Temporal Variance Research Articles

Exploring the trend effects of diffuse anthropogenic pollution in a large river passing through a densely populated area

The detection of non-point pollution in large rivers requires high-frequency sampling over a longer period of time, which, however presumably provides data with large spatial and temporal variance. Variability may mean that data sets recorded upstream and downstream from a densely populated area overlap, suggesting at first glance that the urban area did not affect water quality. This study presents a simple way to explore trend-like effects of non-point pollution in the Danube based on data that varied strongly in space and time. For one year, biweekly sampling was carried out upstream and downstream from a large city with negligible emission of untreated wastewater and the surrounding settlements, industrial and agricultural areas.Although most of the values of the 34 examined physicochemical characteristics fell within the range of data previously published for the Danube, and the mean values of all parameters indicated unpolluted surface water, different water quality was revealed upstream and downstream from the metropolitan area at each sampling time. Since the physicochemical characteristics causing the separation also differed from time to time, univariate tests and consensus ordination were used to determine which variables changed similarly during most of the examined period. With this evaluation method, several diffuse pollutants of anthropogenic origin contaminating the Danube in the long term were identified, such as nitrogen, phosphorus, sulphate, chloride, potassium and vanadium. The results demonstrated that trend-like effects of non-point pollution can be detected even in a large river, where physicochemical measurements can vary strongly in space and time.

SPATIO-TEMPORAL ANALYSIS OF TRAFFIC FLOW ON SARGODHA-JHANG ROAD (PUNJAB) PAKISTAN

Traffic flow in Pakistan has become more complex in recent years due to rapid population growth and rapidly generating economic activities. The main objective of this research was to explore the various factors and attributes related to road network configuration contributing to an irregular traffic flow and congestion at second class inter-district national highway section Sargodha-Jhang road, Punjab, Pakistan. It is a national highway road section in central Punjab extending 124 km. Primary data related to traffic flow and volume was collected manually in mode of time and space while secondary data about annual volume of traffic (3 years) gathered from Punjab highway department, southern Zone Lahore. Global positioning system (GPS) is used to mark the different locations of collecting samples with latitude, longitude and elevation. The analysis is performed in MS Excel by applying descriptive statistics and analytical tendencies and results are geographically presented by various graphs and maps prepared by using ArcGIS 10.5 software. Traffic flow was analyzed on hourly, weekly and yearly basis by comparing spatial and temporal average and variance. Results determined the hourly fluctuations as peak from 8:00AM – 9:00AM (1,103 vehicles) and 2:00PM - 3:00PM (1,215 vehicles), medium from 11:00AM – 12:00AM (1,000 vehicles), and low hours from 5:00PM – 6:00PM (800 vehicles). Thursday and Saturday were the busiest days in a weekly flow of the traffic while the Friday was least busy day regarding the weekly flow of traffic. Likewise, the yearly results of traffic flow also shown slight increase in volume of traffic vehicles over the last three years. Lastly few suggestions were proposed to bring the regularity, road safety, less traffic jam and congestion and to avoid the traffic accidents in the study area.

Accelerated vegetative growth measured by gross primary productivity in China from 1980 to 2018

Vegetation response to climate change can be characterized as long-term trends, but with periodic oscillations. Revealing vegetation response trends, with oscillations, is essential for formulating climate change mitigation and adaptation planning, as well as adopting ecological protection and restoration measures. Here, the long-term trend and periodic oscillations in vegetation changes were explored through a measure of gross primary production (GPP), an indicator of vegetation photosynthesis, using data of both high spatial resolution (1 km2) and a long-term of nearly 4 decades, from 1980 to 2018. The GPP data were estimated for China’s terrestrial ecosystem and validated against the eddy covariance-based observations of 70 site-year in China and compared to other GPP data products. The results showed the estimated GPP can explain 76% of spatial–temporal variance with the same data accuracy as other products, but with the advantage of both high spatial resolution and long time periods that other products do not have. From the estimated GPP data, Chinese terrestrial vegetation growth was found to have a periodic oscillation of close to 2.79 years and a long-term increasing trend of 59.8 Mt C per year over 92% of all vegetated land in China from 1980 to 2018. Moreover, it was found that vegetation growth accelerated significantly in the two recent decades over the previous two decades. The growth rate was 83.5 Mt C a−1 from 2000 to 2018, which was 2.31 times faster than the rate from 1980 to 2000. The underlying mechanism of the vegetation responses was analyzed by ridge regression, a method to decrease the effects of multicollinearity. Long-term variability and trends in vegetation could be attributed to a warming minimum air temperature when examined with precipitation, maximum air temperature, and solar shortwave radiation. Though more influences, such as those from freezing and thawing and human activities, should be explored in the future, this study offers insight that warming, with increasing daily minimum temperatures, is likely to be a vital contributor to ecosystem changes, and will be essential knowledge for mitigation of effects and adaptation to future climate.

Analyzing community wastewater in sub-sewersheds for the small-scale detection of SARS-CoV-2 variants in a German metropolitan area

Wastewater surveillance of SARS-CoV-2 proved useful, including for identifying the local appearance of newly identified virus variants. Previous studies focused on wastewater treatment plants (WWTP) with sewersheds of several hundred thousand people or at single building level, representing only a small number of people. Both approaches may prove inadequate for small-scale intra-urban inferences for early detection of emerging or novel virus variants. Our study aims (i) to analyze SARS-CoV-2 single nucleotide variants (SNVs) in wastewater of sub-sewersheds and WWTP using whole genome sequencing in order to (ii) investigate the potential of small-scale detection of novel known SARS-CoV-2 variants of concern (VOC) within a metropolitan wastewater system. We selected three sub-sewershed sampling sites, based on estimated population- and built environment-related indicators, and the inlet of the receiving WWTP in the Ruhr region, Germany. Untreated wastewater was sampled weekly between October and December 2021, with a total of 22 samples collected. SARS-CoV-2 RNA was analyzed by RT-qPCR and whole genome sequencing. For all samples, genome sequences were obtained, while only 13 samples were positive for RT-qPCR. We identified multiple specific SARS-CoV-2 SNVs in the wastewater samples of the sub-sewersheds and the WWTP. Identified SNVs reflected the dominance of VOC Delta at the time of sampling. Interestingly, we could identify an Omicron-specific SNV in one sub-sewershed. A concurrent wastewater study sampling the same WWTP detected the VOC Omicron one week later. Our observations suggest that the small-scale approach may prove particularly useful for the detection and description of spatially confined emerging or existing virus variants circulating in populations. Future studies applying small-scale sampling strategies taking into account the specific features of the wastewater system will be useful to analyze temporal and spatial variance in more detail.

Analysis of the riverbed backscattered signal registered by ADCPs in different bedload transport conditions – field application

Acoustic Doppler current profilers (ADCP) were deployed to investigate the backscattering (BS) signal in three navigable rivers, in different bedload transport conditions. This study aims to demonstrate that the BS strength, as an additional variable to the apparent bedload velocity, improves the characterization of the bedload transport using ADCPs. The M9 –3 MHz and the vertical beam M9 – 0.5 MHz showed decline of the BS strength as the bedload intensity increased, whereas the RDI –1.2 MHz was relatively insensitive. The correlation between the median grain size and the BS strength for the 0.5 MHz was linear, for the 3 MHz the BS strength was attenuated in the active layer, and for 1.2 MHz, it revealed a parabolic distribution. Moreover, the analyses of the ADCP measured variables, using wavelet transformations and unsupervised machine learning, highlighted the importance of the spatial and temporal variance and transient nature of the bedload transport.

CRYSTAL: A novel and effective method to remove clouds in daily nighttime light images by synergizing spatiotemporal information

Satellite nighttime light (NTL) images offer a valuable depiction of the rapidly changing world by revealing the presence of artificial illumination. Thus, daily NTL images are increasingly applied to monitor human dynamics and environmental events. However, data gaps caused by cloud contamination and low-quality observations inevitably impair the effectiveness of such applications. Although a temporal gap-filling method is employed in recent Black Marble NTL products to produce seamless images, the filled images are unsuitable for quantitative analysis. Therefore, we developed an effective method, named as Cloud Removing bY Synergizing spatio-TemporAL information (CRYSTAL), to generate cloud-free NTL images with satisfactorily accurate pixel brightness and spatial continuity. Simulation experiments show that CRYSTAL can produce more accurate results than the temporal gap-filling method in fifteen cities worldwide, with an average RMSE reduction of 33.69%. Images generated by CRYSTAL restore temporal variances in NTL and are thus suitable for multi-temporal quantitative analysis. CRYSTAL can reconstruct daily NTL time series by filling gaps using available partially clear images. Experiments in two cities demonstrated that the reconstructed time series had 31.85% more valid values than the original time series and effectively revealed urban dynamics during the early stages of the coronavirus disease 2019 pandemic. In summary, CRYSTAL is a novel and effective gap-filling method for the restoration of invalid NTL observations and has the potential to generate high-quality NTL data for use in future applications.

Combining Heart Rate Variability and Oximetry to Improve Apneic Event Screening in Non-Desaturating Patients.

In this paper, we thoroughly analyze the detection of sleep apnea events in the context of Obstructive Sleep Apnea (OSA), which is considered a public health problem because of its high prevalence and serious health implications. We especially evaluate patients who do not always show desaturations during apneic episodes (non-desaturating patients). For this purpose, we use a database (HuGCDN2014-OXI) that includes desaturating and non-desaturating patients, and we use the widely used Physionet Apnea Dataset for a meaningful comparison with prior work. Our system combines features extracted from the Heart-Rate Variability (HRV) and SpO2, and it explores their potential to characterize desaturating and non-desaturating events. The HRV-based features include spectral, cepstral, and nonlinear information (Detrended Fluctuation Analysis (DFA) and Recurrence Quantification Analysis (RQA)). SpO2-based features include temporal (variance) and spectral information. The features feed a Linear Discriminant Analysis (LDA) classifier. The goal is to evaluate the effect of using these features either individually or in combination, especially in non-desaturating patients. The main results for the detection of apneic events are: (a) Physionet success rate of 96.19%, sensitivity of 95.74% and specificity of 95.25% (Area Under Curve (AUC): 0.99); (b) HuGCDN2014-OXI of 87.32%, 83.81% and 88.55% (AUC: 0.934), respectively. The best results for the global diagnosis of OSA patients (HuGCDN2014-OXI) are: success rate of 95.74%, sensitivity of 100%, and specificity of 89.47%. We conclude that combining both features is the most accurate option, especially when there are non-desaturating patterns among the recordings under study.

Cueing in Theatre: Timing and Temporal Variance in Rehearsals of Scene Transitions

This video-ethnographic study explores how professional actors and a director at the end of a theatrical rehearsal process coordinate transitions between rehearsed scenes. This is done through the development and use of cues, that is, ‘signals for action’. The aim is to understand how cues are developed and how timing in transitions is achieved by using the designed cues. Work on three different scene transitions is analysed using multimodal Conversation Analysis. The results show that cueing is a central tool for developing well-timed transitions, and how cues serve different purposes in the developing performance. There is no prior plan for how to achieve timely transitions. In all the analysed examples, it is an actor who must produce or act on the given cue who insists on its precise definition, followed by a negotiation on candidate cues, confirmation and specifying the cue. It is also actors who are primarily responsible for the timing of transitions, and the timing is solved through an interplay of clear-cut and embodied actions that allow for temporal variance. Cues are reflexively linked to actors’ observation and interpretation of other actors’ actions, which prevents a mechanical determination of timing in scene transition.

Exploring Multiscale Variability in Groundwater Quality: A Comparative Analysis of Spatial and Temporal Patterns via Clustering

Defining homogeneous units to optimize the monitoring and management of groundwater is a key challenge for organizations responsible for the protection of water for human consumption. However, the number of groundwater bodies (GWBs) is too large for targeted monitoring and recommendations. This study, carried out in the Provence-Alpes-Côte d’Azur region of France, is based on the intersection of two databases, one grouping together the physicochemical and bacteriological analyses of water and the other delimiting the boundaries of groundwater bodies. The extracted dataset contains 8627 measurements from 1143 observation points distributed over 63 GWB. Data conditioning through logarithmic transformation, dimensional reduction through principal component analysis, and hierarchical classification allows the grouping of GWBs into 11 homogeneous clusters. The fractions of unexplained variance (FUV) and ANOVA R2 were calculated to assess the performance of the method at each scale. For example, for the total dissolved load (TDS) parameter, the temporal variance was quantified at 0.36 and the clustering causes a loss of information with an R2 going from 0.63 to 0.4 from the scale of the sampling point to that of the GWB cluster. The results show that the logarithmic transformation reduces the effect of outliers and improves the quality of the GWB clustering. The groups of GWBs are homogeneous and clearly distinguishable from each other. The results can be used to define specific management and protection strategies for each group. The study also highlights the need to take into account the temporal variability of groundwater quality when implementing monitoring and management programs.

Semantics-enhanced early action detection using dynamic dilated convolution

This paper proposes a new pipeline to perform early action detection from skeleton-based untrimmed videos. Our pipeline includes two new technical components. The first is a new Dynamic Dilated Convolutional Network (DDCN), which supports dynamic temporal sampling and makes feature learning more robust against temporal scale variance in action sequences. The second is a new semantic referencing module, which uses identified objects in the scene and their co-existence relationship with actions to adjust the probabilities of inferred actions. Such semantic guidance can help distinguish many ambiguous actions, which is a core challenge in the early detection of incomplete actions. Our pipeline achieves state-of-the-art performance in early action detection in two widely used skeleton-based untrimmed video benchmarks. The source codes are available at: https://github.com/Powercoder64/DDCN_SRM.

Emission and environmental distribution of decabromodiphenyl ethane (DBDPE) in China from 2006 to 2026: Retrospection, forecasting, and implications for assessment and management

Decabromodiphenyl ethane (DBDPE) is the main alternative to decabromodiphenyl ether (deca-BDE) in commercial use. However, there is increasing evidence show that DBDPE is a potential persistent organic pollutant, and it has been found ubiquitously in environmental media across China in recent years. Monitoring studies have not been able to determine the overall levels and temporal trends of DBDPE contamination in China, and have been unable to explain how emission patterns can affect their environmental distribution. Therefore, this study estimated the temporal variance of DBDPE emissions and environmental concentrations in five regions of China from 2006 to 2026 using the PROduction-To-EXposure (PROTEX) mass balance model. The results showed that Guangdong Province was the greatest DBDPE pollution hotspot in China due to emissions from plastics manufacturing and e-waste disposal; there was also severe pollution in Shandong Province, where almost all the DBDPE in China is produced. The DBDPE concentrations in indoor and outdoor environments increased substantially in all regions during 2006–2021. Furthermore, in Guangdong Province and Shandong Province, the ratio of indoor/outdoor air concentrations was greater than or close to 1, indicative of significant outdoor emission sources of DBDPE. In contrast, the ratios for the Beijing–Tianjin–Hebei region, East China, and Southwest China were below 1 due to the indoor use of electronic equipment containing DBDPE. The temporal trends of these ratios indicated that DBDPE contamination has gradually spread from high-concentration environments with strong emission sources to low-concentration environments. The outcomes of this study have important implications for the risk assessment of DBDPE use in China and can be used to establish contamination-mitigation actions.

Wave field predictions using a multi-layer perceptron and decision tree model based on physical principles: A case study at the Pearl River Estuary

In this study, a novel data-driven model is proposed for efficient spatial–temporal forecasting of the significant wave height (SWH), mean wave period (MWP), and mean wave direction (MWD) in the Pearl River Estuary, located in Southern China, using machine learning. The model utilizes machine learning techniques and employs a multi-layer perceptron (MLP) and decision tree (DT) based architecture following a matrix-to-matrix mapping strategy to establish the underlying spatial–temporal relationship between the wind and wave fields for four-day and ocean-scale forecasts. The training and testing of the model were carried out using long-term hindcast wave and wind datasets released by ERA5. A rectangular sea region covering an area of 5° × 5°in the Pearl River Estuary was regarded as the modeling domain. The data-driven model was validated against a 10 years of historical hindcast data, and the global errors were analyzed. The forecasting performance of the established model for tropical cyclone wave events was investigated in terms of both the spatial pattern and temporal variance of the SWH, MWP, and MWD. Extreme waves were generated using the proposed model and compared with the original hindcast results. The data-driven model showed promising potential for accurately capturing ambiguous patterns and features that are variant in both spatial and temporal dimensions, and exhibited obvious superiority over numerical wave modeling in terms of computational efficiency.

Altered static and dynamic spontaneous neural activity in patients with ischemic pontine stroke.

The purpose of the study was to investigate the abnormality both of static spontaneous brain activity and dynamic temporal variances following a pontine infarction. Forty-six patients with chronic left pontine infarction (LPI), thirty-two patients with chronic right pontine infarction (RPI), and fifty healthy controls (HCs) were recruited for the study. The static amplitude of low-frequency fluctuations (sALFF), static regional homogeneity (sReHo), dynamic ALFF (dALFF), and dynamic ReHo (dReHo) were employed to detect the alterations in brain activity induced by an infarction. The Rey Auditory Verbal Learning Test and Flanker task were used to evaluate the verbal memory and visual attention function, respectively. Receiver operating characteristic curve analysis was used to reveal the potential capacity of these metrics to distinguish the patients from HCs. There were significant variations of these static and dynamic metrics in patients with chronic pontine infarction. The altered regions involved the supratentorial regions, including cortex and subcortical. Moreover, the altered metrics were significantly correlated with verbal memory and visual attention. In addition, these static and dynamic metrics also showed potential in distinguishing stroke patients with behavior deficits from HCs. The pontine infarction-induced cerebral activation changes are observed in both motor and cognitive systems, indicating the functional damage and reorganization across the global cerebral level in these patients with subtentorial infarction, and there is a reciprocal effect between motor and cognitive impairment and repair.

Effects of different handling methods on the behavior of adult zebrafish

The zebrafish is an important biomedical research organism. In most research, zebrafish are removed from their home tank and subsequently their phenotype is measured. The method of handling the fish, however, may significantly affect a variety of phenotypes. This is particularly problematic for studies of brain function that measure behavioral or neuronal responses. Nevertheless, the potential effects of handling have not been analyzed, and in fact are usually ignored. Here, we explore the effects of two usual and two rarely or never-before employed handling methods on the behavior of adult zebrafish. We exposed each fish to one of four handling methods, a between subject experimental design: (1) net chasing followed by air-suspension, (2) gentle net catching (without chasing) followed by air-suspension, (3) gentle net catching followed by being placed in a beaker (no chasing and very short air-suspension), (4) transportation in home tank and pouring the fish directly into the test tank (no chasing, netting or air-suspension). With these handling methods, the fish were placed in a test tank and their swim path was videorecorded and analyzed. Handling significantly affected swim path parameters, duration and frequency of immobility, absolute turn angle and its temporal variance and velocity, but not the distance to bottom. The behavioral effects confirmed that chasing and netting induce robust behavioral changes, and that pouring the fish from its home to its test tank is least aversive for zebrafish. We recommend using this latter method to reduce experimental error variation and increase reproducibility of results.

The spatial-temporal risk profiling of Clonorchis sinensis infection over 50 years implies the effectiveness of control programs in South Korea: a geostatistical modeling study

Over the past 50 years, two national control programs on Clonorchis sinensis infection have been conducted in South Korea. Spatial-temporal profiles of infection risk provide useful information on assessing the effectiveness of the programs and planning spatial-targeted control strategies. Advanced Bayesian geostatistical joint models with spatial-temporal random effects were developed to analyze disease data collecting by a systematic review with potential influencing factors, and to handle issues of preferential sampling and data heterogeneities. Changes of the infection risk were analyzed. We presented the first spatial-temporal risk maps of C.sinensis infection at 5×5km2 resolution from 1970 to 2020 in South Korea. Moderate-to-high risk areas were shrunk, but temporal variances were shown in different areas. The population-adjusted estimated prevalence across the country was 5.99% (95% BCI: 5.09-7.01%) in 1970, when the first national deworming campaign began. It declined to 3.95% (95% BCI: 2.88-3.95%) in 1995, when the campaign suspended, and increased to 4.73% (95% BCI: 4.00-5.42%) in 2004, just before the Clonorchiasis Eradication Program (CEP). The population-adjusted prevalence was estimated at 2.77% (95% BCI: 1.67-4.34%) in 2020, 15 years after CEP started, corresponding to 1.42 (95% BCI: 0.85-2.23) million infected people. The first nationwide campaign and the CEP showed effectiveness on control of C.sinensis infection. Moderate-to-high risk areas identified by risk maps should be prioritized for control and intervention. The National Natural Science Foundation of China (project no. 82073665) and the Natural Science Foundation of Guangdong Province (project no. 2022A1515010042).

Toyota Smarthome Untrimmed: Real-World Untrimmed Videos for Activity Detection.

Designing activity detection systems that can be successfully deployed in daily-living environments requires datasets that pose the challenges typical of real-world scenarios. In this paper, we introduce a new untrimmed daily-living dataset that features several real-world challenges: Toyota Smarthome Untrimmed (TSU). TSU contains a wide variety of activities performed in a spontaneous manner. The dataset contains dense annotations including elementary, composite activities and activities involving interactions with objects. We provide an analysis of the real-world challenges featured by our dataset, highlighting the open issues for detection algorithms. We show that current state-of-the-art methods fail to achieve satisfactory performance on the TSU dataset. Therefore, we propose a new baseline method for activity detection to tackle the novel challenges provided by our dataset. This method leverages one modality (i.e. optic flow) to generate the attention weights to guide another modality (i.e RGB) to better detect the activity boundaries. This is particularly beneficial to detect activities characterized by high temporal variance. We show that the method we propose outperforms state-of-the-art methods on TSU and on another popular challenging dataset, Charades.

Scale Critique for the Anthropocene, Part Two

Scale critique applies to time as well as space. Thinking with temporal scale variance can help us avoid seeing deep time as sublime, numerical, and continuous. By reading Stephen Jay Gould with Michael Madsen's documentary Into Eternity, this essay argues that the second concept of scale critique is about the fit, good or bad, between a temporal scale domain and a narrative timeframe. Because temporal scale domains have material finitude and specificity, they counteract the timelessness that accrues to ideologies of extinction.

Towards learned emulation of interannual water isotopologue variations in General Circulation Models

Abstract Simulating abundances of stable water isotopologues, that is, molecules differing in their isotopic composition, within climate models allows for comparisons with proxy data and, thus, for testing hypotheses about past climate and validating climate models under varying climatic conditions. However, many models are run without explicitly simulating water isotopologues. We investigate the possibility of replacing the explicit physics-based simulation of oxygen isotopic composition in precipitation using machine learning methods. These methods estimate isotopic composition at each time step for given fields of surface temperature and precipitation amount. We implement convolutional neural networks (CNNs) based on the successful UNet architecture and test whether a spherical network architecture outperforms the naive approach of treating Earth’s latitude-longitude grid as a flat image. Conducting a case study on a last millennium run with the iHadCM3 climate model, we find that roughly 40% of the temporal variance in the isotopic composition is explained by the emulations on interannual and monthly timescale, with spatially varying emulation quality. The tested CNNs outperform simple baseline models such as random forest and pixel-wise linear regression substantially. A modified version of the standard UNet architecture for flat images yields results that are as good as the predictions by the spherical CNN. Variations in the implementation of isotopes between climate models likely contribute to an observed deterioration of emulation results when testing on data obtained from different climate models than the one used for training. Future work toward stable water-isotope emulation might focus on achieving robust climate–oxygen isotope relationships or exploring the set of possible predictor variables.

Who helped yesterday will help tomorrow: Trait and state components in prosocial goal realisation.

The present study aims to disentangle the state and trait components of prosocial goal realisation, defined as a set of personally meaningful prosocial actions undertaken in natural settings. Based on a diary study with seven daily measurements from 180 participants (a total of 1005 data points), we performed temporal variance decomposition using the STARTS model. The observed individual scores are explained by: ST-a common trait factor, capturing variance stable across days; ART-a unique autoregressive trait factor, capturing variance changing from 1 day to the next; and S-state factors, a series of uncorrelated factors reflecting occasion-specific variance. The results demonstrate the relative stability of prosocial goal realisation, extending the knowledge on the state/trait distinction in actual prosocial behaviour.

Improving robustness of 3D multi-shot EPI by structured low-rank reconstruction of segmented CAIPI sampling for fMRI at 7T

Three-dimensional (3D) encoding methods are increasingly being explored as alternatives to two-dimensional (2D) multi-slice acquisitions in fMRI, particularly in cases where high isotropic resolution is needed. 3D multi-shot EPI acquisition, as the workhorse of 3D fMRI imaging, is susceptible to physiological fluctuations which can induce inter-shot phase variations, and thus reducing the achievable tSNR, negating some of the benefit of 3D encoding. This issue can be particularly problematic at ultra-high fields like 7T, which have more severe off-resonance effects. In this work, we aim to improve the temporal stability of 3D multi-shot EPI at 7T by improving its robustness to inter-shot phase variations. We presented a 3D segmented CAIPI sampling trajectory (“seg-CAIPI”) and an improved reconstruction method based on Hankel structured low-rank matrix recovery. Simulation and in-vivo results demonstrate that the combination of the seg-CAIPI sampling scheme and the proposed structured low-rank reconstruction is a promising way to effectively reduce the unwanted temporal variance induced by inter-shot physiological fluctuations, and thus improve the robustness of 3D multi-shot EPI for fMRI.