Mango is an important cash crop in the tropics and subtropics.  Determining the yield gap of mango and production constraints can potentially promote the sustainable development of the mango industry.  In this study, boundary line analysis based on survey data from 103 smallholder farmers and a yield gap model were used to determine the yield gap and production constraints in mango plantations in the northern mountain, central valley and southern mountains regions of Tianyang County, Guangxi, China.  The results indicated that the yield of mango in three representing regions of Tianyang County, Northern Mountains, Central Valley and Southern Mountains, was 18.3, 17.0 and 15.4 t ha^–1 yr^–1, with an explainable yield gap of 10.9, 6.1 and 14.8 t ha^–1 yr^–1, respectively.  Fertilization management, including fertilizer N, P₂O₅ and K₂O application rates, and planting density were the main limiting factors of mango yield in all three regions.  In addition, tree age influenced mango yield in the Northern Mountains (11.1%) and Central Valley (11.7%) regions.  Irrigation time influenced mango yield in the Northern Mountains (9.9%) and Southern Mountains (12.2%).  Based on a scenario analysis, the predicted yield would increase by up to 50%, and fertilizer N use would be reduced by as much as approximately 20%.  An improved understanding of production constraints will aid in the development of management strategy measures to increase mango yield.

Apple occupies a dominant position in fruit production globally, and has become the main income source of local smallholder farmers in Luochuan County in the Loess Plateau area, one of the largest apple production areas in China.  However, the annual productivity of apple orchards in this region remains low and has gradually declined over the years.  The distinction and correlation of production constraints can contribute to the promotion of apple orchard productivity and the development of a sustainable orchard system.  In the present study, survey data from 71 smallholder farmers were analyzed using a yield gap model to distinguish the production constraints and determine their correlation with the yield gap based on the structural equation model (SEM).  The results indicated that the average apple yield in Luochuan County was 29.9 t ha^–1 yr^–1, while the attainable yield (Yatt; the highest yield obtained from the on-farm surveys) was 58.1 t ha^–1 yr^–1.  The average explained and unexplainable yield gaps were 26.3 and 1.87 t ha^–1 yr^–1.  According to the boundary line analysis, crop load, number of sprayings and base fertilizer N were the top three constraints on apple production in 9.8, 7.8 and 7.8% of the plots, respectively.  Among the production constraints, crop load and fruit weight affected apple yield through direct pathways, whereas other constraints influenced apple yield through an indirect pathway based on the SEM, explaining 51% of the yield variance by all the main production constraints.  These results can improve the current understanding of production constraints and contribute to the development of management strategies and policies for improving apple yield.

Smallholders in developing countries commonly use intercropping to produce crops with higher yield and value.  Many intercropping studies have been conducted under experimental conditions, but few studies have been performed in farmers’ fields.  We conducted a 4-year study using data from real farms to examine the relationships between yield and yield components of intercropped maize in the North China Plain.  Three field experiments were conducted to compare the suitability of different maize varieties in intercropping.  In the farm study, the grain yield of maize intercropped with watermelon was reduced by more than one third as compared to maize in wheat-maize double cropping, mainly due to lower ear density and lower 100-grain weight.  Under real farm conditions, the yield of intercropped maize increased with increasing ear density and 100-grain weight, while yield of sole maize increased with increasing grain number per ear and 100-grain weight.  In the field experiments, the maize cultivars commonly used in double cropping gave similar yields when grown in the intercropping system and their yields were closely related to ear density and 100-grain weight.  Our results demonstrated that ear density, rather cultivar, was a key factor affecting the productivity of intercropped maize.  Therefore, maintaining high ear density is a practical way for promoting productivity of maize in farmers’ intercropping practices.

Sustainable feeding of the growing population in China without ecological destabilization is a grand challenge.  In this populous country where agriculture is dominated by smallholder farming, developing innovative technology and translating scientific knowledge into action for smallholder farmers is a crucial step in addressing this challenge.  Here, we present a novel approach for technology innovation and dissemination to achieve sustainable intensification in the fields of smallholder farmers.  The Science and Technology Backyard (STB) is a hub in a rural area that links knowledge with practices to promote technology innovation and exchange.  In this study, the framework and functions of STB are introduced, and the key implications for sustainable intensification across millions of smallholder farmers are explicitly stated: (i) develop innovative technology based on stated demands of farmers; (ii) disseminate technology by innovative social service models though combined top-down approaches with bottom-up measures to enable smallholders in rural areas.  This paper provides a perspective on transformation of small-scale agriculture toward sustainable intensification in China and useful knowledge applicable to other developing countries.