Kayla Henriques-Dunlap & Amaris Millington ENGL 21003: Writing for the Sciences Professor Kevaughn Hunter
December 18th, 2023
Introduction:
As urban environments across the United States face more urban challenges, and the issue of climate change shows itself as the root cause. Three distinctive studies venture into the professions of environmental science, gentrification, and flood management. Zhu et al. (2022) explored the urban heat island (UHI) phenomenon in Phoenix and utilized a methodology that combined high-resolution land use and land cover (LULC) data, and socioeconomic variables at the U.S. Census block group level, and satellite-derived 70 m Land Surface Temperature (LST) measurements. Utilizing a machine learning approach (Random Forest), the analysis covered the relationship between land surface temperature (LST) and the fractions of LULC, with an emphasis on the role of vegetation features. The findings underscored the effectiveness of trees in reducing surface temperatures during summer daytime, where trees emerged as the most impactful contributors. Contrarily, elevated daytime LST was frequently noted near roads.
Additionally, the study observed higher summer daytime temperatures on land with unmanaged soil compared to the built environment. This comprehensive analysis lays the groundwork for comprehending the intricate dynamics of urban heat island effects and the varied influence of different land features.
Supplementary to Zhu et. Al. (2022), Cucca et al. (2023) embarked on an extensive review of gentrification and assessed 212 peer-reviewed articles. Implementing strong selection criteria, the review focused on English-language, peer-reviewed academic articles, employing specific search strings in titles, abstracts, and keywords. The shift from traditional grey strategies to green programs was emphasized, with a focus on green infrastructure, urban parks, and trees. These changes not only attracted new residents and businesses but also acted as cooling areas
against urban heatwaves and provided flood storage, contributing to overall urban resilience. The transition from classical technical engineering solutions to green programs underscored a commitment to improving economic resilience and individual well-being while restoring ecosystem services. However, the key challenge lied in mitigating the possible trade-offs associated with the shift, specifically the risk of displacing vulnerable populations with historical precedents. This underscored the complex dynamics between green urbanism, climate change adaptation, and socio‐spatial justice.
Lastly, Pedersen et al. (2012) explored flood management in urban environments and developed four scenarios that simulated the co-evolution of settlements. The assessment of societal evolution, measured through combined wealth and the Gini coefficient, shed light on inequality in wealth distribution. Urban areas, specifically coastal regions like Greater Copenhagen, faced an increased risk of flooding due to anthropogenic climate change. This risk was made worse by a combination of natural hazards, which included heavy rainstorms, fluvial flooding from rivers, coastal flooding from storm surges, and rising groundwater levels.
Additionally, the limited conveyance capacity of urban channels and drains, blockages, and outdated drainage The study utilized a risk analysis methodology, using copula functions to assess the interdependence between hazards. Notably, the results for Greater Copenhagen indicate a weak dependence between rainfall and sea surge, and climate change is not expected to significantly increase this correlation.
These three studies, allow for a broader understanding of urban complexities and offer valuable insights for informed decision-making and sustainable urban development. Climate change worsens existing social disparities of at-risk communities in urban areas through
heightened temperatures, extreme weather, and economic vulnerability, posing even more health risks to citizens in these environments.
Methods:
Zhu et al.’s study (2022) utilised three main types of data: high-resolution land use and land cover (LULC) data, socioeconomic variables at the U.S. Census block group level, and satellite-derived 70 m Land Surface Temperature (LST) measurements. The research aimed to understand the spatial distribution of UHI across Phoenix by exploring correlations among LST, LULC fractions, and socioeconomic/demographic variables. Random Forest (RF) algorithms were developed to model LST changes based on variations in LULC compositions. Zhu et al. (2022) focused on the entire extent of Phoenix and its urban core areas, identified by manually portraying regions with high environmental change caused or influenced by people. Scenarios of LST changes were simulated by replacing percentages of unmanaged soil with different vegetation types (trees, shrubs, grass) using RF models to consider government interventions and budget constraints. The data pre-processing involved calculating mean LST from three distinct scenes representing each season and time of day combination to minimise errors. Fractional cover of LULC classes and proximity to buildings and roads within 100 m were computed for each 70-m LST pixel. The research simulated LST changes under different intervention scenarios and assessed the impact on UHI mitigation. Regression analyses were performed to explore the relationship between household variables and LST at the block group level. The RF algorithm was used to determine the most influential factors affecting UHI mitigation effectiveness, quantified using the mean decrease in accuracy (MDA).
Secondly, Cucca et al. (2023) extensively reviewed 212 peer-reviewed articles on gentrification in relation to the creation of green and blue spaces in urban regions. The selection criteria were limited to articles published in English, peer-reviewed academic articles, and specific search strings in titles, abstracts, and keywords. Cucca et al. (2023) used articles published between 1977 and April 2021. The search period extended from July 2021 to August 2021, with the review conducted from October 2021 to February 2022. The PRISMA format guided the overall process, ensuring transparency and adherence to systematic review standards (Moher et al., 2009). The research analysis was organised quantitatively based on full-text assessment and utilised Microsoft Excel to classify and quantify results based on key themes.
These included the year of publication, geography/location of study sites, methodology, typology of interventions, motivation for using NbS, dimensions of justice, assessment of impacts, and policies/tools to avoid green gentrification. The analysis conducted by Cucca et al. (2023) considered the evolution of terms over time, geographical distribution of study sites, methodological insights, types of interventions studied, motivations for NbS implementation, justice dimensions reflected in the articles, assessment of impacts on various factors, and policies/tools employed to counteract green gentrification.
Lastly, Pederson et al. (2012) created four scenarios to explore disparities in flood management in urban environments. Pederson et al. (2012) created four scenarios, simulating the co-evolution of two settlements and evaluating them based on the evolution of crucial state variables. Societal evolution is assessed in terms of combined wealth and the Gini coefficient (Gini, 1912), a widely used indicator of inequality that measures wealth distribution across a population. The parameters for the scenarios include efficiency (ρE) for economic growth and equity considerations related to exposure and vulnerability disparities between planned and
unplanned settlements. Pederson et al. (2012) introduced parameters αH to account for intracity disparities in the vulnerability of communities to flooding, considering physical and social factors such as the quality of housing, existence of
Drainage infrastructure, access to healthcare, and λP to account for the fact that the urban poor tend to exhibit greater risk-taking behaviour concerning location-based risks such as flood risk (e.g., Ezeh et al., 2017; Olthuis et al., 2015; van Voorst, 2015). Flood risk management incorporates technological and green measures, with γE controlling the unit cost of flood protection. Power inequality is modelled through the safety factor for raising flood protection (ϵT), distinguishing between technological and community-based measures. Wealth redistribution is considered in high equity scenarios. The second stage explores model sensitivity to vulnerability (αH) and risk appetite (1/λP) variations, elucidating potential flood risk management strategies within each scenario. This approach allowed for a more complex understanding of the impacts of different parameters on societal dynamics and flood risk outcomes.
Results/Data:
Zhu et al. (2022) results focused on three sections: “Regression Analysis Between LULC. Fractions and LST,” “Distribution of UHI Impacts by Socioeconomic Status,” and the “Projected Effect of Tree, Shrub, and Grass Plantings for UHI Mitigation”.
FIGURE 4 | Relationships between summer daytime LST and LULC fractions in Phoenix, including their content cover of (A) road, (B) building, (C) active cropland,
(D) orchard, (E) fallow, (F) shrub, (G) tree, (H) grass, (I) unmanaged soil, (J) permanent river,
(K) canal, (L) seasonal river, and (M) swimming pool, determined by linear regression.
FIGURE 5 | Relationship between summer night-time LST and selected LULC fractions in Phoenix, including (A) building, (B) road, (C) tree, (D) grass, (E) shrub, and
(F) soils, determined by linear regression.
FIGURE 6 | Relationship between summer daytime LST and (A) proximity to the building, (B) proximity to the road, (C) interaction of tree fraction and proximity to the building, and (D) interaction of tree fraction and proximity to the road, determined by linear regression.
Section 1.1 – During summer daytime, all Land Use and Land Cover (LULC) fractions exhibited significant correlations with surface Land Surface Temperature (LST) (Figure 4). Road fraction had the most substantial association, followed by unmanaged soil, while shrub fractions showed a weak positive association. Tree and grass fractions had the strongest Urban Heat Island (UHI) mitigating effect, lowering daytime temperatures. Other LULC classes involving vegetation or water, such as a lake, cropland, fallow, orchard, canal, and pool, are also negatively associated with summer daytime LST.
For summer night-time, selected LULC classes (building, road, tree, grass, shrub, and soils) were analysed, revealing distinct relationships with LST (Figure 5). Road fractions showed a weak positive correlation, while building fractions had a soft cooling effect. Vegetation cover (trees, shrubs, and grass) exhibited significant negative correlations with LST, with trees and grass effectively lowering night-time temperatures. However, shrub fractions had a weaker mitigating influence, and soils contributed to a slight temperature reduction.
Figure 6 illustrates the impact of buildings and roads on summer daytime LST, with vegetation cover and distance moderating these effects. For summer daytime LST, the combined impact of tree fractions and distance to anthropogenic features were stronger predictors than individual LULC fractions alone. The percentage of unmanaged soil was explored based on terrain conditions, revealing that it mainly occurs on flat terrain, with slopes having a minimal role in mitigating the urban warming effect.
Figures 4 and 5 depict the relationships between LULC fractions and summer daytime and night-time temperatures, while Figure 6 illustrates the combined effects of vegetation cover and distance on summer daytime LST.
Section 1.2 – Zhu et al. (2022) reveal strong correlations between the spatial distribution of summer daytime and night-time Land Surface Temperature (LST) and socioeconomic disparities. Socioeconomic indicators such as house property values, median household income, per capita income, and the percentage of households below the poverty level regarding surface temperatures in the urban core during summer were analysed. Lower property values, median household income, and per capita income negatively correlated with summer day and night temperatures. This implies that communities with quieter socioeconomic status experience higher exposure to Urban Heat Island (UHI) effects, while higher-status communities have lower
exposure. Additionally, there was a positive correlation between the percentage of households below the poverty level and LST in both summer day and night, indicating a more robust relationship in economically disadvantaged areas.
Section 1.3 – Land Use and Land Cover (LULC) on Land Surface Temperature (LST) distribution in Phoenix during summer, considering different times of the day. The Relative Importance analysis using Random Forest (RF) indicates that unmanaged soil is the most crucial factor influencing LST across the city. At the same time, shrubs play a significant role in suburban areas. Buildings have a more substantial impact on the urban core. Scenario-based analysis with RF shows that replacing unmanaged soil with trees, shrubs, and grass reduces daytime LST. Trees have the most significant cooling effect in Phoenix’s total extent and the urban core. Night-time LST decreases with the gradual replacement of unmanaged soil by grass, trees, or shrubs, with grass providing the best cooling effect across Phoenix and the urban core. Overall, grass plantings are the most effective in cooling, surpassing trees and shrubs in Phoenix’s total extent and the urban core.
The results section of Cucca et al. (2023) literature analyses indicate a growing number of articles (112 out of 212 studies) identifying a relationship between greening initiatives and gentrification. The unintended effects of nature-based solutions (NbS) and climate change measures fall into four categories: displacement, changing socio-demographics, rising housing prices, and qualitative housing upgrades. Approximately one-third of studies (41) report multiple impacts of green gentrification, with displacement often linked to housing market effects and socio-demographic changes. Some studies focus solely on displacement, while others examine the influence of greening on housing prices without framing it within a displacement context.
Regarding NbS interventions, parks, trees, water, and green facades are associated with
displacement and housing price increases, while tree planting and waterfront developments are more strongly linked to changes in the housing stock. Motivations behind NbS interventions, mainly restoring the natural environment, show associations with rising housing costs. Few articles focus on tools and policies addressing green gentrification, with most analysing community engagement against rising housing costs or displacement and limited discussion of planning tools or housing policy interventions.
The results analysis by Pederson et al. (2012) utilises a time series of high-water levels.
In low-efficiency scenarios, both planned and unplanned settlements experience a long-term decline. Unplanned settlements suffer an initial collapse due to a significant flood event around time t = 20. In the “low efficiency, low equity” scenario, this event causes an immediate collapse. In contrast, in the high equity scenario, collapse results from a secondary flood event shortly after the initial one. The planned settlement survives these events but undergoes a terminal decline in wealth, gradually depleted by successive floods from t = 20 to 40. The levee effect is observed in the planned settlement’s evolution until t = 20, during which flood protection is installed. However, during the non-flooding period (t = 10 to 20), the settlement moves closer to the riverfront, and protection declines. Subsequent flood events cause significant damage, leading to rapid migration away from the floodplain and economic decline.
Both planned and unplanned settlements survive the simulation period in high-efficiency scenarios. The planned settlement shows marginal differences in state variables between low and high equity scenarios, but unplanned settlements exhibit significant variations. In the common equity scenario, the inability to install flood protection and the economic imperative to reside near the riverbank result in frequent flood damage, particularly from t = 20 to 50. After a flood event, the unplanned community temporarily retreats from the floodplain before quickly
returning, driven by the economic incentive to live close to the riverbank. This behavior is interpreted as a temporary migration until floodwaters recede and necessary repairs are made.
Discussion:
The data and results of the aforementioned sources prove that climate change worsens existing social disparities of at-risk communities in urban areas through heightened temperatures, extreme weather, and economic vulnerability. What is provided emphasizes the urgency for comprehensive strategies addressing both climate resilience and social equity. Zhu et al.’s (2022) scenario change of LST changes simulated by replacing percentages of unmanaged soil with different vegetation types (trees, shrubs, grass) using RF models to consider government interventions and budget constraints and its effect on LST reveal strong correlations between the spatial distribution of summer daytime and night-time Land Surface Temperature (LST) and socioeconomic disparities. Lower property values, median household income, and per capita income negatively correlated with summer day and night temperatures, cementing the fact that the issue of climate change worsened living circumstances for those with lower socioeconomic prowess compared to their counterparts.
Moreover, Cucca et al’s. (2023) literature analyses indicate a growing number of articles that identify a relationship between greening initiatives and gentrification. The effects of nature- based solutions (NbS) and climate change measures fell into four categories of displacement, changing socio-demographics, rising housing prices, and qualitative housing upgrades.
Approximately one-third of studies (41) report multiple impacts of green gentrification, with displacement often linked to housing market effects and socio-demographic changes. Some studies focus solely on displacement, while others examine the influence of greening on housing
prices without framing it within a displacement context. Regarding NbS interventions, parks, trees, water, and green facades are associated with displacement and housing price increases, while tree planting and waterfront developments are more strongly linked to changes in the housing stock. Motivations behind NbS interventions, mainly restoring the natural environment, show associations with rising housing costs.
Lastly, Pederson et al. (2012) utilized a time series of high-water levels. In low-efficiency scenarios, both planned and unplanned settlements experience a long-term decline. In the “low efficiency, low equity” scenario, such events caused an immediate collapse. But collapse results from a secondary flood event shortly after the initial one in high equity scenarios. The planned settlement survives these events but undergoes a terminal decline in wealth, gradually depleted by successive floods from t = 20 to 40. The levee effect is observed in the planned settlement’s evolution until t = 20, during which flood protection is installed. However, during the non- flooding period (t = 10 to 20), the settlement moves closer to the riverfront, and protection declines. Subsequent flood events cause significant damage, leading to rapid migration away from the floodplain and economic decline. Both planned and unplanned settlements survive the simulation period in high-efficiency scenarios. The planned settlement shows marginal differences in state variables between low and high equity scenarios, but unplanned settlements exhibit significant variations. In the common equity scenario, the inability to install flood protection and the economic imperative to reside near the riverbank result in frequent flood damage, particularly from t = 20 to 50. After a flood event, the unplanned community temporarily retreats from the floodplain before quickly returning, driven by the economic incentive to live close to the riverbank. This highlights the vulnerability of planned and unplanned settlements and their impact on climate change in low-efficiency scenarios. The
economic decline and frequent flood damage experienced by these communities further underscore the health risks faced by residents in at-risk urban areas.
Conclusion:
All in all, Climate change worsens existing social disparities of at-risk communities in urban areas through heightened temperatures, extreme weather, and economic vulnerability, posing even more health risks to citizens in these environments. Zhu et al. (2022), Cucca et al. (2023), and Pederson et al. (2012) findings paint a picture of the complex dance between environmental, social, and economic factors in urban environments. Without the proper measures by governments to ensure equality and mitigate struggle among underprivileged members of communities facing the issues mentioned in this body of writing, urban communities will have to continue to prepare and adapt for hard decisions because of climate change.
References:
Zhu, Y., Myint, S. W., Schaffer‐Smith, D., Rebecca Logsdon Muenich, Tong, D., & Li, Y. (2022). Formulating Operational Mitigation Options and Examining Intra-Urban Social Inequality Using Evidence-Based Urban Warming Effects. Frontiers in Environmental
Science, 9.
https://doi.org/10.3389/fenvs.2021.795474
Cucca, R., Friesenecker, M., & Thaler, T. (2023). Green Gentrification, Social Justice, and Climate Change in the Literature: Conceptual Origins and Future Directions. Urban
Planning, 8(1).
https://doi.org/10.17645/up.v8i1.6129
Pedersen, A. N., Mikkelsen, P. S., & Arnbjerg-Nielsen, K. (2012). Climate change-induced impacts on urban flood risk influenced by concurrent hazards. Journal of Flood Risk
Management, 5(3), 203–214.
