Tag: big data

Profitability under different fertilization recommendation scenarios in Ethiopia and Tanzania, measured in U.S. dollars per hectare.

How the data revolution could help design better agronomic investments

Written by Jérôme Bossuet on April 2, 2019. Posted in Features.

What fertilizer application will give me the best returns? What maize crop variety should I use?

Each farmer faces constraints related to weather uncertainty, soil fertility management challenges, or access to finance and markets. To improve their yields and incomes, African smallholder farmers need agronomic advice adapted to their specific circumstances. The challenge is even greater in sub-Saharan Africa, where agricultural production landscapes are highly diverse. Yet traditional agronomic research was not designed to fit with complex agroecological regions and farming systems. Compounding the problem, research organizations often have limited resources to develop the necessary experiments to generate farm- and site-specific agronomic advice at scale.

“Agronomic research is traditionally not equipped to consider spatial or socio-economic diversity among the millions of farmers it targets,” said Sebastian Palmas, data scientist at the International Maize and Wheat Improvement Center (CIMMYT) in Nairobi, Kenya.

Palmas presented some of the learnings of the Taking Maize Agronomy to Scale in Africa (TAMASA) project during a science seminar called “A spatial ex ante framework for guiding agronomic investments in sub-Saharan Africa” on March, 4, 2019.

The project, funded by the Bill & Melinda Gates Foundation, has used data to improve the way agronomic research for development is done. Researchers working on the TAMASA project addressed this challenge by using available geospatial information and other big data resources, along with new data science tools such as machine learning and Microsoft’s AI for Earth. They were able to produce and package information that can help farmers, research institutions and governments take better decisions on what agronomic practices and investments will give them the best returns.

By adapting the Quantitative Evaluation of the Fertility of Tropical Soils (QUEFTS) model to the conditions of small farmers in TAMASA target countries (Ethiopia, Nigeria and Tanzania), using different layers of information, CIMMYT and its partners have developed a versatile geospatial tool for evaluating crop yield responses to fertilizer applications in different areas of a given country. Because calculations integrate spatial variation of fertilizer and grain prices, the tool evaluates the profitability — a key factor influencing farmers’ fertilizer usage — for each location. The project team can generate maps that show, for instance, the estimated agronomic and economic returns to different fertilizer application scenarios.

The TAMASA team plans to publish the code and user-friendly interface of this new geospatial assessment tool later this year. (Photo: CIMMYT)

Making profits grow

These tools could potentially help national fertilizer subsidy programs be more targeted and impactful, like the ambitious Ethiopia’s Fertilizer Blending initiative which distributes up to 250,000 tons of fertilizer annually. Initial calculations showed that, by optimizing diammonium phosphate (DAP) and urea application, the profitability per hectare could improve by 14 percent on average, compared to the current fertilizer recommendations.

Such an approach could generate farm-specific advice at scale and boost farmers’ incomes. It could also provide insights on many different issues, like estimating market demand for a new fertilizer blend, or the estimated quantity of additional fertilizer required to bring about a targeted maize yield increase.

Future extensions of the framework may incorporate varietal differences in nutrient management responses, and thus enable seed companies to use the framework to predict where a new maize hybrid would perform best. Similarly, crop breeders could adapt this ex ante assessment tool to weigh the pros and cons of a specific trait and the potential impact for farmers.

The TAMASA team plans to publish the code and user-friendly interface of this new geospatial assessment tool later this year.

New digital maps to support soil fertility management in Nepal

Written by on December 4, 2018. Posted in News.

KATHMANDU, Nepal (CIMMYT) — The International Maize and Wheat Improvement Center (CIMMYT) is working with Nepal’s Soil Management Directorate and the Nepal Agricultural Research Council (NARC) to aggregate historic soil data and, for the first time in the country, produce digital soil maps. The maps include information on soil PH, organic matter, total nitrogen, clay content and boron content. Digital soil mapping gives farmers and natural resource managers easy access to location-specific information on soil properties and nutrients, so they can make efficient and localized management decisions.

As part of CIMMYT’s Nepal Seed and Fertilizer (NSAF) project, researchers used new satellite imagery that enabled the resolution of the maps to be increased from 1×1 km to 250×250 m. They have updated the web portal to make it more user friendly and interactive. When loaded onto a smartphone, the map can retrieve the soil properties information from the user’s exact location if the user is within areas with data coverage. The project team is planning to produce maps for the whole country by the end of 2019.

CIMMYT scientist David Guerena talks about the role of the new digital maps to combat soil fertility problems in Nepal.

At a World Soil Day event in Nepal, CIMMYT soil scientist David Guerena presented the new digital soil maps to scientists, academics, policymakers and other attendees. Guerena explained the role this tool can play in combatting soil fertility problems in Nepal.

These interactive digital maps are not simply visualizations. They house the data and analytics which can be used to inform site-specific integrated soil fertility management recommendations.

The first high-resolution digital soil maps for the Terai region have been produced with support from the data assets from the National Land Use Project, developed by Nepal’s Ministry of Agriculture and Livestock Development. These maps will be used to guide field programming of the NSAF project, drive the development of market-led fertilizer products, and inform and update soil management recommendations. The government of Nepal can use the same information to align policy with the needs of farmers and the capacity of local private seed and fertilizer companies.

In 2017, 16 scientists from Nepal’s Soil Management Directorate, NARC and other institutions attended an advanced digital soil mapping workshop where they learned how to use different geostatistical methods for creating soil maps. This year, as part of the NSAF project, four NARC scientists attended a soil spectroscopy training workshop and learned about digitizing soil data management and using advanced spectral methods to convert soil information into fertilizer recommendations.

Soil data matters

Soil properties have a significant influence on crop growth and the yield response to management inputs. For farmers, having access to soil information can make a big difference in the adoption of integrated soil fertility management.

Farmer motivation and decision-making relies heavily on the perceived likeliness of obtaining a profitable return at minimized risk. This largely depends on the yield response to management inputs, such as improved seeds and fertilizers, which depends to a large extent on site-specific soil properties and variation in agro-ecological conditions. Therefore, quantitative estimates of the yield response to inputs at a given location are essential for estimating the risks associated with these investments.

The digital soil maps can be accessed at https://nsafmap.github.io/.

The Nepal Seed and Fertilizer project is funded by the United States Agency for International Development (USAID) and is a flagship project in Nepal. The objective of the NSAF is to build competitive and synergistic seed and fertilizer systems for inclusive and sustainable growth in agricultural productivity, business development and income generation in Nepal.

View of the interface of the Maize-Seed-Area app on mobile phones and tablets. (Photo: CIMMYT)

Are advisory apps a solution for collecting Big Data?

Written by Rodrigo Ordóñez on October 2, 2018. Posted in Blogs.

Big Data is transforming the way scientists conduct agricultural research and helping smallholder farmers receive useful information in real time. Experts and partners of the CGIAR Platform for Big Data in Agriculture are meeting on October 3-5, 2018, in Nairobi, Kenya, to share their views on how to harness this data revolution for greater food and nutrition security.

Jordan Chamberlin, Spatial Economist at CIMMYT, will give his insights on best practices on electronic data capture on October 4, 2018.

NAIROBI (Kenya) — Agronomic researchers face several challenges and limitations related to data. To provide accurate predictions and useful advice to smallholder farmers, scientists need to collect many types of on-farm data; for example, field size, area devoted to each crop, inputs used, agronomic practices followed, incidence of pests and diseases, and yield.

These pieces of data are expensive to obtain by traditional survey methods, such as sending out enumerators to ask farmers a long list of questions. Available data is often restricted to a particular geographical area and may not capture key factors of production variability, like local soil characteristics, fertilizer timing or crop rotations.

As a result, such datasets cannot deliver yield predictions at scale, one of the main expectations of Big Data. Digital advisory apps may be part of the solution, as they use crowdsourcing to routinize data collection on key agronomic variables.

The Taking Maize Agronomy to Scale in Africa (TAMASA) project has been researching the use of mobile apps to provide site-specific agronomic advice to farmers through agro-dealers, extension workers and other service providers.

At CIMMYT, one of the research questions we were interested in was “Why are plant population densities in farmers fields usually well below recommended rates?” From surveys and yield estimates based on crop-cut samples at harvest in Ethiopia, Nigeria and Tanzania, we observed that yields were correlated with plant density.

What was making some farmers not use enough seeds for their fields? One possible reason could be that farmers may not know the size of their maize field. In other cases, farmers and agro-dealers may not know how many seeds are in one packet, as companies rarely indicate it and the weight of each seed variety is different. Or perhaps farmers may not know what plant population density is best to use. Seed packets sometimes suggest a sowing rate but this advice is rather generic and assumes that farmers apply recommended fertilizer rates. However, farmers’ field conditions differ, as does their capacity to invest in expensive fertilizers.

To help farmers overcome these challenges, we developed a simple app, Maize-Seed-Area. It enables farmers, agro-dealers and extension workers to measure the size of a maize field and to identify its key characteristics. Then, using that data, the app can generate advice on plant spacing and density, calculate how much seed to buy, and provide information on seed varieties available at markets nearby.

Maize-Seed-Area is developed using the Open Data Kit (ODK) format, which allows to collect data offline and to submit it when internet connection becomes available. In this case, the app is also used to deliver information to the end users.

Advisory apps usually require some input data from farmers, so advice can be tailored to their particular circumstances. For example, they might need to provide data on the slope of their field, previous crops or fertilizer use. Some additional information may be collected through the app, such as previous seed variety use. All this data entered by the user, which should be kept to a minimum, is routinely captured by the app and retrieved later.

Hello, Big Data!

As the app user community grows, datasets on farmer practices and outcomes grow as well. In this case, we can observe trends in real time, for instance on the popularity of different maize varieties.

In a pilot in western Kenya, in collaboration with Precision Agriculture for Development (PAD), some 100 agro-dealers and extension workers used the app to give advice to about 2,900 farmers. Most of the advice was on the amount of seed to buy for a given area and on the characteristics of different varieties.

Data showed that the previous year farmers grew a wide range of varieties, but that three of them were dominant: DK8031, Duma43 and WH505.

Preferred variety of maize for sample farmers in western Kenya (Bungoma, Busia, Kakamega and Siaya counties), February-March 2018.

A phone survey among some 300 of the farmers who received advice found that most of them anticipated to do things differently in the future, ranging from asking for advice again (37 percent), growing a different maize variety (31 percent), buying a different quantity of seed (19 percent), using different plant spacing (18 percent) or using more fertilizer (16 percent).

Most of the agro-dealers and extension workers have kept the app for future use.

The dataset was collected in a short period of time, just two months, and was available as soon as app users got online.

The Maize-Seed-Area pilot shows that advisory apps, when used widely, are a major source of new Big Data on agronomic practices and farmer preferences. They also help to make data collection easier and cheaper.

TAMASA is supported by the Bill and Melinda Gates Foundation and is implemented by the International Maize and Wheat Improvement Center (CIMMYT), the International Institute of Tropical Agriculture (IITA), the International Plant Nutrition Institute (IPNI) and Africa Soil Information Service (AfSIS).

A farm landscape in Ethiopia. (Photo: Apollo Habtamu/ILRI)

Suitcase-sized lab speeds up wheat rust diagnosis

Written by on September 10, 2018. Posted in News.

Despite her unassuming nature, the literary character Miss Marple solves murder mysteries with her keen sense of perception and attention to detail. But there’s another sleuth that goes by the same name. MARPLE (Mobile And Real-time PLant disEase) is a portable testing lab which could help speed-up the identification of devastating wheat rust diseases in Africa.

Rust diseases are one of the greatest threats to wheat production around the world. Over the last decade, more aggressive variants that are adapted to warmer temperatures have emerged. By quickly being able to identify the strain of rust disease, researchers and farmers can figure out the best course of action before it is too late.

The Saunders lab of the John Innes Centre created MARPLE. In collaboration with the Ethiopian Institute of Agricultural Research (EIAR) and the International Maize and Wheat Improvement Center (CIMMYT), researchers are testing the mobile diagnostic kit in Holeta, central Ethiopia.

“These new pathogen diagnostic technologies … offer the potential to revolutionize the speed at which new wheat rust strains can be identified,” says Dave Hodson, a CIMMYT rust pathologist in Ethiopia. “This is critical information that can be incorporated into early warning systems and result in more effective control of disease outbreaks in farmers’ fields.”

Hodson and his colleagues will be presenting their research at the CGIAR Big Data in Agriculture Convention in Nairobi, on October 3-5, 2018.

Read more about the field testing of the MARPLE diagnostic kit on the ACACIA website.