Please cite this article in press as: Magaia, E., et al., Modelling maize yield response to plant density and water and nitrogen supply in a semi-arid region. Field Crops Res. (2017), http://dx.doi.org/10.1016/j.fcr.2016.12.025 ARTICLE IN PRESS G Model FIELD-6889; No. of Pages 12 Field Crops Research xxx (2017) xxx–xxx Contents lists available at ScienceDirect Field Crops Research journal homepage: www.elsevier.com/locate/fcr Modelling maize yield response to plant density and water and nitrogen supply in a semi-arid region E. Magaia a,b,∗ , S. Famba a , I. Wesström b , R. Brito a , A. Joel b a Faculty of Agronomy and Forestry Engineering-University Eduardo Mondlane, Rural Engineering Department, P.O. Box 257, Maputo, Mozambique b Department of Soil and Environment, Swedish University of Agricultural Sciences (SLU), P.O. Box 7070, SE-750 07 Uppsala, Sweden a r t i c l e i n f o Article history: Received 30 June 2016 Received in revised form 12 December 2016 Accepted 28 December 2016 Available online xxx Keywords: APSIM Fertiliser rate Multiple seasons Crop modelling Planting density Semi-arid a b s t r a c t Nitrogen (N) fertiliser use in Mozambique is very low, but blanket recommendations on higher fertiliser application rates may not be appropriate for small-scale rainfed farms in different edaphic and climate regions of the country. Data from field experiments with two water levels, three tillage methods (only one tillage method was used in the present study) and two fertiliser N application rates to maize (Zea mays L.) were used to calibrate and validate the Agricultural Production System Simulator (APSIM) model. The model was subsequently used to simulate multiple-season scenarios (25 years) and determine adequate fertiliser rates for rainfed and supplemental irrigation conditions on sandy loam soil in Southern Mozam- bique. For this, APSIM was configured to simulate long-term maize response to varying N application (0–120 kg ha −1 ) and planting density (1, 4.2 and 8.4 plants m −2 ) in a rainfed or supplemental irrigation system. The APSIM maize model proved capable of producing fairly accurate predictions of observed trends in soil moisture under rainfed and irrigation conditions in a semi-arid region. The model was also able to predict maize grain and biomass yield well. Multiple-season analysis of grain yield revealed that, compared with the medium (recommended) plant density (4.2 plants m −2 ), in rainfed conditions low plant density and high plant density (8.4 plants m −2 ) reduced overall grain yield by 1281 kg ha −1 (40%) and 242 kg ha −1 (8%), respectively. With high plant density, grain yield was below 1000 kg ha −1 in 20% of simulated years, whereas with low planting density the lowest grain yield was around 500 kg ha −1 (4% of simulated years). Irrigation stabilised grain yield in most simulated years, except in some years with high planting density. Fertiliser and supplemental irrigation increased yield for all scenarios except those with low plant density. The optimal N fertiliser rate was identified as 33–102 kg ha −1 in the rainfed system and 38–86 kg ha −1 in the supplemental irrigation system, compared with the national blanket recommen- dation of 120 kg N ha −1 . The simulation results suggested that the national recommended rate is only appropriate for years and regions with adequate rainfall distribution (>300 mm per cropping season). © 2016 Elsevier B.V. All rights reserved. 1. Introduction Agricultural systems in semi-arid regions of Africa that rely on rainfall as a main source of water supply are prone to a high risk of variable production levels (Vlek et al., 2012). According to Whitbread et al. (2010), smallholder farmers in semi-arid regions face serious challenges in maintaining food security, exacerbated by low soil fertility, limited resources to purchase inputs and high ∗ Corresponding author at: Faculty of Agronomy and Forestry Engineering- University Eduardo Mondlane, Rural Engineering Department, P.O. Box 257, Maputo, Mozambique. E-mail addresses: emilio.magas@gmail.com, emilio.magaia@slu.se (E. Magaia). variable rainfall. Rainfall is a major driver of enhanced efficient response to fertilisers applied in small-scale farming systems under arid and semi-arid conditions (Shamudzarira & Robertson, 2002). It is known that fertiliser efficiency and agricultural produc- tivity are low in most parts of Africa and Mozambique (Kisaka et al., 2015; Silici et al., 2015). The efficiency with which soil and water resources are utilised in sub-Saharan African cropping sys- tems is especially critical, as these resources are generally scarce in that region (Chikowo et al., 2008). In this regard, Kloss et al. (2012) highlighted the need for strategies to improve crop growth, make irrigation more efficient and sustainable and conserve farm- lands. In addition, grain yield is influenced by inter-row spacing and sowing density (Testa et al., 2016). In Mozambique, the standard recommended plant density for maize is 4.4 plants m −2 (Roxburgh http://dx.doi.org/10.1016/j.fcr.2016.12.025 0378-4290/© 2016 Elsevier B.V. All rights reserved.