From the Field
The Burleigh Dodds Science Blog Series
From the Field
The Burleigh Dodds Science Blog Series
FROM THE FIELD
The Burleigh Dodds Science Blog Series
Utilising artificial intelligence (AI) for effective decision making in agriculture
The major advancements that have been made in computing and technology, coupled with societies’ trust in machine learning, has provided a basis for vast amounts of data to be utilised to improve the rate of production and sustainability within the agricultural sector.
Artificial intelligence (AI) in agriculture
The implementation of artificial intelligence (AI) as a means of improvement has been trialled in several sectors over the last decade. However, only recently has it become apparent that AI can be used to improve decision making in agriculture. In particular, the implementation of AI technologies could provide farmers with the opportunity to make better decisions that increase efficiency in crop and livestock production. Read more
Potential worsening of food insecurity: What to do?
With the recent outbreak of Coronavirus (COVID-19), health systems globally have felt the impact first-hand. However, equally as alarming is the food crisis that could potentially worsen due to the interrupted services in the economic and production sectors. This of course includes food production and food supply chain services.
Read Professor Elhadi M. Yahia's full blog on the impact of Coronavirus (COVID-19) on global food security here.
Proximal crop sensing
Growers have always monitored crops during the growing season to predict yields and to identify signs of stress. However, as modern crop producers we typically manage hundreds or thousands of hectares using modern machines, herbicides and irrigation techniques.
Proximal and remote crop sensing, based on the principles of site-specific management throughout the growing season, can enhance crop yields. Accurate and timely information helps to improve sustainability and reduce the environmental impact of crop production. Crop inputs and the choice of crop cultivar or species can all be informed by our improved understanding of the capacity of the underlying soil.
Remote sensing using aircraft or satellites will inevitably only provide a snapshot of conditions and results can be affected by weather or clouds. Proximal sensing at the crop’s actual location, using unmanned aerial systems (UAS), for example, improves our ability to respond quickly to variations in local conditions and requirements.
Use of proximal crop sensors
Most proximal crop sensors are either passive, using optical reflectance from the canopy and the sun as a light source, or active, using internal modulated light sources.
Soil and water management, crop inputs such as fertilizer, pesticides and tillage can be adjusted when issues are detected. This allows unnecessary generalized actions to be minimized, reducing the risks of degradation in the environment.
Factors that will influence crop yield or quality that can be detected and managed during the growing season are of particular interest. These can include biomass accumulation, a crop’s water status, nutrient deficiencies (particularly nitrogen), disease onset ,and weed or insect infestation. Depending on the crop type, certain points in the growing season are especially critical.
The evolution of crop sensors
Contact or in situ – Typically, crop sensors provide information that allows conclusions to be drawn, rather than measuring specific conditions, although some, such as sap flow detectors, can be attached directly to the plants. However, collecting wider, spatial data using this approach will be time consuming.
Vehicle-mounted crop meters can be used to collect wider predictive information
that can be useful for managing fertilizers, growth regulators or fungicides at the
Ranging sensors – Acoustic, laser or radar range-finding sensors can be used to estimate biomass or crop height. Laser scanners have been used on citrus trees to predict yield, water consumption, health and long-term productivity. Acoustic sensors have been used to estimate the canopy height. Thermal and multispectral sensors have been helpful in detecting water stress in maize crops.
Electromagnetic sensors – Most modern sensors use the electromagnetic (EM) spectrum. Passive sensors can detect reflected, scattered or emitted energy whilst active sensors can use pulsed or modulated energy to detect reflectance or fluorescence from the source.
Crop canopy sensors work in ranges from visible light to infra-red to identify water content and the condition of plant pigments such as chlorophyll. This information can be used to predict specific plant properties or stresses.
One of the most common types of EM sensor is the chlorophyll meter which is useful in predicting nitrogen levels that might require interventions and which can be usefully calibrated to the cultivar type and location. Other applications have combined EM and fluorescent sensors to evaluate leaf pigments as indicators for nutrient and other abiotic stresses. These types of contact sensor can be useful to create averages in small plots to identify the effects of a treatment on chlorophyll levels, for example.
Active and passive mobile electromagnetic sensors – The first mobile EM sensors for field-scale commercial use were based on passive tractor-mounted spectrometers scanning the crop canopy and the ambient light. An algorithm allowed the correct fertilizer rate to be calculated and applied in real time. This approach has been successfully used for wheat crops and the early growth stages of corn.
Cloud cover, the angle of light and time of day can affect the results of this type of passive system. As an alternative, active sensors have been developed using an internal modulated light source, which can be calibrated to the specific cultivar, field conditions and seasonal information to improve the efficiency of nitrogen application. Further active systems with three sensors in the red edge of the light spectrum have proved useful with high biomass canopies such as corn.
Crop canopy sensor limitations – Errors can occur with active EM sensors due to the wavelengths used and vegetation index (VI), the sensitivity of the sensors, height and seasonal variations. It can also be difficult to distinguish between nitrogen and water stress.
Other stresses - The majority of research and commercial applications of proximal crop sensing is for nitrogen fertilizer management for grain crops. However, there is a commercially available system for detecting and managing weeds in real time using an active canopy sensor.
Detecting and responding to diseases and insects are further areas for development using high-resolution multispectral imagery. This has been successfully applied to wheat crops. Portable fluorometers have also been used to detect yellow fluorescence that is indicative of specific diseases in citrus canopies. Although little-used with proximal sensors to date, remote sensors have also been used to detect several varieties of pests.
Multiple sensors - A further area of development is the use of multiple proximal sensors to give greater insight, known as ‘sensor fusion’. Currently used to study soil properties, the information can be integrated with other data to create variable-rate application maps for lime, fertilizer and seed density. The approach has also been used to assess the canopy reflectance, temperature and crop height to manage herbicide applications and stress factors.
An area of significant interest is the use of multiple sensors in laboratory, greenhouse and field investigations to speed up the identification of genetic traits for improved yield and stress tolerance.
Case study: Improved environmental stewardship
In the United States the farmers of Nebraska have steadily improved the efficiency of their use of nitrogen fertilizer since 1968. One measure of success is the mass of grain produced for the mass of nitrogen applied. However, there has been little improvement since 2000 and uptake of active proximal sensing has been slow.
In 2015 a multi-agency in-farm research project was launched to improve the use of nitrogen during the growing season. Active proximal sensors were used for the corn crop canopy. This allowed large-scale comparisons of standard approaches with sensor-based nitrogen management. For the test areas 40kg less nitrogen was used per hectare for the same or better crop yields and improved marginal net returns of between $17 and $19 per hectare.
There is significant potential for research and development of active canopy sensors for use with unmanned aerial systems (UAS). This could provide rapid, more timely spatial surveys of crops and greater use of proximal sensing to identify water, nutrient, disease and insect stresses.
Multispectral fluorescence also has significant potential as an indicator of specific stresses. The mid-infrared and thermal EM regions have not yet been widely studied as indicators of stress, other than water stress, primarily due to the lack of lower cost sensors in these wavelength ranges.
Other types of sensors, for example, pheromone or spore detectors could also be used to highlight insects or diseases. These seem especially suited to UAS use, if detection of these indicators can be easily coupled with their position in a field, allowing rapid mapping.
‘Proximal crop sensing’ features in the new book Precision agriculture for sustainability: Edited by John Stafford, 2019, Burleigh Dodds Science Publishing, Cambridge, UK (ISBN: 9781786762047; www.bdspublishing .com)
Major resistance genes in cereals
Cereal crops, such as wheat, maize, rice, barley, sorghum and millets, account for more than half of the global harvest and provide staple foods around the world.
However, viruses, bacteria, water moulds and fungi can limit access to nutrients, reduce yields and can even cause entire crops to fail. Some diseases can also produce toxins that are harmful to humans and animals. To protect food security, identifying disease resistant genes is crucial.
Major resistance genes
The category of disease resistance genes known as ‘major’ are important because they completely block pathogen development and are evident at the seedling stage. These can be more easily mapped than partially resistant or quantitative resistance genes.
Cloning these major resistance genes can be used to identify diagnostic molecular markers for durable disease resistance in the field. This provides a starting point to study the molecular basis of disease resistance that can be transferred across cultivars or species to improve crop yield.
Cereal plants have naturally evolved genes to restrict or slow down diseases. Plant breeders have used these traits to develop high levels of resistance in cereal plants.
Descriptions and genetic maps of these cultivars are large-scale with hundreds of candidates for resistance. Cloning and in-depth study at the molecular level has only taken place in recent years to identify specific genetic codes for disease resistance.
The challenge of gene cloning in cereals
The DNA of many cereals is larger and more complex than the human genome, so recent technological advances have been needed to make cloning faster, cheaper, and more practical.
The first disease resistant gene to be cloned (HM1) was for resistance to a fungus affecting maize (HCTR). In this case, researchers were already aware of the existence of the fungal toxin and a detoxifying maize enzyme.
In contrast, a resistance gene (Xa21) for a bacterial disease in rice was identified without any prior knowledge. Resistance for all Indian and Philippine rice species was provided by a resistant gene from African wild rice, but the process took 10 years.
This illustrates the challenges of time and complexity that have been inherent in gene cloning for cereals to date.
Electromagnetic sensors – Most modern sensors use the electromagnetic (EM) spectrum. Passive sensors can detect reflected, scattered or emitted energy whilst active sensors can use pulsed or modulated energy to detect reflectance or fluorescence from the source.
Current gene cloning approaches
New methods for sequencing, gene capture and gene complexity reduction are making the identification, cloning, and effective use of resistance genes more effective.
Faster trials - Mapping and growing multiple generations of plants with possible resistant traits has typically been a long process. It can now take place considerably more quickly in controlled greenhouse conditions that allow several generations to be grown in a year rather than in seasonal field conditions.
Improved reference material – Although genome reference sequences for rice and maize have been available for several years, more complex sequences for wheat and barley have only recently been completed. This will make mapping and isolation of resistant genes much faster. However, gene content differs between different genotypes of the same cereal species, so a resistance gene might not be present in the reference cultivars.
Reduced complexity - Messenger ribonucleic acid, or mRNA, can be used to identify parts of the genome that will not contribute to identifying resistance genes. This allows studies to focus on differences between resistant and susceptible species, with and without the target disease present.
Bulk sequencing and mapping – An approach known as MutMap uses whole genome resequencing to quickly isolate genes responsible for specific mutations. Plants with specific characteristics can then be produced by crossing interesting mutations with other cultivars. These can be grown in bulk and compared with the original wild species. The MutMap+ approach doesn’t rely on crossing and is especially helpful in identifying mutants that impact reproduction, including sterility or other mutations that prevent crossing.
These approaches are successful with rice, which has a relatively small genome, but aren’t currently suitable for the more complex genomes of wheat or barley, for example.
Targeted mapping - Chromosome flow sorting has been used to simplify identification of resistance genes in more complex cereals, although its accuracy can be limited. To overcome this limitation, high-resolution long-range sequence assembly can now be used to allow additional markers to be identified quickly and cost-effectively. When used with other techniques, some time-consuming processes will no longer be needed.
Resistant gene enrichment - The natural ability of cereal plants to detect diseases and trigger immune responses is a growing area of interest. Gene enrichment sequencing allows genes that are likely to be associated with diseases to be identified without mapping the whole gene. This is a faster process, although currently limited to certain types of naturally occurring resistance genes.
Wider diversity - Association genetics and complexity reduction techniques can be used to isolate resistant genes in a wider population without a reference sequence. This can significantly reduce the time needed to clone a resistance gene, although significant time is needed to create a suitably diverse population of plants.
Cloning most of the major cereal disease resistance genes could be possible in the near future. Further work is needed to understand the most efficient use of resources, looking at how different techniques can be combined and how resistance genes can be more effectively isolated. However, comprehensive breeding programmes are needed to create new crop cultivars with greater disease resistance.
Case study: Combining classical and novel resources
Septoria tritici blotch (STB) is one of the most important fungal diseases of wheat in Europe and other wheat-growing areas. Because of the sexual reproduction of the fungus, genetic variability allows fungicide resistant strains to evolve.
Many wheat cultivars have already been bred with a specific resistance to the fungus. These can now be cloned and used much more quickly thanks to recently developed techniques such as target interval mapping.
Proteins have been identified in this resistant gene that are now recognised as important components of disease resistance in cereals. Another gene has also been discovered that affects the virulence of the disease. It is not yet known whether these genes effectively work together, but they offer valuable options for further development of a plant disease resistance gene.
‘Mapping and isolation of major resistance genes in cereals’ features in the new book Advances in breeding techniques for cereal crops: Frank Ordon and Wolfgang Friedt (eds), 2019, Burleigh Dodds Science Publishing, Cambridge, UK (ISBN: 978-1-78676-244-3; www.bdspublishing .com)
Spray technologies for field crops
Our approach to the use of crop protection products has changed very little in the last decade, despite significant research, even though they can amount to as much as half the overall cost of production.
With growing concerns about their potential environmental impact, using these products efficiently and effectively has never been more important.
A new paper* Spray technology in precision agriculture from the award-winning crop spraying expert, Dr Paul Miller, examines the findings of recent research and looks to future trends that could offer farmers more options in the use of these products.
The precision agriculture approach
While Dr Miller’s review focusses on field crops, the principles of precision agriculture can also apply to other bush, tree, glasshouse and specialist crops.
Precision agriculture depends on controlled product applications to maximise effectiveness and minimise losses in non-target areas; effectively applying different treatments only where they are needed and successfully targeting single plants or small groups of plants.
The application equipment and its speed, product application pressure, droplet size and weather conditions can all affect results.
Controlling the delivered dose
In most conventional crop spraying systems nozzle size and flow rate are used to control the dose volume and concentration. Changes in speed are usually managed by adjustments to the pressure at the nozzle, but only limited adjustment options are available.
To improve accuracy at a wider range of speeds a number of approaches have been tested. These include on/off valves with rapid response times that allow more precise targeting. To give a wider range of outputs, multiple nozzles in a single holder can allow single or multiple switching.
More sophisticated arrangements can be used to support a wider range of adjustment to both nozzle output and droplet size at lower flow rates. This can help to reduce the risk of products drifting into neighbouring areas.
Another approach to varying dose concentrations is to store water and protection products separately until the point of application.
When specific areas of weeds, diseases or pests need to be treated in a wider crop, spatial targeting is needed. For stable patches of grass in arable crops, straightforward mapping can allow accurate use of herbicides. However, for more varied problems, investment in accurate mapping and complex equipment can outweigh the benefits, meaning that the current minimum resolution is at least 4m.
With spraying boom lengths increasing the need for sophisticated solutions that allow variable pressure control or dose concentrations for different boom sections are likely to be needed in the future.
Characteristics of sprays for different crop types
Various nozzle types will change spray quality and droplet size. This can affect the amount of product retained on different crop and weed types. For example, trials have shown more effective control of grass weeds in cereal crops when treated with finer sprays. However, the increased risk of drift could restrict this approach to patches of grass weeds only.
Minimising drift in the wider environment
Buffer zones at the edges of cropped areas are now a well-established control measure for unwanted exposure to pesticides.
One way to reduce the width of buffer zones is to use nozzles and spraying arrangements that have been shown to reduce drift. Implementing treatment zones across a field or plot will allow drift-reduction approaches to be focused in the margins to meet environmental protection requirements without compromising wider crop treatment.
Case study: Spot treatment to control volunteers
Volunteer potatoes remaining from the previous year’s planting in crops such as onions, leeks and carrots can seriously affect yield, quality and harvesting.
European regulations now mean that the herbicides traditionally used for repeated overall spraying to control this problem are no longer available. Spot treatment is a possible alternative.
A project to develop an effective spot spraying approach used a camera system to identify the outline of weeds. This information allowed spray nozzle schedules to be defined. The nozzles in the prototype system were able to apply existing herbicide formulations to spots less than 100mm square.
Overall results from field trials in a range of crop, weed and weather conditions showed that the prototype results were comparable with alternative approaches.
Dr Miller concludes that the costs and complexity of equipment and mapping have delayed the uptake of precision agriculture systems. However, in the last decade, technical advances have reduced the costs and improved performance and reliability.
Options for the future include the use of autonomous robots to apply chemical and non-chemical treatments to individual plants. This could be suitable for high-value crops grown in small areas.
Real-time detection of weeds using ultrasonic detectors to identify height differences could also allow more effective spot spraying.
Use of computer-based control systems is likely to increase, managing a wide range of parameters to produce treatment maps for individual fields and control algorithms for delivery systems. Drift risk could be managed automatically, taking account of both crop and weather conditions. As an additional benefit, compliance requirements could all be met automatically.
*The paper Spray technologies in precision agriculture by Paul Miller, Silsoe Spray Applications Unit Ltd, UK, appears in the new book Precision agriculture for sustainability, Stafford, J. (ed.), published by Burleigh Dodds Science Publishing Limited, 2019.
Climate change and cocoa cultivation
Climate change is likely to affect the delicate balance of growing conditions for cocoa. The possible implications of global changes in temperature and rainfall for cocoa production have now been examined by a team of international researchers*. Their work is featured in the new book Achieving sustainable cultivation of cocoa.
*Christian Bunn, Fabio Castro and Mark Lundy, International Center for Tropical Agriculture (CIAT), Colombia and Peter Läderach, International Center for Tropical Agriculture (CIAT), Vietnam
The study looked at almost 3000 carefully defined ‘cocoa occurrence’ locations. Its findings are the basis for informed discussion about sector-wide adaptations to protect future cocoa production. Areas where policy-makers should expect to look for alternative crops in the future are also highlighted.
The economic importance of cocoa
Suitable growing conditions for cocoa are concentrated in some of the world’s poorest countries. Around half of the world’s cocoa is produced in just two West African nations; Ivory Coast and Ghana. Other African countries produce a further 20% and most of the rest of the world’s production comes from the tropical Americas and Southern Asia.
Globally about ten million hectares are devoted to cocoa growing. The annual yield of 4.5 million metric tons is mostly exported to North America and Europe, making a significant contribution to national economies.
While global circulation models (GCMs) are unable to give firm predictions of future climate conditions, change is increasingly certain. A ‘do nothing’ response assumes that these changes will not have economic consequences. There is also resistance to investment in adaptive changes that would mitigate any effects.
The study highlights that anticipated climate changes are likely to have a significant impact on cocoa production, making cost-effective adaptation a vital challenge for the sector.
The impact of climate change on global cocoa production
There is a narrow temperature range and a fixed upper limit for cocoa growing. Recorded global temperatures have increased 0.85°C in the last 130 years.
However, the pace of change is increasing. By the end of this century temperatures are expected to rise by a further 1.4°C to 3.1°C and there will be an increased risk of drought.
While GCM projections suggest that total rainfall will also increase during this period, higher temperatures are likely to outweigh any benefits.
In addition, much of the available literature examines historic climate patterns. It does not take in to account the increasing frequency of currently rare temperature changes. These could affect the resilience of the cocoa crop and have unpredictable implications for pests and diseases of the cocoa plant.
Overall, the study concludes that there will be a decline in areas with suitable conditions for viable cocoa production in the coming decades.
How climate change could affect cocoa production
The latest available figures are used in the study to examine the impact of climate change on the most significant aspects of cocoa production:
• the length and maximum temperature of the dry season
• total water deficit during the dry season due to an imbalance between precipitation, evaporation and plant transpiration
• rainfall and average temperature during the growing season.
The impact of climate change will vary regionally and throughout the growing season, so the study specifically looks at the implications for the world’s largest cocoa producer, Ivory Coast. Here drought and water deficit were more pronounced than in other cocoa producing regions. This highlights the urgent need for adaptive measures, crop migration and a requirement for new crop types to be identified.
In contrast, South American regions could be less affected thanks to greater
precipitation along the Andes. In these areas cocoa growing could be migrated to higher ground to mitigate the effect of higher temperatures.
Cocoa growers’ responses to climate change
Climate smart cocoa (CSC) shows how sustainable practices can increase income and production on existing farms whilst also adapting to the effects of climate change and mitigating the risks.
One solution is to improve soil health with organic material. This would boost disease resistance and increase productivity with better nutrients. It could also enhance moisture retention during the dry season and improve crop resilience.
Another option is to evaluate the type of shade trees and level of shade needed. This could change over time and vary in different areas, depending on how climate change progresses.
In conclusion, the authors call for good practices to be shared amongst larger growers that will ultimately support the smallholders who rely on income from cocoa production. They also highlight the importance of a coordinated approach throughout the cocoa value chain to deliver a system-wide response to climate change.
To find out more see Chapter 19 of the new book from Burleigh Dodds Science Publishing: P. Umaharan (ed.), Achieving sustainable cultivation of cocoa, 2018, Burleigh Dodds Science Publishing, Cambridge, UK (ISBN: 978-1-78676-168-2; www.bdspublishing.com).
Sustainable responses to non-infectious disorders
Each year the potato harvest is affected by non-infectious disorders that are not passed between the plants or tubers. They have a significant impact on product quality and profitability. However, we still have a lot to learn about detecting these disorders, predicting when they will occur and how they can be prevented.
The research of Dr Andy Robinson of North Dakota State University and the University of Minnesota focuses on science-based solutions to address real-world problems like this. His aim is to help producers increase economic and environmental sustainability through improved crop management.
In his new article* Non-infectious disorders affecting potatoes he has created a comprehensive review of the most common disorders. He highlights research to date and looks at ways to improve our understanding and management of these conditions.
He acknowledges that one of the major difficulties is that similar combinations of conditions do not always produce the same results. Multiple research studies will be needed to identify the causes and potential solutions. Here are some of the key points he discusses, where further research could deliver important benefits.
Bruising and skinning
Bruising and skinning can happen throughout the harvesting, sorting and transportation of the potato crop. Once they are damaged we know that tubers are more vulnerable to infection.
Research is helping us to understand how mechanical processing, such as soil carried through the harvesting equipment, can cause this damage.
It is also known that hydration affects the likelihood of bruising and skinning. Genetic techniques and carefully selected plant types can improve the ability of potatoes to heal wounds, enhancing product quality. Optimizing the desiccation process through carefully controlled fertilization and improvements to the plant’s use of nutrients could also be helpful.
Enlarged sprouts below the soil surface, known as coiled sprout, are often split, cracked or indented. There is evidence that this can be the result of early planting, soil type and plant selection. Research indicates that ethylene is a probable cause, alongside the use of herbicides.
High soil temperatures that shock the potato plant can lead to heat crinkle in leaves, sometimes looking like herbicide damage. This leads to multiple shoots part way along the stem that can affect overall yield. Management of this condition could be improved through better understanding of the relevant temperature ranges and the plant types that are most likely to be affected.
Cracking, when the core of the potato grows more quickly than the skin, is one of the most common disorders, especially for potatoes grown without irrigation. Causes include water stress during early growth, excessive or badly timed nitrogen fertilization during mid-growth, the use of herbicides, physical damage and disease. Improved field diagnostic tools and a better understanding of the combined impact of soil preparation, plant spacing, irrigation, fertilization and the use of herbicides could improve management of this condition.
Some conditions make potatoes more vulnerable to disease. For example, research into lentical spots, which can provide access to other diseases, has focused on the bacteria known to cause soft rot, which is most commonly found around the lenticals. A better understanding of optimal growing and storage conditions is also needed to tackle this problem.
Other conditions reduce the visual appeal of the product. For example, greening could be reduced by using unmanned drones to visually identify tubers that are close to the surface, where this disorder is likely to occur. Identifying improvements to bedding and ridging techniques and suitable storage materials could also be helpful.
Like greening, purple or pink discoloration is also thought to be caused when tubers grow close to the soil’s surface. More detailed work is needed to understand the impact of climate conditions on popular red and purple potato types. However, consumers should be aware that the substance causing this condition is naturally found in the skin and flesh of these potatoes and is harmless.
Discoloration due to stem end disorder can often be seen during cooking. It is typically due to heat stress, drought stress or fertilization management. Plant breeding programs and the addition of native potato DNA have already successfully reduced stem end problems.
Tubers with tan to brown spots caused by high temperatures or low moisture, will often have firm tissue even after cooking. New resistant cultivars are being developed and genetic options are being evaluated.
Rotten tissue and cavities in larger potatoes leads to significant waste in potato production. Lack of oxygen during growth or storage, harvesting in muddy conditions and temperature are contributory factors. The timing of vine removal to allow skin setting is also critical. Incubating potato samples to induce blackheart could help to identify techniques to prevent this problem.
Recent studies have shown how we can improve our understanding and develop new ways to manage non-infectious disorders so that we can maintain sustainable potato production.
Research to understand genetic traits and the effects of silencing certain genes in different environments is also showing promise. This could become part of a potato breeding program to help growers choose the best potato type for their conditions.
Unmanned drones could also become useful tools to identify the visual signatures of non-infectious disorders, allowing early intervention.
*The article Non-infectious disorders affecting potatoes by Andrew P. Robinson of North Dakota State University and University of Minnesota, USA, is published in Volume 2 of the book Achieving sustainable cultivation of potatoes published by Burleigh Dodds Science Publishing Limited, 2018.
Understanding the banana industry: monoculture and beyond
A decade ago Dan Koeppel’s book ‘Banana: The Fate of the Fruit That Changed the World’ gave us a new perspective on the history of cultivation and politics of the banana. He has now written a new insightful article that challenges us to look carefully at the future of one of our favourite fruits
At the time Koeppel was writing ‘Banana’, the world’s leading marketer, Chiquita, was expanding its operations in Mozambique. However, a decade later the business is all but abandoned due to the arrival of a disease called fusarium wilt tropical race 4 (Foc-TR4).
Foc-TR4 emerged in Malaysia in the late 1980s and spread to Asia, the Chinese mainland, the Philippines, and then Australia. In all of these areas it has devastated production of commercial banana variety, the Cavendish. It hasn’t yet arrived in Latin America where most of our bananas are grown, but scientists predict that it will.
A short history of the banana industry
A ‘banana equation’ has existed in the industry since the nineteenth-century. It demands high productivity, a short production cycle, ease of shipping, slowness of ripening and the ability to grow on large plantations.
The original commercial banana, grown in Central and South America, the Gros Michel, became affected by a fungus named the ‘Panama disease’, where it first appeared. Gros Michel could no longer be grown in areas affected by the disease. Banana companies quickly started looking for new land, which had far-reaching consequences.
For example, in Colombia, striking banana workers asked for improved conditions. United Fruit, now known as Chiquita, launched a media campaign. Identifying the strikers as a ‘subversive movement’. In Colombia, martial law was eventually declared. The following day, banana workers and their families were shot by soldiers as they left church and one thousand died. These events are described in the Nobel-prize-winning novel by Gabriel Garcia, ‘One Hundred Years of Solitude’.
Although bananas are more popular than apples and oranges combined, the market has not diversified in the same way as apples or citrus fruits. Even from the 1880s the commercial fruit companies focused on a single banana variety so that growing and shipping methods could be standardised and consumers would receive a consistent product.
Although Panama disease relentlessly progressed, destroying plantations and local economies, new land was simply taken to continue banana cultivation.
Many governments in banana growing countries cooperated and, in some cases, force was used to grab the necessary land. The only pause was during World War 2, when the banana transport ships were commandeered.
A new era
After the war Chiquita and Standard Fruit, now known as Dole, were the major banana growers and exporters. Both knew that their current banana crop was under threat.
A new, resistant banana variety to replace Gros Michel was needed. It would involve changing how plantations were organised and finding new ways to harvest, ship, distribute and ripen the bananas.
Despite these changes, the commercial companies didn’t want to give up their land. In Guatemala, Jacobo Arbenz was elected President in 1950 with the goal of making it an economically independent country. With Chiquita’s high-level links throughout the United States, the company started a three-year campaign to oust Arbenz and replace him a more suitable alterative.
By 1960 Dole had adopted a new banana variety called the Cavendish, which was resistant to Panama disease.
Chiquita believed the new variety was inferior. It was more delicate and needed to bagged and boxed for transport. It ripened poorly and needed special gases in temperature-controlled cargo holds to keep the fruit green during transport. It also had a different taste to Gros Michel.
However, there was wasn’t another viable alternative. Plantations were replaced with production lines where women were employed for the first time to clean and separate the fruit and to add branded stickers to every bunch.
With convenience becoming the top priority for modern consumers, the shortcomings of the new variety weren’t the major problems Chiquita had anticipated.
Today, the only choice is generally the Cavendish because it meets the requirements of the banana equation and is resistant to Panama disease. In many African countries it’s a staple food.
Banana growing today
The banana industry has migrated from Central America to more southern countries because they are bigger and less susceptible to crop-destroying hurricanes.
Labour and environmental standards have improved, and plantation ownership has been replaced by a network of subcontractors and independent suppliers.
Unfortunately, the Cavendish is not immune to the new strain of Panama disease, Foc-TR4, which already affects plantations in China, the Philippines, Australia and Mozambique.
On multiple plots in Mozambique farms, shared water supplies were identified as one of the primary ways Panama wilt destroyed the Gros Michel plantations in the mid-twentieth century. However, despite modern quarantine and clean farming measures, the new disease continues to spread.
Because cultivated bananas are seedless and can only be reproduced by cuttings, they can’t be archived in ‘seed vaults’. With the increasing risk of disease, banana plant collections growing in fields are at risk.
Continued monoculture - Chiquita and Dole are working on Cavendish replacement strategies, but so far there isn’t an obvious replacement that has the ‘banana equation’ qualities needed for large-scale monoculture.
Chiquita is experimenting again in Mozambique, reportedly for the Middle Eastern market. With little commercial banana growing in Africa it could become a new growing zone if crops in Latin America become affected by Foc-TR4.
Although the Philippines is affected by Panama disease it has an alternative variety called Lacatan. The Pilipino Banana Growers and Exporters Association says that shipments of this alternative variety could be available before 2020.
Experimental work in the Philippines aims to reselect Cavendish plants that have survived Panama disease and could retain tolerance. Modified Cavendish plants are also being conventionally bred in Brazil and Japan.
Ironically Chiquita has contracted with the Fundación Hondureña De Investigación Agrícola (FHIA), which it once owned, to begin developing new disease-resistant Cavendish-like varieties.
Dole, operating with a grant from the Gates Foundation, has developed Foc-resistant cultivars of ladyfinger, which is popular in Australia. A very different fruit, ladyfinger is unlikely to be a straightforward replacement for the Cavendish. Rather, it is hoped that the technology used to create this resistance could be transferred to the Cavendish or Gros Michel varieties.
New varieties - There is a growing understanding that replacing one monoculture with another may not be the way forward.
Alternative sources for new varieties could include India, where more than half of the world’s banana varieties grow, although some of these are also susceptible to Foc-TR4.
Many of these varieties look and taste very different and might not survive the export conditions and journey lengths required for commercial crops. However, they could contain genetic material for breeding disease-resistant fruits.
Some of these alternative varieties might be genetically modified, but there could also be multiple ‘gourmet’ or heirloom types sold at higher prices with greater profit margins.
Modern containerisation techniques could provide the differing conditions needed and make transport of new varieties around the world feasible. For example, Cavendish bananas were shipped from Mindanao in the Philippines to the port of Los Angeles in 2013, which is the longest journey commercial bananas have made. Despite an extra week at sea the consignment was in good condition.
Finally, news that Amazon bought the US-based Whole Foods supermarket chain could add a whole new dimension to the market. Whole Foods was already promoting a more sustainable (although still Cavendish-based) banana business model. Now, being owned by the world’s largest online distributor of goods, the challenge of shipping more banana varieties from around the world could be solved.
Understanding the banana industry: monoculture and beyond by Dan Koeppel, Independent Journalist and Researcher, USA is taken from: Kema, G. H. J. and Drenth, A. (eds.), Achieving sustainable cultivation of bananas Volume 1: Cultivation techniques, Burleigh Dodds Science Publishing, Cambridge, UK, 2018, (ISBN: 978 1 78676 156 9; www.bdspublishing.com)