Estimating global-local dynamics of land use systems by downscaling from GLOBIOM model

Ermoliev, Y., Ermolieva, T., Havlik, P. ORCID: https://orcid.org/0000-0001-5551-5085, Mosnier, A., Leclère, D., Obersteiner, M. ORCID: https://orcid.org/0000-0001-6981-2769, & Kostyuchenko, Y. (2014). Estimating global-local dynamics of land use systems by downscaling from GLOBIOM model. In: Integrated Management, Security and Robustness. Eds. Zagorodny, AG, Ermoliev, YM, & Bogdanov, VL, NASU & Innovation Center of the NASU. ISBN 978-966-02-7376-4

Full text not available from this repository.

Abstract

High spatial resolution land use and cover change projections are one of the crucial inputs required by Global Circulation Models. They can be effectively used for the assessment of local carbon and greenhouse-gas (GHG) stocks and fluxes in various ecosystems. However to produce these future land cover maps directly with a global economic land use planning model remains a challenge. Although GIS provides detailed geo-physical information, the socio-economic data on driving forces usually exist only on aggregate level. We propose a model fusion involving two interlinked stages: calculation of regional projections by using a global dynamic model GLOBIOM and proper dynamic downscaling method allowing to obtain the required spatially resolved land use and cover change projections. The proposed procedure allows incorporating data derived from satellite images, statistics, expert opinions, as well as model-derived data from global land use models. The two interlinked models bring consistent results between large scale land use change processes and local dynamics, as illustrated by projections for China, Canada, Brazil, US, Ukraine, Russia. There are connections of proposed entropy-based downscaling approach with a fundamental maximum likelihood method proposed for traditional statistical estimation problems. In many practical applications, available prior distributions may not be known exactly, therefore, we develop a new general probabilistic approach to achieve downscaling results robust with respect to a set of prior distributions. This method generalizes ideas of robust statistics for new estimation problems without real repetitive observations of uncertainties.

Item Type:	Book Section
Research Programs:	Ecosystems Services and Management (ESM)
Bibliographic Reference:	In: AG Zagorodny, YM Ermoliev, VL Bogdanov (Eds); Integrated Management, Security and Robustness; NASU & Innovation Center of the NASU, Kiev, Ukraine pp.228-240
Depositing User:	IIASA Import
Date Deposited:	15 Jan 2016 08:51
Last Modified:	27 Aug 2021 17:24
URI:	https://pure.iiasa.ac.at/11067

Actions (login required)

View Item