Cangela, Arménio
- Department of Forest Ecology and Management, Swedish University of Agricultural Sciences
- Eduardo Mondlane University (UEM)
Shifting cultivation under high population pressure often results in shortened fallow periods and land degradation. Multistrata agroforestry, improved fallow, and woodlots are promoted as sustainable alternatives, but comparative evidence of their benefits remains limited. We conducted a four-year field experiment in Mozambique to evaluate whether these systems improve land productivity, crop yields, and soil fertility compared to shifting cultivation, and to assess the impact of using native trees or eucalyptus. Multistrata system involved maize, pigeon peas, and bananas with either native trees (MS-N) or eucalyptus (MS-E). Improved fallow combined pigeon peas and maize sequentially with or without native trees, while woodlots included maize with native trees or eucalyptus. Productivity was assessed using land equivalent ratios (LER) based on crop, firewood, and pole yields over three seasons. Soil organic carbon, total nitrogen, available phosphorus, and bulk density were also measured. We found that multistrata systems improved productivity compared with shifting cultivation. LER values of MS-N and MS-E rose from 1.0 in the first season to 2.3 and 1.83, respectively, reflecting productivity gains of 130% and 83% by the third season. Woodlots outperformed shifting cultivation only in the third season after firewood and pole harvests, whereas improved fallows showed no significant productivity gains. No significant changes in soil fertility were observed within three years. Choosing native or eucalyptus trees did not affect maize yields or soil nutrient levels. Our findings highlight that multistrata systems can rapidly improve land productivity compared to shifting cultivation. However, detecting soil fertility changes may require long-term monitoring.
Multistrata agroforestry; Improved fallow; Land equivalent ratio; Shifting cultivation
Agroforestry Systems
2026, volume: 100, number: 4, article number: 104
Publisher: SPRINGER
Forest Science
Agricultural Science
https://res.slu.se/id/publ/146610