1 INTRODUCTION

African and Asian food production relies to 70% (of calories provided) on smallholder production systems (Fanzo, 2017). More than 475 million small farms produce food often under marginal conditions, with little input or market access (Lowder, Skoet, & Raney, 2016). The need to satisfy an ever‐growing demand for food production, especially under the reality of climate change, poses new challenges and opportunities to smallholder farmers worldwide. On one hand, marginal lands are being increasingly brought into production, on the other hand, better market opportunities promote diversification of production systems. Non‐traditional crops and varieties are being introduced to existing systems, challenging farmers´ knowledge and experience in handling these new opportunities, as well as the resource base. Cropping calendars need to be adapted or newly developed to include these newcomers and research is needed to understand how they react to the environmental factors of these surroundings. Climate change poses additional risks from both increasingly erratic rainfall patterns and rising temperatures as well as from newly emerging biotic stressors. Droughts and heavy rains are very likely to increase in frequency in large parts of the world (Olesen et al., 2011), resulting in a need for extended scope of and improved access to state of the art and timely weather forecasts also for smallholder farmers, especially in the developing world. But climate change‐related adaptation and mitigation measures do also offer new possibilities to (sustainably) intensify agricultural production particularly in temperature limited environments such as high altitudes or seasonally hot regions (van Oort & Zwart, 2018).

Adaptable solutions are needed to maintain and improve yields and farmers’ livelihoods in these systems. National Extension and Agricultural Research Services are working together with international agricultural research organizations, such as CGIAR Centers and Universities, on a vast variety of related topics, challenges and solutions. This special issue aims at putting a spotlight on some of these approaches.

Often, Agriculture 4.0 is associated with high‐tech solutions for the automatization of agricultural activities, ranging from precision farming approaches, sensor technology, self‐driving machinery to the use of drones. All of these technologies can be found in the agricultural sectors of developing countries, albeit to a much smaller degree than in the global north. Mobile and smartphone technology and related apps are well established in developing countries. For more than a decade, this technology has supported money transfer in structurally less developed regions of the world and has helped communicate, for example weather forecasts, along the value chain from smallholder farmers to merchants and entrepreneurs to trading hubs. More and more apps are being developed that service this line of activities: certification along a products life cycle, from seed to manufactured good, or the teaming of buyers and sellers in rural environments. ICRISATs iHub, as an example, aims at supporting exactly this, by bringing together agricultural entrepreneurs, scientists, and technology experts (ICRISAT Innovation Hub, 2019). Especially when focusing on smallholder targeted approaches, the investment needed for an extensive machine outfit is causing farmers to look for other solutions. “Hello Tractor” is an app supporting agricultural mechanization in Africa by linking equipment owners (such as tractors) to equipment users and thus creating a network among smallholder farmers via the app (Hello Tractor, 2019). Crop production system depends on crop productivity and good management. The app Rice Advice from the Africa Rice Center aims at delivering decision support to rice farmers and extension personnel in Africa, based on farmers input on crop varieties, the availability of input and investment resources, labour and the location of the farmer´s fields (Rice Advice, 2019; Saito, Diack, Dieng, & Ndiaye, 2015). Changes to the original set‐up due to fluctuating fertilizer prices, erratic weather patterns or disturbances in water supply can be accounted for on time, and the recommendations to farmers can be adapted considering the latest developments. Similar, server‐based growth models can improve their performance by having access to the latest regional weather forecasts, allowing not just to suggest optimal cropping calendars to the farmers, but also to adjust crop management strategies such as fertilizer application and weed management.

In order to support this kind of app‐based extension approaches, the models behind have to be calibrated and validated using a wide range of input data. This is especially the case when even well‐known varieties of staple crops such as rice are being introduced to new environments, for example irrigated paddy rice to the highlands of East Africa. How do these varieties respond to the thermal characteristics (cold or hot spells, Abera et al., 2020, Stuerz et al., 2020, this issue), how to adapt cropping calendars to the respective speed and timing of the varieties phenological development (Razafindrazaka et al., 2020, this issue), and do the plants respond to external influences such as fertilizer application in the same way as in their “native” environment (Boshuwenda et al., 2020, Senthilkumar et al., 2020, this issue)? Answering questions like these is crucial for the success of the modelling approaches behind these apps. Without these data, it will not be possible to apply the apps to the wide range of agricultural environments found in smallholder farms throughout the developing world, especially when facing the challenge of securing food security in a world of changing climate (Cotter et al., 2020, this issue).

On larger scales digitalization does not come in the form of apps but requires GIS based, spatially and temporally explicit models. For this, conceptual models need to be established first that scrutinize the current situation and propose a way forward. An example of such approaches applies to Asian Mega Deltas that are under pressure from sea level rise and salt intrusion into rice production systems. These deltas are so vast that only well‐calibrated models are able to yield recommendations for land‐use adaptation to the climate change‐induced changes (Schneider et al., 2020, this issue).

The articles published in this issue are based on research that was presented during the conference “Smallholder targeted Agriculture 4.0 in temperature limited cropping systems (STATCROPS)” organized by the Hans‐Ruthenberg‐Institute for Tropical Agricultural Sciences at the University of Hohenheim and the Africa Rice Center, 20–21 September 2018. The STATCROPS conference addressed the possibilities that modern digitalization and modelling approaches offer to develop tools targeted to smallholder farmers, especially in temperature‐limited environments. The conference aimed at linking research, modelling and application by discussing ideas, concepts, and experiences to build feedback‐ and feed‐forward loops targeted at future options for such cropping systems. The authors would like to thank the Deutsche Forschungsgemeinschaft (DFG) for their support for this conference.