Three types of cropping systems, cassava-based intercropping, hedgerow intercropping and legume cover crop rotations, were evaluated in 1994-1997 in Northern Lampung, Sumatra (Indonesia). The C and N flows returned within and transported out of plots and crop yields of different cropping systems were quantified. Cassava-based systems were not stable and yields declined over time. Intercropping cassava with rice increased cassava fresh tuber weight by 5-48% compared to the monocropping system. The hedgerow intercropping gave lower maize, rice, groundnut and cowpea yields than could be obtained in a crop rotation with legume cover crops. Maize grain yields in the 80-20 rice/maize mixture were ~ 0.4 t ha-1 in the rice - groundnut rotation and about half as much when intercropped with cassava or hedgerows. Rice yields intercropped with cassava or with hedgerows were ~1 t ha-1 less in year 2 and 3 than those grown in rotation with groundnut. The rice yield in the first cropping season was only ~1 t ha-1, but in the second and third year yields in the rice - legume rotation increased to ~2 and 3 t ha-1, respectively. This increase occurred despite a decline in soil organic matter content. The cassava-based systems removed much more C (7 t ha-1 yr-1) than the other systems, while less was returned (~0.5-2 t ha-1) to the soil. In the hedgerow intercropping system ~2.5 t C ha-1 yr-1 was returned to the plot as biomass pruning and crop residues and ~1.5 t C ha-1 yr-1 was removed from the plot as yield. In the cover crop rotation 2.6 t C ha-1 yr-1 was returned to the plot as crop residues plus Mucuna (only the 2nd year) and cowpea biomass, and ~1.1 t ha-1 yr-1 was removed from the plot. The hedgerow intercropping systems gave an N surplus of ~15-50 kg ha-1 yr-1 returned to the soil; while the balance was 10-20 kg ha-1 yr-1 for the cover crop rotation systems, and the cassava-based systems showed a negative N budget of ~60 kg ha-1 yr-1.
Netherlands Journal of Agricultural Science (2000) 48 (1) 3-17