Impact and mechanism of the COVID-19 on the spatio-temporal patterns of Zhangjiajie tourism network attention

Based on the spatial-temporal perspective of geography, this paper quantitatively measures the impact of COVID-19 on the spatial-temporal pattern of tourism network attention in Zhangjiajie, and finally summarizes the influencing factors and mechanisms. The results show as follows. (1) From the perspective of time, the online attention of tourism in Zhangjiajie shows a trend of "decline to rebound, and to stability", which reflects the temporal mobility of the effect of COVID-19 on the tourism. (2) From the spatial dimension, the scale-order of attention to the Zhangjiajie' s tourism network is relatively stable, and the effect of COVID-19 on the tourism shows a trend of "distance decay" on the whole. (3) The adjustment of tourists' perception of tourism risk, destination familiarity and location, tourists' risk tolerance and authority restriction are the influencing factors of tourism net-work attention. These factors interact with each other to drive the spatio-temporal change of tourism network attention.

Different approaches to explore the impact of COVID-19 lockdowns on carbonaceous aerosols at a European rural background site.

To prevent the fast spread of COVID-19, worldwide restrictions have been put in place, leading to a reduction in emissions from most anthropogenic sources. In this study, the impact of COVID-19 lockdowns on elemental (EC) and organic (OC) carbon was explored at a European rural background site combining different approaches: - "Horizontal approach (HA)" consists of comparing concentrations of pollutants measured at 4 m a.g.l. during pre-COVID period (2017-2019) to those measured during COVID period (2020-2021); - "Vertical approach (VA)" consists of inspecting the relationship between OC and EC measured at 4 m and those on top (230 m) of a 250 m-tall tower in Czech Republic. The HA showed that the lockdowns did not systematically result in lower concentrations of both carbonaceous fractions unlike NO2 (25 to 36 % lower) and SO2 (10 to 45 % lower). EC was generally lower during the lockdowns (up to 35 %), likely attributed to the traffic restrictions whereas increased OC (up to 50 %) could be attributed to enhanced emissions from the domestic heating and biomass burning during this stay-home period, but also to the enhanced concentration of SOC (up to 98 %). EC and OC were generally higher at 4 m suggesting a greater influence of local sources near the surface. Interestingly, the VA revealed a significantly enhanced correlation between EC and OC measured at 4 m and those at 230 m (R values up to 0.88 and 0.70 during lockdown 1 and 2, respectively), suggesting a stronger influence of aged and long distance transported aerosols during the lockdowns. This study reveals that lockdowns did not necessarily affect aerosol absolute concentrations but it certainly influenced their vertical distribution. Therefore, analyzing the vertical distribution can allow a better characterization of aerosol properties and sources at rural background sites, especially during a period of significantly reduced human activities.

Identifying, exploring, and interpreting time series shapes in multivariate time intervals

We introduce a concept of episode referring to a time interval in the development of a dynamic phenomenon that is characterized by multiple time-variant attributes. A data structure representing a single episode is a multivariate time series. To analyse collections of episodes, we propose an approach that is based on recognition of particular patterns in the temporal variation of the variables within episodes. Each episode is thus represented by a combination of patterns. Using this representation, we apply visual analytics techniques to fulfil a set of analysis tasks, such as investigation of the temporal distribution of the patterns, frequencies of transitions between the patterns in episode sequences, and co-occurrences of patterns of different variables within same episodes. We demonstrate our approach on two examples using real-world data, namely, dynamics of human mobility indicators during the COVID-19 pandemic and characteristics of football team movements during episodes of ball turnover. © 2023 The Author(s)

Forecast of beef production for Brazilian market between 2021 and 2022

Cattle is one of the main items in the Brazilian productive guideline and an important export product. During the covid-19 pandemic, the price of beef occupied a prominent position in agricultural sector analyzes due to the prices increases. The objective of this research is to observe the national production behavior, exports, and domestic supply. Therefore, a domestic supply forecast was made for January 2021 to December 2022 (24 months). Based on the results obtained, it was found that the beefs supply available to the Brazilian market will not present an expressive upward behavior that compensates the evolution in beef export to international markets. Thus, a shift in the price of beef in the domestic market to higher levels may be observed.

COVID-19 and employment relief programs: A tale of spatially blind policies for a spatially driven pandemic

We use the case of Chile to analyze the effectiveness of a spatially blind employment relief program (hereafter referred to as the LPE program) established by the Chilean government and implemented during the COVID-19 pandemic. Chile is an interesting case because on the one hand its nonpharmaceutical interventions were spatially driven by health indicators based on small geographical areas;hence, producing sizeable regional and temporal variation of the local conditions induced by the COVID-19 pandemic. On the other hand, the LPE program was designed and implemented nationally without distinction of local labor market or pandemic conditions, and each firm could decide whether to enroll in the program. By exploiting the spatial-temporal variation of exogenously imposed lockdowns and using a difference-in-differences panel data framework, we find that the LPE program was only effective for a group of regions in the country but, more importantly, that the LPE program was less effective during lockdowns. Moreover, the requirements of the LPE program were vague and did not target specific populations or entities. Consequently, our results suggest that women, informal and small firm workers, and most economic sectors throughout the country were less able to take advantage of the benefits of this program. © 2023 Wiley Periodicals LLC.

On growth patterns and mechanisms in arctic-alpine shrubs. (Special Issue: The uneven geographies of the COVID-19 pandemic.)

Arctic-alpine ecosystems are considered hot-spots of environmental change, with rapidly warming conditions causing massive alterations in vegetational structure. These changes and their environmental controls are highly complex and variable across spatial and temporal scales. Yet, despite their numerous implications for the global climate system, the underlying physiological processes and mechanisms at the individual plant scale are still little explored. Using hourly recordings of shrub stem diameter change provided by dendrometers, paired with on-site environmental conditions, enabled us to shed light on these processes. In this way, growth patterns in three widely distributed shrub species were assessed and linked to thermal and hygric conditions. We started our analysis with a close examination of one evergreen species under extreme environmental conditions, followed by a comparison of evergreen and deciduous species, and, finally, a comparative look at growth patterns across local micro-habitats. The results revealed distinct growth strategies, closely linked to species-specific water-use dynamics and cambial rhythms. Within the heterogenous alpine landscape these conditions were mainly attributed to the variation in local micro-habitats, defined by fine-scale topography and consequent variation in snow conditions and exposure. Thus, the overall growth success was mainly controlled by complex seasonal dynamics of soil moisture availability, snow conditions, and associated freeze-thaw cycles. It was therefore in many cases decoupled from governing regional climate signals. At the same time, exceedingly high summer temperatures were limiting shrub growth during the main growing season, resulting in more or less pronounced bimodal growth patterns, indicating potential growth limitation with on-going summer warming. While shrubs are currently able to maximize their growth success through a high level of adaptation to local micro-site conditions, their continued growth under rapidly changing environmental conditions is uncertain. However, our results suggest a high level of heterogeneity across spatial and temporal scales. Thus, broad-scale vegetational shifts can not be explained by a singular driver or uniform response pattern. Instead, fine-scale physiological processes and on-site near-ground environmental conditions have to be incorporated into our understanding of these changes.

Characteristics of Surface Ozone and Nitrogen Oxides over a Typical City in the Yangtze River Delta, China

The Yangtze River Delta (YRD) is the most developed region in China. Influenced by intensive and complex anthropogenic activities, atmospheric pollution in this region is highly variable, and reports are sparse. In this study, a seven-year history of the atmospheric O3 and NOx mixing ratios over a typical city, Hangzhou, was presented to enrich the studies on air pollution in the YRD region. Our results revealed that the diurnal variation in NOx corresponded to traffic rush hours, while O3 was mainly impacted by photochemical reactions in the daytime. The weekend effect was significant for NOx, but inapparent for O3. Two O3 peaks in May and September were caused by seasonal atmospheric stability and climatic conditions. The lower NOx and higher O3 levels observed suggested direct effects from traffic restrictions and large-scale industrial shutdowns during the COVID-19 lockdown in 2020 compared with those in the periods before and after lockdown. The model simulation results showed that O3 mixing ratios were not only related to regional anthropogenic emissions but were impacted by air mass transportation from surrounding provinces and the China shelf seas. The NOx mixing ratios showed a decreasing trend, while the O3 mixing ratios showed the opposite trend from 2015 to 2021, which is indicative of the implementation of the Air Pollution Prevention and Control Acton Plan issued by the Chinese government in 2013. [ABSTRACT FROM AUTHOR] Copyright of Atmosphere is the property of MDPI and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Variation of carbon dioxide mole fraction at a typical urban area in the Yangtze River Delta, China

The observations of atmospheric CO2 mole fraction in urban area in China are relative sparse. Here, we present the first-hand observation of atmospheric CO2 mole fraction from 2016 to 2020 at a city station (Hangzhou, abbreviated as HZ) in the Yangtze River Delta, which is one of the strongest CO2 source regions in China. The CO2 mole fraction at an adjacent World Meteorological Organization / Global Atmospheric Watch (WMO/GAW) programme site (Lin'an, LAN) are also presented and compared. The temporal variations, seasonal variations, and influence of COVID-19 pandemic are analyzed. Our results show that, the variations of CO2 mole fraction in Hangzhou are mainly driven by the local emissions, both atmospheric dilution conditions (i.e., wind speed, visibility) and topography, and the temporal variations are apparently different with the suburb site of LAN, although the distance between the two stations is only 50 km. During the observation period, the CO2 mole fraction at HZ is on average 15.6 +or- 0.2 ppm higher than LAN, with two distinct peaks observed at 9:00 and 17:00-18:00, corresponding to traffic rushing hours. The growth rate of atmospheric CO2 mole fraction is 11.2 +or- 0.1 ppm yr-1 before the COVID-19 pandemic (from 2016 to 2019), which is much higher than the suburb site of LAN, 5.4 +or- 0.1 ppm yr-1. The COVID-19 pandemic has led to a plunge of atmospheric CO2 mole fraction at HZ in 2020, with a value of 15.7 +or- 0.7 ppm, corresponding to 3.5% lower than the year of 2019. But at LAN, the annual average CO2 mole fraction in 2020 is 1.5 +or- 0.5 ppm higher than the previous year, similar to the trend in the northern hemisphere. The different annual CO2 mole fraction growth rate at HZ indicates that the CO2 mole fraction at Hangzhou may be dominated by local anthropogenic emissions, despite the transport of airmass from the north and southwest.

Epidemiological investigation of bovine viral diarrhea virus infection in swinery

From 2017 to 2020, 1 078 piglet diarrhea samples were collected from 6 pig farms in different districts of Shanghai. Multiple RT-PCR method was used for detection and analysis to study the infection status of bovine viral diarrhea virus (BVDV) in swinery in Shanghai. The results showed that the overall detection rate of BVDV in swinery in Shanghai was 7.14% (77/1 078), and showed an increasing trend year by year. The mixed infection rate of BVDV and other diarrhea pathogens was high, with the highest dual infection rate (65%, 26/40), mainly BVDV/PASTV (61.54%, 16/26). On this basis, the triple infection rate was 25% (10/40), mainly BVDV/PAStV/PKoV (40%, 4/10) infection mode;The quadruple infection rate was 10% (4/40), which was dominated by BVDV/PAStV/PEDV/PSV (50%, 2/4) infection. The BVDV prevalence in swinery was seasonal, and the prevalence in spring (10.36%) and autumn (13.59%) was higher than that in summer (6.8%) and winter (2.66%). The positive rate of BVDV in different pig farms was significantly different by 0-24.07%. In view of the detection rate of diarrhea virus dominated by PEDV in pig farm 2 had been high in recent years, this study further monitored the infection of BVDV in this pig farm, and found that the detection rate of BVDV in this pig farm was increasing year by year from 2017 to 2019, with the highest detection rate in 2019 (8.61%, 42/488);The mixed infection of BVDV and other diarrhea pathogens was also serious, with the dual infection rate of 57.58% (19/32), triple infection rate of 21.21% (7/32), quadruple infection rate of 21.21% (7/32), respectively. This study enriched the epidemic data of BVDV in swinery in Shanghai, and could provide reference for the prevention and control of pig epidemics.

[Impacts of Emission and Meteorological Conditions on Air Pollutants at Various Sites Around the COVID-19 Lockdown in Wuhan].

The random forest algorithm was used to separate the mass concentrations of six air pollutants (SO2, NO2, CO, PM10, PM2.5, and O3) contributed by emissions and meteorological conditions. Their variations for five types of sites including Wuhan's central urban, suburb, industrial, the third ring road traffic, and urban background sites were investigated. The results showed that the values of PM2.5/CO, PM10/CO, and NO2/CO during the lockdown period decreased by 10.8-21.7, 9.34-24.7, and 14.4-22.1 times compared with the period before the lockdown, indicating that the contributions of emissions to PM2.5, PM10, and NO2 were reduced. O3/CO increased by 50.1-61.5 times, implying that the secondary formation increased obviously. The contributions of emissions to various types of pollutants all increased after the lockdown. During the lockdown period, affected by the operation of some uninterrupted industrial processes, PM2.5 concentrations in industrial areas dropped the least (20.5%). Compared with the lockdown period, residential activities, transportation, and industrial production were basically restored after the lockdown, resulting in the alleviation of the reduction in PM2.5 emission-related concentrations. The increase in emission-related O3 concentrations could be associated with the decreased NO and PM2.5 concentrations during the lockdown period. The elevated O3 partially offset the improved air quality brought by the reduced NO2and PM2.5 concentrations. After the lockdown, ρ(O3) related with meteorology at the suburban and urban background sites increased by 16.2 µg·m-3 and 16.1 µg·m-3, respectively, which could be attributed to the increased ambient temperature and decreased relative humidity. The decrease in PM2.5 and increase in O3 concentrations caused by reduced traffic and industrial emissions at the third ring road traffic and central urban regions can provide reference for the current coordinated and precise control of PM2.5 and O3 in subregions.

A social media-based framework for quantifying temporal changes to wildlife viewing intensity: Quantifying changes to wildlife viewing intensity

Documenting how human pressure on wildlife changes over time is important to minimise potential adverse effects through implementing appropriate management and policy actions;however, obtaining objective measures of these changes and their potential impacts is often logistically challenging, particularly in the natural environment. Here, we developed a modular stochastic model that infers the ratio of actual viewing pressure on wildlife in consecutive time periods (years) using social media, as this medium is widespread and easily accessible. Pressure was calculated from the number of times individual animals appeared in social media in pre-defined time windows, accounting for time-dependent variables that influence them (e.g. number of people with access to social media). Formulas for the confidence intervals of viewing pressure ratios were rigorously developed and validated, and corresponding uncertainty was quantified. We applied the developed framework to calculate changes to wildlife viewing pressure on loggerhead sea turtles (Caretta caretta) at Zakynthos island (Greece) before and during the COVID-19 pandemic (2019–2021) based on 2646 social media entries. Our model ensured temporal comparability across years of social media data grouped in time window sizes, by correcting for the interannual increase of social media use. Optimal sizes for these windows were delineated, reducing uncertainty while maintaining high time-scale resolution. The optimal time window was around 7-days during the peak tourist season when more data were available in all three years, and >15 days during the low season. In contrast, raw social media data exhibited clear bias when quantifying changes to viewing pressure, with unknown uncertainty. The framework developed here allows widely-available social media data to be used objectively when quantifying temporal changes to wildlife viewing pressure. Its modularity allowed viewing pressure to be quantified for all data combined, or subsets of data (different groups, situations or locations), and could be applied to any site supporting wildlife exposed to tourism. © 2022 The Author(s)

Spatio-temporal variations of night-time lights at early stages of the COVID-19 epidemic in the United States

The COVID-19 epidemic swept the world and continued to spread. Without effective medical treatments and vaccine during the early stage of the pandemic, local governments in various countries had to lock down cities and adopt non-pharmaceutical interventions (NPIs), such as the stay-at-home order, social distancing, and so on. NPIs against the COVID-19 epidemic have significantly changed socioeconomic activities in cities. However, characteristics and patterns of urban socio-economic activities under this influence are still unclear. Benefiting from the development of earth observation technologies, such large-scale changes in socioeconomic activities are enough to be captured by satellites through remotely sensed night-time lights (NTL). In this study, we selected 20 major cities in the United States including New York, Chicago and Los Angeles to analyze spatio-temporal variations of NTL caused by the lockdown of cities. The first round of COVID-19 epidemic occurred in the United States in mid-March 2020. Since March 2020, American cities have successively issued stay-at-home orders, but there are differences in the time and strictness of policy implementation. Large cities have a higher population density and a higher intensity of social activities, so they are more susceptible to infectious diseases. The diversity of lockdown dates and strictness of lockdowns in cities in the United States are conducive to investigating the spatio-temporal variations of NTL. We acquired monthly averaged NPP VIIRS products of February, March and April, 2020, which are from Visible Infrared Imaging Radiometer Suite (VIIRS) Day/Night Band (DNB) onboard the Suomi National Polar-orbiting Platform (NPP). We further analyzed the spatial pattern, distance decay and disparities in land use types of changes in NTL. Results show that NTL generally dimmed by 5-8% in U.S. cities caused by the lockdown of cities. There are 6 cities where the luminous brightness has dropped by more than 10%: Chicago, Dallas, Denver, Detroit, Minneapolis, and St. Louis. Among them, Minneapolis has the largest decrease in luminous brightness, with a decrease of about 40% in March. The spatial change of NTL shows obvious "core-periphery" pattern that the reduction of NTL declines with the distance from the city center. This is mainly because the central area of the city is a concentrated commercial area. After the closure of the city, commercial activities have dropped significantly, resulting in an obvious reduction in NTL around city centers. The reduction of NTL varies among diverse urban land use types. In New York, NTL decreased the most on land for residence and aviation facilities by 12% and 11%, respectively. In Chicago, NTL generally decreased by 20% in all types of urban land, and NTL recovered after one month of the lockdown of cities in other urban land except sports facilities land. This study only analyzes the spatio-temporal changes of NTL. In the future, it can be combined with multi-source data to explain the driving force of NTL changes. Nighttime light remote sensing effectively reflects urban socio-economic dynamics with an important application in monitoring and assessing socio-economic impacts of emergencies. © 2022 National Remote Sensing Bulletin. All rights reserved.

Unsupervised Contrastive Learning of Image Representations from Ultrasound Videos with Hard Negative Mining

Rich temporal information and variations in viewpoints make video data an attractive choice for learning image representations using unsupervised contrastive learning (UCL) techniques. State-of-the-art (SOTA) contrastive learning techniques consider frames within a video as positives in the embedding space, whereas the frames from other videos are considered negatives. We observe that unlike multiple views of an object in natural scene videos, an Ultrasound (US) video captures different 2D slices of an organ. Hence, there is almost no similarity between the temporally distant frames of even the same US video. In this paper we propose to instead utilize such frames as hard negatives. We advocate mining both intra-video and cross-video negatives in a hardness-sensitive negative mining curriculum in a UCL framework to learn rich image representations. We deploy our framework to learn the representations of Gallbladder (GB) malignancy from US videos. We also construct the first large-scale US video dataset containing 64 videos and 15,800 frames for learning GB representations. We show that the standard ResNet50 backbone trained with our framework improves the accuracy of models pretrained with SOTA UCL techniques as well as supervised pretrained models on ImageNet for the GB malignancy detection task by 2–6%. We further validate the generalizability of our method on a publicly available lung US image dataset of COVID-19 pathologies and show an improvement of 1.5% compared to SOTA. Source code, dataset, and models are available at https://gbc-iitd.github.io/usucl. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

The spatial-temporal distribution and transmission route of COVID-19 in Shenzhen in the early stage

Objective: To provide reference for future epidemiological investigation, prevention and control of infectious diseases by analyzing the spatial-temporal distribution and transmission characteristics of corona virus diseases 2019 (COVID- 19) in the early stage in Shenzhen.

A statistical analysis of COVID-19 pandemic based on the temporal evolution of entropy in different countries

Objective: Currently the Covid-19 pandemic is studied with great expectations by several epidemiological models with the aim of predicting the future behaviour of the pandemic. Determining the level of disorder in the pandemic can give us insight into the societal reactions to the pandemic the socio-economic structures and health systems in different countries.

Rapid global spread of the SARS-CoV-2 Delta (B.1.617.2) variant: spatiotemporal variation and public health impact

The COVID-19 pandemic has already affected human society for more than 1.5 years. As of August 8, 2021, this pandemic had caused more than 203 million infected and 4.3 million deaths worldwide. As an RNA virus, SARS-CoV-2 is prone to genetic evolution, thus resulting in development of mutations over time. Numerous variants of SARS-CoV-2 have been described globally, four of which are considered variants of concern (VOCs) by the WHO: Alpha (B.1.1.7), Beta (B.1.351), Gamma (P1) and Delta (B.1.617.2). The Delta VOC was first reported in India in December of 2020 and has since affected approximately 130 different countries and regions. Herein, the spatiotemporal spread of the Delta VOC during April to July 2021 in 20 selected countries with available data were analyzed. The prevalence of the Delta VOC sequences was maintained at low levels in the beginning of April, increased rapidly in the following 3 months and is now becoming the predominant viral strain in most regions of the world. We also discuss the effects of the Delta VOC on transmissibility, clinical severity and vaccine effectiveness according to the latest data. The Delta VOC has greater transmissibility and risk of hospitalization than the ancestral SARS-CoV-2 strains and the other three VOCs. The Delta VOC places partially or unvaccinated sub-populations at high risk. Currently authorized vaccines, regardless of vaccine type, still have reliable effectiveness against symptomatic infections and hospitalizations due to the Delta VOC.

Epidemic characteristics and spatial analysis of COVID-19 in Zhejiang Province. (Special Issue: Spatio-temporal dynamics analysis, risk assessment and emergency management for COVID-19 epidemic.) [Chinese]

Objective: To explore the temporal and spatial distribution characteristics of confirmed cases of coronavirus disease(COVID-19)in Zhejiang Province and to determine the correlation between number of confirmed cases and geographical demographic factors, so as to provide theoretical basis for the prevention and control of COVID-19.

Significant reduction in atmospheric organic and elemental carbon in PM2.5 in 2+26 cities in northern China.

To better understand the change characteristics and reduction in organic carbon (OC) and elemental carbon (EC) in particulate matter (PM) with a diameter of ≤2.5 µm (PM2.5) driven by the most stringent clean air policies and pandemic-related lockdown measures in China, a comprehensive field campaign was performed to measure the carbonaceous components in PM2.5 on an hourly basis via harmonized analytical methods in the Beijing-Tianjin-Hebei and its surrounding region (including 2 + 26 cities) from January 1 to December 31, 2020. The results indicated that the annual average concentrations of OC and EC reached as low as 6.6 ± 5.7 and 1.8 ± 1.9 µg/m3, respectively, lower than those obtained in previous studies, which could be attributed to the effectiveness of the Clean Air Action Plan and the impact of the COVID-19-related lockdown measures implemented in China. Marked seasonal and diurnal variations in OC and EC were observed in the 2 + 26 cities. Significant correlations (p < 0.001) between OC and EC were found. The annual average secondary OC levels level ranged from 1.8-5.4 µg/m3, accounting for 37.7-73.0% of the OC concentration in the 2 + 26 cities estimated with the minimum R squared method. Based on Interagency Monitoring of Protected Visual Environments (IMPROVE) algorithms, the light extinction contribution of carbonaceous PM to the total amount reached 21.1% and 26.0% on average, suggesting that carbonaceous PM played a less important role in visibility impairment than did the other chemical components in PM2.5. This study is expected to provide an important real-time dataset and in-depth analysis of the significant reduction in OC and EC in PM2.5 driven by both the Clean Air Action Plan and COVID-19-related lockdown policies over the past few years, which could represent an insightful comparative case study for other developing countries/regions facing similar carbonaceous PM pollution.

Special Issue: Spatio-temporal dynamics analysis, risk assessment and emergency management for COVID-19 epidemic. (Special Issue: Spatio-temporal dynamics analysis, risk assessment and emergency management for COVID-19 epidemic.) [Chinese]

This special issue contains 20 papers focusing on the epidemiological and spatio-temporal dynamic characteristics of COVID-19, the clinical characteristics and risk factors in COVID-19 patients, and the diagnosis and emergency management of COVID-19 infections.

Spatio-temporal variation of Covid-19 health outcomes in India using deep learning based models.

Deep learning methods have become the state of the art for spatio-temporal predictive analysis in a wide range of fields, including environmental management, public health, urban planning, pollution monitoring, and so on. Despite the fact that a variety of powerful deep learning-based models can address various problem-specific issues in different research domain, it has been found that no single optimal model can outperform everywhere. Now, in the last two years, various deep learning-based studies have provided a variety of best-performing techniques for predicting COVID-19 health outcomes. In this context, this study attempts to perform a case study that investigates the spatio-temporal variation in the performance of deep-learning-based methods for predicting COVID-19 health outcomes in India. Various widely applied deep learning models namely CNN (convolutional neural network), RNN (recurrent neural network), Vanilla LSTM (long short-term memory), LSTM Autoencoder, and Bidirectional LSTM are considered to investigate their spatio-temporal performance variation. The effectiveness of the models is assessed using various metrics based on COVID-19 mortality time-series from 36 states and union territories of India.