Logo image
Back
A Multiscalar Approach to Mapping Marginal Agricultural Land: Smallholder Agriculture in Malawi
Journal article   Peer reviewed

A Multiscalar Approach to Mapping Marginal Agricultural Land: Smallholder Agriculture in Malawi

Brad G. Peter, Joseph P. Messina and Sieglinde S. Snapp
Annals of the American Association of Geographers, Vol.108(4), pp.989-1005
07/04/2018
DOI: https://doi.org/10.1080/24694452.2017.1403877

Abstract

Geography Social Sciences
Marginal agricultural lands are defined here by suboptimal biophysical conditions and historically variable or low agricultural production. We characterize these areas using remotely sensed information to disentangle the biophysical and possible social factors driving marginality. Considering both the modifiable areal unit problem and the ecological fallacy problem, the heuristic we propose is generalizable across geographies and scales and provides information at multiple decision-making levels through a multiscalar interannual variability model. We present results from our study of Malawi, where the landscape is densely cultivated and smallholder farmers frequently occupy marginal lands, to illustrate the potential of a multiscalar analysis in a place where food insecurity alleviation is needed and where remote sensing can provide necessary information. Our framework for identifying marginal agricultural lands consists of (1) locating long-term agricultural land, (2) measuring interannual productivity of long-term farmed locations, and (3) assessing marginal biophysical land characteristics and the fundamental climate niche for the dominant crop (in this case maize). Productivity and marginality in Malawi are spatially organized, and an assessment of productivity at multiple scales highlights the importance of presenting both global and local spatiotemporal variability for managing agroecological variance. By disaggregating broad classes of historically marginal production and the underlying drivers of marginality, different intervention efforts can intelligently target areas most likely to receive maximum benefit. These methodologies can be applied by both policymakers and scholars to identify and target marginal agricultural areas for improved productivity and for the support of smallholder farmer livelihoods.

Metrics

1 Record Views

Details

Logo image