integration of internet of things (iot), artificial intelligence (ai), and gis for adaptive management of urban green infrastructure: a case study of rasht, iran

Urban green infrastructure (ugi) in humid cities such as rasht, iran, faces persistent hydrological stress and ecosystem instability. this research develops an eco digital twin framework (edtf) that integrates the internet of things (iot), artificial intelligence (ai), and geographic information systems (gis) for adaptive management of ugi. a network of nine nb iot sensor nodes continuously monitored soil moisture, conductivity, and micro climate parameters, achieving 99.8 % data uptime despite high rainfall. tree health was diagnosed using a resnet 50 convolutional neural network trained on 4 200 tomograms and multispectral indices (auc = 0.94). gis based multi criteria analysis (ahp) synthesized moisture, canopy stress, and surface impermeability into a vulnerability index (vi = 0.40–0.83), which correlated strongly with cnn derived health scores (r = 0.72, p < 0.01). the integrated eco cognitive equilibrium (ece = (thi – vi)/smnorm) quantified ecological stability across the city, revealing negative ece clusters in baharestan and saheli under prolonged saturation. the edtf’s predictive feedback loop reduced irrigation demand by ≈ 15 % and identified high risk tree zones for preventive maintenance. methodologically, the study unites real time sensing, machine vision, and spatial analytics into a replicable model for smart green governance. the framework demonstrates how iot–ai–gis convergence can transform urban ecology from reactive maintenance to cognitive, data driven adaptation—a blueprint for sustainable digital twin ecosystems in humid urban landscapes.