Scales Of Temporal Resolution Research Articles

Co-occurrence models fail to infer underlying patterns of avoidance and aggregation when closure is violated.

Advances in multi‐species monitoring have prompted an increase in the use of multi‐species occupancy analyses to assess patterns of co‐occurrence among species, even when data were collected at scales likely violating the assumption that sites were closed to changes in the occupancy state for the target species. Violating the closure assumption may lead to erroneous conclusions related to patterns of co‐occurrence among species. Occurrence for two hypothetical species was simulated under patterns of avoidance, aggregation, or independence, when the closure assumption was either met or not. Simulated populations were sampled at two levels (N = 250 or 100 sites) and two scales of temporal resolution for surveys. Sample data were analyzed with conditional two‐species occupancy models, and performance was assessed based on the proportion of simulations recovering the true pattern of co‐occurrence. Estimates of occupancy were unbiased when closure was met, but biased when closure violations occurred; bias increased when sample size was small and encounter histories were collapsed to a large‐scale temporal resolution. When closure was met and patterns of avoidance and aggregation were simulated, conditional two‐species models tended to correctly find support for non‐independence, and estimated species interaction factors (SIF) aligned with predicted values. By contrast, when closure was violated, models tended to incorrectly infer a pattern of independence and power to detect simulated patterns of avoidance or aggregation that decreased with smaller sample size. Results suggest that when the closure assumption is violated, co‐occurrence models often fail to detect underlying patterns of avoidance or aggregation, and incorrectly identify a pattern of independence among species, which could have negative consequences for our understanding of species interactions and conservation efforts. Thus, when closure is violated, inferred patterns of independence from multi‐species occupancy should be interpreted cautiously, and evidence of avoidance or aggregation is likely a conservative estimate of true pattern or interaction.

STIR-Net: Spatial-Temporal Image Restoration Net for CT Perfusion Radiation Reduction

Abstract Computed Tomography Perfusion (CTP) imaging is a cost-effective and fast approach to provide diagnostic images for acute stroke treatment. Its cine scanning mode allows the visualization of anatomic brain structures and blood flow; however, it requires contrast agent injection and continuous CT scanning over an extended time. In fact, the accumulative radiation dose to patients will increase health risks such as skin irritation, hair loss, cataract formation, and even cancer. Solutions for reducing the radiation exposure include reducing the tube current and/or shortening the X-ray radiation exposure time. However, images scanned at lower tube currents are usually accompanied by higher levels of noise and artifacts. On the other hand, a shorter X-ray radiation exposure time with longer scanning intervals will lead to image information that is insufficient to capture the blood flow dynamics between frames. Thus, it is critical for us to seek a solution that can preserve the image quality when the tube current and the temporal frequency are both low. We propose STIR-Net in this paper, an end-to-end spatial-temporal convolutional neural network structure, which exploits multi-directional automatic feature extraction and image reconstruction schema to recover high-quality CT images effectively. With the inputs of low-dose and low-resolution patches at different cross-sections of the spatio-temporal data, STIR-Net blends the features from both spatial and temporal domains to reconstruct high-quality CT volumes. In this study, we finalize extensive experiments to appraise the image restoration performance at different levels of tube current and spatial and temporal resolution scales. The results demonstrate the capability of our STIR-Net to restore high-quality scans at as low as 11% of absorbed radiation dose of the current imaging protocol, yielding an average of 10% improvement for perfusion maps compared to the patch-based log likelihood method.

Multiproxy evidence for leaf-browsing and closed habitats in extinct proboscideans (Mammalia, Proboscidea) from Central Chile

Proboscideans are so-called ecosystem engineers and are considered key players in hypotheses about Late Pleistocene megafaunal extinctions. However, knowledge about the autoecology and chronology of the proboscideans in South America is still open to debate and raises controversial views. Here, we used a range of multiproxy approaches and new radiocarbon datings to study the autoecology of Chilean gomphotheres, the only group of proboscideans to reach South America during the Great American Biotic Interchange (∼3.1 to 2.7 million years before present). As part of this study, we analyzed stable isotopes, dental microwear, and dental calculus microfossils on gomphothere molars from 30 Late Pleistocene sites (31° to 42°S). These proxies provided different scales of temporal resolution, which were then combined to assess the dietary and habitat patterns of these proboscideans. The multiproxy study suggests that most foraging took place in relatively closed environments. In Central Chile, there is a positive correlation between lower δ13C values and an increasing consumption of arboreal/scrub elements. Analyses of dental microwear and calculus microfossils have verified these leaf-browsing feeding habits. From a comparative perspective, the dietary pattern of South American gomphotheres appears to be constrained more by resource availability than by the potential dietary range of the individual taxa. This multiproxy study is aimed at increasing knowledge of the life history of gomphotheres and thus follows an issue considered one of the greatest challenges for paleontology in South America, recently pointed out by the need to thoroughly understand the role of ecological engineers before making predictions about the consequences of ecosystem defaunation.

The adaptive significance of crustacean hyperglycaemic hormone (CHH) in daily and seasonal migratory activities of the Christmas Island red crab Gecarcoidea natalis

The Christmas Island red crab Gecarcoidea natalis undergoes extreme changes in metabolic status, ranging from inactivity during the dry season, to a spectacular annual breeding migration at the start of the wet season. The dramatic change in metabolic physiology that this polarisation entails should be reflected in changes in endocrine physiology, particularly that of the crustacean hyperglycaemic hormone (CHH), of which we know relatively little. CHH levels were measured using a novel ultrasensitive time-resolved fluoroimmunoassay (TR-FIA), together with metabolites (glucose, lactate), in the field at several scales of temporal resolution, during migratory activities (wet season) and during the inactive fossorial phase (dry season). Release patterns of CHH were measured during extreme (forced) exercise, showing for the first time an unexpectedly rapid pulsatile release of this hormone. A seasonally dependent glucose-sensitive negative-feedback loop was identified that might be important in energy mobilisation during migration. Haemolymph lactate levels were strongly correlated with CHH levels in both field and experimental animals. During migration, CHH levels were lower than during the dry season and, during migration, daytime CHH levels (when most locomotor activity occurred) increased. However, the intense dawn activity in both dry and wet seasons was not always associated with repeatable hyperglycaemia or CHH release. The results obtained are discussed in relation to the life history and behaviour of G. natalis.

Response of daily and annual shell growth patterns of the intertidal bivalve Phacosoma japonicum to Holocene coastal climate change in Japan

It is well known that the latitudinal distribution of shallow-marine mollusks around the Japanese Islands has changed episodically during the Holocene in response to climate change; however, few studies have investigated the biotic response of individual species to Holocene climate change at daily to annual scales of temporal resolution. Holocene shells of the venerid Phacosoma japonicum recovered from various sites in Japan preserve lunar daily and annual growth increments in the outer shell layer and hinge plate, as also observed in modern shells. Sclerochronological analysis of 29 shells with known 14C ages revealed that the life history traits (e.g., length of the growth period and the mean lunar daily growth rate at a given age) have changed markedly over the past 8000 years. The intra-annual patterns of lunar daily increments and shell oxygen isotope records of the fossil shells can be compared with those of modern specimens from various locations, which are constrained by annual patterns of seawater temperature and summer–autumn precipitation rates. Furthermore, the temporal trends recorded in the fossil shells show a strong correlation with climate change in mid-latitude parts of East Asia over the past 8000 years, indicating that P. japonicum has changed its life history traits in response to Holocene terrestrial and coastal climate change.

Three scales of temporal resolution from automated soil respiration measurements

Soil respiration ( R s) is a combination of autotrophic and heterotrophic respiration, but it is often modeled as a single efflux process, influenced by environmental variables similarly across all time scales. Continued progress in understanding sources of variation in soil CO 2 efflux will require development of R s models that incorporate environmental influences at multiple time scales. Coherence analysis, which requires high temporal frequency data on R s and related environmental variables, permits examination of covariation between R s and the factors that influence it at varying temporal frequencies, thus isolating the factors important at each time scale. Automated R s measurements, along with air, soil temperature and moisture were collected at half hour intervals at a temperate forest at Harvard Forest, MA in 2003 and a boreal transition forest at the Howland Forest, ME in 2005. As in other temperate and boreal forests, seasonal variation in R s was strongly correlated with soil temperature. The organic and mineral layer water contents were significantly related to R s at synoptic time scales of 2–3 days to weeks, representing the wetting and drying of the soils as weather patterns move across the region. Post-wetting pulses of R s were correlated with the amount of precipitation and the magnitude of the change from pre-wet-up moisture content to peak moisture content of the organic horizon during the precipitation events. Although soil temperature at 8–10 cm depth and R s showed strong coherence at a 24-h interval, calculated diel Q 10 values for R s were unreasonably high (6–74) during all months for the evergreen forest and during the growing season for the deciduous forest, suggesting that other factors that covary with soil temperature, such as canopy assimilatory processes, may also influence the diel amplitude of R s. Lower diel Q 10 values were obtained based on soil temperature measured at shallower depths or with air temperature, but the fit was poorer and a lag was needed to improve the fit (peak R s followed peak air temperature by several hours), suggesting a role for delayed substrate supply from aboveground processes to affect diel patterns of R s. High frequency automated R s datasets afford the opportunity to disentangle the temporal scales at which environmental factors, such as seasonal temperature and phenology, synoptic weather events and soil moisture, and diel variation in temperature and photosynthesis, affect soil respiration processes.

Macrostratigraphy and Its Promise for Paleobiology

Macrostratigraphy is the study and statistical analysis of sediment packages that formed continuously at a specified scale of temporal resolution and that are bound by gaps recognizable at that same scale. The temporal ranges of gap-bound packages, compiled separately for different geographic locations, permit area-weighted, survivorship-based measures of rock quantity and spatio-temporal environmental continuity to be measured. Analytical basin fill models suggest that the parameters controlling sedimentation and sequence stratigraphic architecture, such as base level and sediment supply, can be detected quantitatively by macrostratigraphy.Macrostratigraphic analysis of the marine sedimentary rock record in the United States at a temporal resolution of ~106 years reproduces most of the well-known Sloss sequences, but it also identifies two prominent megasequences, the Paleozoic and Modern megasequences, which are separated by a Permian-Triassic discontinuity and Phanerozoic minimum in rock quantity. Many short- and long-term features of the macroevolutionary history of marine animals are reproduced by macrostratigraphy, including 1) many patterns in genus richness, 2) patterns in rates of genus extinction and, to a lesser degree, rates of origination, and 3) patterns of extinction selectivity and the shifting relative richness of Sepkoski's Paleozoic and Modern evolutionary faunas. The extent to which macrostratigraphy reproduces the macroevolutionary history of marine animals transcends what is expected by geologically-controlled sampling biases. Instead, the processes which control the spatio-temporal dynamics of shelf sedimentation, including expansions and contractions of shallow epicontinental seas, have probably exerted a consistent influence on the macroevolutionary history of marine animals. Exploring the common cause hypothesis by putting fossils back into rocks and rocks into a new quantitative framework for physical environmental change holds considerable promise for paleobiology.

How does pattern of feeding and rate of nutrient delivery influence conditioned food preferences?

Ruminant herbivores have been shown to learn about food properties by associating food flavours with the food's post-ingestive consequences. Previous experimentation supporting the conditioned food aversion/preference hypothesis has generally employed very simple diet learning tasks which do not effectively represent the wide range of foods selected within single bouts typical of wild, free-ranging ruminant herbivores. We tested the ability of a ruminant herbivore to associate a food with artificially administered nutrient rewards in a designed experiment where we altered the temporal pattern of encounter with the food as well as the nature (fast or slow reward) of the post-ingestive outcome. Twenty-four goats were offered branches of Sitka spruce (SS) and Norway spruce (NS) for 4 h per day on two days per week for five weeks. The pattern of feeding varied with treatment such that the species on offer changed every hour (short) or every 2 h (long). The energy treatment altered the reward delivered during Sitka consumption so that animals were dosed either with predominantly sugar (rapidly fermented), predominantly starch (slower fermentation rate), or with water (placebo). Preference was measured on the day following each learning day. We expected that goats would find it easier to associate SS with post-ingestive rewards when the duration of encounter was longest, and that associations would be stronger with the most rapidly digested post-ingestive reward. In the event, goats did not alter their consumption of SS in response to the treatments. Our results suggest that at the scale of temporal resolution of encounters with different plant species (1-2 h), and at the different rates of experiencing post-ingestive consequences tested in this experiment, ruminants do not appear to discriminate the nutritive properties of foods predominantly through a post-ingestive feedback mechanism. They must, instead, use a range of cues-including post-ingestive consequences-to assess food properties.

High-resolution carbon isotope stratigraphy and mammalian faunal change at the Paleocene-Eocene boundary in the Honeycombs area of the southern Bighorn Basin, Wyoming

The Paleocene-Eocene boundary is defined chemostratigraphically by the onset of a 100 kyr global carbon isotope excursion (CIE) that is recognized in marine and continental strata in both carbonate and dispersed organic carbon. The CIE is important in continental sections as a proxy for environmental change during the Paleocene-Eocene thermal maximum (PETM), which clearly affected faunal succession in Paleocene-Eocene mammals. At Polecat Bench in the northern Bighorn Basin, Wyoming, the CIE is in the Willwood Formation, where it spans four mammalian biozones from part of latest Clarkforkian Cf-3, through early Wasatchian Wa-M (<b><i>Meniscotherium</i></b>), Wa-0, and part of Wa-1. Here we report a new high-resolution study of the CIE recorded in dispersed organic carbon of the basal Willwood Formation in the Honeycombs area of the southern Bighorn Basin. In the Honeycombs area, 58 isotope sample sites span a vertical section of 71 meters, including the Wa-M locality ‘Halfway Hill North’ with both <b><i>Meniscotherium</i></b> and a <b><i>Heptodon</i></b>-like perissodactyl, and the richly fossiliferous Wa-0 wash site ‘Castle Gardens’ yielding a diverse mammalian fauna. Total organic carbon in the Honeycombs section is in the range of 0.1 to 0.4 percent by weight, and CaCO₃ is notably lacking except in the Castle Gardens interval where it reaches ca. 3 percent. Pre- and post-excursion values of δ¹³C_org (PDB) range from −26.2 to −24.1 permil. The Honeycombs CIE is about 42 m thick, and CIE excursion values range from −28.8 to −26.2 permil. The Honeycombs area is important because the Halfway Hill North locality is the first to yield a perissodactyl from the Wa-M <b><i>Meniscotherium</i></b> zone, and the Castle Gardens wash site is the first to yield abundant evidence of Wa-0 zone microvertebrates. As at Polecat Bench, the <b><i>Meniscotherium</i></b> zone in the Honeycombs area is in the lower part of the CIE, and Castle Gardens and other Wa-0 sites are in the longer middle and upper parts, starting near or after the maximum negative excursion. The CIE at Polecat Bench includes a series of 5 to 6 negative-excursion landmarks that are spaced, on average, about 8.5 m apart. These are tentatively interpreted as 21-kyr orbital precession cycles like those inferred for paleosol color. Negative-excursion landmarks are present in the Honeycombs CIE as well, but these are fewer and part of the Honeycombs section may be condensed or missing. Full understanding of the Paleocene-Eocene biotic transition will require investigation at even finer scales of temporal resolution, and results reported here suggest that finer-scale studies should be possible.

Macrostratigraphy of North America

Abstract The geological record is a three‐dimensional mosaic of gap‐bound rock bodies that, at any given scale of temporal resolution, were emplaced more or less continuously. At any geographic location, the irregular alternation of processes responsible for the formation and destruction of rock bodies results in the preservation of hiatus‐bound rock packages that have a distinct time of first occurrence (base, or oldest portion), a time of last occurrence (top, or youngest portion), and a suite of defining characters (e.g., lithologies, thickness, fossils, etc.). Many important aspects of the geologic record can be quantified by compiling the stratigraphic ranges of such gap‐bound rock packages. These include the quantity and spatial and temporal distribution of preserved rock, rates of rock formation, sequence stratigraphic architecture, and area‐weighted rates of expansion and contraction/erosional truncation of rock emplacement settings. This approach to characterizing the rock record is scalable, per...

Web traffic demand forecasting using wavelet‐based multiscale decomposition

In this paper we propose an experimental forecasting strategy taking into account the long-range dependence of aggregate network traffic, and we apply it to provide one-minute-ahead World-Wide Web (Web) traffic demand forecasts in terms of average number of bytes transferred. Recently, statistical examination of Web traces have shown evidence that Web traffic arising from file transfers exhibits a behavior that is consistent with the notion of self-similarity. Essentially, self-similarity indicates that significant burstiness is present on a wide range of time scales (i.e., the process is long-range dependent). Hence the idea of exploiting this multiscale property with a view towards discovering and capturing regularities underlying the time series which may prove useful for short-term traffic load forecasting. We carry out a wavelet transform decomposition of the original series to decompose the traffic time series into varying scales of temporal resolution, with the aim of making the underlying temporal structures more tractable. In a second step, each individual wavelet series—supposed to capture some features of the series—is fitted with a dynamical recurrent neural network (DRNN) model to output the wavelet forecast. The latter are afterwards recombined to form the next-minute Web Traffic demand. The method is applied on a large set of HTTP logs and is shown to yield good results. © 2001 John Wiley & Sons, Inc.

Web traffic demand forecasting using wavelet‐based multiscale decomposition

In this paper we propose an experimental forecasting strategy taking into account the long-range dependence of aggregate network traffic, and we apply it to provide one-minute-ahead World-Wide Web (Web) traffic demand forecasts in terms of average number of bytes transferred. Recently, statistical examination of Web traces have shown evidence that Web traffic arising from file transfers exhibits a behavior that is consistent with the notion of self-similarity. Essentially, self-similarity indicates that significant burstiness is present on a wide range of time scales (i.e., the process is long-range dependent). Hence the idea of exploiting this multiscale property with a view towards discovering and capturing regularities underlying the time series which may prove useful for short-term traffic load forecasting. We carry out a wavelet transform decomposition of the original series to decompose the traffic time series into varying scales of temporal resolution, with the aim of making the underlying temporal structures more tractable. In a second step, each individual wavelet series—supposed to capture some features of the series—is fitted with a dynamical recurrent neural network (DRNN) model to output the wavelet forecast. The latter are afterwards recombined to form the next-minute Web Traffic demand. The method is applied on a large set of HTTP logs and is shown to yield good results. © 2001 John Wiley & Sons, Inc.

Linking ocean and continental records of paleoclimate: determining accord and discord in the records

Climate forcing at Milankovitch and sub-Milankovitch frequencies are evaluated by linking climate proxy data from terrestrial and ocean records at two scales of temporal resolution: (1) relatively coarse resolution over the last 500 ka, and (2) higher resolution over the last 200 ka. We determine that climate proxy data from Hole 810C represent regional atmospheric and oceanic processes by comparison to core V21-146, both from the northwest Pacific Ocean. We then compare records that represent regional deposition from China and the western Pacific Ocean to focus on determining Milankovitch-type periodicities. Grain-size and magnetic susceptibility data from Hole 810C and Baoji, China are contrasted using a time-sensitive wavelet analysis. The analyses indicate a general correspondence to Milankovitch frequencies, although there is no specific frequency that dominates throughout the records. Additionally, the dominant 100 000 year frequency does not occur at the same time in the records. These factors suggest that differences in the physical mechanisms of deposition may influence the preserved apparent periodicities in these environments. Lastly, climate events over the last 200 000 years in marine and terrestrial records are correlated at a high resolution in the depth domain. The last interglacial-glacial boundary is consistently placed at the base of a cold period (C20) in the terrestrial and marine records. However, the cold period in the ocean data (δ18O) is very subdued relative to terrestrial records. Additionally, over the last 200 ka, periods of minimum insolation intensity do not correspond linearly to periods of maximum grain-size and ice volume variations revealed in the loess and ocean records, respectively. However, during the warm interstadials of the last two glaciations, peaks in insolation intensity compare favourably with the times of strongest pedogenic development in the continental records and minimum ice volume in the ocean records.

Combining Neural Network Forecasts on Wavelet-transformed Time Series

We discuss a simple strategy aimed at improving neural network prediction accuracy, based on the combination of predictions at varying resolution levels of the domain under investigation (here: time series). First, a wavelet transform is used to decompose the time series into varying scales of temporal resolution. The latter provides a sensible decomposition of the data so that the underlying temporal structures of the original time series become more tractable. Then, a dynamical recurrent neural netork is trained on each resolution scale with the temporal-recurrent backpropagation algorithm. By virtue of its internal dynamic, this general class of dynamic connections network approximates the underlying law governing each resolution level by a system of non-linear difference equations. The individual wavelet scale forecasts are afterwards recombined to form the current estimate. The predictive ability of this strategy is assessed with the sunspot series.