Smart Farming in sub-Saharan Africa:

Challenges and Opportunities

Pamela Abbott

, Alessandro Checco

and Davide Polese

Information School, The University of Sheffield, U.K.

Istituto per la Microelettronica e Microsistemi, Consiglio Nazionale delle Ricerche, Roma, Italy

Keywords: Global South, Smart Farming.

Abstract: Smallholder farmers provide the majority of food production in sub-Saharan Africa. They will be severely

impacted by climate change, especially because they are dependent on rain-fed irrigation. We provide a

summary of challenges and opportunities in designing smart farming infrastructure in this context. We

observe that innovation in technology and knowledge production is necessary to increase the efficacy of water

usage and land management. Such solutions must take into account the technological constraints and their

regional variability to be able to provide sustainable and scalable solutions. Such solutions also need to

embrace the notion of openness, encouraging collaborative endeavour and avoiding proprietary

implementations.

1 INTRODUCTION

Whilst it has been acknowledged that smallholder

farmers may be contributing to environmental

degradation through unsustainable agricultural

practices, it is increasingly recognised that they may

also hold the key to alleviating these problems (IFAD,

2013). Smallholder farmers are thought to provide as

much as 80% of food production in sub-Saharan

Africa (Stewart et al., 2014). They are, however,

vulnerable to environmental conditions presumably

linked to climate change such as unpredictable

variation in rainfall leading to droughts and floods

(Morton, 2007). Disruption to these farming

practices thus threaten food security due to the

dependence on smallholders for food production in

countries already deemed vulnerable to climate

shocks (IPCC, 2014). Solutions that may address this

issue and lead to more sustainable farming practices

are difficult to envisage and implement not the least

because of the complex environments in which

smallholder farmers operate (Morton, 2007). Their

own situation is complex, often perceived as

marginalised and lacking adequate resources (IFAD,

2013), while national level policymaking around

https://orcid.org/0000-0002-4680-0754

https://orcid.org/0000-0002-0981-3409

https://orcid.org/0000-0002-6332-5051

agriculture tries to find a balance between the rhetoric

of sustainability and competing agendas of market-

driven production (Beddington et al., 2012;

Busingye, 2017).

Many of these smallholder farmers in sub-

Saharan Africa are dependent on rain-fed irrigation to

grow their crops (Nahayo et al. 2018; Kinda &

Badolo, 2019), thus increasing their exposure to crop-

yield risk. In Eastern Africa, estimates of yield

reductions in some crops are as much as 72% (wheat)

and 45% (maize, rice, soybean) projected by the end

of the century (Adhikari et al., 2015). Farmer-led

initiatives to increase the efficacy of water usage and

land management are thus being investigated along

with innovations in technology and knowledge

production (Woodhouse et al., 2017). One such

innovation has been dubbed ‘smart farming’, which

is described as the application of Information and

Communication technologies (ICTs) to agricultural

production, especially more advanced forms of

technology such as the Internet of Things (IoT) and

artificial intelligence (AI) (Wolfert et al., 2017).

In the last decades, an intensive investigation has

been developed in order to maximize crop production

trying to reduce resource waste. To this purpose, a

Abbott, P., Checco, A. and Polese, D.

Smart Farming in sub-Saharan Africa: Challenges and Opportunities.

DOI: 10.5220/0010416701590164

In Proceedings of the 10th International Conference on Sensor Networks (SENSORNETS 2021), pages 159-164

ISBN: 978-989-758-489-3

159

new research line, called precise agriculture has been

created (Liaghat et al., 2010; Brisco et al., 1998; Ge

et al., 2011; Jawad et al., 2017). Many efforts have

been done to improve the state of health of the crops,

reduce the waste of water, and reduce fertilizer usage.

In terms of the implemented crop monitoring

infrastructure, the main goal has been to improve its

durability and to reduce its power consumption.

Nevertheless, the vast majority of past work had a

tacit prerequisite: the possibility of using the latest

developed technology; to have access to a modern

networking infrastructure, to satellite data, and

modern hardware to visualise and process the

information. This prerequisite is not always true in

sub-Saharan Africa, where precise agriculture has to

face strict constraints in terms of the aforementioned

dimensions. Moreover, open hardware and software

practices (including in the licencing context) are

fundamental to reduce the dependencies to external

economic factors (e.g. royalties).

In this work, we will briefly describe the most

pressing needs of sub-Saharan Africa agriculture,

which technological constraints are in place, what

solutions have been proposed and what could be

improved.

The rest of the paper is structured as follows. In

Section 2 we provide a summary of the literature in

the field of smart agriculture, especially in the context

of sub-Saharan Africa. In Section 3, we present the

main technological constraints that should be

considered when designing smart farming solutions

in this context. In Section 4 we discuss what we

learned and provide a summary of our vision for a

smart farming solution in sub-Sahar Africa.

2 RELATED WORK

2.1 Smart Farming in sub-Saharan

Africa

Some sub-Saharan African countries are already

embracing and implementing agricultural policies

around smart farming, e.g. in Rwanda (Musoni, 2020;

MYICT, 2015). More common, though, are solutions

that attempt to go beyond pilot projects to implement

more sustainable solutions that are scalable. Hence,

many ICT-related agricultural initiatives involve the

use of mobile phones often integrated with

Unstructured Supplementary Service Data (USSD)

platforms for widespread usage and uptake (Wouters

et al., 2009), since mobile telephony is one of the

more widely available ICTs due to relatively high

penetration rates in the sub-continent. Some

examples we are aware of include Farm-SMS in

Tanzania (Makoye, 2013) and the M’chikumbe

project in Malawi (Palmer & Darabian, 2017). These

types of solutions often attempt not only to push

information towards farmer participants but to

encourage other forms of information and knowledge

sharing such as building communities around

particular kinds of knowledge.

Figure 1: Map of Tanzania, Eastern Africa, highlighting the

Dodoma district and Tabora region, in central and western

Tanzania, respectively, where the Farm-SMS initiative is

reported to be sited (image source: Perry-Castañeda Library

Map Collection [PCLMC], 2016).

According to Makoye (2013), the Farm-SMS

initiative was founded in 2010 through a

collaboration between the World Meteorological

Organization (WMO) and the Tanzanian

Meteorological Agancy (TMA). The pilot project has

two sites, one located near Dodoma, the legislative

capital of Tanzania in the central region of the country

and the other in the Tabora region, both of which are

Key facts about Tanzania:

 Population: approx. 56 million (2018 est.)

 Size: approx. 947,303 km

 GDP per capita: $USD 3,574 (2019 est.)

 HDI: 0.529 (2019)

 Agricultural sector:

o 24.5% GDP (2013 est.)

o 85% of exports

o Employs approx. 50% workforce

o Largest food crop: maize

Source:

https://en.wikipedia.org/wiki/Tanzania

WSN4PA 2021 - Special Session on Wireless Sensor Networks for Precise Agriculture

160

reported to be drought-prone (see Figure 1). The

service provided by Farm-SMS allows farmers in the

region to receive real-time weather forecasts and tips

from agricultural experts at a nearby research centre

through SMS text messaging and emails. Farmers

who participated in the pilot reported between 50%

and 125% improvement in yields compared to

traditional methods of forecasting the weather. It is

unclear what is the current status of the project and

whether it has had a sustainable impact beyond the

pilot (some further information seems to be available

via Tall et al., 2014).

The M’chikumbe project in Malawi (See Figure

2) was started in 2015 as a pilot project through

collaborations with various international donor

agencies, Airtel Malawi and Malawian government

agricutlural services (Palmer & Darabian, 2017).

According to these authors, the project gained

siginficant support among its user base acquiring

400,000 users by December 2016. The service

provides access to agricultural information and

educational materials from both the agricultural

extension services and partner content providers.

SMS messaging and interactive voice recognition

(IVR) through mobile technology underpin these

services. In addition, a network of ‘lead farmers’ and

other trusted information providers enable the spread

of content to farmers. The report provides evidence

of ‘power users’, those most actively engaged in the

service, claiming increases in crop yields as a result

of using the service (about 53% of those users).

While it started as a pilot in 2015, the project is

ongoing with services branching out into mobile

money, which helps to keep the platform host, Airtel

Malawi, on board. Further plans are being put into

place to maintain the sustainability of the project.

Figure 2: Map of Malawi, in Eastern Africa.

2.2 Smart Irrigation and Flood

Prevention

In the last 20 years, an extensive analysis has been

carried out on the technical challenges of automated

irrigation, together with multiple attempts at

providing the design of the hardware and of the

network infrastructure (Kim et al., 2009; Gutiérrez et

al., 2013). The majority of these works make use of

densely located wireless humidity and in-field water

sensors connected to the internet.

Another related field is smart flood disaster

prediction. It has been shown that using a sensor

network monitoring humidity, temperature, pressure,

rainfall, and water level it is possible to perform

accurate flood prediction tasks (Bande et al., 2017).

In both these fields, the constraints described in

the introduction and discussed in the rest of this work

are not taken into account, and thus they are not easily

transferable in the context of sub-Saharan Africa.

3 CONSTRAINTS

In this section, we discuss the main constraints that

need to be considered when designing smart farming

solutions in sub-Saharan Africa.

Key facts about Malawi:

 Population: approx. 19 million (2020 est.)

 Size: approx. 118,484 km

 GDP per capita: $USD 1,234 (2019 est.)

 HDI: 0.483 (2019)

 Agricultural sector:

o 27% GDP (2013 est.)

o 90% of exports

o 80% of population are subsistence

farmers

o Largest export crop: tobacco

o Large food export crops: tea, sugar,

coffee

Source: https://en.wikipedia.org/wiki/Malawi

Smart Farming in sub-Saharan Africa: Challenges and Opportunities

161

3.1 GSM Network & Mobile Hardware

Both the user application and sensor network

interface should be implemented on top of the GSM

network, because of the limitations on mobile

hardware availability in sub-Saharan Africa. This in

turns will introduce additional constraints: (i) low

throughput, (ii) USSD/WAP protocol rather than

modern Internet connection, (iii) need of a concise

(often only textual or in any case in low resolution)

representation of the processed data.

In practice, an USSD interface could allow a

simple interface to: (i) the sensor network, (ii) the

cloud application, (iii) the crowdsourcing interface.

The users would not need a smartphone, nor will need

to install any application to use the interface.

However, this would limit the complexity of the

interaction with the sensor network and of the

processed data, and would require significant design

work to translate data, e.g. to visualise a trend or

describe a map using only textual information.

3.2 Reliance on Electricity

The sensor network should be robust to lack of

electricity, and thus be based on self-charging sensors

equipped with a solar panel and a battery. This in

turns will require the use of LoRA networks (Lavric

et al., 2017; Wixted et al., 2016), with a constraints of

the order of hundreds of mW per square centimetre.

Moreover, the polling frequency of the sensor should

be low, of the order of hours rather than minutes.

3.3 Low Cost of the Sensor Network

The need of limit the cost of the sensor network and

the need of reducing the dependencies from external

economic influences will require the need of using

open hardware design and open network protocols,

and the use of a limited number of sensors. For this

reason, we believe that the nodes should be equipped

with basic sensors like humidity, temperature, and

pressure, leaving more complex measurement to

crowd sourcing (e.g. for river water level) or via post-

processing inference.

The need of an open protocol and the

aforementioned constraints suggests that an ideal

choice for the networking protocol would be

OpenThread (Kim et al., 2019, Checco & Polese,

2020).

4 CONCLUSIONS

Building sustainability and scalability, often

competing concepts, into ICT solutions has often

been an issue in many Information and

Communication Technologies for Development

(ICT4D) projects in lower middle income countries

(LMICs). The contexts of use of these ICTs tend to

exhibit an uneven distribution of resources, with

considerable regional variability. Thus, what may

work in an urban centre may fail to provide the same

results in a village or other rural community. ICT

solutions in such environments tend to rely on

contingent conditions and often the cause of failure is

rooted in a lack of attention to contextual

particularities (Davison & Martinsons, 2016). One

group of ICT4D researchers in health innovations in

LMICs have reported some success in addressing

both the sustainability and scalability issue by

adopting what they refer to as flexible standards (Braa

et al., 2007). The concept is embedded in theory from

science and technology studies, but nonetheless

useful in acknowledging that sustainability and

scalability need to consider ICT innovations as

encompassing both people and technology acting

within a particular context. Such solutions also

embrace the notion of openness, thus encouraging

collaborative endeavour and avoiding proprietary

software/systems or platforms.

From a technical standpoint, we envision an

infrastucture as depicted in Figure 3, where the use of

low cost (mainly humidity, temperature, and

pressure), long range and low-power sensors on top

of an OpenThread infrastructure are connected to

GSM supernodes able to provide a local USSD API

Figure 3: Low cost LoRa Infrastructure example.

WSN4PA 2021 - Special Session on Wireless Sensor Networks for Precise Agriculture

162

and a cloud USSD user interface. Such API would

also allow the collection of crowd sourced data, e.g.

for river water level measurements.

REFERENCES

Adhikari, U., Nejadhashemi, A. P., & Woznicki, S. A.

(2015). Climate change and eastern Africa: A review of

impact on major crops. Food and Energy Security, 4(2),

110–132. https://doi.org/10.1002/fes3.61

Bande, S., & Shete, V. V. (2017, August). Smart flood

disaster prediction system using IoT & neural networks.

In 2017 International Conference On Smart

Technologies For Smart Nation (SmartTechCon) (pp.

189-194). IEEE.

Beddington, J., Asaduzzaman, M., Clark, M., Fernández,

A., Guillou, M., Jahn, M., Erda, M., Mamo, T., Van Bo,

N., Nobre, C., Scholes, R., Sharma, R., & Wakhungu,

J. (2012). Achieving food security in the face of climate

change: Final report from the Commission on

Sustainable Agriculture and Climate Change. (p. 64).

CGIAR Research Program on Climate Change,

Agriculture and Food Security (CCAFS).

www.ccafs.cgiar.org/commission

Braa, J., Hanseth, O., Heywood, A., Mohammed, W., &

Shaw, V. (2007). Developing Health Information

Systems in Developing Countries: The Flexible

Standards Strategy. MIS Quarterly, 31(2), 381–402.

https://doi.org/10.2307/25148796

Brisco, B., Brown, R. J., Hirose, T., McNairn, H., & Staenz,

K. (1998). Precision agriculture and the role of remote

sensing: a review. Canadian Journal of Remote

Sensing, 24(3), 315-327.

Busingye, J. D. (2017). Smallholder farming and food

sovereignty in Uganda: An in-depth analysis of policy

vis-a-vis farmers’ realities. Net Journal of Agricultural

Science, 5(4), 131–140.

Checco, A., & Polese, D. (2020). Internet of trees: A vision

for advanced monitoring of crops. In SENSORNETS

2020 - Proceedings of the 9th International Conference

on Sensor Networks.

Davison, R. M., & Martinsons, M. G. (2016). Context is

king! Considering particularism in research design and

reporting. Journal of Information Technology, 31(3),

241–249. https://doi.org/10.1057/jit.2015.19

Ge, Y., Thomasson, J. A., & Sui, R. (2011). Remote sensing

of soil properties in precision agriculture: A review.

Frontiers of Earth Science, 5(3), 229-238.

Gutiérrez, J., Villa-Medina, J. F., Nieto-Garibay, A., &

Porta-Gándara, M. Á. (2013). Automated irrigation

system using a wireless sensor network and GPRS

module. IEEE transactions on instrumentation and

measurement, 63(1), 166-176.

IFAD. (2013). Smallholders, Food Security, and the

Environment. International Fund for Agricultural

Development.

https://www.ifad.org/documents/38714170/39135645/

smallholders_report.pdf/133e8903-0204-4e7d-a780-

bca847933f2e

IPCC. (2014). Climate Change 2014: Synthesis Report.

Contribution of Working Groups I, II and III to the Fifth

Assessment Report of the Intergovernmental Panel on

Climate Change [Core Writing Team, R.K. Pachauri

and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland,

151 pp.

Jawad, H. M., Nordin, R., Gharghan, S. K., Jawad, A. M.,

& Ismail, M. (2017). Energy-efficient wireless sensor

networks for precision agriculture: A review. Sensors,

17(8), 1781.

Kim, H. S., Kumar, S., & Culler, D. E. (2019).

Thread/OpenThread: A compromise in low-power

wireless multihop network architecture for the Internet

of Things. IEEE Communications Magazine, 57(7), 55-

61.

Kim, Y., Evans, R. G., & Iversen, W. M. (2009). Evaluation

of closed-loop site-specific irrigation with wireless

sensor network. Journal of irrigation and drainage

engineering, 135(1), 25-31.

Lavric, A., & Popa, V. (2017, July). Internet of things and

LoRa™ low-power wide-area networks: a survey. In

2017 International Symposium on Signals, Circuits and

Systems (ISSCS) (pp. 1-5). IEEE.

Liaghat, S., & Balasundram, S. K. (2010). A review: The

role of remote sensing in precision agriculture.

American journal of agricultural and biological

sciences, 5(1), 50-55.

Makoye, K. (2013, November 4). SMS weather advice

cushions Tanzanian farmers from drought [Thompson

Reuters Foundation News]. News.Trust.Org.

https://news.trust.org/item/20131104120841-2fm9a/

Morton, J. F. (2007). The impact of climate change on

smallholder and subsistence agriculture. Proceedings of

the National Academy of Sciences of the United States

of America, 104(50), 19680–19685.

https://doi.org/10.1073/pnas.0701855104

Musoni, M. (2020, May 28). How to: Smart Farming in

Rwanda with an IoT-based irrigation system. Digital

Transformation Center Kigali.

https://digicenter.rw/smart-farming-in-rwanda-with-

an-iot-based-irrigation-system/

Nahayo, L., Habiyaremye, G., Kayiranga, A., Kalisa, E.,

Mupenzi, C., & Nsanzimana, D. F. (2018). Rainfall

variability and its impact on rain-fed crop production in

Rwanda. American Journal of Social Science Research,

4(1), 9-15.

Palmer, T., & Darabian, N. (2017). M’chikumbe 212: A

mobile agriculture service by Airtel Malawi. GSMA.

https://www.gsma.com/mobilefordevelopment/wp-

content/uploads/2017/07/M%E2%80%99chikumbe-

212-A-mobile-agriculture-service-by-Airtel-

Malawi.pdf

Perry-Castañeda Library Map Collection (2016). [online]

Available at:

https://legacy.lib.utexas.edu/maps/cia16/tanzania_sm_

2016.gif [Accessed 17 December 2020].

Stewart, R., Erasmus, Y., Zaranyika, H., Silva, N. R. D.,

Muchiri, E., Korth, M., Langer, L., Madinga, N.,

Smart Farming in sub-Saharan Africa: Challenges and Opportunities

163

Randall, N., & Wet, T. (2014). PROTOCOL: The

Effects of Training, Innovation and New Technology

on African Smallholder Farmers’ Wealth and Food

Security: A Systematic Review. Campbell Systematic

Reviews, 10(1), 1–87. https://doi.org/10.1002/CL2.129

Tall, A., Hansen, J., Jay, A., Campbell, B., Kinyangi, J.,

Aggarwal, P. K., & Zougmoré, R. (2014). Scaling up

climate services for farmers: Mission Possible (No. 13;

p. 44). CGIAR Research Program on Climate Change,

Agriculture and Food Security (CCAFS).

Wolfert, S., Ge, L., Verdouw, C., & Bogaardt, M.-J. (2017).

Big Data in Smart Farming – A review. Agricultural

Systems, 153, 69–80.

https://doi.org/10.1016/j.agsy.2017.01.023

Woodhouse, P., Veldwisch, G. J., Venot, J.-P.,

Brockington, D., Komakech, H., & Manjichi, . (2017).

African farmer-led irrigation development: Re-framing

agricultural policy and investment? The Journal of

Peasant Studies, 44(1), 213–233.

https://doi.org/10.1080/03066150.2016.1219719

Wixted, A. J., Kinnaird, P., Larijani, H., Tait, A.,

Ahmadinia, A., & Strachan, N. (2016, October).

Evaluation of LoRa and LoRaWAN for wireless sensor

networks. In 2016 IEEE SENSORS (pp. 1-3). IEEE.

Wouters, B., Barjis, J., Maponya, G., Maritz, J., & Mashiri,

M. A. M. (2009). Supporting home based health care in

South African rural communities using USSD

technology. Association for Information Systems.

WSN4PA 2021 - Special Session on Wireless Sensor Networks for Precise Agriculture

164