Multi-Index Ecoacoustics Analysis for Terrestrial Soundscapes: A New Semi-Automated Approach Using Time-Series Motif Discovery and Random Forest Classification

High rates of biodiversity loss caused by human-induced changes in the environment require new methods for large scale fauna monitoring and data analysis. While ecoacoustic monitoring is increasingly being used and shows promise, analysis and interpretation of the big data produced remains a challenge. Computer-generated acoustic indices potentially provide a biologically meaningful summary of sound, however, temporal autocorrelation, difficulties in statistical analysis of multi-index data and lack of consistency or transferability in different terrestrial environments have hindered the application of those indices in different contexts. To address these issues we investigate the use of time-series motif discovery and random forest classification of multi-indices through two case studies. We use a semi-automated workflow combining time-series motif discovery and random forest classification of multi-index (acoustic complexity, temporal entropy, and events per second) data to categorize sounds in unfiltered recordings according to the main source of sound present (birds, insects, geophony). Our approach showed more than 70% accuracy in label assignment in both datasets. The categories assigned were broad, but we believe this is a great improvement on traditional single index analysis of environmental recordings as we can now give ecological meaning to recordings in a semi-automated way that does not require expert knowledge and manual validation is only necessary for a small subset of the data. Furthermore, temporal autocorrelation, which is largely ignored by researchers, has been effectively eliminated through the time-series motif discovery technique applied here for the first time to ecoacoustic data. We expect that our approach will greatly assist researchers in the future as it will allow large datasets to be rapidly processed and labeled, enabling the screening of recordings for undesired sounds, such as wind, or target biophony (insects and birds) for biodiversity monitoring or bioacoustics research.

Population-level inference for home-range areas

· Home-range estimates are a common product of animal tracking data, as each range informs on the area needed by a given individual. Population-level inference on home-range areas—where multiple individual home-ranges are considered to be sampled from a population—is also important to evaluate changes over time, space, or covariates, such as habitat quality or fragmentation, and for comparative analyses of species averages. Population-level home-range parameters have traditionally been estimated by first assuming that the input tracking data were sampled independently when calculating home ranges via conventional kernel density estimation (KDE) or minimal convex polygon (MCP) methods, and then assuming that those individual home ranges were measured exactly when calculating the population-level estimates. This conventional approach does not account for the temporal autocorrelation that is inherent in modern tracking data, nor for the uncertainties of each individual home-range estimate, which are often large and heterogeneous. · Here, we introduce a statistically and computationally efficient framework for the population-level analysis of home-range areas, based on autocorrelated kernel density estimation (AKDE), that can account for variable temporal autocorrelation and estimation uncertainty. · We apply our method to empirical examples on lowland tapir (Tapirus terrestris), kinkajou (Potos flavus), white‐nosed coati (Nasua narica), white-faced capuchin monkey (Cebus capucinus), and spider monkey (Ateles geoffroyi), and quantify differences between species, environments, and sexes. · Our approach allows researchers to more accurately compare different populations with different movement behaviors or sampling schedules, while retaining statistical precision and power when individual home-range uncertainties vary. Finally, we emphasize the estimation of effect sizes when comparing populations, rather than mere significance tests.

Internet Search Effort on Covid-19 and the Underlying Public Interventions and Epidemiological Status

This study aimed to quantify the relative importance of indices of personal freedom, economy, and epidemiology on the public interest on Covid-19 expressed by internet searches on the topic. The relationship between the effective reproduction rate Rt, news media cover, and web search effort was also quantified. Data of online search in Greece on Covid-19 topic for one year were analyzed using indices of social distancing, financial measures, and epidemiological variables using machine learning. Temporal autocorrelation of web search effort was quantified and control charts of web search, Rt, new cases and new deaths were employed. Results indicated that the trained model exhibited a fit of R2 = 91% between the actual and predicted web search effort. The top five variables for predicting web search effort were new deaths, the opening of international borders to non-Greek nationals, new cases, testing policy, and restrictions in internal movements. Web search had negligible temporal autocorrelation between weeks. Web search peaked during the same weeks that the Rt was peaking although new deaths or new cases were not peaking during those dates. The extent to which online searches may reflect the actual epidemiological situation is discussed.

Spatial and temporal autocorrelation weave human brain networks

Correlations are a basic object of analysis across neuroscience, but multivariate patterns of correlations can be difficult to interpret. For example, correlations are fundamental to understanding timeseries derived from resting-state functional magnetic resonance imaging (rs-fMRI), a proxy of brain activity. Networks constructed from regional correlations in rs-fMRI timeseries are often interpreted as brain connectivity, yet the links between brain networks and neurobiology have until now been largely speculative. Here, we show that the topology of rs-fMRI brain networks is caused by the spatial and temporal autocorrelation of the timeseries used to construct them. Spatial and temporal autocorrelation show high test-retest reliability, and are correlated with popular measures of network topology. A generative model of spatially and temporally autocorrelated timeseries exhibits similar network topology to brain networks, and when fit to individual subjects, it captures near the reliability limit of subject and regional variation. We demonstrate why spatial and temporal autocorrelation induce network structure, and highlight their ability to link graph properties to neurobiology during healthy aging. These results offer a reductionistic account of brain network complexity, explaining characteristic patterns in brain networks using timeseries statistics.

Identifying and Decoding Temporally Correlated Neural Patterns in the Brain to Accurately Predict Individual Finger Movement and Tactile Stimuli of the Human Hand Using Stereoelectroencephalography

Millions of people worldwide suffer motor or sensory impairment due to stroke, spinal cord injury, multiple sclerosis, traumatic brain injury, diabetes, and motor neuron diseases such as ALS (amyotrophic lateral sclerosis). A brain-computer interface (BCI), which links the brain directly to a computer, offers a new way to study the brain and potentially restore function in patients living with debilitating conditions. One of the challenges currently facing BCI technology, however, is how to minimize surgical risk. Minimally invasive techniques, such as stereoelectroencephalography (SEEG) have become more widely used in clinical applications since they can lead to fewer complications. SEEG electrodes also give access to sulcal and white matter areas of the brain but have not been widely studied in brain-computer interfaces. We therefore investigated the viability of using SEEG electrodes in a BCI for recording and decoding neural signals related to movement and the sense of touch and compared its performance to electrocorticography electrodes (ECoG) placed on gyri. Initial poor decoding performance and the observation that most neural modulation patterns were variable trial-to-trial and transient (significantly shorter than the sustained finger movements studied), led to the development of a feature selection method based on temporal autocorrelation, a repeatability metric. An algorithm based on temporal autocorrelation was developed to isolate features that consistently repeated (required for accurate decoding) and possessed information content related to movement or touch-related stimuli. We subsequently used these features, along with deep learning methods, to automatically classify various motor and sensory events for individual fingers with high accuracy. Repeating features were found in sulcal, gyral, and white matter areas and were predominantly phasic or phasic-tonic across a wide frequency range for both HD (high density) ECoG and SEEG recordings. These findings motivated the use of long short-term memory (LSTM) recurrent neural networks (RNNs) which are well-suited to handling both transient and sustained input features. Combining temporal autocorrelation-based feature selection with LSTM yielded decoding accuracies of up to 92.04 +/- 1.51% for hand movements, up to 91.69 +/- 0.49% for individual finger movements, and up to 80.64 +/- 1.64% for focal tactile stimuli to the finger pads and palm while using a relatively small number of SEEG electrodes. These findings may lead to a new class of minimally invasive brain-computer interface systems in the future, increasing its applicability to a wider variety of conditions.

Quantifying community responses to environmental variation from replicate time series

Time-series data for ecological communities are increasingly available from long-term studies designed to track species responses to environmental change. However, classical multivariate methods for analyzing community composition have limited applicability for time series, as they do not account for temporal autocorrelation in community-member abundances. Furthermore, traditional approaches often obscure the connections between responses at the community level and those for individual taxa, limiting their capacity to infer mechanisms of community change. We show how linear mixed models that account for group-specific temporal autocorrelation and observation error can be used to infer both taxon- and community-level responses to environmental predictors from replicated time-series data. Variation in taxon-specific responses to predictors is modeled using random effects, which can be used to characterize variation in community composition. Moreover, the degree of autocorrelation is estimated separately for each taxon, since this is likely to vary due to differences in their underlying population dynamics. We illustrate the utility of the approach by analyzing the response of a predatory arthropod community to spatiotemporal variation in allochthonous resources in a subarctic landscape. Our results show how mixed models with temporal autocorrelation provide a unified approach to characterizing taxon- and community-level responses to environmental variation through time.

Size and shape assortative mating in Japanese beetles (Popillia japonica)

Assortative mating is hypothesized to be a product of sexual selection, mating constraints, or temporal autocorrelation. I test these hypotheses in the Japanese beetle (Popillia japonica Newman, 1841), a sexually size dimorphic invasive insect pest in North America, by measuring the size and shape of bodies and wings of pair members in a wild population. Because male P. japonica prefer to mate with larger females and larger males outcompete rivals for mating opportunities, sexual selection is expected to produce size-related assortative mating. The current study did not support this hypothesis. The mating constraints hypothesis was also not supported because beetle pairs did not have similar body shapes. I, however, did find support for the temporal autocorrelation hypothesis as the wing size and shape of pair members were significantly correlated. This mating pattern likely arises due to individuals with larger and more slender wings arriving earlier at aggregation sites and pairing according to their arrival sequence. Although I found less support for the sexual selection hypothesis, I argue that mate choice might play an important, but secondary, role to temporal autocorrelation in explaining assortative mating in Japanese beetles.

Mining the Spatial-Temporal Characteristics of Population Flow and Material Flow Using Big Data

Population flow and material flow are the core elements of the “space of flows.” Studying the temporal and spatial distribution characteristics of these two types of flows (TToF) can provide us a greater understanding of the research area. Most of the existing works related to TToF only focus on exploring the difference characteristics of one of the members of TToF in a certain time or space scale in the research area. Different from these related works, the spatial-temporal characteristics of the population flow and material flow in Taiwan Province and the spatial-temporal autocorrelation of Taiwan’s expressway network are explored by means of multimembership and layer-by-layer refinement. The research work carried out in this paper includes the following: (1) studying the differentiated characteristics of the TToF in different time units; (2) studying the spatial differences among each type of the TToF under different scales; (3) dividing both the population flow and material flow into two subtypes and then analyzing the temporal variation characteristics of the four subtypes of flows; and (4) studying the global and local spatial-temporal autocorrelation of Taiwan’s expressway network. The results show the following. (1) The spatial-temporal differentiation characteristics of the TToF are obvious in different time units and on different scales. (2) The contribution of the population flow to the TToF in flow quantities is far greater than that of the material flow. (3) The population flow and material flow are dominated by the “minority population flow” and “small-scale material flow,” respectively. (4) Meanwhile, in Taiwan’s expressway network, there is a significant spatial-temporal positive correlation mainly reflected in the spatial first-order adjacent road sections.