Forecasting Municipal Solid Waste and Landfill-Gas Generation in Sulaimani City, Iraq: An Integrated Support Vector Regression First-Order Decay Framework with International Benchmarking

Abstract Municipal solid waste forecasting and landfill-gas assessment are essential for planning collection, disposal capacity, and methane mitigation in rapidly growing cities. This study develops an integrated data-driven framework for Sulaimani City, Iraq, using monthly waste records from 2016–2025 and support vector regression (SVR) to forecast seven municipal waste series. Stationarity was evaluated before supervised lag construction, and linear or radial-basis-function SVR kernels were tuned by chronological cross-validation. January–December 2025 was retained as a strict holdout, after which final models were refitted through December 2025 to forecast 2026. The reported Monthly Total was additionally translated into landfill-gas generation using a first-order decay model with a base methane potential of 100 m³ CH₄/Mg and a sensitivity value of 170 m³ CH₄/Mg. The base case estimates 6.46 million m³ of methane and 12.93 million m³ of landfill gas in 2025. International benchmarking places Sulaimani’s modeled methane generation above the Oaxaca case and below the peak reported for Okhla, South Delhi, while parameter comparisons highlight strong sensitivity to local waste composition and decay conditions. Results support SVR for operational forecasting but show that modeled gas generation should directly guide mitigation planning rather than be interpreted as direct ambient-air pollution measurement for evidence-based urban governance.