Beyond the forest transition hypothesis: Uncovering the drivers influencing natural, planted and plantation forest area development using regression-based machine learning approaches

Julia Tandetzki, Christian Morland and Franziska Schier

Land Use Policy, 2025, vol. 158, issue C

Abstract: Given the crucial role forests play in sustainable development and climate mitigation, identifying and understanding the key drivers of global forest area change is essential. However, the importance that the various factors have in driving forest development on a global scale remains unclear. Thus, the objective of this study is to analyze whether different forest types are affected by different drivers of change. Building on the forest transition hypothesis, we examined the effects of 95 pre-selected variables on the development of two different forest type areas: natural forests and planted forests including plantations. For this, we used different machine learning approaches (univariate and multivariate panel regressions as well as the regularized regression Lasso and Elastic Net methods) to identify and cross-validate the most plausible drivers based on panel data from 190 countries between 1990 and 2020. Our analysis highlights the importance of proximate factors, such as land area and forest-based production, that directly influence the development of natural forests in particular. We further identified four underlying supergroups (consumption, demography income, and trade), whose drivers can influence the development of each forest type, but notably – depending on their nature – in opposite directions. In summary, our study emphasizes how sustainable energy consumption patterns appear to be essential for mitigating deforestation and promoting sustainable forest use. Demographic pressures are driving changes in both forest types, but notably in different directions: High population densities promote the deforestation of natural forests, whereas increasing total population levels benefit the expansion of planted forests. Although important, urban and rural population dynamics affect the two forest types differently. The impacts of roundwood trading are mostly, but not always, positive: Roundwood exports seemingly provide economic incentives to maintain natural forests in particular, whereas roundwood imports can hinder planted forest development, thus underscoring trade’s complex effects. The income variable adjusted savings – net forest depletion has a universal negative impact on the development of both forest types. Our findings emphasize the need for integrated, region-specific policies to achieve a balance between economic progress and forest conservation. This study thus offers valuable insights for policymakers and decision-makers and highlights the importance of cohesive strategies to safeguard and sustainably manage global forests.

Keywords: Forward selection; Machine learning; Lasso; Elastic net; Deforestation drivers; Forest area development; Forest transition hypothesis (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0264837725002960
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:lauspo:v:158:y:2025:i:c:s0264837725002960

DOI: 10.1016/j.landusepol.2025.107762

Access Statistics for this article

Land Use Policy is currently edited by Jaap Zevenbergen

More articles in Land Use Policy from Elsevier
Bibliographic data for series maintained by Joice Jiang ().