Temporal Granularity Research Articles

Using spatial video and deep learning for automated mapping of ground-level context in relief camps

BackgroundThe creation of relief camps following a disaster, conflict or other form of externality often generates additional health problems. The density of people in a highly stressed environment with questionable safe food and water access presents the potential for infectious disease outbreaks. These camps are also not static data events but rather fluctuate in size, composition, and level and quality of service provision. While contextualized geospatial data collection and mapping are vital for understanding the nature of these camps, various challenges, including a lack of data at the required spatial or temporal granularity, as well as the issue of sustainability, can act as major impediments. Here, we present the first steps toward a deep learning-based solution for dynamic mapping using spatial video (SV).MethodsWe trained a convolutional neural network (CNN) model on a SV dataset collected from Goma, Democratic Republic of Congo (DRC) to identify relief camps from video imagery. We developed a spatial filtering approach to tackle the challenges associated with spatially tagging objects such as the accuracy of global positioning system and positioning of camera. The spatial filtering approach generates smooth surfaces of detection, which can further be used to capture changes in microenvironments by applying techniques such as raster math.ResultsThe initial results suggest that our model can detect temporary physical dwellings from SV imagery with a high level of precision, recall, and object localization. The spatial filtering approach helps to identify areas with higher concentrations of camps and the web-based tool helps to explore these areas. The longitudinal analysis based on applying raster math on the detection surfaces revealed locations, which had a considerable change in the distribution of tents over space and time.ConclusionsThe results lay the groundwork for automated mapping of spatial features from imagery data. We anticipate that this work is the building block for a future combination of SV, object identification and automatic mapping that could provide sustainable data generation possibilities for challenging environments such as relief camps or other informal settlements.

Geospatial tools and data for health service delivery: opportunities and challenges across the disaster management cycle.

As extreme weather events increase in frequency and intensity, the health system faces significant challenges, not only from shifting patterns of climate-sensitive diseases but also from disruptions to healthcare infrastructure, supply chains and the physical systems essential for delivering care. This necessitates the strategic use of geospatial tools to guide the delivery of healthcare services and make evidence-informed priorities, especially in contexts with scarce human and financial resources. In this article, we highlight several published papers that have been used throughout the phases of the disaster management cycle in relation to health service delivery. We complement the findings from these publications with a rapid scoping review to present the body of knowledge for using spatial methods for health service delivery in the context of disasters. The main aim of this article is to demonstrate the benefits and discuss the challenges associated with the use of geospatial methods throughout the disaster management cycle. Our scoping review identified 48 articles employing geospatial techniques in the disaster management cycle. Most of them focused on geospatial tools employed for preparedness, anticipatory action and mitigation, particularly for targeted health service delivery. We note that while geospatial data analytics are effectively deployed throughout the different phases of disaster management, important challenges remain, such as ensuring timely availability of geospatial data during disasters, developing standardized and structured data formats, securing pre-disaster data for disaster preparedness, addressing gaps in health incidence data, reducing underreporting of cases and overcoming limitations in spatial and temporal coverage and granularity. Overall, existing and novel geospatial methods can bridge specific evidence gaps in all phases of the disaster management cycle. Improvement and 'operationalization' of these methods can provide opportunities for more evidence-informed decision making in responding to health crises during climate change.

Improving the forecast accuracy of wind power by leveraging multiple hierarchical structure

Renewable energy generation is of utmost importance for global decarbonization. Forecasting renewable energies, particularly wind energy, is challenging due to the inherent uncertainty in wind energy generation, which depends on weather conditions. Recent advances in hierarchical forecasting through reconciliation have demonstrated a significant increase in the quality of wind energy forecasts for short-term periods. We leverage the cross-sectional and temporal hierarchical structure of turbines in wind farms and build cross-temporal hierarchies to further investigate how integrated cross-sectional and temporal dimensions can add value to forecast accuracy in wind farms. We found that cross-temporal reconciliation was superior to individual cross-sectional reconciliation at multiple temporal aggregations. Additionally, machine learning based forecasts that were cross-temporally reconciled demonstrated high accuracy at coarser temporal granularities, which may encourage adoption for short-term wind forecasts. Empirically, we provide insights for decision-makers on the best methods for forecasting high-frequency wind data across different forecasting horizons and levels.

Decarbonizing the US Energy System

Recent rapid and unexpected cost reductions in decarbonization technologies have accelerated the cost-effective decarbonization of the US economy, with greenhouse gas (GHG) emissions falling by 20% from 2005 to 2020. The literature on US economy-wide decarbonization focuses on maximizing long-term GHG emissions reduction strategies that rely mostly on renewable energy expansion, electrification, and efficiency improvements to achieve net-zero GHG emissions by 2050. While these studies provide a valuable foundation, further research is needed to properly support decarbonization policy development and implementation. In this review, we identify key decarbonization analysis gaps and opportunities, including issues related to cross-sectoral linkages, spatial and temporal granularity, consumer behavior, emerging technologies, equity and environmental justice, and political economy. We conclude by discussing the implications of these analysis gaps for US decarbonization pathways and how they relate to challenges facing major global emitters.

Innovative approaches in imaging photoplethysmography for remote blood oxygen monitoring

Peripheral Capillary Oxygen Saturation (SpO2) has received increasing attention during the COVID-19 pandemic. Clinical investigations have demonstrated that individuals afflicted with COVID-19 exhibit notably reduced levels of SpO2 before the deterioration of their health status. To cost-effectively enable individuals to monitor their SpO2, this paper proposes a novel neural network model named “ITSCAN” based on Temporal Shift Module. Benefiting from the widespread use of smartphones, this model can assess an individual’s SpO2 in real time, utilizing standard facial video footage, with a temporal granularity of seconds. The model is interweaved by two distinct branches: the motion branch, responsible for extracting spatiotemporal data features and the appearance branch, focusing on the correlation between feature channels and the location information of feature map using coordinate attention mechanisms. Accordingly, the SpO2 estimator generates the corresponding SpO2 value. This paper summarizes for the first time 5 loss functions commonly used in the SpO2 estimation model. Subsequently, a novel loss function has been contributed through the examination of various combinations and careful selection of hyperparameters. Comprehensive ablation experiments analyze the independent impact of each module on the overall model performance. Finally, the experimental results based on the public dataset (VIPL-HR) show that our model has obvious advantages in MAE (1.10%) and RMSE (1.19%) compared with related work, which implies more accuracy of the proposed method to contribute to public health.

Predicting urban signal-controlled intersection congestion events using spatio-temporal neural point process

ABSTRACT The urban traffic signal-controlled intersections are of great significance for solving the problem of urban road congestion. Previous research on congestion prediction mainly aggregated data at the level of road segments or traffic flow at a coarse regulated time interval. Fine-grained prediction of congestion events at the lane-level and cycle-level enables detailed a understanding of spatio-temporal dependencies, leading to congestion reduction, improved efficiency. This paper presents a Spatio-Temporal Neural Point Process (STNPP) model that combines Graph Neural Networks and Neural Temporal Point Process to predict congestion events at urban intersections. The proposed model allows for complete prediction of congestion events, including their occurrence, development, dissipation. In the process of spatial correlation modeling, graph neural networks are used to model the spatial relationships between both region and intersections. The current intersection and its upstream/downstream areas are modeled separately. To model the temporal correlations at individual intersections, we focus on a specific lane and capture the evolution of congestion events using the Neural Point Process Gated Recurrent Unit (NPPGRU), which captures the temporal granularity changes of signal-controlled cycles in congestion events. Using actual traffic speed and signal-controlled data from Hangzhou city, we validate that the proposed method achieves stable predictive performance.

System-wide benefits of temporal alignment of wholesale–retail electricity prices

Exposing residential consumers to real-time pricing (RTP) can yield significant efficiency gains, but may also pose challenges due to the inherent complexity. Using an extensive spectrum of temporal granularities for retail electricity pricing, we analyze to what extent these capture the benefits of RTP while concurrently considering the influence of three distribution tariff types in distorting the pricing signal. Via a mixed complementarity model, we consider both operational and investment decisions at retail as well as wholesale levels, accounting for their interdependencies. Our analysis reveals three key findings. Firstly, we find that decreasing temporal granularity of retail electricity prices increases total system cost due to an increasingly inefficient generation mix, however, three-hourly or six-hourly pricing can approximate RTP benefits. Secondly, decreasing temporal granularity of retail electricity prices raises average offtake and lowers average injection prices for consumers due to changes in wholesale market prices as well as consumption patterns of consumers. Thirdly, capacity-based tariffs reduce system offtake peaks and associated price increases by incentivizing battery discharge during peak consumption while volumetric tariffs, over-incentivizing solar PV investments, result in highest offtake prices and injection peaks.

Accelerating decarbonization digitally: Status quo and potentials of greenhouse gas emission tracking and trading

To effectively mitigate climate change, policymakers worldwide established various GHG tracking and trading systems. In the light of ambitious climate goals, stricter regulations, and increasing demand for climate action, various groups such as researchers and governmental institutions suggested additional approaches. This paper addresses the complexity that arises from the breadth of suggested approaches and implemented systems for GHG tracking and trading. By doing so, it synthesizes relevant dimensions in a way that is understandable to enterprises and policymakers, enabling them to design meaningful systems incorporating the reduction of GHG emissions and advance cleaner production. Therefore, this paper presents a first-of-its-kind taxonomy of GHG tracking and trading approaches through a systematic literature review. It illustrates ten main design and implementation dimensions with 30 corresponding characteristics. To accelerate decarbonization, this paper sets impulses for future GHG tracking in the electricity sector based on semi-structured expert interviews. Consecutively, it provides policy directions for CO2-adaptive decision-making for enterprises, formulated as a Call for Action with seven prospective questions. These include, for example, questions concerning technical aspects like data management, legal issues like the sufficiency of existing data security and privacy regulations, as well as economic topics like the calculation of an appropriate local and temporal granularity.

Exploring impacts of thermal and lighting conditions on office workers’ subjective evaluations, cognitive performance and EEG features in multi-person offices

Modern office workers spend lots of time in multi-person offices. For human-centered building design and operation, it is essential to investigate the influence of coupled physical environments on occupants' well-being and task performance. In this study, to assess impacts of thermal and lighting environments on brainworkers, experiments were conducted under 64 conditions of different temperature, correlated color temperature, and illumination in a multi-person office. Electroencephalogram (EEG) signals were collected to measure the mental state of brainworkers and neurobehavioral tests were utilized to replicate cognitive tasks. The results indicated that only the interaction between the temperature and illumination regarding the sensation of CCT was significant. Primarily, lower illumination in colder environments made subjects perceive the lighting was warmer. Furthermore, cross-domain effects influenced office workers' sensation via synesthetic effects. And the validity of the hue-heat hypothesis on thermal sensation depended on whether subjects received enough stimulation. Besides, for brainworkers, visual comfort peaked at 5000 K and 750 lux but the optimal temperature varied based on specific requirements, with 24 °C promoting overall comfort and 26 °C enhancing executive cognitive performance. The study preliminarily explored the effects among acceptable levels of T, CCT, and I on EEG features, revealing that temperature could adjust occupants’ mental load. However, the further investigation requests equipment with finer temporal granularity, cost-effective techniques and advanced analytical methods to guide office building design and operation.

Evaluating rainfall erosivity on the Tibetan Plateau by integrating high spatiotemporal resolution gridded precipitation and gauge data

High-precision rainfall erosivity mapping is crucial for accurately evaluating regional soil erosion on the Tibetan Plateau (TP) under the backdrop of climate warming and humidification. Although high spatiotemporal resolution gridded precipitation data provides the foundation for rainfall erosivity mapping, the increasing spatial heterogeneity of rainfall with decreasing temporal granularity can lead to greater errors when directly computing rainfall erosivity from gridded precipitation data. In this study, a site-scale conversion coefficient was established so that rainfall erosivity calculated using hourly data can be converted to rainfall erosivity calculated using per-minute data. A revised model was established for calculating the rainfall erosivity based on high-resolution hourly precipitation data from the Third Pole gridded precipitation dataset (TPHiPr). The results revealed a notable underestimation in the original calculation results obtained using the TPHiPr, but strong correlation was observed between the two sets of results. There was a significant improvement in the Nash–Sutcliffe coefficient of efficiency (from −0.39 to 0.80) and the Percent Bias (from −63.95 % to 0.37 %) after model revision. The TPHiPr effectively depict the spatial characteristics of rainfall erosivity on the TP. It accurately reflected the rain shadow area on the northern flank of the Himalayas and the dry-hot valley in the Hengduan Mountains. It also showed high rainfall erosivity values in the tropical rainforest area on the southern flank of the eastern Himalayas. The overall trend of rainfall erosivity has increased on the TP during the period 1981 to 2020, with 65.91 % of the regions exhibiting an increasing trend and 22.25 % showing significant increases, indicating an intensified risk of water erosion. These findings suggest that the 40-year-high spatial resolution rainfall erosivity dataset can provide accurate data support for a quantitative understanding of soil erosion on the TP.

A forecast-driven stochastic optimization method for proactive activation of manual reserves

Reducing operating balancing costs is paramount for an affordable transition towards renewable-dominated power systems. In European balancing markets, operating balancing costs are driven by the activation of automatic and manual frequency restoration reserves, respectively aFRR and mFRR. An inadequate combination of both products for resolving grid imbalances may result in economic inefficiencies where, e.g., saturated aFRR can lead to balancing price spikes. To avoid such situation, we propose a proactive activation policy of manual reserves, aiming at an optimal trade-off between aFRR and mFRR products via a stochastic optimization method. The tool is fed with 1-min time trajectories of system imbalances covering the next quarter hour. The one minute temporal granularity allows modeling the ramping phenomena of mFRR products, while keeping track of the faster activation of aFRR products. The proposed balancing energy activation methodology is tested on Belgian market data, which currently adopts a reactive balancing strategy. Ex-post comparisons of the proposed balancing strategy with a reactive one show that our methodology allows a decrease of the balancing activation costs. This is expected as early activation of mFRR, when appropriately provided, allows avoiding the activation of extremely high aFRR bids.

Examining time-varying dynamics of co-occurring depressed mood and anxiety

BackgroundDimensional frameworks of psychopathology call for multivariate approaches to map co-occurring disorders to index what symptoms emerge when and for whom. Ecological momentary assessment (EMA) offers a method for assessing and differentiating the dynamics of co-occurring symptoms with greater temporal granularity and naturalistic context. The present study used multivariate mixed effects location-scale modeling to characterize the time-varying dynamics of depressed mood and anxiety for women diagnosed with social anxiety disorder (SAD) and major depression (MDD). MethodsWomen completed five daily EMA surveys over 30 days (150 EMA surveys/woman, T ≈ 5250 total observations) and two clinical diagnostic and retrospective self-report measures administered approximately two months apart. ResultsThere was evidence of same-symptom lagged effects (bs = 0.08–0.09), but not cross-symptom lagged effects (bs < 0.01) during EMA. Symptoms co-varied such that momentary spikes from one's typical level of anxiety were associated with increases in momentary depressed mood (b = 0.19) and greater variability of depressed mood (b = 0.06). Similarly, spikes from one's typical levels of depressed mood were associated with increases in momentary anxiety (b = 0.19). Furthermore, the presence and magnitude of effects demonstrated person-specific heterogeneity. LimitationsOur findings are constrained to the dynamics of depressed and anxious mood among cisgender women with primary SAD and current or past MDD. ConclusionsFindings from this work help to characterize how daily experiences of co-occurring mood and anxiety fluctuate and offer insight to aid the development of momentary, person-specific interventions designed to regulate symptom fluctuations.

Power prediction using high-resolution SCADA data with a farm-wide deep neural network approach

Accurate loss estimation methods with a high level of temporal granularity are necessary to enable the implementation of efficient and adaptable control strategies for wind farms. Predictive models for the power of wind turbines within a wind farm are investigated using high-resolution SCADA data and deep learning methodologies. Traditional physical models offer detailed insights but are computationally expensive. Statistical models face limitations in handling wind energy variability. In this study, deep learning models are explored to capture spatial and temporal dynamics, with four models developed: Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory network (LSTM), and a hybrid CNN-LSTM model. SCADA data from an offshore wind farm is preprocessed, anomalies removed, and annotated based on operational regions. The models are trained, validated, and tested, demonstrating improved accuracy over baseline methods. The hybrid model, incorporating spatial and temporal information, yields the highest predictive performance, showcasing the significance of these dimensions in wind power prediction.

A data-driven approach for estimating travel-related carbon emissions with high spatial and temporal granularity

A data-driven approach for estimating travel-related carbon emissions with high spatial and temporal granularity

Contextual operational energy performance indexing of district heating consumers

Energy performance indices are widely used methods for determining and assessing the efficiency of buildings in operation. However, methodologies for indices that compare heat consumption among buildings and at higher temporal granularity than yearly are needed. This makes it possible for consumers to compare themselves to others to encourage them to improve their energy efficiency. For a district heating system, it also provides the district heating company with an overview of consumers’ energy efficiency (and consumers’ efficiency must be comparable). This is critical due to the interconnectedness of district heating systems, as consumers can indirectly have negative impacts on each other’s operations. Having an overview of the efficiency of consumers can provide useful information for the district heating companies to detect unusual consumption and possible faults. This paper therefore has created a framework for the comparison of heat consumption among buildings, by taking into account various contextualities to make comparisons possible. The framework has been tested on a case study, and its capabilities are showcased.

Exploring drivers of overnight stays and same-day visits in the tourism sector

We employ mobile network data referred to the area of Lombardy in Italy to investigate alternative touristic behaviours, such as same-day visits and overnight stays in Italy. We show that larger availability of tourism accommodations, cultural and natural endowments are relevant factors explaining overnight stays. Conversely, temporary entertainment and transportation facilities increase municipalities attractiveness for same-day visits. The results also highlight a trade-off in the capability of municipalities of being attractive in connection to both the tourism behaviours. For instance, higher tourists arrivals are observed in areas receiving limited visitors, coming from municipalities with low same-day visits outflows. We highlight mobile data offer an adequate level of spatial and temporal granularity and can be thus employed to support policy makers in the design of effective tourist management strategies.

GGNet: A novel graph structure for power forecasting in renewable power plants considering temporal lead-lag correlations

Power forecast for each renewable power plant (RPP) in the renewable energy clusters is essential. Though existing graph neural networks (GNN)-based models achieve satisfactory prediction performance by capturing dependencies among distinct RPPs, the static graph structure employed in these models ignores crucial lead-lag correlations among RPPs, resulting from the time difference of the air flow at spatially dispersed RPPs. To address this problem, this paper proposes a novel dynamic graph structure using multiple temporal granularity groups (TGGs) to characterize the lead-lag correlations among RPPs. A granular-based GNN called GGNet is designed to generate an optimal adjacency matrix for the proposed graph structure. Specifically, a two-dimensional convolutional neural network (2D-CNN) is used to quantify the uncertain lead-lag correlations among RPPs; secondly, a gate mechanism is used to calculate a dynamic adjacency matrix; Finally, a graph attention network (GAT) is used to aggregate the information on RPPs based on the well-learned adjacency matrix. Case studies conducted using real-world datasets, with wind power plants and photovoltaic power plants, show our method outperforms state-of-the-art (SoTA) ones with better performance. Compared with the SoTA models, the RMSEN and MAEN of wind power plants for 1–4 h forecast steps decreased on average by 22.925% and 13.18%, respectively; the RMSEN and MAEN of photovoltaic power plants for 1–4 h forecast steps decreased on average by 48.95% and 18.75%, respectively. The results show that the proposed framework can generate improved performance with accuracy and robustness.

Leaving terrorism behind? The role of terrorist attacks in shaping migration intentions around the world

ABSTRACT Terrorism globally yields severe consequences for individuals and societies, potentially driving migration within and across borders. Yet, empirical evidence on its causal impact remains limited. The contribution of our paper is twofold. First, we construct various indicators of terrorist activity at a fine level of spatial and temporal granularity, which allow to accurately identify individuals’ exposure to terrorist threat. Second, we use these geo-localised indicators to empirically analyse the role of terrorist attacks in shaping internal and international migration intentions for 133 countries between 2007 and 2015. Our results indicate that terrorist attacks spur both internal and international migration intentions, though the effect is stronger for the latter. The effect on international migration intentions is linked to the intensity of attacks, rather than their frequency. Furthermore, the impact varies based on individual and country characteristics.

GMP-AR: Granularity Message Passing and Adaptive Reconciliation for Temporal Hierarchy Forecasting

Time series forecasts of different temporal granularity are widely used in real-world applications, e.g., sales prediction in days and weeks for making different inventory plans. However, these tasks are usually solved separately without ensuring coherence, which is crucial for aligning downstream decisions. Previous works mainly focus on ensuring coherence with some straightforward methods, e.g., aggregation from the forecasts of fine granularity to the coarse ones, and allocation from the coarse granularity to the fine ones. These methods merely take the temporal hierarchical structure to maintain coherence without improving the forecasting accuracy. In this paper, we propose a novel granularity message-passing mechanism (GMP) that leverages temporal hierarchy information to improve forecasting performance and also utilizes an adaptive reconciliation (AR) strategy to maintain coherence without performance loss. Furthermore, we introduce an optimization module to achieve task-based targets while adhering to more real-world constraints. Experiments on real-world datasets demonstrate that our framework (GMP-AR) achieves superior performances on temporal hierarchical forecasting tasks compared to state-of-the-art methods. In addition, our framework has been successfully applied to a real-world task of payment traffic management in Alipay by integrating with the task-based optimization module.

CityPulse: Fine-Grained Assessment of Urban Change with Street View Time Series

Urban transformations have profound societal impact on both individuals and communities at large. Accurately assessing these shifts is essential for understanding their underlying causes and ensuring sustainable urban planning. Traditional measurements often encounter constraints in spatial and temporal granularity, failing to capture real-time physical changes. While street view imagery, capturing the heartbeat of urban spaces in a pedestrian point of view, can add as a high-definition, up-to-date, and on-the-ground visual proxy of urban change. We curate the largest street view time series dataset to date, and propose an end-to-end change detection model to effectively capture physical alterations in the built environment at scale. We demonstrate the effectiveness of our proposed method by benchmark comparisons with previous literature and implementing it at the city-wide level. Our approach has the potential to supplement existing dataset and serve as a fine-grained and accurate assessment of urban change.