Agriculture, Ecosystems and Environment 306 (2021) 107170 Available online 15 October 2020 0167-8809/© 2020 Elsevier B.V. All rights reserved. Maize yields from rotation and intercropping systems with different legumes under conservation agriculture in contrasting agro-ecologies W. Mupangwa a, *, I. Nyagumbo b , F. Liben c , L. Chipindu b , P. Craufurd d , S. Mkuhlani e a International Maize and Wheat Improvement Centre, ILRI Shoalla Campus, P.O. Box 5689, Addis Ababa, Ethiopia b International Maize and Wheat Improvement Centre, P.O. Box MP 163, Mount Pleasant, Harare, Zimbabwe c Ethiopian Institute of Agricultural Research, P.O. Box 2003, Addis Ababa, Ethiopia d International Maize and Wheat Improvement Centre, Agricultural Botany Division, NARC Research Station Khumaltor, Lalitpur, P.O. Box 5186, Kathmandu, Nepal e Climate System Analysis Group, Department of Environmental and Geographical Science, University of Cape Town, Cape Town, South Africa A R T I C L E INFO Keywords: Crop rotation Eastern and Southern Africa Forage Intercropping Monocropping ABSTRACT Maize (Zea mays L.) is a key food security crop in Eastern and Southern Africa (ESA). Maize yield gaps are large in the smallholder farming systems of ESA but can be closed by a combination of improved crop varieties and sustainable cropping systems. On-farm trials were conducted in fve countries of ESA over a seven-year period to assess the effects of different cropping systems on maize productivity. Cropping systems tested were conventional practice (Conv_sole) compared with variants of conservation agriculture (CA) such as sole maize (CA_sole), intercropping (CA_intercropping) and rotation (CA_rotation). Maize yield varied with cropping system and agro- ecological conditions. In highlands, CA_intercropping (3709 kg ha 1 ) outperformed Conv_sole (3456 kg ha 1 ), CA_sole (3596 kg ha 1 ) and CA_rotation (3545 kg ha 1 ) with similar legumes in the cropping system. The CA_rotation (3050 kg ha 1 ) out-yielded the other treatments (26452864 kg ha 1 ) in lowlands where similar legumes were grown in the cropping system. Without rotation practices, CA_sole (3596 kg ha 1 ) outperformed the conventional and CA_intercropping treatments (29543036 kg ha 1 ) in highlands with similar legumes in the cropping system. In lowlands, CA_intercropping (2802 kg ha 1 ) out yielded the other treatments (24852658 kg ha 1 ). The highest maize yield was from CA systems with groundnut (3609 kg ha 1 ) and common bean (33073576 kg ha 1 ) under rotation and intercropping practices, respectively. Maize-pigeon pea intercropping (35 %) and maize-groundnut rotation (31–43 %) under CA had the highest maize yield advantages over the conventional practices. The most stable maize yields were from the maize-common bean systems under CA. There is scope for promoting CA cropping systems integrated with grain and forage legumes to address maize productivity challenges on smallholder farms of ESA. 1. Introduction Maize (Zea mays L.) productivity remains low in Eastern and Southern Africa (ESA) and yields vary with agro-ecology, crop hus- bandry, cropping systems and cultivars (Sileshi et al., 2010). Low pro- ductivity is attributed to poor and declining soil fertility, inappropriate agronomic practices including cereal monoculture, among other rea- sons. Conservation agriculture (CA) is being advocated for, to address some of the crop production constraints including poor soil fertility and low crop productivity, and improve adaptation of smallholder farming systems to the changing climatic conditions (Hobbs et al., 2008; Thier- felder and Wall, 2010; Wall et al., 2013). Conservation agriculture is hinged on minimizing soil disturbance, judicious choice of crop associations involving cereals and legumes, as well as having a semi-permanent to permanent soil cover in the cropping system (FAO, 2001). Various CA based cropping systems have been tested in different countries of the ESA region and these include manual, animal traction and small engine-tractor powered options (Ngwira et al., 2012; Thier- felder et al., 2014). Crop yields from CA systems have been variable and differences can be attributed to soil types, agro-ecological conditions and agronomic management in the cropping systems (Thierfelder et al., 2015). Maize yield depression has been observed in some growing sea- sons and this has been attributed to waterlogging and nitrogen immo- bilization in CA systems with crop residues as soil cover in high rainfall areas (Nyagumbo et al., 2016; Kitonyo et al., 2018; Mupangwa et al., * Corresponding author. E-mail address: w.mupangwa@cgiar.org (W. Mupangwa). Contents lists available at ScienceDirect Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee https://doi.org/10.1016/j.agee.2020.107170 Received 29 February 2020; Received in revised form 30 August 2020; Accepted 13 September 2020