Scientific Overview & Analytical Framework
Purpose and Scope
This application provides a satellite-driven analytical framework for quantifying urban thermal dynamics, mapping green infrastructure deficits, and prioritizing climate adaptation interventions. By processing multi-temporal Landsat Collection 2 observations, the system generates high-resolution geospatial metrics that support evidence-based urban planning, thermal risk assessment, and strategic greening initiatives.
Data Acquisition and Preprocessing
The engine queries the Planetary Computer STAC API to retrieve cloud-filtered Landsat Collection 2 surface reflectance and thermal infrared data. For each analysis period, the system identifies optimal summer acquisition windows (June–August) and applies a user-defined cloud cover threshold to ensure data quality. A temporal composite is generated using median filtering across the top five clearest scenes, which minimizes atmospheric noise, reduces cloud contamination, and stabilizes surface reflectance values for robust index computation.
Spectral Index Computation
The application calculates standardized remote sensing indices to characterize urban surface properties:
- Land Surface Temperature (LST): Derived from Landsat thermal band 11 using the single-channel algorithm and converted to degrees Celsius.
- Normalized Difference Vegetation Index (NDVI): Quantifies photosynthetic canopy density and vegetation vigor.
- Enhanced Vegetation Index (EVI): Provides improved sensitivity in high-biomass urban forest canopies.
- Normalized Difference Built-up Index (NDBI): Maps impervious surfaces, concrete, and heavy urban fabric.
- Modified Normalized Difference Water Index (MNDWI): Detects surface water bodies using green and shortwave infrared reflectance.
Analytical Framework and Risk Modeling
The core methodology employs a Multi-Criteria Decision Analysis (MCDA) approach to synthesize thermal stress, vegetation deficit, and built-up intensity into a unified priority map. Each index is normalized to a 0–1 range and weighted according to user-defined parameters that reflect local adaptation priorities. A bivariate compound risk model is simultaneously computed to identify zones experiencing concurrent high thermal stress and low vegetation cover.
Risk Stratification
The urban landscape is classified into four thermal-ecological strata using percentile thresholds and vegetation benchmarks:
- Critical Risk (High Heat + Vegetation Deficit): Requires urgent intervention and targeted green infrastructure deployment.
- High Thermal Stress (High Heat + Partial Canopy): Indicates areas where existing vegetation provides limited cooling capacity and requires canopy enhancement.
- Sparse Green Cover (Cooler Temperatures + Low Vegetation): Represents preventive zones where proactive greening can mitigate future heat accumulation.
- Thermally Stable (Cooler Temperatures + Dense Canopy): Identifies well-vegetated areas that provide ecosystem services and should be preserved.
Statistical and Spatial Analytics
The system computes Pearson correlation coefficients to quantify relationships between thermal stress and land cover indices, validating expected inverse relationships between LST and vegetation density, and positive correlations between LST and built-up intensity. Hotspot delineation is performed using the 90th percentile temperature threshold, with spatial extent calculated at 30-meter pixel resolution. A full statistical summary (mean, median, standard deviation, percentiles, and interquartile range) is generated for all primary indices to support reproducible analysis.
Applications and Significance
This framework supports municipal climate resilience planning, urban forestry optimization, and environmental justice assessments by transforming raw satellite observations into actionable spatial intelligence. The modular weighting system allows stakeholders to adjust analytical priorities based on local policy goals, budget constraints, or equity-focused adaptation strategies. Outputs include interactive risk maps, correlation diagnostics, thermal distribution histograms, and exportable analytical reports, enabling transparent, data-driven decision-making for urban heat mitigation and sustainable city development.