Integration Window Research Articles

A new function of offset response in the primate auditory cortex: marker of temporal integration

Offset responses are traditionally viewed as indicators of sound cessation. Here, we investigate offset responses to auditory click trains, examining how they are modulated by inter-click intervals (ICIs) and train duration. Using extracellular recordings and electrocorticography (ECoG) in non-human primates, alongside electroencephalography (EEG) in humans, we show that offset responses are significantly influenced by both ICI and train length, thereby establishing them as markers of temporal integration. We introduce the concept of the ‘Neuronal Integrative Window’ (NIW), defined as the temporal span during which neurons integrate stimuli to produce or modulate the temporal integration signal. Our data reveal that on the neuronal level, the auditory cortex (AC) exhibits a more expansive NIW than the medial geniculate body (MGB), integrating stimuli over longer durations and showing a preference for larger ICIs. Furthermore, our results indicate that offset responses could serve as potential biomarkers for neurological and psychiatric conditions, highlighted by their sensitivity to pharmacological modulation with ketamine. This study advances our understanding of auditory temporal processing and proposes a novel approach for assessing and monitoring brain health.

Strategies for motion- and respiration-robust estimation of fMRI intrinsic neural timescales

Abstract Intrinsic neural timescales (INT) reflect the time window of neural integration within a brain region and can be measured via resting-state fMRI (rs-fMRI). Despite the potential relevance of INT to cognition, brain organization, and neuropsychiatric illness, the influences of physiological artifacts on rs-fMRI INT have not been systematically considered. Two artifacts, head motion and respiration, pose serious issues in rs-fMRI studies. Here, we described their impact on INT estimation and tested the ability of two denoising strategies for mitigating these artifacts, high-motion frame censoring and global signal regression (GSR). We used a subset of the HCP Young Adult dataset with runs annotated for breathing patterns (Lynch et al., 2020) and at least one “clean” (reference) run that had minimal head motion and no respiration artifacts; other runs from the same participants (n = 46) were labeled as “non-clean.” We found that non-clean runs exhibited brain-wide increases in INT compared to their respective clean runs and that the magnitude of error in INT between non-clean and clean runs correlated with the amount of head motion. Importantly, effect sizes were comparable to INT effects reported in the clinical literature. GSR and high-motion frame censoring improved the similarity between INT maps from non-clean runs and their respective clean run. Using a pseudo-random frame-censoring approach, we uncovered a relationship between the amount of censored frames and both the mean INT and mean error, suggesting that frame censoring itself biases INT estimation. A group-level correction procedure reduced this bias and improved similarity between non-clean runs and their respective clean run. Based on our findings, we offer recommendations for rs-fMRI INT studies, which include implementing GSR and high-motion frame censoring with Lomb-Scargle interpolation of censored data, and performing group-level correction of the bias introduced by frame censoring.

Encoding of vibrotactile stimuli by mechanoreceptors in rodent glabrous skin.

Somatosensory coding in rodents has been mostly studied in the whisker system and hairy skin, whereas the function of low-threshold mechanoreceptors (LTMRs) in rodent glabrous skin has received scant attention, unlike in primates where glabrous skin has been the focus. The relative activation of different LTMR subtypes carries information about vibrotactile stimuli, as does the rate and temporal patterning of LTMR spikes. Rate coding depends on the probability of a spike occurring on each stimulus cycle (reliability) whereas temporal coding depends on the timing of spikes relative to the stimulus cycle (precision). Using in vivo extracellular recordings in male rats and mice of either sex, we measured the reliability and precision of LTMR responses to tactile stimuli including sustained pressure and vibration. Similar to other species, rodent LTMRs were separated into rapid-adapting (RA) or slow-adapting (SA) based on their response to sustained pressure. However, unlike the dichotomous frequency preference characteristic of RA1 and RA2/Pacinian afferents in other species, rodent RAs fell along a continuum. Fitting generalized linear models (GLMs) to experimental data reproduced the reliability and precision of rodent RAs. The resulting model parameters highlight key mechanistic differences across the RA spectrum; specifically, the integration window of different RAs transitions from wide to narrow as tuning preferences across the population move from low to high frequencies. Our results show that rodent RAs can support both rate and temporal coding, but their heterogeneity suggests that co-activation patterns play a greater role in population coding than for dichotomously tuned primate RAs.Significance Statement Our sense of touch starts with activation of nerve fibres in the skin. Although response properties of various fibre types are well-established in other species (e.g. primates), quantitative characterization in rats and mice is limited. To fill this gap, we performed a comprehensive electrophysiological investigation into the coding properties of tactile fibres in rodent non-hairy skin and then simulated these fibres to explain differences in their responses. We show that rodent tactile fibres resemble those from other species, but that their heterogeneity at the population level may differ, with potentially important implications for encoding of touch. Simulations reveal intrinsic mechanisms that support this heterogeneity and provide a useful tool to explore somatosensation in rodents.

Recommendations on bioacoustical metrics relevant for regulating exposure to anthropogenic underwater sounda).

Metrics to be used in noise impact assessment must integrate the physical acoustic characteristics of the sound field with relevant biology of animals. Several metrics have been established to determine and regulate underwater noise exposure to aquatic fauna. However, recent advances in understanding cause-effect relationships indicate that additional metrics are needed to fully describe and quantify the impact of sound fields on aquatic fauna. Existing regulations have primarily focused on marine mammals and are based on the dichotomy of sound types as being either impulsive or non-impulsive. This classification of sound types, however, is overly simplistic and insufficient for adequate impact assessments of sound on animals. It is recommended that the definition of impulsiveness be refined by incorporating kurtosis as an additional parameter and applying an appropriate conversion factor. Auditory frequency weighting functions, which scale the importance of particular sound frequencies to account for an animal's sensitivity to those frequencies, should be applied. Minimum phase filters are recommended for calculating weighted sound pressure. Temporal observation windows should be reported as signal duration influences its detectability by animals. Acknowledging that auditory integration time differs across species and is frequency dependent, standardized temporal integration windows are proposed for various signal types.

Temporal integration in human auditory cortex is predominantly yoked to absolute time, not structure duration.

Sound structures such as phonemes and words have highly variable durations. Thus, there is a fundamental difference between integrating across absolute time (e.g., 100 ms) vs. sound structure (e.g., phonemes). Auditory and cognitive models have traditionally cast neural integration in terms of time and structure, respectively, but the extent to which cortical computations reflect time or structure remains unknown. To answer this question, we rescaled the duration of all speech structures using time stretching/compression and measured integration windows in the human auditory cortex using a new experimental/computational method applied to spatiotemporally precise intracranial recordings. We observed significantly longer integration windows for stretched speech, but this lengthening was very small (~5%) relative to the change in structure durations, even in non-primary regions strongly implicated in speech-specific processing. These findings demonstrate that time-yoked computations dominate throughout the human auditory cortex, placing important constraints on neurocomputational models of structure processing.

The continuous adjoint to the incompressible (D)DES Spalart-Allmaras turbulence models

This article formulates the continuous adjoint method for the gradient-based shape optimization of fluid flows governed by the incompressible Detached Eddy Simulation (DES) and Delayed-DES (DDES) models, based on the Spalart-Allmaras turbulence model. As both flow models are inherently unsteady, challenges arise regarding the availability of flow fields during the backward in time integration of the unsteady adjoint equations. To minimize both the computational cost and the memory demands, the computed flow fields are compressed using the iPGDZ+ lossy compression technique, recently developed by the authors. Its application in the context of turbulence-resolving flows, where the compression of flow fields poses an intricate challenge, is a second original contribution of this article. Everything is implemented as an extension to the publicly available adjointOptimisation library of OpenFOAM, which is used to solve the flow and adjoint equations and conduct the optimization. Using two shape optimization problems in external aerodynamics, it is demonstrated that including the adjoint to the turbulence model equation is crucial for the computation of accurate sensitivity derivatives. In contrast to sensitivities computed under the “frozen turbulence” assumption, which neglects variations in turbulent viscosity due to changes in the design variables, the proposed adjoint method yields sensitivities that align with those obtained using Finite Differences. This is due to the Think-Discrete Do-Continuous adjoint method which, inspired by hand-differentiated discrete adjoint, gives rise to consistent discretization schemes of the terms involved in the equations derived by continuous adjoint. Furthermore, it is demonstrated that the proposed adjoint method can significantly benefit from the iPGDZ+ algorithm, by reducing memory requirements by more than two orders of magnitude, eliminating the need for flow recomputations, while maintaining the accuracy of the computed derivatives. Ways to handle large integration windows of the objective function with this type of flow models are beyond the scope of this article.

A shared temporal window of integration across cognitive control and reinforcement learning paradigms: A correlational study.

Cognitive control refers to the ability to override prepotent response tendencies to achieve goal-directed behavior. On the other hand, reinforcement learning refers to the learning of actions through feedback and reward. Although cognitive control and reinforcement learning are often viewed as opposing forces in driving behavior, recent theories have emphasized possible similarities in their underling processes. With this study, we aimed to investigate whether a similar time window of integration could be observed during the learning of control on the one hand, and the learning rate in reinforcement learning paradigms on the other. To this end, we performed a correlational analysis on a large public dataset (n = 522) including data from two reinforcement learning tasks, i.e., a probabilistic selection task and a probabilistic Wisconsin Card Sorting Task (WCST), and data from a classic conflict task (i.e., the Stroop task). Results showed expected correlations between the time scale of control indices and learning rate in the probabilistic WCST. Moreover, the learning-rate parameters of the two reinforcement learning tasks did not correlate with each other. Together, these findings suggest a reliance on a shared learning mechanism between these two traditionally distinct domains, while at the same time emphasizing that value updating processes can still be very task-specific. We speculate that updating processes in the Stroop and WCST may be more related because both tasks require task-specific updating of stimulus features (e.g., color, word meaning, pattern, shape), as opposed to stimulus identity.

Prestimulus Alpha Phase Modulates Visual Temporal Integration.

When presented shortly after another, discrete pictures are naturally perceived as continuous. The neuronal mechanism underlying such continuous or discrete perception is not well understood. While continuous alpha oscillations are a candidate for orchestrating such neuronal mechanisms, recent evidence is mixed. In this study, we investigated the influence of prestimulus alpha oscillation on visual temporal perception. Specifically, we were interested in whether prestimulus alpha phase modulates neuronal and perceptual processes underlying discrete or continuous perception. Participants had to report the location of a missing object in a visual temporal integration task, while simultaneously MEG data were recorded. Using source reconstruction, we evaluated local phase effects by contrasting phase angle values between correctly and incorrectly integrated trials. Our results show a phase opposition cluster between -0.8 and -0.5 s (relative to stimulus presentation) and between 6 and 20 Hz. These momentary phase angle values were correlated with behavioral performance and event-related potential amplitude. There was no evidence that frequency defined a window of temporal integration.

Dopamine biases decisions by limiting temporal integration.

Motivations bias our responses to stimuli, producing behavioural outcomes that match our needs and goals. Here we describe a mechanism behind this phenomenon: adjusting the time over which stimulus-derived information is permitted to accumulate towards a decision. As a Drosophila copulation progresses, the male becomes less likely to continue mating through challenges1-3. We show that a set of copulation decision neurons (CDNs) flexibly integrates information about competing drives to mediate this decision. Early in mating, dopamine signalling restricts CDN integration time by potentiating Ca2+/calmodulin-dependent protein kinase II (CaMKII) activation in response to stimulatory inputs, imposing a high threshold for changing behaviours. Later into mating, the timescale over which the CDNs integrate termination-promoting information expands, increasing the likelihood of switching behaviours. We suggest scalable windows of temporal integration at dedicated circuit nodes as a key but underappreciated variable in state-based decision-making.

Abnormal Temporal Window of Integration in Auditory Sensory Memory in Schizophrenia.

Mismatch negativity (MMN) is automatically elicited by incoming sound deviation compared to the neural representation of preceding homogenous sounds stored in the brain's auditory sensory memory. This study aimed to assess time-functional deviation sensitivity in auditory sensory memory associated with a temporal window of integration (TWI) of 160-170 msec in patients with schizophrenia. To this end, we measured the magnetic counterpart of the MMN (MMNm) in 20 patients with schizophrenia on medication and 20 healthy age-matched adults as a control group responding to an omitted tone segment incorporated into a complex sound of 176 ms duration corresponding to the TWI duration. Overall, the magnitude of the MMNm was smaller in the patients with schizophrenia than in the healthy control group. The peak latency of the MMNm was prolonged in the latter omitted segments for both groups, but to a greater extent in patients with schizophrenia. These results indicate that deviation detection is impaired in the later part of the TWI, corresponding to the duration of auditory sensory memory in patients with schizophrenia. Thus, the specific impairment of MMN in response to duration deviants (duration MMN), as previously reported, might result from a damaged mechanism in the later part of the TWI of sensory memory, suggesting that a decline in sensory memory causes distorted perception or disturbances in cognitive function in patients with schizophrenia.

3D Lagrangian Particle Tracking In A Model Of The Human Airways

The three-dimensional flow field in the trachea of a physiologically correct model of the human respiratory tract from the nasal/oral cavities up to the 6th bronchi generation is investigated. The investigations comprise steady inflow conditions, i.e., steady inhalation and oscillatory flow combining inspiration and expiration to simulate calm breathing. A realistic breathing pattern is approximated by a sinusoidal waveform for two Womersley numbers of 3 and 4.5 and two Reynolds numbers of 400 and 1200. Due to the inherent three-dimensional and asymmetric nature of the flow field, 3D particle-tracking velocimetry (3D-PTV) measurements are performed using the Shake-The-Box (StB) algorithm to determine the influence of oral and nasal inhalation and exhalation on the flow field in the lower human airways. By using a refractive-index matched fluid consisting of water and glycerin, the complex flow structures inside the trachea are fully resolved. During oscillatory inhalation/exhalation, an integration window size of 6/40 pi was chosen as it can be considered short enough to assume a steady state. The PTV measurements confirm that the nasal and/or oral cavity must be taken into account when analyzing the flow field in the lower respiratory tract. Characteristic sets of velocity profiles in the sagittal and coronal plane of the trachea as well as contour plots evidence nearly identical flow structures for both oral and nasal inhalation, but also reveal that the presence of both cavities possesses a huge impact on the flow field compared to ideal, i.e., fully developed inflow conditions

A hierarchy of processing complexity and timescales for natural sounds in human auditory cortex.

Efficient behavior is supported by humans' ability to rapidly recognize acoustically distinct sounds as members of a common category. Within auditory cortex, there are critical unanswered questions regarding the organization and dynamics of sound categorization. Here, we performed intracerebral recordings in the context of epilepsy surgery as 20 patient-participants listened to natural sounds. We built encoding models to predict neural responses using features of these sounds extracted from different layers within a sound-categorization deep neural network (DNN). This approach yielded highly accurate models of neural responses throughout auditory cortex. The complexity of a cortical site's representation (measured by the depth of the DNN layer that produced the best model) was closely related to its anatomical location, with shallow, middle, and deep layers of the DNN associated with core (primary auditory cortex), lateral belt, and parabelt regions, respectively. Smoothly varying gradients of representational complexity also existed within these regions, with complexity increasing along a posteromedial-to-anterolateral direction in core and lateral belt, and along posterior-to-anterior and dorsal-to-ventral dimensions in parabelt. When we estimated the time window over which each recording site integrates information, we found shorter integration windows in core relative to lateral belt and parabelt. Lastly, we found a relationship between the length of the integration window and the complexity of information processing within core (but not lateral belt or parabelt). These findings suggest hierarchies of timescales and processing complexity, and their interrelationship, represent a functional organizational principle of the auditory stream that underlies our perception of complex, abstract auditory information.

Distinct Cortical Networks Subserve Spatio-temporal Sampling in Vision through Different Oscillatory Rhythms.

Although visual input arrives continuously, sensory information is segmented into (quasi-)discrete events. Here, we investigated the neural correlates of spatiotemporal binding in humans with magnetoencephalography using two tasks where separate flashes were presented on each trial but were perceived, in a bistable way, as either a single or two separate events. The first task (two-flash fusion) involved judging one versus two flashes, whereas the second task (apparent motion: AM) involved judging coherent motion versus two stationary flashes. Results indicate two different functional networks underlying two unique aspects of temporal binding. In two-flash fusion trials, involving an integration window of ∼50 msec, evoked responses differed as a function of perceptual interpretation by ∼25 msec after stimuli offset. Multivariate decoding of subjective perception based on prestimulus oscillatory phase was significant for alpha-band activity in the right medial temporal (V5/MT) area, with the strength of prestimulus connectivity between early visual areas and V5/MT being predictive of performance. In contrast, the longer integration window (∼130 msec) for AM showed evoked field differences only ∼250 msec after stimuli offset. Phase decoding of the perceptual outcome in AM trials was significant for theta-band activity in the right intraparietal sulcus. Prestimulus theta-band connectivity between V5/MT and intraparietal sulcus best predicted AM perceptual outcome. For both tasks, phase effects found could not be accounted by concomitant variations in power. These results show a strong relationship between specific spatiotemporal binding windows and specific oscillations, linked to the information flow between different areas of the where and when visual pathways.

Audiovisual speech asynchrony asymmetrically modulates neural binding

Previous psychophysical and neurophysiological studies in young healthy adults have provided evidence that audiovisual speech integration occurs with a large degree of temporal tolerance around true simultaneity. To further determine whether audiovisual speech asynchrony modulates auditory cortical processing and neural binding in young healthy adults, N1/P2 auditory evoked responses were compared using an additive model during a syllable categorization task, without or with an audiovisual asynchrony ranging from 240 ms visual lead to 240 ms auditory lead. Consistent with previous psychophysical findings, the observed results converge in favor of an asymmetric temporal integration window. Three main findings were observed: 1) predictive temporal and phonetic cues from pre-phonatory visual movements before the acoustic onset appeared essential for neural binding to occur, 2) audiovisual synchrony, with visual pre-phonatory movements predictive of the onset of the acoustic signal, was a prerequisite for N1 latency facilitation, and 3) P2 amplitude suppression and latency facilitation occurred even when visual pre-phonatory movements were not predictive of the acoustic onset but the syllable to come. Taken together, these findings help further clarify how audiovisual speech integration partly operates through two stages of visually-based temporal and phonetic predictions.

The impact of memory on learning sequence-to-sequence tasks

The recent success of neural networks in natural language processing has drawn renewed attention to learning sequence-to-sequence (seq2seq) tasks. While there exists a rich literature that studies classification and regression tasks using solvable models of neural networks, seq2seq tasks have not yet been studied from this perspective. Here, we propose a simple model for a seq2seq task that has the advantage of providing explicit control over the degree of memory, or non-Markovianity, in the sequences—the stochastic switching-Ornstein–Uhlenbeck (SSOU) model. We introduce a measure of non-Markovianity to quantify the amount of memory in the sequences. For a minimal auto-regressive (AR) learning model trained on this task, we identify two learning regimes corresponding to distinct phases in the stationary state of the SSOU process. These phases emerge from the interplay between two different time scales that govern the sequence statistics. Moreover, we observe that while increasing the integration window of the AR model always improves performance, albeit with diminishing returns, increasing the non-Markovianity of the input sequences can improve or degrade its performance. Finally, we perform experiments with recurrent and convolutional neural networks that show that our observations carry over to more complicated neural network architectures.

Temporal integration by multi-level regularities fosters the emergence of dynamic conscious experience.

The relationship between integration and awareness is central to contemporary theories and research on consciousness. Here, we investigated whether and how information integration over time, by incorporating the underlying regularities, contributes to our awareness of the dynamic world. Using binocular rivalry, we demonstrated that structured visual streams, constituted by shape, motion, or idiom sequences containingperceptual- or semantic-level regularities, predominated over their nonstructured but otherwise matched counterparts in the competition for visual awareness. Despite the apparent resemblance, a substantial dissociation of the observed rivalry advantages emerged between perceptual- and semantic-level regularities. These effects stem from nonconscious and conscious temporal integration processes, respectively, with the former but not the latter being vulnerable to perturbations in the spatiotemporal integration window. These findings corroborate the essential role of structure-guided information integration in visual awareness and highlight a multi-level mechanism where temporal integration by perceptually and semantically defined regularities fosters the emergence of continuous conscious experience.

Tactile cues are more intrinsically linked to motor timing than visual cues in visual-tactile sensorimotor synchronization.

Many tasks require precise synchronization with external sensory stimuli, such as driving a car. This study investigates whether combined visual-tactile information provides additional benefits to movement synchrony over separate visual and tactile stimuli and explores the relationship with the temporal binding window for multisensory integration. In Experiment 1, participants completed a sensorimotor synchronization task to examine movement variability and a simultaneity judgment task to measure the temporal binding window. Results showed similar synchronization variability between visual-tactile and tactile-only stimuli, but significantly lower than visual only. In Experiment 2, participants completed a visual-tactile sensorimotor synchronization task with cross-modal stimuli presented inside (stimulus onset asynchrony 80 ms) and outside (stimulus-onset asynchrony 400 ms) the temporal binding window to examine temporal accuracy of movement execution. Participants synchronized their movement with the first stimulus in the cross-modal pair, either the visual or tactile stimulus. Results showed significantly greater temporal accuracy when only one stimulus was presented inside the window and the second stimulus was outside the window than when both stimuli were presented inside the window, with movement execution being more accurate when attending to the tactile stimulus. Overall, these findings indicate there may be a modality-specific benefit to sensorimotor synchronization performance, such that tactile cues are weighted more strongly than visual information as tactile information is more intrinsically linked to motor timing than visual information. Further, our findings indicate that the visual-tactile temporal binding window is related to the temporal accuracy of movement execution.

Development of a comprehensive method to estimate the optical, thermal and electrical performance of a complex PV window for building integration

Increasing concerns over energy consumption and greenhouse gas emissions in buildings have contributed to the emerging of innovative PV glazing technologies to improve the building energy performance. However, some of these glazing systems have complex structures, making it challenging to investigate their optical, thermal and electrical performance for estimating their energy saving potential in buildings. In this research, a validated Computational Fluid Dynamics (CFD) combined with a ray-tracing model has been developed to accurately predict the optical, thermal and electrical performance of complex PV glazing systems under varying incident angles. A ray-tracing model is developed to calculate the light transmittance of the window and the solar energy absorbed by each solid element and PV cells. To estimate temperature profiles (e.g., PV temperature and window temperature) and secondary heat within the window, the results from the ray-tracing analysis, which detail the solar flux absorbed by each layer, are inputted into a validated CFD model as boundary conditions. Using the CFD combined ray-tracing calculation illustrated above, the Solar Heat Gain Coefficient (SHGC) of these complex PV window systems can be obtained. Furthermore, a PV modelling algorithm is developed to predict the power output based on the simulated PV temperature. This procedure is implemented to investigate a Crossed Compound Parabolic Concentrator Photovoltaic (CCPC-PV) window, which serves as an example of a complex PV glazing system in this study. The developed optical, thermal and electrical models have been validated through experimental tests. Additionally, new configurations have been designed to explore the impact of the pitch between adjacent optics on the SHGC and power output of the window. The results show that the original window (1.77 mm-pitch) possesses the maximum PV temperature of 64.73 °C and the maximum window inside surface temperature of 61.58 °C under National Fenestration Rating Council (NFRC) standard. Meanwhile the PV efficiency is 15.21 % and the SHGC is 0.463. The SHGC value of this innovative PV window is notably lower than that of a conventional double-glazed window, which has a SHGC value of 0.813. This reduction in SHGC decreases the likelihood of overheating issues, especially during the summer months.

Information-Restricted Neural Language Models Reveal Different Brain Regions' Sensitivity to Semantics, Syntax, and Context.

A fundamental question in neurolinguistics concerns the brain regions involved in syntactic and semantic processing during speech comprehension, both at the lexical (word processing) and supra-lexical levels (sentence and discourse processing). To what extent are these regions separated or intertwined? To address this question, we introduce a novel approach exploiting neural language models to generate high-dimensional feature sets that separately encode semantic and syntactic information. More precisely, we train a lexical language model, GloVe, and a supra-lexical language model, GPT-2, on a text corpus from which we selectively removed either syntactic or semantic information. We then assess to what extent the features derived from these information-restricted models are still able to predict the fMRI time courses of humans listening to naturalistic text. Furthermore, to determine the windows of integration of brain regions involved in supra-lexical processing, we manipulate the size of contextual information provided to GPT-2. The analyses show that, while most brain regions involved in language comprehension are sensitive to both syntactic and semantic features, the relative magnitudes of these effects vary across these regions. Moreover, regions that are best fitted by semantic or syntactic features are more spatially dissociated in the left hemisphere than in the right one, and the right hemisphere shows sensitivity to longer contexts than the left. The novelty of our approach lies in the ability to control for the information encoded in the models' embeddings by manipulating the training set. These "information-restricted" models complement previous studies that used language models to probe the neural bases of language, and shed new light on its spatial organization.

Adjoint Data Assimilation of the Flow on the Southern Flank of Georges Bank: March–June 1999

An adjoint data assimilation method was incorporated into the ECOM-si coastal ocean circulation model and applied to assimilate the flow field on the southern flank of Georges Bank from March to June 1999. The model was driven by tidal forcing consisting of ten tidal constituents at the open boundary, observed winds, and surface heat fluxes. Numerical experiments were conducted following a strategy to understand the critical issues affecting the efficiency and accuracy of the assimilated flow field. The adjoint data assimilation method significantly improved the computational accuracy of subtidal currents, especially for the along-isobath velocity. The integration window length and iteration number were two parameters affecting the assimilation convergence rate toward the observations. In such a nonlinear dynamical system, using a window length close to the M2 tidal period could make the adjoint model difficult to converge, no matter how many iterations were made. Reducing the time interval for the window length speeded up the convergence rate, but it was paid out with the sacrifice of statistical confidence. This assimilation experiment used a 6-h window length, which led to a faster convergence rate in the first ten iterations. The best-assimilated fields that satisfied the error criteria were obtained as the iteration number increased.