Human-induced vegetation shifts drive global declines in urban vegetation transpiration

Zenodo Description Title Source data for: Human-induced vegetation shifts drive global declines in urban vegetation transpiration Description This repository contains the complete source datasets underlying the manuscript "Human-induced vegetation shifts drive global declines in urban vegetation transpiration." The study develops a hybrid physics–deep learning framework (CNN-PM) to reconstruct and project urban vegetation transpiration (VTu) across 3,560 global cities from 1986 to 2100, and attributes the observed and projected VTu decline to urbanization-driven vegetation structural transformation rather than climatic forcing. Dataset overview The repository comprises 10 thematic dataset packages totaling over 30 individual CSV files, organized by their role in the analysis pipeline: 1. Eddy covariance observations and transpiration partitioning Daily vegetation transpiration (VTu) estimates derived from three independent high-frequency turbulence partitioning methods — Flux Variance Similarity (FVS), Conditional Eddy Covariance (CEC), and Modified Relaxed Eddy Accumulation (MREA) — plus a multi-method fusion product, for 20 globally distributed urban eddy covariance flux tower sites. Each method provides a central estimate and 5th–95th percentile uncertainty bounds in mm/day. 2. Flux footprint and land-use characterization Footprint model summaries (equivalent radii at 60%, 70%, 80%, and 90% source-area contours), baseline land-use composition (impervious, vegetation, water, bare/other), and source-area-resolved land-use fractions for the top-eight sites, along with visual encoding lookup tables for figure reproduction. 3. Model training and hyperparameter optimization Generation-by-generation convergence trajectories of the Genetic Algorithm (GA) optimization for the two sub-CNN models (CNN_c for climatic forcing, CNN_a for anthropogenic forcing) across 300 cross-validation iterations and 70 generations, recording validation loss, RMSE, KGE, learning rate, weight decay, and batch size with 5th–95th percentile envelopes. 4. Model validation and performance evaluation Feature importance rankings, extreme-scenario stress test results (RMSE and bias for 10 scenarios × 6 model configurations), observed-versus-predicted scatter data for training (15 sites), validation (3 sites), and independent generalization (2 sites) datasets, and KGE distribution samples. Summary statistics: training KGE = 0.94, validation KGE = 0.93, generalization KGE = 0.90. 5. Anthropogenic heat flux (AHF) estimates City-level AHF for 3,560 cities under historical and future conditions (W/m²), derived from nighttime light remote sensing calibrated with energy consumption statistics (historical) and impervious surface projections (future). Global mean AHF increases from 20.8 to 31.1 W/m², a 50% rise. 6. Urban land surface composition trends Long-term trends in fractional vegetation cover (fc), impervious surface fraction (fimp), grassland fraction (fgrass), and woody vegetation fraction (fwoody) for 3,560 cities (%/year), separately for historical and future periods. All four trends accelerate under future scenarios (19–41% intensification). 7. Urban vegetation transpiration: period means and trends Period-mean annual VTu (mm/year) and long-term linear VTu trends (mm/year²) for 3,560 cities under historical and three future SSP scenarios (SSP1-2.6, SSP2-4.5, SSP5-8.5). Global mean VTu declines from 359 mm/year (historical) to 276 mm/year (SSP5-8.5), with trends steepening from −1.41 to −1.69 mm/year². 8. Urban vegetation transpiration anomaly time series Annual VTu and VTu anomaly time series for 3,560 cities across 7 geographic regions and 4 scenarios (1,901,040 records, 1986–2100), plus period-mean anomaly distributions (28,480 records) for cross-city distributional analysis. 9. Trend attribution: landscape pattern change versus climate change Relative contributions of landscape pattern change (fc, fimp, fgrass, fwoody, LAI, AHF) versus climate change (Ta, Ts, VPD, SM, Ra, AC, u) to the VTu trend at each city, expressed as dimensionless fractions summing to 1.0. Landscape pattern change dominates globally (69–76% across scenarios), with a landuse-to-climate dominance ratio increasing from 2.2:1 (historical) to 3.2:1 (SSP5-8.5). 10. VTu changes by plant functional type Source data for the vegetation type analysis (Figure 5), including: (a) a 9 × 9 plant functional type transition matrix quantifying vegetation conversion pathways during urbanization; (b) continental VTu change rates across 6 continents and 9 PFTs; (c) VTu anomaly time series (1985–2100) with confidence intervals for each PFT; and (d) 13-variable SHAP-based trend attribution across 7 regions and 4 scenarios. Spatial and temporal coverage Spatial: 3,560 cities across 104 countries and 6 continents (Asia: 1,345; North America: 554; Europe: 549; Africa: 540; South America: 536; Oceania: 36), plus 20 urban eddy covariance flux tower sites for model calibration Temporal: Historical period (1986–2020) and future projections (2021–2100) under SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios; flux tower observations spanning 2002–2020 File format All data files are provided as Comma-Separated Values (CSV) with UTF-8 encoding. Column headers include units in parentheses for unambiguous interpretation. Each thematic package is accompanied by a detailed README file documenting column descriptions, summary statistics, methodology notes, and usage guidance. Key findings supported by these data Urban vegetation transpiration has declined across the majority of global cities since 1986, and this decline is projected to accelerate through 2100 under all SSP scenarios. The dominant driver of VTu decline is urbanization-induced vegetation structural transformation — specifically, the replacement of deep-rooted woody vegetation by shallow-rooted grassland — rather than direct climatic forcing. Landscape pattern change accounts for 69–76% of the VTu trend globally, with the dominance intensifying under higher-emission scenarios and in rapidly urbanizing regions of Asia, Africa, and South America. Deciduous forests (DNF, DBF) experience the steepest transpiration declines after urbanization, while grasslands and croplands show stable or increasing transpiration, reflecting the functional consequences of the woody-to-grass structural shift. How to cite If you use these data, please cite both this Zenodo repository and the accompanying publication: Chen, H. et al. Human-induced vegetation shifts drive global declines in urban vegetation transpiration. Keywords urban vegetation transpiration, evapotranspiration partitioning, urban heat island, eddy covariance, deep learning, CNN-PM, SHAP attribution, urban greening, plant functional types, SSP scenarios, global cities, land use change, vegetation structural transformation, impervious surface, anthropogenic heat flux License Creative Commons Attribution 4.0 International (CC BY 4.0)