Temporal Variance Research Articles

Real-time subsurface scattering with single pass variance-guided adaptive importance sampling

In real-time applications, it is difficult to simulate realistic subsurface scattering with differing degrees translucency. Burley's reflectance approximation by empirically fitting the diffusion profile as a whole makes it possible to achieve realistic looking subsurface scattering for different translucent materials in screen space. However, achieving a physically correct result requires real-time Monte Carlo sampling of the analytic importance function per pixel per frame, which seems prohibitive to achieve. In this paper, we propose an approximation of the importance function that can be evaluated in real-time. Since subsurface scattering is more pronounced in certain regions (e.g., with light gradient change), we propose an adaptive sampling method based on temporal variance to lower the required number of samples. We propose a one phase adaptive sampling pass that is unbiased, and able to adapt to scene changes due to motion and lighting. To further improve the quality, we explore temporal reuse with a guiding pass prior to the final temporal anti-aliasing (TAA) phase that further improves the quality. Our local guiding pass does not constrain the TAA implementation, and only requires one additional texture to be passed between frames. Our proposed variance-guided algorithm has the potential to make stochastic sampling algorithm effective for real-time rendering.

Macroecological dynamics of gut microbiota.

The gut microbiota is now widely recognized as a dynamic ecosystem that plays an important role in health and disease. Although current sequencing technologies make it possible to explore how relative abundances of host-associated bacteria change over time, the biological processes governing microbial dynamics remain poorly understood. Therefore, as in other ecological systems, it is important to identify quantitative relationships describing various aspects of gut microbiota dynamics. In the present study, we use multiple high-resolution time series data obtained from humans and mice to demonstrate that, despite their inherent complexity, gut microbiota dynamics can be characterized by several robust scaling relationships. Interestingly, the observed patterns are highly similar to those previously identified across diverse ecological communities and economic systems, including the temporal fluctuations of animal and plant populations and the performance of publicly traded companies. Specifically, we find power-law relationships describing short- and long-term changes in gut microbiota abundances, species residence and return times, and the correlation between the mean and the temporal variance of species abundances. The observed scaling laws are altered in mice receiving different diets and are affected by context-specific perturbations in humans. We use the macroecological relationships to reveal specific bacterial taxa, the dynamics of which are substantially perturbed by dietary and environmental changes. Overall, our results suggest that a quantitative macroecological framework will be important for characterizing and understanding the complex dynamics of diverse microbial communities.

Investigation of controls on the regional soil moisture spatiotemporal patterns across different climate zones

Understanding soil moisture spatiotemporal variability at regional scales is of great importance for studying various ecohydrological and land surface processes. In this study, long-term soil moisture data (5 years) were obtained from three regional monitoring networks across the continental United States with contrasting climatic conditions, including the Enviro-weather Automated Weather Station Network in Michigan, the Nebraska Mesonet, and the Soil Climate Analysis Network in Utah. Both soil moisture spatial variance and temporal variance were decomposed into time-invariant and time-variant components. To evaluate the impacts of different environmental factors on soil moisture spatiotemporal variability and its contribution components, static (e.g., soil properties) and non-static (e.g., climatic variables) environmental factors were also compiled for the stations of each network. The results showed that the time-invariant component was the leading factor for controlling the soil moisture spatial variance in all study regions with marked seasonal variations due to changes in soil moisture wetness conditions. More importantly, the soil moisture spatial variance and its contribution components (in absolute values and relative contributions) were shown to be affected by both soil properties (e.g., soil texture) and climatic conditions (e.g., precipitation) with varying degrees of impacts among the study regions. Meanwhile, the results further revealed that depending on the region under consideration, static and non-static environmental factors could play important roles in determining soil moisture temporal dynamics and its contribution components at regional scales. Overall, this study provided additional observational evidence, which underscored the importance of local factors (e.g., soil properties) in determining soil moisture spatiotemporal variability at regional scales.

Typical pesticides diffuse loading and degradation pattern differences under the impacts of climate and land-use variations

Riverine sediment can reconstruct the history of organic pollution loads and can provide reliable temporal information for pesticide metabolite dynamics in watershed. Sediment core samples were collected from two riverine sections of a cold watershed base in the presence land use change under agricultural development, and the vertical concentrations of four pesticides (atrazine, prometryn, isoprothiolane, and oxadiazon) and two atrazine metabolites (deisopropyl-atrazine and deethyl-atrazine) were determined by gas chromatography–mass spectrometry. The presence of pesticides and metabolites was detected at different depths (11–17 cm) at 1-cm intervals along the two sediment cores, and the flux was calculated with a constant rate of supply model based on the observed concentrations and 210Pb isotope radioactivity chronology. By comparing the concentrations and fluxes of pesticides between the two sediment sections, significant differences in accumulation under different land-use patterns were found. Redundancy analysis further indicated that temporal watershed farmland variance was the dominant factor for pesticide loading. The lower concentration of atrazine and the higher concentration of the other pesticides in the estuarine sediment was closely related to the decreasing upland in the upstream area and the increase in paddy fields in the downstream area. The analysis of atrazine and the metabolites indicated that atrazine is more likely degraded to deethyl-atrazine and the metabolites have similar migration processes in the sediments, which can easily migrate downward. Moreover, the ratio of metabolites to atrazine showed that atrazine degradation was intensive during the transport process, but the metabolites efficiency was lower in this area due to the cold temperature. The results provide insights for the management of pesticide pollution control in watersheds and the potential effects of low temperature on the degradation of pesticides.

Land Cover Mapping in Cloud-Prone Tropical Areas Using Sentinel-2 Data: Integrating Spectral Features with Ndvi Temporal Dynamics

Accurate remote sensing and mapping of land cover in the tropics remain difficult tasks since data gaps and a heterogenic landscape make it challenging to perform land cover classification. In this paper, we proposed a multi-feature classification method to integrate temporal statistical features with spectral and textural features. This method is designed to improve the accuracy of land cover classification in cloud-prone tropical regions. Sentinel-2 images were used to construct an NDVI stack for a time-series statistical analysis to characterize the temporal variance of land cover. Two statistical indices were calculated and used to represent the variation in annual vegetation. These indices included the mean (NDVI_mean) and coefficient of variation (NDVI_cv) for the NDVI time series. The temporal statistical features were then integrated with spectral and textural features extracted from high-quality Sentinel-2 imagery for Random Forest classification. The performance and contribution of different combinations were assessed based on their classification accuracies. Our results show that the time-series statistical analysis is an effective way to represent land cover category information contained in annual NDVI variance. The method uses clear pixels from dense low-quality images to obtain the NDVI statistical characteristics, thus, to reduce the influence of random factors such as weather conditions on single-date image. The addition of NDVI_mean and NDVI_cv can improve the separability among most types of land cover. The overall accuracy and the kappa coefficient reached values of 0.8913 and 0.8514 when NDVI_mean and NDVI_cv were integrated. Furthermore, the time-series statistical analysis has less stringent requirements regarding image quality and features a high computational efficiency, which shows its great potential to improve the overall accuracy of land cover classification at regional scales in cloud-prone tropical regions.

Robustness of prediction for extreme adaptive optics systems under various observing conditions

Context. For high-contrast imaging systems, such as VLT/SPHERE, the performance of the system at small angular separations is contaminated by the wind-driven halo in the science image. This halo is a result of the servo-lag error in the adaptive optics (AO) system due to the finite time between measuring the wavefront phase and applying the phase correction. One approach to mitigating the servo-lag error is predictive control. Aims. We aim to estimate and understand the potential on-sky performance that linear data-driven prediction would provide for VLT/SPHERE under various turbulence conditions. Methods. We used a linear minimum mean square error predictor and applied it to 27 different AO telemetry data sets from VLT/SPHERE taken over many nights under various turbulence conditions. We evaluated the performance of the predictor using residual wavefront phase variance as a performance metric. Results. We show that prediction always results in a reduction in the temporal wavefront phase variance compared to the current VLT/SPHERE AO performance. We find an average improvement factor of 5.1 in phase variance for prediction compared to the VLT/SPHERE residuals. When comparing to an idealised VLT/SPHERE, we find an improvement factor of 2.0. Under our 27 different cases, we find the predictor results in a smaller spread of the residual temporal phase variance. Finally, we show there is no benefit to including spatial information in the predictor in contrast to what might have been expected from the frozen flow hypothesis. A purely temporal predictor is best suited for AO on VLT/SPHERE. Conclusions. Linear prediction leads to a significant reduction in phase variance for VLT/SPHERE under a variety of observing conditions and reduces the servo-lag error. Furthermore, prediction improves the reliability of the AO system performance, making it less sensitive to different conditions.

What, if Anything, Can the History of Emotions Learn from the Neurosciences?

The last decade has seen a wave of scholarship on both the history of emotions and the affective and social neurosciences. Both disciplines promote the view that emotions serve partly cognitive and goal-driven functions and are thus susceptible to change. Differences are, however, marked in understanding the nature of the change in question. Is it rooted in social constructivism or neuroplasticity? Are these two paradigms mutually exclusive or compatible? Is it even possible or intellectually sound to utilize ‘laboratory-produced’ and isolated data in historical studies? Nevertheless, in light of recent research on the connections between the history of emotions and the neurosciences the article suggests that the interplay between the historical and brain-scientific knowledge may be more fruitful than has previously been understood. Thus far the implications of the critical and social brain sciences for the history of emotions remain to be worked out. This article proposes an intellectual move from linguistically defined constructivism towards a more corporeal understanding of emotions with a constructivist element. This leads us theoretically towards an understanding of emotions as biocultural, historically changing experiences and intellectual concepts, which, however, possess certain transcultural continuities. The article highlights the intrinsic and dynamic relationship between pan-human characteristics and cultural, normative and temporal variance in emotions. However, the article explores the methodological difficulties and dangers of applying specific brain-scientific data to historical study. The difficulty arises mainly from the academic distance between the history of emotions and the neurosciences.

The Limits to Estimating Population-Genetic Parameters with Temporal Data.

The ability to obtain genome-wide sequences of very large numbers of individuals from natural populations raises questions about optimal sampling designs and the limits to extracting information on key population-genetic parameters from temporal-survey data. Methods are introduced for evaluating whether observed temporal fluctuations in allele frequencies are consistent with the hypothesis of random genetic drift, and expressions for the expected sampling variances for the relevant statistics are given in terms of sample sizes and numbers. Estimation methods and aspects of statistical reliability are also presented for the mean and temporal variance of selection coefficients. For nucleotide sites that pass the test of neutrality, the current effective population size can be estimated by a method of moments, and expressions for its sampling variance provide insight into the degree to which such methodology can yield meaningful results under alternative sampling schemes. Finally, some caveats are raised regarding the use of the temporal covariance of allele-frequency change to infer selection. Taken together, these results provide a statistical view of the limits to population-genetic inference in even the simplest case of a closed population.

The changing physical and ecological meanings of North Pacific Ocean climate indices

Climate change is likely to change the relationships between commonly used climate indices and underlying patterns of climate variability, but this complexity is rarely considered in studies using climate indices. Here, we show that the physical and ecological conditions mapping onto the Pacific Decadal Oscillation (PDO) index and North Pacific Gyre Oscillation (NPGO) index have changed over multidecadal timescales. These changes apparently began around a 1988/1989 North Pacific climate shift that was marked by abrupt northeast Pacific warming, declining temporal variance in the Aleutian Low (a leading atmospheric driver of the PDO), and increasing correlation between the PDO and NPGO patterns. Sea level pressure and surface temperature patterns associated with each climate index changed after 1988/1989, indicating that identical index values reflect different states of basin-scale climate over time. The PDO and NPGO also show time-dependent skill as indices of regional northeast Pacific ecosystem variability. Since the late 1980s, both indices have become less relevant to physical-ecological variability in regional ecosystems from the Bering Sea to the southern California Current. Users of these climate indices should be aware of nonstationary relationships with underlying climate variability within the historical record, and the potential for further nonstationarity with ongoing climate change.

TVENet: Temporal variance embedding network for fine-grained action representation

With the breakthroughs in general action understanding, it has become an inevitable trend to analyze the actions in finer granularity. However, related researches have been largely hindered by the lack of fine-grained datasets and the difficulty of capturing subtle differences between fine-grained actions that are highly similar overall. In this paper, we address the above challenges by constructing a fine-grained action dataset, i.e., Figure Skating, which can be used for end-to-end network training and presenting a framework for the joint optimization of classification and similarity constraints. We propose to incorporate the triplet loss into the training of Convolutional Neural Network, which learns a mapping from fine-grained actions to a compact Euclidean space where distances directly correspond to a measure of action similarity. Triplet loss compels actions of distinct classes to have larger distances than actions of the same class. Besides, to boost the discrimination of the fine-grained actions, we further propose a temporal variance embedding network (TVENet) embedding temporal context variances into the feature embeddings during the joint network training. The experimental results on Figure Skating dataset, HMDB51 dataset as well as UCF101 dataset demonstrate the effectiveness of TVENet representation for fine-grained action search.

A Comparative Analysis of Women’s Political Rights, 1981–2004: The Role of Legal Traditions

ABSTRACT In this article, we focus on the legal system to explain cross-national and temporal variance in women’s political rights. Compared to alternative legal systems, we find that common law is correlated with less political rights for women. The concepts of political discontinuity and legal memory are central to our theoretical framework. Political discontinuity occurs in times of deep political disruptions; for instance, during revolutionary periods. Whereas typically revolutions did not readily yield lasting improvements in women’s political rights, individual and systemic forms of legal memory meant that later progress toward political equality was facilitated. It is hard to overestimate the influence of legal systems on women’s rights around the world. Using available data for 148 countries from 1981 to 2004, we found that legal systems’ effect is robust to inclusion of more recent periods of upheaval, various model specifications and functional forms, disparate datasets, and different outcome variables.

Robustness to bad frames in angular differential imaging: a local weighting approach

Context. The detection of exoplanets by direct imaging is very challenging. It requires an extreme adaptive-optics (AO) system and a coronagraph as well as suitable observing strategies. In angular differential imaging, the signal-to-noise ratio is improved by combining several observations. Aims. Due to the evolution of the observation conditions and of the AO correction, the quality of the observations may vary significantly during the observing sequence. It is common practice to reject images of comparatively poor quality. We aim to decipher when this selection should be performed and what its impact on detection performance is. Methods. Rather than discarding a full image, we study the local fluctuations of the signal at each frame and derive weighting maps for each frame. These fluctuations are modeled locally directly from the data through the spatio-temporal covariance of small image patches. The weights derived from the temporal variances can be used to improve the robustness of the detection step and reduce estimation errors of both the astrometry and photometry. The impact of bad frames can be analyzed by statistically characterizing the detection and estimation performance. Results. When used together with a modeling of the spatial covariances (PACO algorithm), these weights improve the robustness of the detection method. Conclusions. The spatio-temporal modeling of the background fluctuations provides a way to exploit all acquired frames. In the case of bad frames, areas with larger fluctuations are discarded by a weighting strategy and do not corrupt the detection map or the astrometric and photometric estimations. Other areas of better quality are preserved and are included to detect and characterize sources.

Decoupling seasonal and temporal dynamics of macroalgal canopy cover in seagrass beds

Many seagrass habitats are susceptible to undesirable macroalgal overgrowth in response to growing watershed development, but the mechanisms leading to overgrowth remain unresolved. Partitioning the influences of intermittent seasonal cycles and directional human stressors is one main challenge. We examined the dynamics of macroalgal canopies within Halodule uninervis beds across a 10-year period in a tropical lagoon and first hypothesized that seasonal and temporal variances were distinguishable. We found that cooler and dryer winter months were consistently associated with blue-green algal canopies, mainly Lyngbya sp. and Phormidium sp., with an inverse relationship between sea-surface temperatures and coverage. For example, given that winter temperatures during the coolest month of February ranged between ±1.0 °C of the mean over the past decade, the expected coverage of blue-green algae shifted by ±15%. In contrast, the warmer and wetter summer months were associated with red algae canopies, predominantly Acanthophora spicifera, which were positively related to rainfall. Last, a weaker seasonal trend existed between some green and brown algae and a tidal-height proxy to groundwater discharge where karst watersheds existed. We next hypothesized that removing the variances associated with seasonal growth cycles in the differing phyla would reveal temporal trends in total macroalgal persistence associated with watershed development that were otherwise masked. Following seasonal adjustments, we found that persistent macroalgal canopies existed in the central, urbanized lagoon over the past decade; however, when moving to the north or south of the urban center, significant increases in macroalgal canopies were revealed. Watershed size and development were the strongest predictors of seasonally-adjusted macroalgal canopies through time. No similar trends were found when using unadjusted macroalgal cover data. More importantly, predictions were used to determine thresholds in watershed development associated with the transition between seagrass-versus-macroalgal dominance. The results offered a novel approach for ecological studies that may be limited by financial and logistical constraints to partition seasonal versus temporal change, and better appreciate their differences. For our study lagoon, the results unmasked a growing human footprint into seagrass beds and identified predictive thresholds for management to consider.

Deep Learning for User Interest and Response Prediction in Online Display Advertising

User interest and behavior modeling is a critical step in online digital advertising. On the one hand, user interests directly impact their response and actions to the displayed advertisement (Ad). On the other hand, user interests can further help determine the probability of an Ad viewer becoming a buying customer. To date, existing methods for Ad click prediction, or click-through rate prediction, mainly consider representing users as a static feature set and train machine learning classifiers to predict clicks. Such approaches do not consider temporal variance and changes in user behaviors, and solely rely on given features for learning. In this paper, we propose two deep learning-based frameworks, {hbox {LSTM}}_{mathrm{cp}} and {hbox {LSTM}}_{mathrm{ip}}, for user click prediction and user interest modeling. Our goal is to accurately predict (1) the probability of a user clicking on an Ad and (2) the probability of a user clicking a specific type of Ad campaign. To achieve the goal, we collect page information displayed to the users as a temporal sequence and use long short-term memory (LSTM) network to learn features that represents user interests as latent features. Experiments and comparisons on real-world data show that, compared to existing static set-based approaches, considering sequences and temporal variance of user requests results in improvements in user Ad response prediction and campaign specific user Ad click prediction.

Seasonality of Taylor\u2019s law of fluctuation scaling in all-India daily rainfall

Here we show that the variance over time of all-India daily rainfall (AIR) can be related to the mean over time of AIR by a seasonally varying power law. Outside of the peak monsoon months of July and August, AIR variance increases in proportion to a positive power of mean daily rainfall. During July and August, monthly averages of AIR show little association with the corresponding variances. This power-law relationship of temporal variance to temporal mean is known in biological sciences as Taylor’s law (TL) and in physical sciences as fluctuation scaling. We explain the seasonal variation in TL qualitatively by independent and identically distributed random sampling. Accounting for day-to-day correlation in AIR is sufficient to match quantitatively the observed power-law behavior. Our findings provide a quantitative month-specific assessment of the variability of AIR that could prove useful for the design of crop insurance and social safety nets for the large fraction of the population of the Indian subcontinent that depends on rainfed agriculture.

Global Attribution of Runoff Variance Across Multiple Timescales

AbstractPrevious findings on the relative importance of driving factors to runoff have been controversial, and the relative contribution of driving factors behind the temporal variance of runoff (Rv) still remains unquantified at a global scale. Here, we present a new framework to quantify the driving factors behind Rv globally by decomposing Rv into the temporal variance of precipitation (P), potential evapotranspiration (E0), total water storage change (∆S), and other factors (represented by n in Budyko equation), such as climate seasonality, land cover, and human impact, at intra‐annual, interannual, and multiannual timescales. The results reveal that P controls Rv at different timescales in most parts of the globe. P has distinctly larger impacts on Rv in humid basins at the multiannual scale, while other factors have larger impacts in arid basins. We highlight the different impacts of driving factors on Rv at multiple timescales, which contradicts previous global assessments that assess the relative importance by sensitivity.

Understanding the bike sharing travel demand and cycle lane network: The case of Shanghai

As a sustainable transportation alternative, cycling has been developed by many large cities as one essential measure for addressing the “last mile problem” in urban areas with high mobility demand. Developing a safe and friendly lane network for cycling has become an urgent task for governments, especially for many Chinese cities encountering a rapid increase in the use of “dockless” shared bikes, such as Mobike and Ofo. The emergence of such kind of app-driven dockless bike sharing systems results in a fast growth in the cycling mobility demand, and leads to a gap between the growing demand and the existing cycling infrastructure. Therefore, it is imperative to have a good understanding of the cycling travel demand and its infrastructure, and in turn to bridge the supply-demand gap. In this paper, we employ data mining techniques including graphic clustering algorithm, Louvain Method, on a large data set from one of the largest bike sharing companies in China. Then we applied the methodology in two cases in Shanghai, including a campus and an urban area. Typical cycling patterns in the spatial and temporal dimensions are identified automatically by the cluster analysis. It is also found that the construction of the cycling infrastructure is closely related to three factors: non-negligible impact of points of interest (POIs), geographical barriers, and temporal variance of the network. Managerial insights and policy measures are proposed accordingly for improving the cycle lane network.

An Assessment of the Dynamic Global Warming Impact Associated with Long-Term Emissions from Landfills.

Landfills are a major contributor of anthropogenic CH4 emissions. Since the greenhouse gas (GHG) emissions associated with landfilling waste can occur over decades to centuries, the standard static approach to estimating global warming impacts may not accurately represent the global warming impacts of landfills. The objective of this study is to assess the implications of using 100 yr and 20 yr static and dynamic global warming potential (GWP) approaches to estimate the global warming impacts from municipal solid waste landfills. A life-cycle model was developed to estimate GHG emissions for three gas treatment cases (passive venting, flare, CH4 conversion to electricity) and four decay rates. For the 100 yr GWP, other model uncertainties (e.g., static GWP values, decay rate, moisture content, or gas collection efficiency) generally had a larger effect on the estimated global warming impact than the choice of static versus dynamic GWP methods. This shows that when comparing single-point GWP values, the choice of static versus dynamic is relatively unimportant for most landfills. While dynamic GWPs consider temporal variance and provide useful estimates for the warming over a set time horizon, for most comparative analyses, static values provide reasonable bounds for the actual 100 yr warming impact.

Bayesian inference to partition determinants of community dynamics from observational time series

Ecological communities are shaped by a complex interplay between abiotic forcing, biotic regulation and demographic stochasticity. However, community dynamics modelers tend to focus on abiotic forcing overlooking biotic interactions, due to notorious challenges involved in modeling and quantifying inter-specific interactions, particularly for species-rich systems such as planktonic assemblages. Nevertheless, inclusive models with regard to the full range of plausible drivers are essential to characterizing and predicting community response to environmental changes. Here we develop a Bayesian model for identifying, from in-situ time series, the biotic, abiotic and stochastic factors underlying the dynamics of species-rich communities, focusing on the joint biomass dynamics of biologically meaningful groups. We parameterize a multivariate model of population co-variation with an explicit account for demographic stochasticity, density-dependent feedbacks, pairwise interactions, and abiotic stress mediated by changing environmental conditions and resource availability, and work out explicit formulae for partitioning the temporal variance of each group in its biotic, abiotic and stochastic components. We illustrate the methodology by analyzing the joint biomass dynamics of four major phytoplankton functional types namely, diatoms, dinoflagellates, coccolithophores and phytoflagellates at Station L4 in the Western English Channel using weekly biomass records and coincident measurements of environmental covariates describing water conditions and potentially limiting resources. Abiotic and biotic factors explain comparable amounts of temporal variance in log-biomass growth across functional types. Our results demonstrate that effective modelling of resource limitation and inter-specific interactions is critical for quantifying the relative importance of abiotic and biotic factors.

Spatial and Temporal Variation in an Apple Orchard

One important goal of precision horticulture (PH), as well as precision agriculture (PA), is to measure and manage spatial and temporal variation in orchards. In this study, temporal and spatial analysis of yields were carried out over 2 years for a 0.5-ha apple orchard (at the Haymana Research Station of Ankara University, Turkey, from 2017 to 2018) to determine the variability of yields over time and included seven apple varieties: ‘Royal Gala’, ‘Red Chief’, ‘Braeburn’, ‘Mondial Gala’, ‘Jonagold’, ‘Fuji’, and ‘Mitch Gala’. To achieve this, yield data for two different years were analyzed for mean yield, temporal variance, and cv in terms of spatial and temporal stability, and their yield maps were produced. The results showed that ‘Jonagold’, ‘Braeburn’, and ‘Red Chief’ varieties yielded less than the average yield, whereas the other varieties produced average yields when the yield from 2 years was taken into account. Calculation of the values for determining temporal stability over time resulted in all existing varieties being identified as stable over time. For example, the ‘Jonagold’ and ‘Red Chief’ varieties showed 100% stability in terms of temporal variance. Results also showed that the ‘Gala’ varieties were stable for 2 years and produced high yields, whereas the other varieties were specified as stable and low yielding when spatial and temporal variability was considered in combination.