As scientists and policy-makers struggle to understand the complex global carbon (C) system, it has become apparent that a better understanding of regional-scale C dynamics is needed. Fundamental questions regarding C storage and flux in arid ecosystems remain unanswered and the size of existing C pools and potential for future sequestration are unknown. A collaborative research project with different local and international organizations entitled "Carbon Sequestration Potential of Arid Rangelands" was launched to measure and model C storage in plants and soils of arid rangelands in Uzbekistan ; a pilot site was set up in soil-climatic conditions representative of the main farming areas of Central Asia. The research approach consisted of direct field measurement of CO2 fluxes using Bowen Ratio systems, improved analytical approaches to soil C assessment, development of new models for soil C calculation, landscape-level computer simulation modelling to scale-up field-based measurements into regional-scale estimates of C sequestration. Based on laboratory and field investigations it was possible to identify and quantify C storage potentials of soils in both irrigated and rainfed conditions. A detailed database was created on climate, biomass, soil types and land use. The GIS and ecosystem modelling methodologies allowed integration of much available data on soil organic matter, vegetation, climate, with new approaches to provide managers and decision-makers at the local, regional and international scales an adequate tool and information on C sequestration potential of soils in Uzbekistan.