Data-Driven Multi-Scale Integration of Transportation Networks and Greenhouse Gas Emissions for Landscape Infrastructure Planning

Abstract

This study introduces a novel multi-scale framework for analyzing transportation-related greenhouse gas emissions across Illinois from a landscape infrastructure perspective. By integrating county-level, census tract, and high-resolution (30 × 30 m) data of road, rail, and aviation emissions, the research reveals distinct landscape corridor effects where major transportation routes create linear zones of elevated rural emissions connecting urban centers. While urban areas demonstrate higher total emissions, rural regions show higher tract-level and per capita emission intensities along transportation corridors. The extensive available land in rural areas, particularly along these high-emission corridors, presents significant opportunities for strategic green infrastructure deployment and mitigation. The study establishes a technical foundation for data-driven green infrastructure planning by identifying optimal locations for landscape interventions based on emission patterns. This research advances the integration of transportation and landscape planning by providing actionable insights for policymakers and designers working to mitigate climate impacts through strategic green infrastructure and nature-based solutions.

1. Introduction

Meeting climate targets demands aggressive cuts in greenhouse gas emissions across sectors. The transformation of transportation emissions is particularly crucial to achieving these reductions. Transportation stands as the second-largest source of U.S. greenhouse gas emissions [1]. Driven by an entrenched system of fossil fuel-powered vehicles and sprawling land use infrastructure, car and truck dependence has cemented itself into American life [2]. Logistically, Illinois plays an important role in this network, with approximately 30% of the nation’s freight originating, terminating, or passing through this region [3]. This makes the Illinois transportation system an essential component for navigating a transition towards a climate-conscious transportation system [4].

The dominance of highways, rail networks, and intermodal freight systems in structuring American land use patterns reflects historic priorities of efficiency in economic expansion, often at the expense of environmental sustainability [5]. Transportation infrastructure influences both spatial growth patterns [6] and economic activities [7,8] while creating long-term emission “lock-in” effects (high-emission assets dominate despite the low emission alternatives) [9] due mainly to the difficulty in transitioning into such a large and monopolistic system. This structural inertia presents significant challenges in decarbonizing the sector [4].

Addressing these challenges requires an integrated approach that not only improves vehicle efficiency but also rethinks infrastructure design and spatial planning to incorporate flexible, dynamic, and sustainable low-carbon alternatives [5]. This type of argument does not focus solely on either green or gray infrastructure but on the synergies of both and the need for this synergy to significantly reduce the impact of climate change [10].

The interplay between hard (gray) infrastructure and landscape (green) infrastructure is a dynamic area of academic research that continues to evolve. While scholars have examined this relationship through historical and theoretical perspectives [11,12], much of the discourse remains conceptual. Waldheim and Berger demonstrate this through their analysis of logistics infrastructure, showing how distribution networks and transportation systems shape urban landscapes across multiple scales [13]. Similarly, Bélanger explores the connection through case studies, describing landscapes that function as infrastructure while supporting urban economies and ecological processes [5]. However, these explorations, while valuable, primarily remain at the theoretical level and rely heavily on individual case studies.

Current research on transportation emission mitigation primarily focuses on three strategies: (1) improving public transportation systems, (2) implementing economics-based incentives and disincentives to reduce single vehicle use, and (3) advancing technological solutions such as electrification and alternative fuels [14,15,16]. These are proven effective urban strategies toward transportation emissions reductions; they are typically, however, geographically bound and fail to capture larger, system-wide impacts [17]. Moreover, traditional emissions assessments tend to operate at a single spatial scale, limiting their ability to inform policy decisions that require a multi-scalar perspective [17]. Multi-scale approaches provide a more comprehensive assessment of urban sustainability by integrating local, regional, and national-level interventions, thereby capturing the complex interactions between different governance levels and infrastructure networks [18].

Extensive research highlights the potential of vegetation as a nature-based solution for climate mitigation, typically emphasizing its role in carbon sequestration [19,20]. Studies have demonstrated that green spaces can significantly contribute to emissions reduction, while multi-scale assessments reveal the broader sequestration potential of urban vegetation at local and state-wide levels [21,22]. Various studies have also been conducted to investigate the benefits of green infrastructure through the lens of noise pollution [23], energy consumption reduction [24], and climate change mitigation [25].

This study addresses a critical gap in current research by developing a novel multi-scale framework that assesses transportation emissions while incorporating landscape-based solutions. By analyzing emission patterns across multiple scales—from state-wide networks to local street configurations—this research seeks to answer the following questions: How can landscape infrastructure be strategically deployed to mitigate transportation emissions? Where are the optimal locations for green infrastructure interventions based on emission patterns? How do transportation emission patterns vary across urban and rural landscapes, and what implications does this have for mitigation strategies? This integrated approach represents a significant departure from conventional transportation planning by explicitly considering landscape as infrastructure and identifying opportunities for strategic green infrastructure deployment. The findings provide actionable insights for policymakers and planners working to create more sustainable transportation systems through landscape-based interventions.

2. Methods

This study implements a comprehensive framework to quantify greenhouse gas (GHG) emissions across the Illinois transportation network, incorporating on-road vehicles, passenger rail, freight rail, and aviation sectors. The methodology adopts 2006 IPCC guidelines [26], while integrating region-specific, multi-scale data to ensure reliable emission calculations at a range of resolutions for potential landscape design interventions.

2.1. Study Area: Illinois

The state of Illinois serves as an ideal case study due to its strategic role as a national transportation hub, characterized by an extensive network of highways, rail systems, and intermodal logistics centers. Key state transportation statistics include:

• Road: Illinois contains approximately 146,000 miles of roads (3rd nationally) with over 10,000 bridges [27].
• Rail Network: The state’s rail infrastructure includes over 7000 rail miles (2nd nationally) serving both freight (49 freight lines) and passengers (+1.7 million) [27].
• Air and Water: Illinois maintains 17 major airports and 102 public-use landing facilities. There are 3 major water ports and over 1100 miles of navigable waterways across the state [27].
• Economics: The Transportation and Utilities sector is one of the largest employment categories in Illinois, accounting for approximately 20% of total jobs [28].

The Illinois transportation sector also generates substantial environmental impacts. It accounts for nearly a quarter of the state’s total GHG emissions, primarily from on-road vehicles, freight transport, and aviation [3]. Without intervention, these emissions are expected to grow, exacerbating climate risks and air quality concerns.

2.2. Data Sources

The analysis draws from multiple authoritative data sources to ensure comprehensive coverage of transportation activities (see

Table 1. Data sources and parameters used for estimating transportation emissions by mode in Illinois.

Transportation Mode	Data Type	Description	Source
On-road Transportation	Vehicle Miles Traveled (VMT)	Annual VMT by fuel type and vehicle type	Illinois Department of Transportation (IDOT)
	Fuel Economy	Fuel economy by vehicle and fuel type (miles/gallon)	U.S. Department of Energy
Passenger Rail	Passenger Rail Mileage	Mileage data for passenger rail operations	Bureau of Transportation Statistics, U.S. Department of Transportation
	Fuel Economy	Fuel efficiency (ton-miles per gallon)	U.S Department of Transportation
	Typical Train Weight	Standard weight for passenger trains	U.S Department of Transportation
Freight Rail	Freight Rail Mileage	Mileage data for freight rail operations	Bureau of Transportation Statistics, U.S. Department of Transportation
	Total Weight	Total freight rail weight	Association of American of Railroads
	Fuel Economy	Fuel efficiency (ton-miles per gallon)	Bureau of Transportation Statistics, U.S. Department of Transportation
Aviation	Sales of Aviation Gasoline	Daily sales (thousand gallons)	U.S. Energy Information Administration (EIA)
	Sales of Jet Fuel	Daily sales (thousand gallons)	U.S. Energy Information Administration (EIA)
	Airport Operation Data	Operational data for Illinois airports	Illinois Department of Transportation (IDOT)
Emission Factors	Emission Factors	CO2, CH4, N2O Emission factors	U.S. Environmental Protection Agency (EPA)

). For on-road transportation, we utilized Vehicle Miles Traveled (VMT) data from the Illinois Department of Transportation (IDOT) for 2021 [29], complemented by fuel economy data from the U.S. Department of Energy [30]. Rail transportation data were sourced from the Bureau of Transportation Statistics [31] with supplementary operational parameters from industry reports [32]. Aviation data incorporated fuel sales statistics from the U.S. Energy Information Administration [33,34] and operational data from IDOT’s Public Use and Publicly Owned Airport Inventory Report [35]. All emission factors were derived from the U.S. Environmental Protection Agency’s 2018 standards [36].

2.3. Emission Calculations

The study implements mode-specific calculation methodologies to account for the distinct operational characteristics of each transportation sector. With the availability of detailed mileage data, the IPCC 2006 Guideline tier 3 calculation for on-road emission could be used [26,37]. For passenger rail, freight rail, and aviation sectors, the calculations are based on the IPCC 2006 Guideline tier 1 calculation with adaptation on the data variations [26].

On-Road Transportation. On-road emissions were calculated using a modified IPCC tier 3 approach. The total emissions ($E_{\mathrm{total}}$) were computed using the formula

$$E_{\mathrm{total}}=\sum _{i=1}^n\left({\displaystyle \frac{{\mathrm{VMT}}_i\times {\mathrm{EF}}_i}{{\mathrm{MPG}}_i}}\right)$$

(1)

where ${\mathrm{VMT}}_i$ represents Vehicle Miles Traveled for vehicle type $i$, ${\mathrm{EF}}_i$ denotes the emission factor, and ${\mathrm{MPG}}_i$ represents the average fuel economy for each vehicle type. This calculation incorporates CO₂, CH₄, and N₂O emissions, converted to CO₂-equivalent (CO₂e) using U.S. EPA emission factors.

Rail Transportation. Rail emissions were calculated separately for passenger and freight operations. For passenger rail, the calculation incorporated a train weight of 646.392 metric tons and fuel economy of 277 ton-miles per gallon [31], using the formula

$$E_{\mathrm{total}}=\left({\displaystyle \frac{\mathrm{AM}}{\mathrm{FE}}}\right)\times \mathrm{TW}\times \mathrm{EF}$$

(2)

where AM is the annual mileage, FE is the fuel economy, TW is the train weight, and EF is the emission factor. This formula estimates total emissions by accounting for distance traveled, fuel efficiency, and pollutant intensity.

Freight rail calculations utilized a similar approach but incorporated different operational parameters, including a fuel efficiency of 498 miles per gallon of diesel [38] and total freight weight of 118.9 million tons [39].

Aviation. Aviation emissions were calculated by combining aviation gasoline and jet fuel consumption data. The spatial distribution of emissions was weighted according to airport operations using the formula

$$E_{\mathrm{total}}=\mathrm{AFS}\times \mathrm{EF}\times \left({\displaystyle \frac{\mathrm{CO}}{\mathrm{TSO}}}\right)$$

(3)

where AFS represents the annual fuel sales (e.g., jet or aviation fuel in gallons), EF is the emission factor (e.g., kg CO₂/gallon), CO denotes the number of airport operations in the county, and TSO refers to the total number of airport operations across the state. This equation estimates the county-level aviation emissions by proportionally allocating state-level fuel emissions based on the county’s share of total operations.

2.4. Data Integration and Multi-Scale Analysis Framework

The integration of multiple data sources required careful attention to unit consistency and spatial resolution. All calculations were performed using the R programming language, ensuring consistent data manipulation and analysis across transportation modes. County-level emissions were aggregated using operational weighting factors specific to each transportation mode. The analysis assumes that 2021 data represent typical operational patterns, though this period may have been influenced by external factors. The spatial distribution of emissions was modeled within defined administrative boundaries, recognizing that these boundaries may not fully capture the dynamic nature of transportation networks.

2.4.1. County Level

County-level transportation emissions data were sourced from a pre-aligned dataset. Road network data, including Annual Average Daily Traffic (AADT) provided by the Illinois Department of Transportation [40], were integrated to represent transportation activity. Tract-level population data were extracted from the U.S. Census Bureau to reflect demographic distribution [41]. Road segments were attributed with AADT and calculated road lengths. Total AADT-weighted road lengths (AADT × Length) were computed for each road segment.

2.4.2. Tract Level

To accurately allocate transportation-related emissions from the county level to census tracts, a geospatial optimization-based approach was implemented. The methodology integrates road network data, population data, and county-level emissions to produce tract-level emission estimates, as detailed below:

A spatial intersection operation was performed to assign road segments to their corresponding tracts. Emissions were initially prorated to tracts based on the proportion of road segment lengths within each tract. Two primary weights shown in

Table 2. Numerical weights used in tract-level emission allocation.

Weight Type	Description	Range	Mean
Traffic Weight ($W_{\mathrm{traffic}}$)	Proportion of AADT-weighted road lengths in each tract relative to county total	1.09 × 10−6–0.004388	3.07 × 10−4
Population Weight ($W_{\mathrm{population}}$)	Proportion of tract population relative to county population	0–9.57 × 10−4	3.06 × 10−4
Alpha Parameter ($\alpha$)	Optimization parameter balancing traffic and population weights	6.47 × 10−9–1	0.3173
Combined Weight ($W_{\mathrm{combined}}$)	Final weighting factor used for emission allocation	1.48 × 10−5–0.7106	0.03132

were then calculated for each tract: Traffic Weight, based on the total AADT-weighted road lengths in each tract normalized by the county total, and Population Weight, proportional to the tract population relative to the county population. A combined weight was formulated as

$$W_{\mathrm{combined}}=\alpha \cdot W_{\mathrm{traffic}}+\left(1-\alpha \right)\cdot W_{\mathrm{population}}$$

(4)

where α is an adjustable parameter optimized to balance the influence of traffic and population. County-level emissions were distributed to tracts using the combined weights, with the formula

$$E\left(i\right)=E_{\mathrm{county}}\times W_{\mathrm{combined}}\left(i\right)$$

(5)

representing the emission assigned to tract and $E_{\mathrm{county}}$ is the total emission for the county.

The parameter α was optimized by minimizing the difference between estimated and observed emissions at the tract level using the Brent optimization method. Rescaling weights ensured that it summed to 1 within each county, preserving consistency in total emissions. This comprehensive weighting and optimization approach ensured a balanced and accurate distribution of transportation-related emissions across tracts, reflecting both traffic intensity and population density.

2.4.3. Street Level

To provide a high-resolution spatial representation of street-level carbon emissions, emissions data were distributed across a 30 × 30 m raster grid using geospatial and optimization techniques. This approach allowed for granular insights into emission hotspots and ensured consistency with road network and population data.

Road network data, including Annual Average Daily Traffic (AADT), road geometries, and county population data, were preprocessed to align with the study area’s spatial framework. AADT data are collected through IDOT Traffic Monitoring Program using NuMetrics Hi-Star magnetic lane counters to collect AADT data. These devices record traffic volumes and classify vehicles into three length categories (passenger vehicles, single-unit trucks, and multi-unit trucks) with 3–5% accuracy during 24 h weekday counts, which are then seasonally adjusted [42]. Emissions for each road segment were calculated as the product of AADT and road length, creating a traffic-intensity-weighted measure of emissions. These emissions were allocated proportionally to grid cells based on the intersection of road segments with the 30 × 30 m grid. For cells intersected by multiple road segments, emissions were aggregated. A validation step confirmed that the total emissions across all grid cells matched the original road network totals.

Optimization. To refine street-level emissions allocations, a combined weighting approach was implemented. The parameter α was introduced to balance the influence of traffic-based $W_{\mathrm{traffic}}$ and population-based $W_{\mathrm{population}}$ weights in determining the combined weight $W_{\mathrm{combined}}$:

$$W_{\mathrm{combined}}=\alpha \cdot W_{\mathrm{traffic}}+\left(1-\alpha \right)\cdot W_{\mathrm{population}}$$

(6)

where $W_{\mathrm{traffic}}$ represents the fraction of total road traffic assigned to a grid cell based on AADT and road length, $W_{\mathrm{population}}$ accounts for the proportion of the county’s population residing in the grid cell, and $\alpha$ determines the relative importance of traffic and population.

The optimization of α used the Brent method, which minimized the error between observed $E_{\mathrm{observed}}$ and estimated emissions $E_{\mathrm{estimated}}$ across grid cells:

$$\mathrm{Error}=\sum _i\left|E_{\mathrm{observed}}-E_{\mathrm{estimated}}\right|$$

(7)

Starting with $\alpha$ = 0.5, the algorithm iteratively adjusted α to achieve the lowest error. Once the optimal $\alpha$ was determined, combined weights were recalculated, and the total emissions were rescaled to maintain consistency with the road network data.

Normalization. To address the substantial variability in emissions intensities, a logarithmic transformation was applied:

$$E_{\mathrm{normalized}}={\displaystyle \frac{\log \left(E+1\right)-\mathrm{Min}}{\mathrm{Max}-\mathrm{Min}}}$$

(8)

This transformation compressed the data range, reducing the influence of outliers while preserving spatial patterns. Normalized emissions were categorized into discrete classes, enabling clear visual differentiation between low- and high-emission areas.

3. Results

This section examines the distribution and concentration of greenhouse gas emissions across the Illinois transportation network, revealing distinct spatial patterns that highlight critical emission hotspots and regional variations. Not surprisingly, emissions vary significantly across the state’s 102 counties and 3265 census tracts, with particularly high concentrations in northeastern Illinois, where major transportation infrastructure converges in the Chicago metro area. Urban centers generate disproportionate emissions across all transportation modes, while rural areas maintain consistently lower emission profiles except places intersected by major transportation corridors. These patterns illustrate the fundamental relationship between infrastructure density, urbanization, population, and transportation-related greenhouse gas emissions.

Road Emissions. Analysis of on-road transportation emissions as shown in Figure 1 reveals distinct spatial hierarchies across Illinois. County-level data demonstrate emission values ranging from 14,427.51 to 14,709,524.95 MTCO2e. The northeastern region, particularly Cook County and its surrounding metropolitan area, exhibits the highest emission levels. Quantitative analysis indicates that the five counties with the highest emissions (Cook, DuPage, Will, Lake, and Kane) collectively account for 43.4% of the state’s total on-road emissions, underscoring the significant concentration of vehicular activity in urbanized areas.

Rail Emissions. Passenger rail emissions in Figure 2 demonstrate a highly concentrated spatial distribution, ranging from 0.024 to 10.563 MTCO2e. Analysis reveals that only 46.1% (47 of 102) of counties report passenger rail emissions, with the highest levels observed in Cook County (10.563 MTCO2e) and its adjacent metropolitan areas. This concentration corresponds directly with the region’s intensive commuter rail operations and major terminal facilities.

Freight rail emissions in Figure 3 present a broader but still facility-centric distribution pattern, with values ranging from 15,016.64 to 2,039,760.29 MTCO2e. The data indicate that major freight hubs, particularly Cook County (2,039,760.29 MTCO2e) and Will County (565,626.78 MTCO2e), generate substantially higher emissions. These two counties constitute 13.5% of Illinois’s total freight rail emissions, reflecting the presence of significant classification yards and intermodal facilities.

Air Transportation Emissions. Aviation emissions demonstrate the most pronounced facility-dependent pattern among all transportation modes, as shown in Figure 4, with values ranging from 5071.96 to 3,560,872.98 MTCO2e. The data indicate that 63.7% of counties (65 of 102) report aviation emissions. Cook County, hosting O’Hare and Midway International Airports, accounts for 3,560,872.98 MTCO2e, representing 34.8% of the state’s total aviation emissions. This distribution pattern emphasizes the highly centralized nature of air transportation infrastructure.

3.1. County-Level Analysis

The county-level total emissions map in Figure 5 reveals a clear hierarchical pattern in transportation emissions across Illinois. The data show five distinct emission categories, ranging from 14,427.51 to 20,310,168.79 MTCO2e. The spatial distribution demonstrates a pronounced concentration in northeastern Illinois, with Cook County recording the highest total emissions (20,310,168.79 MTCO2e), followed by DuPage County (4,492,467.74 MTCO2e) and Will County (4,241,014.49 MTCO2e). These three counties, all located in the northeastern metropolitan region, collectively constitute 36.2% of Illinois’s total transportation emissions.

A distinct north–south gradient emerges in the total emissions pattern, with the highest values concentrated in the northeastern region. This pattern is particularly evident in the total emissions map, where counties in the Chicago metropolitan area fall within the highest emission categories (7,466,779.20–20,310,168.79 MTCO2e). The gradient extends outward from this core area, with emissions generally decreasing with distance from the metropolitan center.

The spatial distribution of aggregate emissions demonstrates a strong correlation with infrastructure networks. Counties intersected by multiple interstate highways or containing major rail junctions show significantly higher emissions. Counties in the highest emission categories (3,378,744.10–7,466,779.19 MTCO2e) typically contain major transportation junctions or urban areas, as evidenced by the convergence of interstate highways and rail lines in the infrastructure maps. Counties with moderate emissions (1,201,331.20–3,378,744.09 MTCO2e) typically contain significant transportation corridors but less dense infrastructure networks.

Rural counties, particularly in southern Illinois, consistently show lower aggregate emissions (14,427.51–1,201,331.19 MTCO2e). However, even within these regions, counties containing regional transportation hubs or intersected by major highways demonstrate relatively higher emissions compared to their neighbors. This pattern underscores the significant influence of transportation infrastructure on emission levels, even in less densely populated areas. The relationship between infrastructure density and emission levels is further supported by both the road system and railroad maps, which show corresponding patterns of network density.

3.2. Tract Level Analysis

Analysis at the census tract level (n = 3265) reveals complex spatial patterns and significant intra-county variations in emission distribution (Figure 6). The highest tract-level transportation emissions can reach 276,796.02 MTCO2e, with emission patterns showing distinct spatial clustering aligned with transportation infrastructure and urban development patterns.

The tract-level emission map demonstrates that high-emission tracts (61,273.23–111,393.88 MTCO2e) are predominantly concentrated in northeastern Illinois, particularly in the Chicago metropolitan area. These areas correspond to locations with dense transportation infrastructure networks, as evidenced by the road system and railroad maps. The highest emission category tracts (>111,393.89 MTCO2e) frequently occur at the intersection of major interstate highways or near significant rail facilities.

The optimized emission allocation analysis reveals varying influences of population and traffic factors. The combined weight factors demonstrate how local demographic and infrastructure characteristics contribute to emission patterns. This relationship is particularly evident in urban areas, where high population density coincides with major transportation corridors.

3.3. Street-Level Analysis

The high-resolution (30 × 30 m) visualization of transportation emissions reveals distinct spatial patterns at the local scale. The analysis in Figure 7 shows concentrated emission points at major intersections and along primary transportation corridors, with a clear transition from high values (red) in urban cores to lower values (yellow to green) in peripheral areas. The point-level emission distribution aligns with urban development patterns, where dense clusters of high-emission points correspond to areas showing elevated tract-level measurements. Along transportation corridors, the visualization displays connected patterns of higher emission points that extend through census tracts, including those in rural areas intersected by major infrastructure. These fine-scale emission patterns illustrate the direct relationship between infrastructure configuration and emission distribution observed in the tract-level spatial analysis.

4. Discussion

The analysis reveals fundamental insights into how transportation infrastructure shapes emission patterns across Illinois. The significant concentration of emissions in northeastern Illinois, where Cook County alone accounts for 2,031,068.45 MTCO2e of emissions, reflects more than just population density—it demonstrates how the convergence of multiple transportation modes creates compounding effects on emission levels. This concentration becomes even more striking when considering that just three counties (Cook, DuPage, and Will) constitute 36.2% of Illinois’s total transportation emissions.

The relationship between infrastructure and emissions manifests differently across urban and rural contexts. In urban areas, particularly the Chicago metropolitan region, emissions reach peak intensities of 276,796.02 MTCO2e at the tract level, largely due to dense infrastructure networks. Meanwhile, rural areas intersected by major transportation corridors show distinct emission patterns, typically ranging from 14,427.51 to 120,133.11 MTCO2e, demonstrating how infrastructure configuration significantly influences emission patterns, even in less populated areas.

4.1. Urban–Rural Dynamics

Analysis of the spatial distribution of transportation emissions across Illinois reveals a complex relationship between urban development and emission patterns that challenges conventional assumptions. The total emissions map shows that while urban counties have higher total emissions, the per capita analysis as shown in Figure 8 at the tract level reveals that urban core tracts consistently exhibit lower emission intensities (averaging 12,390 MTCO2e) compared to their peripheral areas. This pattern is observable across Illinois’s urban centers, as shown in Figure 9, suggesting a systematic relationship between urban form and transportation emissions. The spatial analysis reveals that despite higher population densities, urban core tracts generally display moderate to lower per capita emission levels, indicating more efficient transportation patterns in these areas.

The research further investigates the relationship between transportation-related emissions and population density within the framework of territorial structure—particularly contrasting urban cores and peripheral areas—by conducting a series of statistical analyses. To determine the appropriate correlation method, the normality of the key variables—population density, total transportation emissions, and per capita emissions—was first assessed using the Shapiro–Wilk test (

Table 3. Shapiro–Wilk test results for assessing the normality of key variables used in the density/emissions analysis.

Variable	W Statistic	p Value
Pop. Density	0.52	2.48768459997053 × 10−69
Total Emission	0.65	2.2466710082327 × 10−63
Emission Per Capita	0.05	1.6878781033421 × 10−83

). All three variables significantly deviated from a normal distribution, with W statistics of 0.52 (population density), 0.65 (total emissions), and 0.05 (emission per capita), and p-values less than 0.001 in each case. These results, along with skewed distribution shapes, justified the use of Spearman’s correlation, a non-parametric method suitable for capturing monotonic relationships without requiring normality.

To explore how the association between density and emissions varies across spatial contexts, population density was log-transformed to reduce skewness and grouped into high- and low-density typologies using k-means clustering. Spearman correlation analysis (

Table 4. Spearman correlation coefficients between population density and transportation emissions, stratified by emission type (total vs. per capita) and density class (high vs. low).

Variable	Density	Spearman ρ	p Value
Total Emissions	High	−0.41	4.02000044983973 × 10−103
Total Emissions	Low	−0.27	3.491314742132 × 10−14
Emissions Per Capita	High	−0.70	0
Emissions Per Capita	Low	−0.47	7.90254642393145 × 10−44

) was then conducted within each density group. In high-density tracts, population density demonstrated a strong negative correlation with per capita transportation emissions (ρ = −0.70, p < 0.001) and a moderate negative correlation with total emissions (ρ = −0.41, p < 0.001). In low-density tracts, the associations remained statistically significant but were weaker (ρ = −0.47 and ρ = −0.27, respectively). These findings support the earlier spatial analysis by providing statistical evidence that compact, high-density environments are substantially more efficient in terms of emissions per capita, while the emission-reduction benefits of density decline in lower-density or peripheral regions.

These findings reveal a nuanced relationship between urban form and transportation emissions that extends beyond simple density metrics. The lower emissions observed in urban cores can be attributed to several key factors inherent to dense urban development. First, compact urban forms typically enable shorter trip distances due to the proximity of diverse land uses, reducing the overall vehicle miles traveled. Second, urban cores generally offer multiple transportation mode choices, including public transit, walking, and cycling infrastructure, which can decrease reliance on private vehicles. This multimodal accessibility, combined with higher density development patterns, creates conditions that naturally support lower per capita transportation emissions. In contrast, the higher emissions observed in peripheral and rural areas (averaging 49,847 MTCO2e) can be explained by several structural factors. These areas often serve as crucial transportation links between urban centers, bearing the burden of both local and inter-regional traffic flows. Additionally, lower-density development patterns in these areas typically necessitate longer trip distances and higher car dependency, as destinations are more dispersed and alternative transportation modes are less viable.

This urban–rural interface demonstrates important transitional patterns. While urban areas maintain the highest total emission concentrations, the data show that rural counties intersected by major transportation infrastructure exhibit notably higher emissions compared to their neighbors. Tract-level emissions range from as low as 300 MTCO2e in some areas to as high as 285,734 MTCO2e in others, with a median of 17,239 MTCO2e. The tract-level analysis further illuminates these transition zones, showing how emission patterns diffuse from urban centers into surrounding areas. This gradient effect creates a complex mosaic of emission intensities that reflects the interplay between urban development patterns and rural landscapes.

4.2. Transportation Corridor Effects

The corridor effect map (Figure 10) reveals distinctive patterns in how transportation infrastructure shapes emission patterns across Illinois. Through spatial analysis, distinct linear patterns of elevated emissions emerge along major transportation routes, creating a network of higher emission zones (orange to red) that connect urban centers. These patterns demonstrate how transportation infrastructure significantly influences emission distribution beyond local population density.

The map illustrates two key patterns in how transportation corridors influence emissions. First, major transportation corridors create linear zones of elevated emissions that radiate from urban centers and connect across the state. This is particularly evident in the north–south corridors and the east–west connections that span Illinois. Second, the intersections of these major routes consistently show heightened emissions (darker red), suggesting a multiplicative effect where multiple transportation corridors converge.

The emissions data reveal a hierarchical pattern in the corridor effects based on the map’s legend values. Primary corridors show the highest emission intensities (70,045–127,705 MTCO2e) and secondary corridors display moderate emission levels (19,644–39,397 MTCO2e), while areas away from major transportation infrastructure show notably lower emissions (<19,644 MTCO2e). This hierarchy demonstrates how transportation infrastructure creates distinct zones of influence on emission patterns.

The significance of these corridor effects extends beyond urban areas, as seen in the continuous patterns of elevated emissions along major routes through rural regions. This indicates that transportation corridors function as critical conduits for both regional and inter-regional movement, generating substantial emissions regardless of local population density. Understanding these corridor effects is crucial for developing targeted emission reduction strategies that address not just urban centers, but also the critical infrastructure networks that connect them.

4.3. Policy and Design Implications

Our analysis of transportation-related emissions across Illinois yields important implications for both policy development and design interventions. The spatial patterns and multi-modal interactions we identified suggest the need for integrated approaches that combine climate action planning with strategic landscape infrastructure for large-scale emission mitigation.

At the megalopolitan scale, Li and Phelps identified the importance of internal coordination and network development [43]. This parallels our finding that the three counties accounting for 36.2% of Illinois’s total transportation emissions require coordinated regional action. Climate action plans should adopt a multi-jurisdictional framework that addresses emission patterns transcending municipal boundaries. This is particularly crucial in the Chicago metropolitan area, where the interaction of multiple transportation modes creates complex emission landscapes. The facility-dependent nature of certain emission sources, notably aviation’s 34.8% concentration in Cook County, calls for targeted intervention strategies. Policy frameworks should recognize the distinct characteristics of different transportation modes while promoting integrated solutions. This could include establishing regional transportation authorities with explicit emission reduction mandates or developing inter-county climate action agreements. This aligns with Illinois’ recent climate policy direction, which emphasizes the strategic importance of coordinated planning in the Chicagoland region, where the convergence of seven interstate highways, six Class I railroads, and North America’s largest inland port creates unique challenges for emissions reduction [3].

From a planning and design perspective, our findings support several strategic interventions. Green infrastructure development should prioritize high-emission zones, particularly where multiple transportation modes intersect. The high-resolution analysis showing emission gradients from urban cores to peripheries indicates optimal locations for green buffers and ecological corridors toward the rural areas, where large areas of land for sequestration potential are available. Figure 11 illustrates a practical application of landscape infrastructure planning at a rural road intersection generating 43 MTCO2e/yr across four corridors. Using Champaign’s sequestration rate of 225.45 gCO₂/m²/yr from the Illinois sequestration study [22], approximately 0.02 mi² of green infrastructure strategically placed in four plots (green) would offset these emissions. This demonstration supports our finding that rural transportation corridors, while creating elevated emission zones, simultaneously offer available land for effective carbon sequestration through targeted landscape interventions.

Land use planning emerges as a critical tool for emission reduction. The tract-level analysis showing emissions up to 276,796.02 MTCO2e in certain areas suggests where land use modifications could help reduce transportation demand. This includes promoting transit-oriented development in high-emission areas and establishing green buffer zones around major transportation hubs. Different street and building conditions require different tree species, density and heights to optimize emission reduction [44]. These interventions should be integrated with transportation infrastructure to maximize both functional efficiency and emission reduction potential.

4.4. Assumptions and Limitations

This study faces several significant limitations that should be considered when interpreting its findings. Our reliance on static, annual emission data from 2021 may not capture important temporal variations in transportation patterns, while the spatial allocation method using the α parameter assumes a consistent relationship between population density and traffic volumes that may not hold true across all urban contexts.

This analysis incorporates several key assumptions across transportation sectors to enable comprehensive emissions calculations. For on-road transportation emission calculation, we exclude cold-start emissions and assume average traffic volumes based on road classifications. The rail emission calculations assume constant train weights and uniform fuel efficiency across routes, while aviation calculations assume a direct correlation between operation counts and fuel consumption. These assumptions allow for consistent comparisons across transportation modes while acknowledging potential variations in actual operational patterns. Additionally, the standardized EPA emission factors used in calculations may not fully reflect regional variations in vehicle fleet composition and operational conditions, potentially affecting the accuracy of our emission estimates. Finally, the limited availability of real-time traffic data and actual fuel consumption measurements at the facility level introduces uncertainty into our mode-specific emission calculations.

5. Conclusions

This research reveals distinct corridor effects in transportation-related emissions across Illinois, where major transportation routes create linear zones of elevated emissions connecting urban centers. While urban areas show higher total emissions in tract level (with Cook County generating 20,310,168.79 MTCO2e), rural regions demonstrate higher per capita emission intensities along transportation corridors, averaging 49,847 MTCO2e compared to urban core averages of 12,390 MTCO2e. The extensive available land in rural areas, particularly along these high-emission corridors where emissions range from 61,273.23 to 111,393.88 MTCO2e, presents significant opportunities for carbon sequestration through green infrastructure deployment. This finding provides new evidence supporting potential strategic landscape interventions in rural areas as an effective approach to transportation emission mitigation.

The study’s methodology has several important limitations that affect the interpretation of our results. The reliance on standardized EPA emission factors, which do not account for regional variations in vehicle fleet composition, likely introduces systematic bias in emission estimates. This could particularly impact rural areas, where older vehicle fleets might generate higher emissions than our calculations suggest. Similarly, the use of static 2021 data excludes temporal variations that could significantly alter emission patterns throughout the year. These limitations mean that the quantitative findings should be interpreted as estimates rather than precise measurements, with potential variation across different regions.

For rail calculations, the assumptions about train weights (646.392 metric tons for passenger rail) and fuel economy (277 ton-miles per gallon) represent industry averages that may not reflect the operational reality of Illinois’ 7000 rail miles. This could affect the accuracy of our corridor effect findings, particularly along routes with non-standard rail operations. Future research should incorporate region-specific emission factors, multi-year data series, and direct fuel consumption measurements to enhance calculation accuracy and better inform mitigation strategies.

Acknowledging the limitations, the study’s significance remains in establishing a technical and theoretical foundation for optimizing green infrastructure deployment in gray infrastructure-dominated environments through multi-scale data analysis frameworks. This approach, integrating county-level (14,427.51 to 20,310,168.79 MTCO2e), census tract (300 to 276,796.02 MTCO2e), and high-resolution (30 × 30 m) data, provides valuable guidance for policymakers and planners working to mitigate transportation emissions in Illinois through strategic landscape interventions.

Future research should develop more sophisticated spatial allocation methods that better account for the complex interactions between different transportation modes, integrate emerging transportation technologies into emission assessments, and establish real-time emission-monitoring protocols. These advancements would address the current methodological limitations and significantly improve the accuracy and applicability of findings for climate mitigation planning across diverse urban–rural landscapes.