Socioeconomic Disparities in the Usage of Urban Opportunities in South Korea During the COVID-19 Pandemic: Using Land Use/Land Cover and Mobile Phone Data

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the cause of coronavirus disease 19 (COVID-19), has resulted in dramatic changes in human lifestyles and the geographic distribution of populations. However, despite the unequal impact of COVID-19 across urban spaces, research on the association between socioeconomic disparities in the usage of various types of urban amenities during the pandemic is limited. Thus, this study utilized mobile phone data and land use/land cover (LULC) data to investigate COVID-19-induced changes in the hot spots of the daytime and nighttime populations of two districts in Seoul, South Korea: Gangnam (a high-income community) and Gangbuk (a low-income community). First, the differences between Gangnam and Gangbuk in the LULC and mobile phone data, before and during the pandemic, were statistically analyzed by age. Second, the areas with significantly increased mobile phone-based populations during COVID-19 were identified using a hot spot analysis method and Welch’s t-test. This study identified that there were significant disparities in the use of green spaces during the pandemic, with a higher percentage of the mobile phone-based population in Gangnam than Gangbuk. Youths and adults in Gangnam were more likely to visit schools and enjoy physical activities in forests and open spaces during the pandemic, whereas there was no such increase in Gangbuk. The findings contribute to the understanding of the impact of COVID-19 on human behaviors and socioeconomic disparities in the quality of urban life.

1. Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the cause of coronavirus disease 19 (COVID-19), has resulted in major changes in human lifestyles and the geographic distribution of populations. Specifically, due to the COVID-19 lockdowns, people started working from home, participating in remote learning, and considering takeout food/food delivery services instead of visiting restaurants. In addition, one of the most noticeable changes was the significant increase in the use of green spaces. For example, in Oslo, Norway, outdoor recreational activities increased by 291% after the COVID-19 outbreak [1]. Since fitness centers and community recreational facilities were closed, green spaces became a viable option for engaging in physical activities. In March 2020, shelter-in-place orders were issued by United States governors and resulted in the closure of schools and recreational facilities [2]. Public health organizations like the American Public Health Association and the Centers for Disease Control and Prevention have emphasized the significance of maintaining physical activity while staying at home during COVID-19, which encourages activities like visiting parks and green spaces [3]. A study also highlighted the increased use of green spaces for physical activities during the COVID-19 pandemic, particularly due to the closure of gyms and other recreational facilities. Even after the reopening of parks post-lockdown, usage often exceeded pre-pandemic levels as people sought outdoor spaces for physical wellbeing [4].

During the COVID-19 pandemic, identifying population hot spots became critical for understanding the transmission of the virus. In this regard, analyzing mobile phone data is a cost-effective way of offering reasonably accurate and detailed information regarding the population distribution over a large geographic area [5]. More recent studies found that mobile phone data are one of the best sources of information on large-scale population behaviors [6,7,8,9,10,11,12]. For example, in Japan, the nation-wide behavior changes during COVID-19 were explored using mobile phone data [13]. In China, the unequal distribution of 15 min PCR COVID-19 test sites was measured based on mobile phone data [14]. In addition, the population movements were analyzed with the COVID-19 virus spread through the outflows of population travel from Wuhan using mobile phone data [15]. Meanwhile, understanding the land use/land cover (LULC) types in population hot spots and the areas with major population changes through mobile phone data can help elucidate the modified lifestyles and use of urban opportunities during the pandemic. However, there is no city-level study that uses such data to examine how the hot spots of daytime and nighttime populations changed due to the pandemic and to determine the composition of LULC in such hot spots. Yet, other than the mobile phone data, multi-source geospatial big data including remote sensing imageries, land cover data, and location-based service data were used to evaluate the disparities in urban green space exposure for 303 cities in China, considering different times of the day [16].

Although some studies have employed mobile phone data to explore the changes in lifestyle patterns, the majority only focused on one LULC type, i.e., green spaces [17,18], rather than assessing urban ecology by considering all LULC types.

Furthermore, many studies that assessed the unequal impact of COVID-19 on the use of urban spaces among different social groups only focused on the use of green spaces and excluded other urban opportunities or infrastructure such as school playgrounds and restaurants [19,20]. Interestingly, a related study revealed that people living in communities with a higher socioeconomic status visited green spaces more frequently during the COVID-19 pandemic through survey questions [21]. While green space inequities have been a long-standing issue that has drawn considerable attention from researchers [22,23,24,25], limited focus has been paid to whether neighborhoods had different types of urban opportunities during the pandemic based on their socioeconomic status. Other than green space, COVID-19 especially exacerbated urban inequalities in different ways, including access to urban infrastructure [26], and we need to examine how people’s behaviors were changed during COVID-19 in different LULC types. The pandemic exacerbated disparities in the use of indoor and outdoor recreational facilities according to the levels of neighborhood disadvantage [27]. Recreational facility use decreased in the neighborhood with high deprivation, but not in the neighborhood with low deprivation.

Therefore, the present study utilizes mobile phone and LULC data to investigate the COVID-19-induced changes in the use of urban spaces in two contrasting areas in terms of socioeconomic status: the Gangnam and Gangbuk Districts in Seoul, South Korea. Gangnam (which has the highest average monthly household income) is located on the south side of the Han River in Seoul, whereas Gangbuk (which has the lowest average monthly household income) is situated on the north side of the Han River. In this study, it is assumed that mobile phone users represent the daytime and nighttime populations of each area because the mobile phone-based population includes different types of floating populations (e.g., tourists, workers, and students) as well as the static population (e.g., local residents).

First, this study examines how the LULC in hot spots of mobile phone users changed before and during COVID-19. Understanding these changes allows us to identify the effect of the pandemic on human behaviors [28,29,30]. Second, this study determines whether there is a significant difference in the change between the high-income (Gangnam) and low-income (Gangbuk) areas by identifying hot spots of mobile phone users by age in these two areas. Gangnam and Gangbuk were chosen because they exhibit stark differences in socioeconomic characteristics compared to other districts in Seoul. This focus facilitates the identification of potential socioeconomic disparities that have been exacerbated due to COVID-19. Further, it allows us to highlight significant socioeconomic disparities that have been exacerbated due to COVID-19 and facilitates the interpretation of data analysis. Emphasizing Gangnam and Gangbuk districts during COVID-19 can offer a strategic and efficient approach to uncovering the most pronounced patterns of inequality. This targeted approach can yield sharper findings than analyzing the entire city. The COVID-19-induced differences in the LULC of the hot spots between Gangnam and Gangbuk can play an important role in understanding the disparities in urban opportunities. Third, this study identifies the areas with significantly increased mobile phone-based populations during COVID-19 and groups them according to their LULC compositions. The findings would provide important insights into the disproportionate impact of COVID-19 on the use of urban spaces among neighborhoods with different socioeconomic characteristics.

The remainder of this study is as follows. Section 2 describes the data used in this study and the methodological approach, while Section 3 reveals the findings. Section 4 discusses how the LULC of the areas that people frequently visited changed during the pandemic, while Section 5 presents the conclusion.

2. Materials and Methods

2.1. Study Area and Data

The study area is the capital city, Seoul in South Korea (see Figure 1). We selected Seoul for this study because Seoul has a variety of urban opportunities and socioeconomically distinct neighborhoods, called Gangnam and Gangbuk. Gangnam is located on the south side of the Han River, whereas Gangbuk is on the north side of the Han River in Seoul. Seocho-gu and Gangnam-gu, which have the highest average monthly household income in Seoul (approximately $3835), were set as the Gangnam area, while Gangbuk-gu and Seongbuk-gu, which have the lowest average monthly household income in Seoul (roughly $2915), were defined as the Gangbuk area [31] (see Figure 2). The taxes collected by the local governments were also the highest in Gangnam-gu ($3615,923,640) and Seocho-gu ($2050,246,818) in 2021, whereas they were the lowest in Gangbuk-gu ($234,923,254) and near the middle in Seongbuk-gu ($448,457,385) [32]. The major age category in Gangnam is 40–49, while in Gangbuk it is 50–59, which is higher than Gangnam (Table 1).

Gangnam is a major business area in Seoul, where a large number of startups and various information technology (IT) companies are located. It is also a hot spot to which many travelers visit, due to its great public transportation, as well as an area with a large daytime and nighttime population. Within this area, Gangnam-gu is recognized as South Korea’s largest affluent village with the highest housing prices and the highest average standard of living. In addition, the financial and service industries have been well developed, and there are many high-rise office buildings. Although Gangbuk-gu includes a relatively low population density, Bukhansan National Park occupies a large portion of the area. In Seongbuk-gu, there is a combination of old 20th-century landscapes and modern high-tech urban landscapes, with many cultural assets and attractions.

This study used two types of datasets: mobile phone and LULC data. The mobile phone data was provided by SK Telecom, one of the largest mobile operators in South Korea. In fact, the total number of SK Telecom users accounts for 44.4% of the population in South Korea [8]. According to a report by Pew Research Center, South Korea had 100% mobile phone ownership and the highest smartphone penetration rate of 95% [34]. The data included the time-weighted monthly average number of mobile phone users in Seoul, which was estimated according to the aggregated locations of the mobile phones as well as other attributes such as gender, age, weekdays/weekend days, and time (see Table 2). Specifically, the averages were calculated by using the monthly average number of people positioned within a 50 m grid and the time that they remained there [8]. For instance, if two people stayed in a specific grid cell for a total of 72 h (3 days) within a 30-day month, then the monthly average value for the cell for that month is two people * 3 days/30 days, or 0.2. Figure 1 presents the mobile phone-based population in April 2020. It should be noted that this study focused on the data in April and August 2019 (before the pandemic) and April and August 2020 (during the pandemic) to determine any differences immediately right after the COVID-19 outbreak (April) and during the following summer (August). Specifically, we can see where people spent their time in the summer break or vacation in August during the COVID-19 pandemic compared to April.

This study used LULC data (released every five years) by the Ministry of Environment in South Korea (see Figure 3). Specifically, we used LULC data from 2020, which initially included 11 LULC types. However, in this study, the LULC types with low percentages (<5%) were merged with similar LULC types with higher percentages, resulting in a total of seven categories. Specifically, Gangnam had higher percentages of industrial and public facilities (10.2%), transportation facilities (10.9%), commercial and business areas (8.1%), mixed residential and business areas (10.3%), and river and lake areas (7.4%), compared to Gangbuk. However, Gangbuk had higher percentages of forest and open spaces (45.0%) and residential areas (29.1%) compared to Gangnam (see Table 3). When we downloaded the LULC data, we selected more grids and a larger coverage than Seoul to avoid the edge effect that might exist near the boundaries of Gangnam and Gangbuk.

2.2. Data Analysis

Hot spots can be used to identify the areas with a significantly increased mobile phone-based population, especially during the COVID-19 pandemic. After subtracting the 2019 population from the 2020 population, only the top 0.5% (z-score: 2.58) of the hot spots were used to identify the differences in LULC types between Gangnam and Gangbuk. Focusing on the top 0.5% of hot spots for the LULC analysis can help identify emerging trends or patterns regarding the use of urban opportunities during the pandemic that may not be apparent when examining larger areas. It can help explore more extreme patterns in the use of urban opportunities that emerged during COVID-19, which better contrast with pre-pandemic trends. Additionally, it highlights areas with the greatest impact, potentially revealing critical insights about the concentration of activity, mobility patterns, and exposure to the pandemic. The Getis-Ord G_i^* statistic [35,36] was also used to identify the increase in the mobile phone-based population in 2020, compared to 2019. The equation of the G_i^* statistic is as follows:

(1)${\mathrm{G}}_{\mathrm{i}}^{\mathrm{*}}={\displaystyle \frac{\sum _{j=1}^n{w}_{ij}{x}_j-\overline{X}{\sum }_{j=1}^n{w}_{ij}}{S\sqrt{\frac{[n\sum _{j=1}^n{w}_{ij}^2-{\left({\sum }_{j=1}^n{w}_{ij}\right)}^2]}{n-1}}}}$

(2)$\overline{X}={\displaystyle \frac{{\sum }_{j=1}^n{x}_j}{n}}$

(3)$S=\sqrt{{\displaystyle \frac{{\sum }_{j=1}^n{x}_j^2}{n}}-{\left(\overline{X}\right)}^2}$

where G_i^* is the statistic that describes the spatial dependency of location i over all n locations, w_ij is the symmetric spatial weight between locations i and j, and x_j is the value of the variable x of location j. The spatial weight was calculated based on the first order Queen’s contiguity method to define adjacent neighbors because Queen’s contiguity provides a more inclusive understanding of spatial relationships regarding hot spots, which considers all adjacent cells, including diagonally touching ones. In the Queen’s contiguity, the spatial weight is set to 1 if the location pair shares either a common edge or a vertex and 0 otherwise. In this study, each location indicates a grid cell in Seoul, while the variable is the time-weighted monthly average mobile phone-based population and is the subtraction of the 2019 population from the 2020 population. Hot spots are calculated over the entire Seoul area; however, we focus only on Gangnam and Gangbuk.

The differences in the percentages of LULC types in hot spots were also statistically analyzed. Welch’s t-test was used in this study since it was assumed that the variances in the percentages of LULC types in the hot spots between Gangnam and Gangbuk were not equal because there were LULC types with unequal variances found by using Levene’s test (p < 0.05). As for Welch’s t-test, it is a two-sample t-test that can determine the mean of the percentage of each LULC type in hot spots. The equation for Welch’s t-test is as follows:

(4)$\mathrm{t}={\displaystyle \frac{{\overline{x}}_1-{\overline{x}}_2}{\sqrt{\frac{s_1^2}{n_1}+\frac{s_2^2}{n_2}}}}$

where x is the mean of the percentage of each LULC type in hot spots, s is the standard deviation, and n is the group size of hot spots. Mobile phone-based populations of different ages, including children (0–14), youths (15–24), adults (25–59), and seniors (60–99), were also considered to identify the hot spots of different population groups. We used R (v.4.2.2, R Foundation for Statistical Computing, Vienna, Austria) programming language for all the analyses.

3. Results

3.1. The Differences Between Gangnam and Gangbuk Based on the Percentages of LULC Types in Hot Spots for Children and Youths

In this study, the differences between Gangnam and Gangbuk according to LULC types in hot spots of the mobile phone-based population during the COVID-19 were statistically analyzed by using Welch’s t-test and considering different age groups. Table 3 presents the statistically significant differences (p < 0.05). Additionally, a positive t-value indicates a larger percentage of a particular LULC type in Gangnam, whereas a negative t-value represents a larger percentage in Gangbuk.

In Table 4, the differences between hot spots in Gangnam and Gangbuk in April can be seen in the percentages of forest and open spaces (t = 3.717, p < 0.001), mixed residential and business areas (t = 6.289, p < 0.001), and residential areas (t = −5.738, p < 0.001) used by the children. Overall, many children in Gangnam spent a significant amount of time in forest and open spaces and mixed residential and business areas, while those in Gangbuk mainly stayed in residential areas. In particular, mixed residential and business areas had the highest t-value (t = 6.289, p < 0.001), whereas residential areas had the lowest (t = −5.738, p < 0.001). Regarding youths, many in Gangnam stayed in mixed residential and business areas (t = 3.953, p < 0.001), whereas those in Gangbuk spent much time in commercial and business areas (t = −2.871, p = 0.006).

In Table 5, in August, the differences between the hot spots in Gangnam and Gangbuk can be seen in the percentages of industrial and public facilities areas (t = 3.671, p < 0.001), forest and open spaces (t = 5.162, p < 0.001), mixed residential and business areas (t = 5.426, p < 0.001), and residential areas (t = −3.850, p < 0.001) used by the children. Overall, many children in Gangnam spent much time in industrial and public facilities areas, forest and open spaces, and mixed residential and business areas, while those in Gangbuk mainly stayed in residential areas. In particular, forest and open spaces and mixed residential and business areas in Gangnam had the highest t-values. Among the youths, industrial and public facilities areas (t = 5.015, p < 0.001), forest and open spaces (t = 3.084, p = 0.002), and mixed residential and business areas (t = 7.990, p < 0.001) had higher percentages in Gangnam than in Gangbuk, whereas residential areas (t = −2.481, p = 0.019) had a higher percentage in Gangbuk. Mixed residential and business areas had the highest t-value. Meanwhile, there were no hot spots in industrial and public facilities areas, forest and open spaces, and mixed residential and business areas in Gangbuk. Figure 4 and Figure 5 show the hot spots of the increased population of children and youths in Gangbuk and Gangnam in April and August during COVID-19.

3.2. The Differences Between Gangnam and Gangbuk Based on the Percentages of LULC Types in Hot Spots for Adults and Seniors

In Table 6, the differences between the hot spots in Gangnam and Gangbuk can be seen in the percentages of forest and open spaces (t = 4.040, p < 0.001), commercial and business areas (t = −2.101, p = 0.038), and mixed residential and business areas (t = 2.102, p = 0.037) used by the adults. Overall, many adults in Gangnam spent much time in forest and open spaces and mixed residential and business areas, while there were no hot spots in forest and open spaces in Gangbuk, since many adults mostly stayed in commercial and business areas. Regarding the seniors in Gangnam, they stayed in transportation facilities areas (t = 2.513, p = 0.012), forest and open spaces (t = 3.144, p = 0.002), and mixed residential and business areas (t = 2.243, p = 0.025), whereas those in Gangbuk spent much time in residential areas (t = −4.180, p < 0.001). Figure 4 and Figure 5 show the hot spots of the children, youths, adults, and seniors in Gangbuk and Gangnam in April, respectively, which were used to identify the differences in LULC. In both Gangbuk and Gangnam, most of the hot spots were located on tall apartment buildings due to the lockdown during COVID-19 in April. Gangbuk showed a few hot spot areas where the tall apartment buildings are located, whereas the hot spots in Gangnam were dispersed because apartment areas are located all over Gangnam.

In August, among the adults, industrial and public facilities areas (t = 6.450, p < 0.001) and forest and open spaces (t = 5.401, p < 0.001) had higher percentages in Gangnam than in Gangbuk, whereas commercial and business areas (t = −2.105, p = 0.042) had a higher percentage in Gangbuk. Industrial and public facilities areas also had the highest t-value, indicating that this type of LULC in Gangnam and Gangbuk had the largest difference among the adults in August (see Table 7). There were also no hot spots in industrial and public facilities areas and forest and open spaces. As for the seniors, many stayed in industrial and public facilities areas (t = 4.171, p < 0.001), transportation facilities areas (t = 3.006, p = 0.003), forest and open spaces (t = 2.905, p = 0.004), and mixed residential and business areas (t = 3.435, p = 0.001) in Gangnam, whereas those in Gangbuk spent much time in residential areas (t = −7.361, p < 0.001), indicating the lowest t-value.

Figure S1 in the supplementary material shows the changes in the t-values among the different age groups in April and August. The t-values for all of the age groups were mostly positive in industrial and public facilities, transportation facilities areas, forest and open spaces, and mixed residential and business areas, while those were mostly negative in commercial and business areas and residential areas. The most noticeable change between April and August was that the t-values in the industrial and public facilities areas increased in August, compared to April.

4. Discussion

This study evaluated the disparities in urban opportunities between socioeconomically different neighborhoods, while considering the differences in age groups. Based on the findings, there were clear disparities in the usage of green spaces between the high- and low-income communities. The results also indicated that, in both April and August, forest and open spaces accounted for a higher percentage in the hot spots of the children, youths, adults, and seniors in Gangnam than in Gangbuk. In other words, more people spent time in green and open spaces in Gangnam, while there was no hot spot of the mobile phone-based population in forest and open spaces in Gangbuk during April and August. This means that more green spaces that people prefer for relaxing or those that have better access were located in Gangnam during the COVID-19 pandemic, even though the percentage of forest and open spaces was higher in Gangbuk (see Table 2). In particular, since Gangnam includes more offices and a better transportation network than Gangbuk, many people working in or visiting Gangnam might have utilized a nearby green space. According to previous studies, low-income people have less access to green space [37], and individuals with low incomes are less likely to secure stable employment, a situation that may have worsened due to COVID-19, resulting in reduced leisure time for them [38]. In addition, one study also found out that people in low-income communities have difficulties accessing green spaces due to the lack of transportation, whereas those in high-income communities have much easier access [39]. Inequalities in access to green spaces have been a major concern in many countries [40,41,42,43]. In related research, during the COVID-19 pandemic, green spaces have played a critical role in preventing mental health problems caused by social distancing and lockdowns [2]. This may worsen the health disparities in low-and high-income communities. In this regard, any disproportionate access to green spaces between low- and high-income neighborhoods during the pandemic must be taken seriously.

Socioeconomic disparities were also identified in industrial and public facilities areas. There is a very small percentage of industrial areas in Gangnam and Gangbuk, and the hot spots were over public facilities, including schools and recreational facilities. In all of the age groups, the population hot spots only had a higher percentage in such facilities in Gangnam during August. In particular, adults had the highest t-value (t = 6.450, p < 0.001). We also found that there were some hot spots in elementary, middle, and high schools in Gangnam, which were more than those in Gangbuk (not present). Moreover, it was revealed that not only the children or youths, but also the adults and seniors, visited schools for purposes other than education. This finding is contrary to the previous study that the COVID-19 pandemic was associated with a 46% and 62% decrease in the number of female and male children at school playgrounds outside of school hours in the U.S. [44]. In addition, children spent 27 min less in playgrounds than they used to before the pandemic in Turkey [45]. Nevertheless, the play and learning value of school playgrounds became important during COVID-19, and adaptive playground design was studied to promote the general wellbeing of students and outside learning space [46].

In South Korea, people mostly used public sports facilities to engage in physical activities, and among the facilities, playgrounds were used the most (57.4%) during the COVID-19 pandemic [47]. This was different from the finding in the U.S. that the popular locations for physical activity in emerging adults during the pandemic period were on roads/sidewalks (87.5% of survey participants) and at/around the home (85.7% of survey participants) [48]. For comparison, the percentage of the use of playgrounds was 47.0% in 2019 before the pandemic [49], but it increased by more than 10% after the COVID-19 outbreak. In this regard, there is a possibility that many people enjoyed physical activities in playgrounds near schools during the pandemic. Meanwhile, the pandemic decreased recreational walking outdoors in the communities with a low socioeconomic status, whereas it increased in the communities with a high socioeconomic status [50]. This indicates that the percentage of industrial and public facilities areas was higher in Gangnam. However, this pattern was only found in August, since playgrounds were open to residents during the summer break. For policy implications, policymakers should promote physical activities, such as educational and social events, in deprived neighborhoods, and provide better accessibility to locations where people in disadvantaged communities can engage in such activities.

One of the interesting findings in this study was the higher percentage of commercial and business areas in Gangbuk in April for the youths and in August for the adults. This result may be related to increased sales of businesses in some areas of Gangbuk and the significant decrease in sales in some areas of Gangnam [51]. According to its press release, there was the largest decrease in sales in the apparel industry and department stores in Gangnam. However, this is contrary to the finding that older adult workers became loyal customers who significantly contributed to the high sales of nearby commercial and business areas during the COVID-19 pandemic [50]. Hence, how the youths and adults were able to contribute to the increased sales in Gangbuk and what specific sectors they contributed to remain unclear.

In April and August, the hot spots of the children and seniors had a higher percentage of mixed residential and business areas in Gangnam, whereas in Gangbuk, the children and seniors had a higher percentage of residential areas. Since these age groups do not typically work in offices, this pattern suggests that many children and seniors in Gangnam and Gangbuk remained at home due to the social distancing and lockdowns. Due to the lockdown, children stayed at home, and parent–child activities at home became important during COVID-19 [52]. Further, older adults were vulnerable to COVID-19, and one of the protective mechanisms was an isolation at home, which led to unforeseen physical, mental, emotional, social, and financial repercussions [53]. Additionally, since Gangnam includes more offices and buildings for residential and commercial purposes than Gangbuk, the mixed residential and business areas showed a higher percentage. Conversely, since Gangbuk has fewer offices and buildings for residential and commercial purposes, a significant difference was shown in residential areas. In August, the only age group that did not have a higher percentage of mixed residential and business areas in Gangnam was the adults, indicating that many still went to work in person and did not remain at home for remote work. Although this is in line with the low percentage of remote work in South Korea before the COVID-19 pandemic, continuous policy efforts have been made to improve the inflexible employment culture. In fact, the utilization of remote work in South Korea was much lower (4.5%) than other countries (U.S.: 43%; France: 29%; England, 27%; and Japan: 13%) before the pandemic [54], and it was most likely to remain low after the outbreak of COVID-19. However, the situation did not improve during the pandemic because the proportion of remote workers (which was less than 0.5% until 2019) only increased to 2.5% in 2020 [55]. Moreover, the percentage of jobs that allowed remote work was relatively low in the manufacturing, wholesale, and retail industries [56]. However, more data are necessary to support the finding that the percentage of remote work was relatively low in South Korea during the pandemic.

The concept of a 15 min city, where all essential services are accessible within a 15 min walk or bike ride, gained new relevance during COVID-19 [57]. The pandemic highlighted the importance of local accessibility and the need for safe and nearby options for working, shopping, and recreation. As people attempted to minimize travel and maintain social distancing, the idea of living in a community where amenities are close at hand became increasingly appealing. Additionally, the focus on public health and outdoor spaces emphasized the value of walkable neighborhoods in supporting both physical and mental wellbeing during lockdowns and beyond [58]. Particularly, researchers revealed disadvantaged neighborhoods for 15 min green and blue space accessibility in Hong Kong [59]. In Italy, regarding the 15 min city concept, researchers found disparities in access to services of general interest including the areas for housing, water and energy supply, public transport, health, and culture [60].

At this point, several limitations in this study should be noted. First, this study focused only on two months (April and August) due to budget constraints for data purchase. However, we believe that the two months that we selected can be representative because in Korea, April is a month with no special holidays and is representative of a month focused on work, while August is representative of a month when people take the most vacations away from work. In addition, we have observed that several previous studies focusing on COVID-19 used short-term data given the time-sensitive nature of the pandemic-related topic, and produced sound and meaningful results [61,62,63,64]. For example, among those studies, Mustafa et al. (2023) conducted analyses of the accessibility and perceptions of urban green space based on survey data from 13 May to 15 June 2020, and the authors found that the size of urban green space had a significant effect on its visitation and usage during the pandemic [61]. However, future research could utilize data from other months during the COVID-19 pandemic when more funds are available to identify the consistency in findings. Moreover, while the data from these two months were sufficient for our analysis, examining longer-term data in the future could help reveal seasonal changes in hot spots and identify disparities in urban opportunities that may occur during specific seasons. Second, while this study focused on the usage of urban opportunities, including green space, future research can evaluate not only the usage but also the accessibility to those amenities by considering transportation options and travel distance from populations (e.g., grid cells or centroids of an administrative unit). This may help further highlight socioeconomic disparities in the quality of urban life. Third, an edge effect could have influenced the results. Because the LULC data this study had access to were made only for Seoul, we could not obtain the data for surrounding areas and thus could not consider the potential influences of LULC near the boundaries of the two districts. Fourth, as the mobile phone data did not provide the exact number of people tracked, we could not explain whether the hot spots indicated people spending more time and/or more people crowded there. We interpreted the patterns in both ways; however, it needs to be clearer with the increased number of people and time usage for future research. Fifth, the SK Telecom mobile phone data do not include sociodemographic information about its users, which makes it challenging to identify data representativeness. However, we believe that SK Telecom mobile phone users are reasonably representative of the general population in South Korea as 44.4% of the population in South Korea uses SK Telecom, which is one of the two major mobile carriers in South Korea, alongside KT Corporation. It is generally considered that the two mobile carriers offer similar quality of service and pricing. Lastly, since this study only explored urban settings in South Korea, people’s modified behaviors in other settings during the pandemic might have differed from the findings of this study. More studies are needed to explore the impact of the pandemic in other settings.

5. Conclusions

This study investigated the changes in the use of urban spaces in Seoul, South Korea, before and during the COVID-19 pandemic, by using mobile phone and LULC data. It also identified how socioeconomic status influenced the changes by comparing two contrasting areas (i.e., Gangnam and Gangbuk) in Seoul in terms of income and the amount of taxes. Additionally, this study contributed to advancing the understanding of the impact of COVID-19 on human behaviors and inequalities in urban opportunities by using geographical methods and mobile phone data.

Overall, this study revealed that disparities in green space usage between high- and low-income communities became greater during the COVID-19 pandemic for all age groups. In particular, there were significant differences in the forest and open spaces among the adults in April and August, with no hot spots of increased population in Gangbuk. There were also socioeconomic disparities in the industrial and public facilities areas in all age groups. For example, people in Gangnam were more likely to visit schools (as an industrial and public facility) and enjoy physical activities, whereas there were no hot spots in industrial and public facilities among the youths and adults in Gangbuk.

The findings of this study shed new light on how the COVID-19 pandemic has exacerbated the socioeconomic disparities in the usage of and variety of urban opportunities between communities with different income levels. This study also proposes the provision of preferable green spaces with better transportation networks in order to relieve the inequalities in usage of green spaces between high- and low-income communities. Moreover, policies that promote health in low-income communities should be developed, which should differ from those in high-income communities. Specifically, future policies should not solely focus on the equal provision of playgrounds to promote physical activities in low-income communities, but they should be more about education and awareness regarding the benefits of physical activities. Furthermore, finding accessible locations in which people in low-income communities can visit for physical activities should be another policy goal.

Author Contributions

Conceptualization, Kangjae Lee; methodology, Kangjae Lee, Yoo Min Park, Yoohyung Joo, Minsoo Joo and Joon Heo; software, Kangjae Lee and Minsoo Joo; validation, Kangjae Lee; formal analysis, Kangjae Lee and Minsoo Joo; investigation, Kangjae Lee; resources, Kangjae Lee; data curation, Kangjae Lee and Minsoo Joo; writing—original draft preparation, Kangjae Lee; writing—review & editing, Kangjae Lee, Yoo Min Park, Yoohyung Joo, Minsoo Joo and Joon Heo; visualization, Kangjae Lee; supervision, Joon Heo; project administration, Joon Heo; funding acquisition, Joon Heo. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

This can be provided upon request.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Gangnam (Gangnam-gu and Seocho-gu) and Gangbuk (Gangbuk-gu and Seongbuk-gu) districts.

Figure 2. Distribution of the mobile phone-based population of Seoul in April 2020 and the locations of subway stations.

Figure 3. LULC in Gangnam and Gangbuk in 2020.

Figure 4. Hot spots (red) of the increased population of all age groups in Gangbuk in April and August during COVID-19.

Figure 5. Hot spots (red) of the increased population of all age groups in Gangnam in April and August during COVID-19.

Table 1

Socioeconomic and demographic characteristics of Gangnam and Gangbuk in August 2020 [33].

		Gangnam	Gangbuk
Age	0–9	71,987 (7.4%)	46,022 (6.0%)
	10–19	104,601 (10.8%)	60,549 (8.1%)
	20–29	130,510 (13.5%)	110,263 (14.7%)
	30–39	147,578 (15.3%)	100,163 (13.3%)
	40–49	177,182 (18.3%)	118,516 (15.8%)
	50–59	142,944 (14.8%)	122,638 (16.3%)
	60–69	105,918 (10.9%)	99.697 (13.3%)
	70–79	61,047 (6.3%)	63,764 (8.5%)
	80–89	21,103 (2.2%)	25,195 (3.4%)
	90–99	4003 (0.4%)	3514 (0.5%)
	Over 100	512 (0.1%)	629 (0.1%)
Sex	Male	462,687 (47.8%)	363,935 (48.5%)
	Female	504,698 (52.2%)	387,015 (51.5%)
Household income		$3835	$2915

Table 2

Description of the mobile phone dataset.

Attribute	Description	Example
X_COORD	Longitude	126.9780
Y_COORD	Latitude	37.5665
GNDR_CD	Gender code (1 or 2)	1
AGE_GR_SCTN_CD	Age group (0–4, 5–9, 10–14, 15–19, 20–24, …, 95–99)	2024
WKDY_CD	Weekday/weekend day code (1, 2, 3, 4, 5, 6, 7)	1
TMST_CD	Time slot code (0, 1, 2, …, 23)	7
FLOW_POP_CNT	Time-weighted monthly average of mobile phone-based population	0.01

Table 3

Types of LULC and percentages in Gangnam and Gangbuk.

LULC Types	Percentage in Gangnam (%)	Percentage in Gangbuk (%)
Industrial and public facilities area	10.2	5.7
Transportation facilities area	10.9	6.1
Forest and open space	33.2	45.0
Commercial and business area	8.1	4.0
Mixed residential and business area	10.3	9.5
Residential area	19.9	29.1
River and lake	7.4	0.6

Table 4

Statistically significant differences in the percentages of LULC types in hot spots among the children and youths in April.

Category	LULC	t	p-Value	Percentages in Gangnam Hot Spots	Percentages in Gangbuk Hot Spots
Children	Industrial and public facilities area	0.291	0.771	8.51 ± 24.52	8.02 ± 23.22
	Transportation facilities area	1.530	0.126	11.91 ± 20.40	9.90 ± 16.45
	Forest and open space	3.717	<0.001	3.22 ± 15.45	0.28 ± 3.32
	Commercial and business area	−0.826	0.409	11.28 ± 27.49	12.95 ± 29.47
	Mixed residential and business area	6.289	<0.001	21.33 ± 37.08	7.39 ± 23.99
	Residential area	−5.738	<0.001	43.73 ± 45.43	61.47 ± 41.47
	River and lake	1.000	0.318	0.02 ± 0.40	0.00 ± 0.00
Youths	Industrial and public facilities area	0.842	0.402	6.08 ± 22.67	3.32 ± 13.51
	Transportation facilities area	−1.024	0.310	14.96 ± 23.62	19.78 ± 23.52
	Forest and open space	1.000	0.320	0.04 ± 0.34	0.00 ± 0.00
	Commercial and business area	−2.871	0.006	16.42 ± 30.71	37.72 ± 39.17
	Mixed residential and business area	3.953	<0.001	23.63 ± 37.82	4.36 ± 17.20
	Residential area	0.496	0.622	38.88 ± 43.69	34.83 ± 39.81
	River and lake	NA	NA	0.00 ± 0.00	0.00 ± 0.00

Bold font: statistically significant.

Table 5

Statistically significant differences in the percentages of LULC types in hot spots among the children and youths in August.

Category	LULC	t	p-Value	Percentages in Gangnam Hot Spots	Percentages in Gangbuk Hot Spots
Children	Industrial and public facilities area	3.671	<0.001	10.12 ± 26.71	3.45 ± 16.63
	Transportation facilities area	0.061	0.952	13.89 ± 22.09	13.77 ± 20.23
	Forest and open space	5.162	<0.001	3.22 ± 15.25	0.25 ± 1.95
	Commercial and business area	−1.971	0.051	15.57 ± 31.21	22.29 ± 34.70
	Mixed residential and business area	5.426	<0.001	21.49 ± 37.25	7.57 ± 23.61
	Residential area	−3.850	<0.001	35.70 ± 43.69	52.67 ± 44.40
	River and lake	1.432	0.153	0.02 ± 0.47	0.00 ± 0.00
Youths	Industrial and public facilities area	5.015	<0.001	9.98 ± 26.25	0.00 ± 0.00
	Transportation facilities area	0.500	0.620	21.26 ± 27.19	18.76 ± 23.65
	Forest and open space	3.084	0.002	3.83 ± 16.39	0.00 ± 0.00
	Commercial and business area	−2.029	0.051	20.31 ± 33.69	36.37 ± 38.91
	Mixed residential and business area	7.990	<0.001	22.04 ± 36.39	0.00 ± 0.00
	Residential area	−2.481	0.019	22.58 ± 38.07	44.87 ± 44.23
	River and lake	1.000	0.319	0.00 ± 0.03	0.00 ± 0.00

Bold font: statistically significant.

Table 6

Statistically significant differences in the percentages of LULC types in hot spots among the adults and seniors in April.

Category	LULC	t	p-Value	Percentages in Gangnam Hot Spots	Percentages in Gangbuk Hot Spots
Adults	Industrial and public facilities area	0.921	0.358	10.83 ± 27.86	7.88 ± 22.31
	Transportation facilities area	0.504	0.615	20.56 ± 27.28	18.95 ± 22.35
	Forest and open space	4.040	<0.001	4.94 ± 18.08	0.00 ± 0.00
	Commercial and business area	−2.101	0.038	22.52 ± 35.71	33.18 ± 38.40
	Mixed residential and business area	2.102	0.037	18.78 ± 35.89	10.53 ± 26.57
	Residential area	−1.391	0.167	22.37 ± 37.61	29.46 ± 38.00
	River and lake	1.000	0.318	0.00 ± 0.02	0.00 ± 0.00
Seniors	Industrial and public facilities area	1.479	0.140	9.95 ± 25.98	6.77 ± 21.15
	Transportation facilities area	2.513	0.012	16.34 ± 23.33	11.602 ± 17.62
	Forest and open space	3.144	0.002	4.27 ± 16.09	0.69 ± 6.54
	Commercial and business area	−0.611	0.542	15.31 ± 30.69	17.06 ± 32.80
	Mixed residential and business area	2.243	0.025	17.80 ± 34.56	11.27 ± 29.34
	Residential area	−4.180	<0.001	36.30 ± 43.63	52.58 ± 42.70
	River and lake	−0.104	0.917	0.04 ± 0.53	0.04 ± 0.47

Bold font: statistically significant.

Table 7

Statistically significant differences in the percentages of LULC types in hot spots among the children and youths in August.

Category	LULC	t	p-Value	Percentages in Gangnam Hot Spots	Percentages in Gangbuk Hot Spots
Adults	Industrial and public facilities area	6.450	<0.001	8.85 ± 24.84	0.00 ± 0.00
	Transportation facilities area	1.012	0.317	21.89 ± 28.07	17.52 ± 23.20
	Forest and open space	5.401	<0.001	5.87 ± 19.70	0.00 ± 0.00
	Commercial and business area	−2.105	0.042	21.72 ± 34.53	36.61 ± 39.14
	Mixed residential and business area	0.896	0.376	19.12 ± 34.76	13.90 ± 31.63
	Residential area	−1.329	0.192	22.24 ± 38.23	31.98 ± 40.29
	River and lake	1.290	0.198	0.31 ± 4.34	0.00 ± 0.00
Seniors	Industrial and public facilities area	4.171	<0.001	7.97 ± 23.11	1.62 ± 10.41
	Transportation facilities area	3.006	0.003	20.17 ± 25.96	14.03 ± 19.46
	Forest and open space	2.905	0.004	5.66 ± 19.68	1.76 ± 10.12
	Commercial and business area	0.697	0.486	21.02 ± 34.32	18.86 ± 33.05
	Mixed residential and business area	3.435	0.001	22.31 ± 36.95	12.26 ± 27.99
	Residential area	−7.361	<0.001	22.84 ± 38.60	51.28 ± 43.92
	River and lake	−0.919	0.359	0.03 ± 0.60	0.19 ± 2.26

Bold font: statistically significant.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijgi13120421/s1, Figure S1: The t-values of the different LULC types among the children, youths, adults, and seniors.

References

1. Venter, Z.S.; Barton, D.N.; Gundersen, V.; Figari, H.; Nowell, M. Urban Nature in a Time of Crisis: Recreational Use of Green Space Increases during the COVID-19 Outbreak in Oslo, Norway. Environ. Res. Lett.; 2020; 15, 104075. [DOI: https://dx.doi.org/10.1088/1748-9326/abb396]

2. Slater, S.J.; Christiana, R.W.; Gustat, J. Recommendations for Keeping Parks and Green Space Accessible for Mental and Physical Health During COVID-19 and Other Pandemics. Prev. Chronic Dis.; 2020; 17, 200204. [DOI: https://dx.doi.org/10.5888/pcd17.200204] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32644919]

3. Centers for Disease Control and Prevention. Visiting Parks and Recreational Facilities. Available online: https://stacks.cdc.gov/view/cdc/88255/cdc_88255_DS1.pdf (accessed on 22 May 2024).

4. Volenec, Z.M.; Abraham, J.O.; Becker, A.D.; Dobson, A.P. Public Parks and the Pandemic: How Park Usage Has Been Affected by COVID-19 Policies. PLoS ONE; 2021; 16, e0251799. [DOI: https://dx.doi.org/10.1371/journal.pone.0251799] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34010353]

5. Deville, P.; Linard, C.; Martin, S.; Gilbert, M.; Stevens, F.R.; Gaughan, A.E.; Blondel, V.D.; Tatem, A.J. Dynamic Population Mapping Using Mobile Phone Data. Proc. Natl. Acad. Sci. USA; 2014; 111, pp. 15888-15893. [DOI: https://dx.doi.org/10.1073/pnas.1408439111]

6. Grantz, K.H.; Meredith, H.R.; Cummings, D.A.T.; Metcalf, C.J.E.; Grenfell, B.T.; Giles, J.R.; Mehta, S.; Solomon, S.; Labrique, A.; Kishore, N. et al. The Use of Mobile Phone Data to Inform Analysis of COVID-19 Pandemic Epidemiology. Nat. Commun.; 2020; 11, 4961. [DOI: https://dx.doi.org/10.1038/s41467-020-18190-5]

7. Wesolowski, A.; Buckee, C.O.; Engø-Monsen, K.; Metcalf, C.J.E. Connecting Mobility to Infectious Diseases: The Promise and Limits of Mobile Phone Data. J. Infect. Dis.; 2016; 214, pp. S414-S420. [DOI: https://dx.doi.org/10.1093/infdis/jiw273]

8. Yun, S.B.; Kim, S.; Ju, S.; Noh, J.; Kim, C.; Wong, M.S.; Heo, J. Analysis of Accessibility to Emergency Rooms by Dynamic Population from Mobile Phone Data: Geography of Social Inequity in South Korea. PLoS ONE; 2020; 15, e0231079. [DOI: https://dx.doi.org/10.1371/journal.pone.0231079]

9. Lim, H.; Kim, S.; Heo, J. Graph Analyses of Phone-Based Origin-Destination Data for Understanding Urban Human Mobility in Seoul, Korea. Proceedings of the 3rd ACM SIGSPATIAL International Workshop on Prediction of Human Mobility; Chicago, IL, USA, 5 November 2019; pp. 62-65.

10. Lee, W.K.; Sohn, S.Y.; Heo, J. Utilizing Mobile Phone-Based Floating Population Data to Measure the Spatial Accessibility to Public Transit. Appl. Geogr.; 2018; 92, pp. 123-130. [DOI: https://dx.doi.org/10.1016/j.apgeog.2018.02.003]

11. Ma, J.W.; Ju, S.; Nguyen, H.M.; Yun, S.B.; Yoon, S.; Heo, J. Spatio-Temporal Factor Analysis of Characterizing Mass Protest Events Using Taxi Trajectory in Seoul, Korea. Proceedings of the 1st ACM SIGSPATIAL Workshop on Analytics for Local Events and News; Redondo Beach, CA, USA, 7 November 2017; pp. 1-7.

12. Yun, S.B.; Hieu, N.M.; Park, S.Y.; Lim, H.; Heo, J. Implementation of Floating Population Analysis for Smart Cities: A Case Study in Songdo Incheon South Korea. Proceedings of the 2nd ACM SIGSPATIAL Workshop on Prediction of Human Mobility; Seattle, WA, USA, 6 November 2018; pp. 32-36.

13. Hara, Y.; Yamaguchi, H. Japanese Travel Behavior Trends and Change under COVID-19 State-of-Emergency Declaration: Nationwide Observation by Mobile Phone Location Data. Transp. Res. Interdiscip. Perspect.; 2021; 9, 100288. [DOI: https://dx.doi.org/10.1016/j.trip.2020.100288]

14. Wang, J.; Kwan, M.-P.; Liu, D.; Peng, X. Assessing the Spatial Distribution of and Inequality in 15-Minute PCR Test Site Accessibility in Beijing and Guangzhou, China. Appl. Geogr.; 2023; 154, 102925. [DOI: https://dx.doi.org/10.1016/j.apgeog.2023.102925]

15. Lu, X.; Tan, J.; Cao, Z.; Xiong, Y.; Qin, S.; Wang, T.; Liu, C.; Huang, S.; Zhang, W.; Marczak, L.B. et al. Mobile Phone-Based Population Flow Data for the COVID-19 Outbreak in Mainland China. Health Data Sci.; 2021; 2021, 9796431. [DOI: https://dx.doi.org/10.34133/2021/9796431] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36405355]

16. Song, Y.; Chen, B.; Ho, H.C.; Kwan, M.-P.; Liu, D.; Wang, F.; Wang, J.; Cai, J.; Li, X.; Xu, Y. et al. Observed Inequality in Urban Greenspace Exposure in China. Environ. Int.; 2021; 156, 106778. [DOI: https://dx.doi.org/10.1016/j.envint.2021.106778] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34425646]

17. Heo, S.; Lim, C.C.; Bell, M.L. Relationships between Local Green Space and Human Mobility Patterns during COVID-19 for Maryland and California, USA. Sustainability; 2020; 12, 9401. [DOI: https://dx.doi.org/10.3390/su12229401]

18. Zhai, W.; Yue, H.; Deng, Y. Examining the Association between Social Characteristics of Urban Green Space Visitors and the Early Outbreak of COVID-19. Appl. Geogr.; 2022; 147, 102768. [DOI: https://dx.doi.org/10.1016/j.apgeog.2022.102768]

19. de Freitas, R.S.G.; Stedefeldt, E. COVID-19 Pandemic Underlines the Need to Build Resilience in Commercial Restaurants’ Food Safety. Food Res. Int.; 2020; 136, 109472. [DOI: https://dx.doi.org/10.1016/j.foodres.2020.109472]

20. Saragih, J.F.B.; Subroto, T.Y.W. Child-Friendly School: Female Students’ Strategies for Equality in Accessibility of School Playground. J. Asian Archit. Build. Eng.; 2022; 22, pp. 2047-2057. [DOI: https://dx.doi.org/10.1080/13467581.2022.2153061]

21. Pipitone, J.M.; Jović, S. Urban Green Equity and COVID-19: Effects on Park Use and Sense of Belonging in New York City. Urban For. Urban Green.; 2021; 65, 127338. [DOI: https://dx.doi.org/10.1016/j.ufug.2021.127338]

22. Liu, D.; Kwan, M.-P.; Kan, Z. Analysis of Urban Green Space Accessibility and Distribution Inequity in the City of Chicago. Urban For. Urban Green.; 2021; 59, 127029. [DOI: https://dx.doi.org/10.1016/j.ufug.2021.127029]

23. Sharifi, F.; Nygaard, A.; Stone, W.M.; Levin, I. Accessing Green Space in Melbourne: Measuring Inequity and Household Mobility. Landsc. Urban Plan.; 2021; 207, 104004. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2020.104004]

24. Shen, Y.; Sun, F.; Che, Y. Public Green Spaces and Human Wellbeing: Mapping the Spatial Inequity and Mismatching Status of Public Green Space in the Central City of Shanghai. Urban For. Urban Green.; 2017; 27, pp. 59-68. [DOI: https://dx.doi.org/10.1016/j.ufug.2017.06.018]

25. Williams, T.G.; Logan, T.M.; Zuo, C.T.; Liberman, K.D.; Guikema, S.D. Parks and Safety: A Comparative Study of Green Space Access and Inequity in Five US Cities. Landsc. Urban Plan.; 2020; 201, 103841. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2020.103841]

26. Boza-Kiss, B.; Pachauri, S.; Zimm, C. Deprivations and Inequities in Cities Viewed Through a Pandemic Lens. Front. Sustain. Cities; 2021; 3, 645914. [DOI: https://dx.doi.org/10.3389/frsc.2021.645914]

27. Lee, S.; Lee, C.; Xu, M.; Li, W.; Ory, M. People Living in Disadvantaged Areas Faced Greater Challenges in Staying Active and Using Recreational Facilities during the COVID-19 Pandemic. Health Place; 2022; 75, 102805. [DOI: https://dx.doi.org/10.1016/j.healthplace.2022.102805]

28. Ashraf, K.; Park, Y.M.; Browning, M.H.E.M.; Wang, J.; Wang, R.; Lee, K. Examining the Joint Effect of Air Pollution and Green Spaces on Stress Levels in South Korea: Using Machine Learning Techniques. Int. J. Digit. Earth; 2024; 17, 2372321. [DOI: https://dx.doi.org/10.1080/17538947.2024.2372321]

29. Karurung, W.S.; Lee, K.; Lee, W. Comparison of Different Green Space Measures and Their Impact on Dementia Cases in South Korea: A Spatial Panel Analysis. ISPRS Int. J. Geo-Inf.; 2024; 13, 126. [DOI: https://dx.doi.org/10.3390/ijgi13040126]

30. Lee, K. Association between Depression and Green Space before and after the COVID-19: Panel Data Evidence in South Korea. Korean J. Geomatics; 2022; 40, pp. 459-467. [DOI: https://dx.doi.org/10.7848/ksgpc.2022.40.6.459]

31. Herald. Monthly Household Income by Districts in Seoul. Available online: http://biz.heraldcorp.com/view.php?ud=20180330000760 (accessed on 22 May 2024).

32. Seoul Metropolitan Government Seoul Metropolitan City Local Tax Collection. Available online: https://data.seoul.go.kr/dataList/177/S/2/datasetView.do# (accessed on 22 May 2024).

33. Seoul. Statistics on the Registered Population of Seoul. Available online: https://data.seoul.go.kr/dataList/10718/S/2/datasetView.do# (accessed on 22 May 2024).

34. KBS World. S. Korea’s Smartphone Penetration Highest in the World at 95%. Available online: https://world.kbs.co.kr/service/news_view.htm?lang=e&Seq_Code=142787#:~:text=According%20to%20a%20report%20published,100%20percent%20mobile%20phone%20ownership (accessed on 22 May 2024).

35. Getis, A.; Ord, J.K. The Analysis of Spatial Association by Use of Distance Statistics. Geogr. Anal.; 2010; 24, pp. 189-206. [DOI: https://dx.doi.org/10.1111/j.1538-4632.1992.tb00261.x]

36. Ord, J.K.; Getis, A. Local Spatial Autocorrelation Statistics: Distributional Issues and an Application. Geogr. Anal.; 2010; 27, pp. 286-306. [DOI: https://dx.doi.org/10.1111/j.1538-4632.1995.tb00912.x]

37. Yu, S.; Zhu, X.; He, Q. An Assessment of Urban Park Access Using House-Level Data in Urban China: Through the Lens of Social Equity. Int. J. Environ. Res. Public Health; 2020; 17, 2349. [DOI: https://dx.doi.org/10.3390/ijerph17072349]

38. Gao, S.; Zhai, W.; Fu, X. Green Space Justice amid COVID-19: Unequal Access to Public Green Space across American Neighborhoods. Front. Public Health; 2023; 11, 1055720. [DOI: https://dx.doi.org/10.3389/fpubh.2023.1055720]

39. Marquet, O.; Aaron Hipp, J.; Alberico, C.; Huang, J.-H.; Fry, D.; Mazak, E.; Lovasi, G.S.; Floyd, M.F. Park Use Preferences and Physical Activity among Ethnic Minority Children in Low-Income Neighborhoods in New York City. Urban For. Urban Green.; 2019; 38, pp. 346-353. [DOI: https://dx.doi.org/10.1016/j.ufug.2019.01.018]

40. Lindsey, G.; Maraj, M.; Kuan, S. Access, Equity, and Urban Greenways: An Exploratory Investigation. Prof. Geogr.; 2001; 53, pp. 332-346. [DOI: https://dx.doi.org/10.1111/0033-0124.00288]

41. Matthew McConnachie, M.; Shackleton, C.M. Public Green Space Inequality in Small Towns in South Africa. Habitat Int.; 2010; 34, pp. 244-248. [DOI: https://dx.doi.org/10.1016/j.habitatint.2009.09.009]

42. Wright Wendel, H.E.; Zarger, R.K.; Mihelcic, J.R. Accessibility and Usability: Green Space Preferences, Perceptions, and Barriers in a Rapidly Urbanizing City in Latin America. Landsc. Urban Plan.; 2012; 107, pp. 272-282. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2012.06.003]

43. Wüstemann, H.; Kalisch, D.; Kolbe, J. Access to Urban Green Space and Environmental Inequalities in Germany. Landsc. Urban Plan.; 2017; 164, pp. 124-131. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2017.04.002]

44. Lanza, K.; Durand, C.P.; Alcazar, M.; Ehlers, S.; Zhang, K.; Kohl, H.W. School Parks as a Community Health Resource: Use of Joint-Use Parks by Children before and during COVID-19 Pandemic. Int. J. Environ. Res. Public Health; 2021; 18, 9237. [DOI: https://dx.doi.org/10.3390/ijerph18179237]

45. Bozkurt, M. Locked-down!: How Children’s Access and Use of Urban Green Spaces and Playgrounds Changed after COVID 19 Outbreak. Adnan Menderes Üniv. Ziraat Fakültesi Derg.; 2021; 18, pp. 303-310. [DOI: https://dx.doi.org/10.25308/aduziraat.1007588]

46. Quinn, A.; Russo, A. Adaptive School Grounds Design in Response to COVID-19: Findings from Six Primary Schools in South East England. Build. Environ.; 2022; 215, 108946. [DOI: https://dx.doi.org/10.1016/j.buildenv.2022.108946]

47. Ministry of Culture, Sports and Tourism. National Living Sports Survey. 2021; Available online: https://policy.nl.go.kr/search/searchDetail.do?rec_key=SH2_PLC20220283455 (accessed on 22 May 2024).

48. Yi, L.; Wang, S.D.; Chu, D.; Ponnada, A.; Intille, S.S.; Dunton, G.F. Examining Whether Physical Activity Location Choices Were Associated with Weekly Physical Activity Maintenance Across 13 Months of the COVID-19 Pandemic in Emerging Adults. J. Phys. Act. Health; 2022; 19, pp. 446-455. [DOI: https://dx.doi.org/10.1123/jpah.2021-0769]

49. Ministry of Culture, Sports and Tourism. National Living Sports Survey. 2019; Available online: https://www.mcst.go.kr/kor/s_policy/dept/deptView.jsp?pSeq=1297&pDataCD=0417000000&pType=07 (accessed on 22 May 2024).

50. Lee, K. Analysis of Changes in Geographical Factors Affecting Sales in Commercial Alleys after COVID-19 Using Machine Learning Techniques. Heliyon; 2022; 8, e10708. [DOI: https://dx.doi.org/10.1016/j.heliyon.2022.e10708]

51. Seoul Metropolitan Government Smart City Policy Bureau. Impact of COVID-19 Varies Greatly by Regions for Living Population and by Industries for Store Sales; Seoul Metropolitan Government Smart City Policy Bureau: Seoul, Republic of Korea, 2020.

52. Lu, J.; Zhang, M.; Yuan, M. Parenting and Parent–Child Home Practice during the COVID-19 Pandemic: A Case in Central China. Sci. Rep.; 2023; 13, 18655. [DOI: https://dx.doi.org/10.1038/s41598-023-45726-8] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37907511]

53. Cocuzzo, B.; Wrench, A.; O’Malley, C. Effects of COVID-19 on Older Adults: Physical, Mental, Emotional, Social, and Financial Problems Seen and Unseen. Cureus; 2022; 14, e29493. [DOI: https://dx.doi.org/10.7759/cureus.29493] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36299954]

54. Ministry of Employment and Labor. Comprehensive Manual for Working from Home. Available online: https://www.moel.go.kr/common/downloadFile.do?file_seq=20201000322&bbs_seq=20200900802&bbs_id=29&file_ext=pdf (accessed on 22 May 2024).

55. Son, Y.-J. Changes in the Way We Work after COVID-19 and Employment Impact Analysis. Employ. Impact Assess. Brief.; 2022; 28, pp. 1-8.

56. Choi, S. Characteristics and Distribution of Teleworkable Jobs Based on Physical Working Conditions. J. Econ. Geogr. Soc. Korea; 2020; 23, pp. 276-291. [DOI: https://dx.doi.org/10.23841/EGSK.2020.23.3.276]

57. Moreno, C.; Allam, Z.; Chabaud, D.; Gall, C.; Pratlong, F. Introducing the “15-Minute City”: Sustainability, Resilience and Place Identity in Future Post-Pandemic Cities. Smart Cities; 2021; 4, pp. 93-111. [DOI: https://dx.doi.org/10.3390/smartcities4010006]

58. Weng, M.; Ding, N.; Li, J.; Jin, X.; Xiao, H.; He, Z.; Su, S. The 15-Minute Walkable Neighborhoods: Measurement, Social Inequalities and Implications for Building Healthy Communities in Urban China. J. Transp. Health; 2019; 13, pp. 259-273. [DOI: https://dx.doi.org/10.1016/j.jth.2019.05.005]

59. Liu, D.; Kwan, M.-P.; Kan, Z.; Wang, J. Toward a Healthy Urban Living Environment: Assessing 15-Minute Green-Blue Space Accessibility. Sustainability; 2022; 14, 16914. [DOI: https://dx.doi.org/10.3390/su142416914]

60. Murgante, B.; Annunziata, A. Application of the 15-Minute City Criteria to a Metropolitan Area: The Case Study of the Metropolitan City of Cagliari, Italy. Proceedings of the Computational Science and Its Applications—ICCSA 2024 Workshops; Hanoi, Vietnam, 1–4 July 2024; Gervasi, O.; Murgante, B.; Garau, C.; Taniar, D.; Rocha, A.M.A.C.; Faginas Lago, M.N. Lecture Notes in Computer Science Springer Nature: Cham, Switzerland, 2024; Volume 14817, pp. 218-233. ISBN 978-3-031-65237-0

61. Mustafa, A.; Kennedy, C.; Lopez, B.; McPhearson, T. Perceived and Geographic Access to Urban Green Spaces in New York City during COVID-19. Cities; 2023; 143, 104572. [DOI: https://dx.doi.org/10.1016/j.cities.2023.104572]

62. Mollalo, A.; Vahedi, B.; Rivera, K.M. GIS-Based Spatial Modeling of COVID-19 Incidence Rate in the Continental United States. Sci. Total Environ.; 2020; 728, 138884. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2020.138884]

63. Martellucci, C.A.; Sah, R.; Rabaan, A.A.; Dhama, K.; Casalone, C.; Arteaga-Livias, K.; Sawano, T.; Ozaki, A.; Bhandari, D.; Higuchi, A. et al. Changes in the Spatial Distribution of COVID-19 Incidence in Italy Using GIS-Based Maps. Ann. Clin. Microbiol. Antimicrob.; 2020; 19, 30. [DOI: https://dx.doi.org/10.1186/s12941-020-00373-z]

64. Nasiri, R.; Akbarpour, S.; Zali, A.R.; Khodakarami, N.; Boochani, M.H.; Noory, A.R.; Soori, H. Spatio-Temporal Analysis of COVID-19 Incidence Rate Using GIS: A Case Study—Tehran Metropolitan, Iran. GeoJournal; 2022; 87, pp. 3291-3305. [DOI: https://dx.doi.org/10.1007/s10708-021-10438-x]