This study aimed at selecting an optimal interpolation method to analyze the spatial pattern of atmospheric N deposition in South China. N deposition was calculated by 259 moss sample data. Four spatial interpolation methods, including inverse distance weighting (IDW), radial basis function (RBF), ordinary kriging (OK), and universal kriging (UK), were utilized for modeling the spatial distribution of N deposition. It is the first time that these methods were applied to analyze N deposition in South China. Validation method was used to evaluate the interpolation precision of the various methods, and the cross-validation method was used to evaluate their interpolation accuracy. Comparison of predicted values with measured values indicated that OK was the optimal method for analyzing the spatial distribution of N deposition in this study; it had the highest precision (mean error (ME)=-0.059, root-mean-square error (RMSE)=5.240, mean relative error (MRE)=0.129, mean absolute error (MAE)=4.007) and the lowest uncertainties (standard deviation (SD)=5.47, coefficient of variation (CV)=0.15). RBF produced similar results as good as OK, while the worst performed interpolation method was UK. By using the OK method for analyzing N deposition, this work revealed systematic temporal and spatial variations in atmospheric N deposition in South China.

The Luanchuan Mo–W–Pb–Zn–Ag magmatic–hydrothermal system in the East Qinling metallogenic belt, China: Constrains on metallogenesis from C–H–O–S–Pb isotope compositions and Rb–Sr isochron ages

The Luanchuan Mo–W–Pb–Zn–Ag polymetallic ore district is located in the East Qinling metallogenic belt on the southern margin of the North China Craton. Two ore fields (Nannihu and Yuku) are recognized in the district, and three types of deposits are identified from the two ore fields as follows: (1) the 6 proximal porphyry–skarn type Mo–W deposits occurring at the inner contact zone of the granite porphyries, (2) the 3 middle skarn–hydrothermal type Zn deposits, and (3) the 8 distal hydrothermal type Pb–Zn–Ag deposits at the periphery of the porphyry. We present C–H–O isotope compositions of hydrothermal quartz and calcite, S–Pb isotope compositions of sulfide minerals, and sphalerite Rb–Sr isochron ages from the 17 deposits. The geochemical and geochronological data from the two ore fields all show systematic temporal and spatial variation, and primarily lead to the following inferences. (1) The temperatures and salinities of the ore-forming fluids decreased during mineralization. The ore-forming fluids gradually evolved from magmatic water to mixed magmatic–meteoric water. (2) The metallogenic components were primarily derived from igneous rocks, with increasing proportions of the materials from the ore-bearing rocks. (3) The mineralization ages of these deposits are close (147–136Ma), which correspond to the emplacement of the granite intrusions. (4) The three types of deposits and the ore-related late Mesozoic intrusives constitute a unified magmatic–hydrothermal–mineralization system. Finally, we also suggest exploration strategies for the Luanchuan ore district.

Earnings sustainability, economic conditions and the value relevance of accounting information

Summary This study demonstrates that the value relevance of accounting information is influenced by the ability to capitalize investments in valuable resources. We use data from Sweden to show that firms that operate in industries in which accounting conservatism limits this capitalization display lower value relevance as a result of more unsustainable earnings components. However, when controlling for the different properties of sustainable and unsustainable earnings components, the difference vanishes. Moreover, we show that firms operating in industries in which more investments are immediately expensed display systematic temporal variations in the level of value relevance. We contend that economic conditions in the form of investment levels and growth expectations explain this variation. Thus, value relevance can be substantially affected by the prevailing economic context.

Radon signals at the Roded site, Southern Israel

Abstract. Temporal variations of radon in the geological environment (upper crust) are frequent and recognized as unique in terms of the signals encountered and for the lack of substantial and generally applicable explanations. The phenomena observed at the Roded site, located in arid southern Israel, illustrate this situation. The monitoring of radon in the last 10 years or more has been carried out in massive meta-diorite of the Precambrian basement block of Roded. The measurement is conducted using an alpha detector at a resolution of 15-min, placed in a borehole at a depth of 9 m, within a PVC casing to that depth. Systematic temporal variation patterns, manifested as large relative signals are composed of sub-diurnal (SDR) radon, multi-day (MD) and annual (AR) signals. The overall variation is dominated by the intense SDR signals which occur in some days, and may vary from background levels (5 counts or less) to peak values (attaining >1000 counts) and back to background at an interval of 6 to 12 h. Intervals of up to several tens of days without significant SDR signals interchange with times of intense daily occurrences of such signals. Their occurrence indicates very fast variations of radiation from radon at the point of measurement. The peak times, within the diurnal 24-h cycle of SDR signals occur preferentially at an interval of 14–16 h (UT+2). Spectral analysis indicates: (a) A diurnal periodicity composed of a primary 24-h and a secondary 12-h periodicity, which are attributed to the solar tide constituents S1 and S2. Tidal constituents indicative for gravity tide (O1, M2) are lacking; (b) An annual periodicity. A compound relation among the diurnal and annual periodicity is indicated by: (a) Continuous Wavelet Transform (CWT) analysis shows an overall annual structure with a modulation of the S1 and S2 periodicities; (b) Moving-time-window Fourier spectral analysis showing that the amplitudes of S1 and S2 vary in an annual pattern, with relatively high values in summer. The phase of S1, S2 and S3 shows a systematic multi-year variation. It is suggested that the significant signatures of the periodic phenomena and their modulations reflect a direct link with the solar radiation tide.

Radon signals in the Gavnunim intrusion, Makhtesh Ramon, Israel

SUMMARY High time resolution monitoring of radon, using alpha detectors is carried out in a massive and jointed syenite located in arid southern Israel. Monitoring, at a resolution of 15-min, is conducted in two boreholes at depths of 1.2 and 85–90 m, resulting in more then 10-yr long time-series. Systematic temporal variation patterns, manifesting large relative signals are composed primarily of annual (AR), multiday (MD), daily (DR) and subdaily (SDR) radon signals. Prominent periodic diurnal variation underlies the DR and probably also the SDR signals. Diurnal (S1; 24-hr) and semidiurnal (S2; 12-hr) constituents characterize the DR signal at 1.2 m, while diurnal constituents typical for gravity related periodicities (M2, O1) are absent. The amplitudes of the S1 and S2 constituents, calculated for consecutive 512-hr-long time intervals, exhibit a similar covarying annual pattern. The periodic DR signal is absent at 85 m depth. The peak time of SDR signals, occurring at this depth in some days, exhibits a 24- and 12-hr periodicity. Time offsets of hours are observed between time-series of the measured signal from the two levels. The lag, investigated for the decomposed MD signal, using consecutive 20-d-long time intervals, shows that the signal at depth systematically lags by around 9 hr the signal at the shallow level. Discrepancy of the patterns of the temporal variation in the time domain, and also in the frequency and time-frequency domains of the diurnal periodic components, indicates that atmospheric influence on the radon signal is insignificant even at a depth of 1.2 m. Combining this conclusion with the compound interrelations occurring among the different radon signals suggests that other geophysical processes are driving the radon variation in the subsurface geogas. It is suggested that the significant S1 and S2 signatures, and their modulations, are imposed on the radon system by means of a direct link with solar radiation tide.

Remote triggering of fault-strength changes on the San Andreas fault at Parkfield

Fault strength is a fundamental property of seismogenic zones, and its temporal changes can increase or decrease the likelihood of failure and the ultimate triggering of seismic events. Although changes in fault strength have been suggested to explain various phenomena, such as the remote triggering of seismicity, there has been no means of actually monitoring this important property in situ. Here we argue that approximately 20 years of observation (1987-2008) of the Parkfield area at the San Andreas fault have revealed a means of monitoring fault strength. We have identified two occasions where long-term changes in fault strength have been most probably induced remotely by large seismic events, namely the 2004 magnitude (M) 9.1 Sumatra-Andaman earthquake and the earlier 1992 M = 7.3 Landers earthquake. In both cases, the change possessed two manifestations: temporal variations in the properties of seismic scatterers-probably reflecting the stress-induced migration of fluids-and systematic temporal variations in the characteristics of repeating-earthquake sequences that are most consistent with changes in fault strength. In the case of the 1992 Landers earthquake, a period of reduced strength probably triggered the 1993 Parkfield aseismic transient as well as the accompanying cluster of four M > 4 earthquakes at Parkfield. The fault-strength changes produced by the distant 2004 Sumatra-Andaman earthquake are especially important, as they suggest that the very largest earthquakes may have a global influence on the strength of the Earth's fault systems. As such a perturbation would bring many fault zones closer to failure, it should lead to temporal clustering of global seismicity. This hypothesis seems to be supported by the unusually high number of M >or= 8 earthquakes occurring in the few years following the 2004 Sumatra-Andaman earthquake.

Comparison of interpolation methods for depth to groundwater and its temporal and spatial variations in the Minqin oasis of northwest China

Severe water shortages and dramatic declines in groundwater levels have resulted in environmental deterioration in the Minqin oasis, an arid region of northwest China. Understanding temporal and spatial variations in the depth to groundwater in the region is important for developing management strategies. Depth to groundwater records for 48 observation wells in the Minqin oasis were available for 22 years from 1981 to 2003, allowing us to compare three different interpolation methods based on three selected years (1981, 1990, 2002) as starting points. The three methods were inverse distance weighting (IDW), radial basis function (RBF), and kriging (including ordinary kriging (OK), simple kriging (SK), and universal kriging (UK)). Cross-validation was applied to evaluate the accuracy of the various methods, and two indices – the correlation coefficient ( R 2) and the root mean squared error ( RMSE) – were used to compare the interpolation methods. Another two indices – deviation of estimation errors ( σ) and 95% prediction interval (95 PPI) – were used to assess prediction errors. Comparison of interpolated values with observed values indicates that simple kriging is the optimal method for interpolating depth to groundwater in this region: it had the lowest standard deviation of estimation errors and smallest 95% prediction interval (95 PPI). By using the simple kriging method and an autoregressive model for depth to groundwater based on the data from 1981 to 2003, this work revealed systematic temporal and spatial variations in the depth to groundwater in the Minqin oasis. The water table has declined rapidly over the past 22 years, with the average depth to groundwater increasing from 4.95 m in 1981 to 14.07 m in 2002. We attribute the decline in the water table to excessive extraction and to decreases in irrigation channel leakage.

Comparing Methods for Accounting for Seasonal Variability in a Biomarker When Only a Single Sample Is Available: Insights From Simulations Based on Serum 25-Hydroxyvitamin D

In biomarker-disease association studies, the long-term average level of a biomarker is often considered the optimal measure of exposure. Long-term average levels may not be accurately measured from a single sample, however, because of systematic temporal variation. For example, serum 25-hydroxyvitamin D (25(OH)D) concentrations may fluctuate because of seasonal variation in sun exposure. Association studies of 25(OH)D and cancer risk have used different strategies to minimize bias from such seasonal variation, including adjusting for date of sample collection (DOSC), often after matching on DOSC, and/or using season-specific cutpoints to assign subjects to exposure categories. To evaluate and understand the impact of such strategies on potential bias, the authors simulated a population in which 25(OH)D levels varied between individuals and by season, and disease risk was determined by long-term average 25(OH)D. Ignoring temporal variation resulted in bias toward the null. When cutpoints that did not account for DOSC were used, adjustment for DOSC sometimes resulted in bias away from the null. Using season- or month-specific cutpoints reduced bias toward the null and did not cause bias away from the null. To avoid potential bias away from the null, using season- or month-specific cutpoints may be preferable to adjusting for DOSC.