In the last five years or so, the total volume of investments on the agricultural sector has grown by a massive ~80%. According to experts, precision agriculture (the technique of optimizing existing inputs and fertilizers, tillage tools, fields and crops, for the purpose of improved control and measurement of farm yields) has the potential of playing a key role in meeting the incremental food demands of the growing population worldwide. A recent report estimated the value of the global precision farming market at the end of this decade at around $4.6 billion – with the CAGR between 2015 and 2020 being just a touch under 12%. In the United States alone, the market for smart agriculture software is likely to jump by more than 14% between now and 2022. However, the actual growth and proliferation of precision farming has not been as robust as was expected earlier. The sector faces several key challenges, and we turn our attentions on them in this post:
Interoperability of different standards
With more and more OEMs coming up with new and innovative agricultural IoT tools and platforms, interoperability is rapidly becoming a point of concern. The various available tools and technologies often do not follow the same technology standards/platforms – as a result of which there is a lack of uniformity in the final analysis done by end users. In many instances, the creation of additional gateway(s) becomes essential, for the translation and transfer of data across standards. As things stand now, precision agriculture (while evolving rapidly) is still, to a large extent, fragmented. The challenge lies in transforming the smart standalone devices and gateways to holistic, farmer-friendly platforms.
The learning curve
Precision farming involves the implementation of cutting-edge technology for bolstering crop growth. For the average farmer, setting up the necessary IoT architecture and sensor network for his/her field(s) can be a big ask. It has to be kept in mind that the room for error in a tech-upgraded ‘smart farm’ is minimal - and faulty management (a wrongly pressed valve here, forgetting to switch off the irrigation tank there, etc.) can be disastrous. Getting farmers thoroughly acquainted with the concept of smart farming, and the tools/devices involved in it, is of the utmost importance – before they can actually proceed with the implementation. Lack of knowledge can be dangerous.
Connectivity in rural areas
In many remote rural locations across the world (particularly in the developing countries, although several locations in the US suffers from this as well), strong, reliable internet connectivity is not available. That, in turn, thwarts the attempts to apply smart agriculture techniques at such places. Unless the network performances and bandwidth speeds are significantly improved, implementation of digital farming will remain problematic. Since many agro-sensors/gateways depend on cloud services for data transmission/storage, cloud-based computing also needs to become stronger. What’s more, in farmlands that have tall, dense trees and/or hilly terrains, reception of GPS signals becomes a big issue.
Making sense from big data in agriculture
The modern, connected agricultural farm has, literally, millions of data points. It is, however, next to impossible to monitor and manage every single data point and reading on a daily/weekly basis, over the entire growing seasons (neither is it necessary). The problem is particularly bigger in large, multi-crop lands and when there are multiple growing seasons. The onus is on the farmers to find out which data points and layers they need to track on a regular basis, and which data ‘noise’ they can afford to ignore. Digital agriculture is increasingly becoming big data-driven – but the technology is helpful only when users can ‘make sense’ of the available information.
Non-awareness of the varying farm production functions
In-depth economic analysis needs to complement internet tools, to ensure higher yields on farms. Users need to be able to define the correct production function (output as a function of key inputs, like nutrients, fertilizers, irrigation, etc.). Typically, the production function is not the same for all crops, differs in the various zones of a farm, and also changes over the crop/plant-growth cycle. Unless the farmer is aware of this varying production function, there will always remain the chance of application of inputs in incorrect amounts (spraying too much of nitrogen fertilizer, for example) – resulting in crop damages. Precision agriculture is all about optimizing output levels by making the best use of the available, limited inputs – and for that, the importance of following the production function is immense.
Size of individual management zones
Traditionally, farmers have considered their entire fields as single farming units. That approach is, however, far from being effective for the application and management of IoT in agriculture. Users have to divide their lands in several smaller ‘management zones’ – and there is quite a lot of confusion regarding the ‘correct’ size of these zones. The zones have to be divided with respect to the soil sampling requirements (different zones have varying soil qualities) and fertilizer requirements. The number of zones on a field, and their respective sizes, should depend on the overall size of the growing area. There is not much of reference work for the farmers to go by, while trying to divide their lands in these zones. As an alternative, many farmers continue to follow uniform fertilizer application and/or irrigation methods for the entire farm – leading to sub-optimal results.
Barriers to entry for new firms
Although precision farming has been a subject of considerable interest for several years now, the concept is still relatively ‘new’. As such, the big hardware/software manufacturers that entered this market at an early stage still have a definite ‘first-mover advantage’. The lowly competitiveness of the market can prevent new firms from entering this domain – with the existing big firms retaining a stranglehold. Farmers can also face problems while trying to migrate data streams from an older platform to a newer one, and there are risks of data loss. The resources and platforms provided by a big player in the agro-IoT sector might not be compatible with those provided by a smaller OEM – and that might prevent the latter from having enough clients.
Lack of scalability and configuration problems
Agricultural farms can be of different sizes. A single owner can have a large crop-growing land, along with several smaller lands. In India, nearly 33% of the total area under agriculture is accounted for by only 5% of the total number of farms – clearly highlighting the uneven nature of farm sizes over here. A farmer needs to be provided IoT tools (access points, gateways, etc.) that are completely scalable. In other words, the same technology should be applicable, and the same benefits should be available, on a large commercial farm as well as a small piece of personal garden/crop land. The need for manually configuring the setup and the devices is yet another probable point of concern. For agriculture to become truly autonomous, the technology should be self-configurable. The recent surges in artificial intelligence and M2M learning opens up the possibility for that.
Energy depletion risks
A lot has already been written about the environmental advantages of switching over to smart agriculture (precision farming is ‘greener’). However, the need for powerful data centers and gateways/hubs for the operation of the smart sensors and other gadgets can lead to heavy energy consumption – and more resources are required to replenish that energy. What’s more, the creation of new agricultural IoT tools also has an effect on the energy sector. Not surprisingly, companies have started to focus on farming technology platforms which do not cause too much of energy depletion…but there is still am long way to go in this regard.
Challenge for indoor farming
Most precision agriculture methods and resources are optimized for conventional outdoor farming. With the value of the global vertical farming industry projected to go beyond $4 billion by 2021, more attention has to be given on technology support for indoor farming. The absence of daily climatic fluctuations and regular seasons have to be taken into account, while coming up with smart indoor farming methods. The nutritional value of the outputs must not get adversely affected in any way either. Farmers need to be able to rely on the technology to create the optimal growing environment (light, temperature, water availability) for indoor plants.
Technical failures and resultant damages
The growing dependence of agriculture (or anything else, for that matter!) on technology comes with a potentially serious downside. If there is a mechanical breakdown in the hardware, or a farming IoT unit/sensor malfunctions – serious crop damages can be the result. For example, in case the smart irrigation sensors are down, plants are likely to be underwatered or overwatered. Food safety can be compromised, if the technological resources in the storage area(s) are not functioning. Even a few minutes of downtime due to a power failure can have serious consequences – particularly when backup power is not available.
Farms powered by smart technology have (in various extents) done away with the problems of runoff, contamination, and other channels of ecological damages. Carbon dioxide emissions have been brought down significantly (~2.0 GHt in a five-year span) as well. A new risk has cropped up though – in the form of electronic wastes (e-wastes). In 2013, the total volume of such wastes was in excess of 52 million metric tons – and the piles of discarded IoT tools and computers and outdated electronic devices are compounding this problem further. In a nutshell, the regular hardware upgrades are making the older units obsolete – and in many areas, dumping them is causing landfills. For things to be sustainable, proper arrangements for the disposal of e-waste have to be made. Soon.
Loss of manual employment
On average, 4 out of every 10 members of the global workforce are employed in the primary sector. The figures are particularly high in Oceania, Africa and Asia. As IoT in agriculture becomes more and more mainstream and things become automated – a large percentage of this agricultural labour will lose their jobs. The other sectors need to have the capacity to absorb this workforce (now rendered jobless) – and in many of the developing/underdeveloped countries, the economy is not strong enough for that to happen. There is no scope for doubting the benefits that precision agriculture brings to the table – but the large-scale displacement of manual workers can lead to dissatisfaction among people.
The security factor
The presence of malware and data thefts is a risk in practically all types of ‘connected systems’, and smart agriculture is not an exception from that. As the count of middleware technology, endpoints and IoT devices in active use in agriculture is increasing, the number of entry-points for malicious third-party programs is going up as well. Since the third-party attacks on a complex IoT system are often decentralized, detecting and removing them emerges as a big challenge. The situation becomes more complicated due to the propensity of many farmowners to opt for slightly cheaper devices and resources, which do not come with the essential safety assurances. The multiple software and API layers can cause problems as well. There is an urgent need for tighter security and provisioning policies for agricultural IoT – to make it more acceptable for users.
Benefits not immediately apparent
To get the motivation to invest on a ‘new technology’ like smart farming, users (understandably) would want to get an idea of the ROI from this technology. Unfortunately though, there is almost no way to guesstimate the benefits of precision farming over the long-run – and the benefits do not become apparent from the very outset. For this very reason, many landowners still view the use of advanced technology in agriculture as ‘risky’ and ‘uncertain’, and stay away from adopting it. With greater familiarity with agritech and comprehensive training, such fears should go away.
Smart gadgets that merely provide information about the extent of crop damages are of little use – and there is need for more ‘predictive maintenance’ tools, that would be able to anticipate damages, and help farmers avoid the same. Customization of the sensors and resources to meet the varying nutrient/water/pest control requirements of different plants is a challenge, as is getting together and comparing data from multiple farms. Farmers need to have a complete knowledge of the correct ‘nutrient algorithms’, so that the platforms/gateways can be configured optimally. There is also room for cutting down the rather frequent ‘yield map errors’, which lead to faulty output estimates.
The concept of precision agriculture is based on four pillars – Right place, Right source, Right quantity and Right time. It has already made a difference to agriculture and farm yield performance worldwide…and once the aforementioned challenges are overcome, its benefits will become more evident, more sustainable.
Latest posts by Hussain Fakhruddin (see all)
- AI-as-a-Service: All That You Need To Know - October 30, 2018
- Top 12 Trends In Digital Transformation To Watch Out For In 2019 - October 23, 2018
- Machine Learning in 2019: Tracing The Artificial Intelligence Growth Path - October 9, 2018