Tag: data management

Harnessing econometric and statistical tools to support climate-resilient agriculture

Written by Julian Bañuelos-Uribe on May 13, 2024. Posted in Blogs.

Globally, climate extremes are adversely affecting agricultural productivity and farmer welfare. Farmers’ lack of knowledge about adaptation options may further exacerbate the situation. In the context of South Asia, which is home to rural farm-based economies with smallholder populations, tailored adaptation options are crucial to safeguarding the region’s agriculture in response to current and future climate challenges. These resilience strategies encompass a range of risk reducing practices such as changing the planting date, Conservation Agriculture, irrigation, stress-tolerant varieties, crop diversification, and risk transfer mechanisms, e.g., crop insurance. Practices such as enterprise diversification and community water conservation are also potential sector-specific interventions.

Atlas of Climate Adaptation in South Asian Agriculture (ACASA) aims to identify hazard-linked adaptation options and prioritize them at a granular geographical scale. While doing so, it is paramount to consider the suitability of adaptation options from a socioeconomic lens which varies across spatial and temporal dimensions. Further, calculation of scalability parameters such as economic, environmental benefit, and gender inclusivity for prioritized adaptation are important to aid climatic risk management and developmental planning in the subcontinent. Given the credibility of econometric and statistical methods, the key tenets of the approach that are being applied in ACASA are worth highlighting.

Evaluating the profitability of adaptation options

Profitability is among the foremost indicators for the feasible adoption of any technology. The popular metric of profitability evaluation is benefit-to-cost ratio. This is a simple measure based on additional costs and benefits because of adopting new technology. A benefit-to-cost ratio of more than one is considered essential for financial viability. Large-scale surveys such as cost of cultivation and other household surveys can provide cost estimates for limited adaptation options. Given the geographical and commodity spread, ACASA must resort to the meta-analysis of published literature or field trials for adaptation options. For example, a recent paper by International Food Policy Research Institute (IFPRI) based on meta-analysis shows that not all interventions result in a win-win situation with improvements in both tradable and non-tradable outcomes. While no-till wheat, legumes, and integrated nutrient management result in an advantageous outcome, there are trade-offs between the tradable and non-tradable ecosystem services in the cases of directed seed rice, organic manure, and agroforestry².

Quantification of adaptation options to mitigate hazards

Past studies demonstrate the usefulness of econometric methods when analyzing the effectiveness of adaptation options such as irrigation, shift in planting time, and crop diversification against drought and heat stress in South Asia. Compared to a simple cost-benefit approach, the adaptation benefits of a particular technology under climatic stress conditions can be ascertained by comparing it with normal weather conditions. The popular methods in climate economics literature are panel data regression and treatment-based models. Subject to data availability, modern methods of causal estimation, and machine learning can be used to ascertain the robust benefits of adaptation options. Such studies, though available in literature, have compared limited adaptation options. A study by the Indian Council of Agricultural Research-National Institute of Agricultural Economics and Policy Research (ICAR-NIAP), based on ‘Situation Assessment Survey of Agricultural Households’ of National Sample Survey Office (NSSO), concluded that though crop insurance and irrigation effectively improve farm income and reduce farmers’ exposure to downside risk, irrigation is more effective than crop insurance¹.

Statistical models for spatial interpolation of econometric estimates

Since ACASA focuses on gridded analysis, an active area of statistical application is the spatial interpolation or downscaling of results to a more granular scale. Many indicators used for risk characterization are available at coarser geographical units or points from surveys. Kriging is a spatial interpolation method where there is no observed data. Apart from spatial interpolation of observed indicators, advanced Kriging methods can be potentially used to interpolate or predict the estimates of the econometric model.

ACASA’s approach involves prioritizing adaptation options based on suitability, scalability, and gender inclusivity. Econometric and statistical methods play a crucial role in evaluating the profitability and effectiveness of various adaptation strategies from real world datasets. Despite challenges such as limited observational data and integration of econometric and statistical methods, ACASA can facilitate informed decision-making in climate risk management and safeguard agricultural productivity in the face of climatic hazards.

¹ Birthal PS, Hazrana J, Negi DS and Mishra A. 2022. Assessing benefits of crop insurance vis-a-vis irrigation in Indian agriculture. Food Policy 112:102348. https://doi.org/10.1016/j.foodpol.2022.102348

² Kiran Kumara T M, Birthal PS, Chand D and Kumar A. 2024. Economic Valuation of Ecosystem Services of Selected Interventions in Agriculture in India. IFPRI Discussion Paper 02250, IFPRI-South Asia Regional Office, New Delhi.

Blog written by Prem Chand, ICAR-NIAP, India and Kaushik Bora, BISA-CIMMYT, India

Unlocking insights from literature: exploring adaptation options in ACASA

Written by Julian Bañuelos-Uribe on May 8, 2024. Posted in Blogs.

To address the vulnerability of increased climate risks which impact agriculture, it is imperative to identify location-specific adaptation options. Atlas of Climate Adaptation in South Asian Agriculture (ACASA) is working on identifying commodity specific hazards at different geographical regions and the key adaptation options aligned with geography and hazards. This has been done for major cereal crops (rice, wheat, and maize), coarse grains (millets), oilseeds (coconut, mustard), legumes and vegetable crops (chickpea, potato), livestock, and fisheries. In ACASA, Systematic Literature Review (SLR) serves as a fundamental tool to identify key climate adaptation options and assess their effectiveness, considering agroecological factors.

Literature reviews are a customary approach for researchers to grasp existing knowledge and findings. The SLR methodically establishes clear research objectives, employs structured search queries to identify relevant literature, applies defined exclusion criteria, and extracts data for scientific analysis. This structured approach facilitates mapping the literature, validating findings, identifying gaps, and refining methodologies thereby minimizing biases, and ensuring comprehensive coverage of evidence.

Commodity-specific research questions, aligned with the problem/population, intervention, comparison/consequences, outcome, and time PICO(T) framework, have been used to guide the search process. By utilizing keywords specific to these questions, ACASA sourced literature from reputable databases such as Web of Science, Scopus, Google Scholar, and local databases of South Asian countries: Bangladesh, India, Nepal, and Sri Lanka. Local databases and gray literature further bolstered the understanding of local conditions and broadened the coverage of studied literature.

The searched literature was then filtered using the well-established Preferred Reporting Items for Systematic Reviews and Meta Analysis (PRISMA) framework. PRISMA provides a minimum set of evidence-based literature to be used for further analysis. Let us look at maize as an example of a commodity under analysis in ACASA. For maize, a total of 1,282 papers were identified and based on four exclusion criteria pertaining to adaptation options, quantitative assessment, hazard, and risk only of which 72 papers were shortlisted. The PRISMA framework supported in getting a manageable dataset for in-depth analysis while ensuring transparency in the overall filtering process.

After filtering through PRISMA, a bibliometric analysis was conducted which contained research trend analysis, regional distribution patterns, adaptation option categorizations, and a co-occurrence analysis. Useful patterns in popularity of studied adaptation options, hazards, and their linkages were observed through this analysis. For instance, drought was the most studied hazard, while pest diseases and economics were major hazard impacts studied for the maize literature. In terms of adaptation options, stress tolerant varieties were the most popular adaptation option. Further, co-occurrence analysis provided linkages between adaptation options and hazards, and demonstrated that researchers have also studied bundled technologies.

SLR helped understand the effectiveness of certain adaptation options. Going ahead, this step will be fully realized through a “meta-analysis” which will be pivotal in quantifying the evidence and prioritizing adaptation options for different agroecologies. SLR has proven to be an effective research method to build a comprehensive database that can be used across different thematic areas of ACASA. Adaptation options enlisted through SLR can be further substantiated through expert elicitations via heurism, crop modelling, cost-benefit analysis, and other important pillars of ACASA to identify efficient and cost-effective options.

SLR also provided the ACASA team with the opportunity to identify certain literature gaps such as uneven geographical coverage and excessive emphasis on certain adaptation options versus the rest. Conceptualization of systematically reviewing climate adaptation options in the South Asian context by integrating bibliometric and meta-analysis adds novelty to the current efforts of ACASA.

Blog written by Aniket Deo, BISA-CIMMYT India; Niveta Jain, ICAR-IARI India; Roshan B Ojha, NARC Nepal; and Sayla Khandoker, BARI Bangladesh

Greater successes through NARS partnerships

Written by Julian Bañuelos-Uribe on May 3, 2024. Posted in Blogs.

Map: BISA works with National Agricultural Research Systems (NARS) of South Asia to develop ACASA.

Atlas of Climate Adaptation in South Asian Agriculture (ACASA) is different from many projects supported by our team. I would love to dive into the promising features of the ACASA platform and the exciting technical advances being made, but I want to focus here on how the Borlaug Institute for South Asia (BISA) has organized this program for greater and longer-term impact.

BISA is a strong regional partner and is the lead institution for the ACASA program. In fact, we could have simply asked BISA to build the ACASA platform and known they would make a great technical product. However, our goal is not just to have great technical products, but also to improve the lives of small-scale producers. For any great technical product to deliver impact, it must be used.

From day one, the ACASA program has not just kept the users’ needs in mind, indeed they have kept the users themselves engaged on the project. By establishing strong, financially supported partnerships with the National Agricultural Research Systems (NARS) in Bangladesh, India, Nepal, and Sri Lanka, they are achieving four key outcomes, among many others:

Benefit from local expertise regarding national agricultural practices, climate risks, and solutions
Leverage NARS connections to national and subnational decision makers to inform product requirements
Establish national ownership with a partner mandated to support users of the product
Strengthen climate adaptation analytics across South Asia through peer-to-peer learning.

These outcomes lead to more accurate and appropriate products, user trust, and the long-term capacity to maintain and update the ACASA platform. The latter being essential given the constantly improving nature of our understanding of and predictions around climate and agriculture.

If this model of working has such advantages over “if you build it, they will come”, you might wonder why we do not use it in all cases. This approach requires divergence from business-as-usual for most researchers and is not without a cost. The BISA team are not only putting deep emphasis on the technical development of this product, but they are also spending considerable time, effort, and budget to create a program structure where the NARS are catalytic partners. The NARS teams are empowered on the project to contribute to methodologies used beyond their national boundaries, they have the task of making the best data available and validating the outputs, the responsibility of understanding and representing stakeholder requirements, and the ownership of their national platform for long-term use. BISA has developed a structure of accountability, provided funding, facilitated team-wide and theme-specific workshops, and shared decision-making power, which all presents additional work.

In the end, we encouraged this approach because we see too many decision support tools and platforms developed by international researchers who merely consult with users a few times during a project. These efforts may result in building captivating products, meeting all the needs brainstormed by the research team, but their future is sitting in a dusty (and unfortunately crowded) corner of the internet. While this approach seems fast and efficient, the efficiency is zero if there is no value gained from the output. So, we look for other ways to operate and engage with partners, to work within existing systems, and to move beyond theoretically useful products to ones that are used to address needs and can be evolved as those needs change. BISA has been an exemplary partner in building and supporting a strong ACASA team, and we are eager to see how each NARS partner leverages the ACASA product to generate impact for small-scale producers.

Tess Russo is a senior program officer at the Bill & Melinda Gates Foundation, based in Seattle, United States.

Scientist urges upgrades to monitor groundwater use for agriculture in low-income countries

Written by Mike Listman on September 27, 2022. Posted in News.

Data collector reading data from offline groundwater level logger – one of the three tested monitoring technologies. (Credit: Subash Adhikari/CIMMYT)

Based on a pilot study regarding the feasibility and cost effectiveness of several groundwater monitoring approaches for agriculture in Nepal’s Terai region, a water and food security specialist who led the research has recommended the use of phone-based systems.

Speaking to diverse experts at the recent World Water Week 2022 in Stockholm, Sweden, Anton Urfels, a systems agronomist at the International Maize and Wheat Improvement Center (CIMMYT), said that manual monitoring with phone-based data uploading is relatively low-cost and effective and could be scaled up across the Terai.

“One alternate monitoring approach studied — online data uploading — has substantially lower staff time requirements and technology costs and higher temporal resolution than phone-based monitoring, but does not provide real-time data and entails high technical skills, capital costs, and risks of theft and damage,” said Urfels in his presentation, ‘Upgrading Groundwater Monitoring Networks in Low-Income Countries’.

Urfels and partners also developed a prototype of an open-source groundwater monitoring dashboard to engage stakeholders, help translate raw data into actionable information, and detect water depletion trends.

Water has become a key part of food research and innovation, critical for sustainable and ecological intensification in agriculture, according to the scientist.

“Collecting groundwater data is difficult and the technology for monitoring is unreliable, which impairs effective modeling, decision-making, and learning,” Urfels explained. “Like other countries in the region, Nepal is increasing its agricultural groundwater consumption, particularly through private investment in irrigation wells and pumps that open irrigation to more farmers. This and climate change have altered groundwater recharge rates and availability, but national data on these trends are incomplete.”

An extensive lowland region bordering India and comprising one-fifth of the nation’s territory, the Terai is Nepal’s breadbasket.

Held yearly since 1991, World Water Week attracts a diverse mix of participants from many professions to develop solutions for water-related challenges including poverty, the climate crisis, and biodiversity loss. The 2022 theme was “Seeing the Unseen: The Value of Water”.

“I’d recommend more pilot studies on phone-based groundwater monitoring for other areas of Nepal, such as the Mid-hill districts,” Urfels said. “We also need to fine-tune and expand the system dashboard and build cross-sectoral coordination to recognize and take into account groundwater’s actual economic value.”

Urfels said the Nepal Ministry of Energy, Water Resources and Irrigation has requested the nationwide scale-out of a digital monitoring system, and CIMMYT and Nepal experts will support this, as well as improving the system, which would be freely available for use and development by researchers and agencies outside of Nepal.

The research described was carried out under the Cereal Systems Initiative in South Asia (CSISA), which is funded by USAID and the Bill & Melinda Gates Foundation, and under the CGIAR integrated research initiative, Transforming Agrifood Systems in South Asia (TAFSSA).

Nilupa Gunaratna (right), statistician at the International Nutrition Foundation, helps a farmer and her daughter to fill in a survey form on quality protein maize (QPM) as part of the QPM Development (QPMD) project in Karatu, Tanzania. (Photo: CIMMYT)

Turning data swamps into data lakes

Written by Rodrigo Ordóñez on April 26, 2022. Posted in Blogs.

Recently, I published the technical description of Ontology-Agnostic Metadata Schema (OIMS) in the journal Frontiers in Sustainable Food Systems, as part of a special issue on “Agile Data-Oriented Research Tools to Support Smallholder Farm System Transformation.”

CGIAR and the International Maize and Wheat Improvement Center (CIMMYT) are dedicated to providing research data information products (RDIP) in open access, following the FAIR data standards. FAIR stands for findable, accessible, interoperable and reusable. Organizations dedicated to open data have made massive progress in making data findable and accessible. A clear example is a free, open-access repository of research studies developed by CIMMYT scientists. Article 4.1.c.i. of the CGIAR data policy states that “Relevant data assets (e.g. datasets) and metadata shall be interoperable and fit for reuse.”

This is easier said than done. There are well-established standards for descriptive metadata such as the Dublin Core and the derived standard used widely across the CGIAR, aptly called CGcore, used in CIMMYT’s Dataverse research data repository. However, these standards are lacking in many domains for describing the actual content of data sets.

At best, idiosyncratic data dictionaries are developed for specific datasets, projects and sometimes even programs. Idiosyncratic data dictionaries help make data interoperable but, in many cases, require a lot of preprocessing before scientists can actually reuse the data. Having a standard for data dictionaries would be a huge leap forward, but is not likely to happen anytime soon.

The next best thing is to standardize the way that you describe data dictionaries. This was recognized by the community of practice on socioeconomic data of the CGIAR Platform for Big data in Agriculture. Over the past few years, efforts led by CIMMYT set to remedy that lack of a standard, resulting in the flexible, extensible, machine-readable, human-intelligible and ontology-agnostic metadata schema (OIMS).

The paper in the journal Frontiers in Sustainable Food Systems describes a lightweight, flexible, and extensible metadata schema. It is designed to succinctly describe data collected for international agricultural research for development, facilitating interoperability. The schema is also meant to make it easier to store, retrieve and link different datasets stored in a data lake.

Agricultural research data comes to the surface

The paper discusses a need for this type of schema. Typically, agricultural research data comes in different formats and from different sources. For example, we can have structured surveys, semi-structured surveys, mobile phone records and satellite data. In the case of socioeconomic data, it can be particularly “messy.” To facilitate interoperability, we need to find methods to describe these datasets, which are machine readable — or actionable.

There have been other attempts to provide a standardized way to make data interoperable. Past approaches have been comprehensive but cumbersome. That could be the reason why they are typically only used by larger-scale projects. OIMS provides a framework that can be used by all data managers and scientists to enhance the interoperability for research data to ensure the data can be reused with much more ease.

The paper provides a detailed description of OIMS, including: the metadata schema, which describes the data dictionary; and the self-describing metadata, which describes the fields in the metadata. The paper then demonstrates the utility of this schema using a small segment of a household survey.

This paper presents an internally consistent approach to providing metadata for data files when standards are missing. It is flexible and extensible, so it will not be obsolete before it is implemented at scale. The approach is based on the concept of data lakes where data is stored as is. To ensure that data lakes do not become swamps, metadata is indispensable. The OIMS metadata schema approach can help to standardize the description of metadata and thus can be considered the fishing gear to extract data from the data lake.

As part of the on-going work started by the community of practice on socioeconomic data of the CGIAR Platform for Big Data in Agriculture, implementation of the OIMS metadata schema approach on datasets that can create indicators highlighted in the 100Q approach with linkages to the nascent socioeconomic ontology SEOnt is envisaged. This will provide datasets with enhanced interoperability.

With more and more datasets using the OIMS approach in the near future, it will become possible to turn what is currently a socioeconomic data swamp into a data lake. This will provide timely actionable information to support agri-food systems transformation — helping smallholders make a living while staying within planetary boundaries.

Implementing OIMS in practice requires data managers and scientists that collect the data to actively engage in providing the relevant metadata. As mentioned before, some of the metadata can be gleaned from the software solutions the scientists use already. As these are structured metadata, they can be extracted by machines. Often it does require curation by the scientist involved, especially when the software solution does not provide key information that the scientist has at hand but is not documented in a machine-readable way.

Read the full paper:
A Flexible, Extensible, Machine-Readable, Human-Intelligible, and Ontology-Agnostic Metadata Schema (OIMS)

Breaking Ground: Gokul Paudel finds the best on-farm practices for South Asia

Written by Madeline Dahm on December 18, 2020. Posted in Features.

Gokul Paudel is an agricultural economist working to streamline farming practices in South Asia. He seeks to understand, learn from and improve the efficiency of on-farm management practices in a vast variety of ways. Although he joined the International Improvement Center for Wheat and Maize (CIMMYT) right after university, Paudel’s on-farm education started long before his formal courses.

“I was born in a rural village in Baglung district, in the mid-hills of Nepal. My parents worked on a small farm, holding less than half a hectare of land,” he says. “When I was a kid, I remember hearing that even though Nepal is an agricultural country, we still have a lot of food insecurity, malnutrition and children who suffer from stunting.”

“I would ask: How is Nepal an agricultural country, yet we suffer from food insecurity and food-related problems? This question is what inspired me to go to an agricultural university.”

Paudel attended Tribhuvan University in Nepal, and through his coursework, he learned about plant breeding, genetic improvement and how Norman Borlaug brought the first Green Revolution to South Asia. “After completing my undergraduate and post-graduate studies, I realized that CIMMYT is the one organization that contributes the most to improving food security and crop productivity in developing countries, where farmers livelihoods are always dependent on agriculture,” he explains.

Approaching the paradox

Paudel is right about the agriculture and food paradox of his home country. Almost two thirds of Nepal’s population is engaged in agricultural production, yet the country still has shockingly high numbers in terms of food insecurity and nutritional deficiency. Furthermore, widespread dissemination of unsustainable agronomic practices, like the use of heavy-tilling machinery, present similar consequences across South Asia.

If research and data support the claim that conservation agriculture substantially improves crop yields, then why is the adoption of these practices so low? That is exactly what Paudel seeks to understand. “I want to help improve the food security of the country,” he explains. “That’s why I joined the agricultural sector.”

Paudel joined CIMMYT in 2011 to work with the Socioeconomics Program (SEP) and the Cereal Systems Initiative for South Asia (CSISA), providing regional support across Bangladesh, India and Nepal.

His work is diverse. Paudel goes beyond finding out which technological innovations increase on-farm yield and profit, because success on research plots does not always translate to success on smallholder fields. He works closely with farmers and policy makers, using surveys and high-tech analytical tools such as machine learning and data mining to learn about what actually happens on farmers’ plots to impact productivity.

Gokul Paudel holds up two bags of wheat crop-cuts in a farmer's field in India. (Photo: CIMMYT) — Gokul Paudel holds up two bags of wheat crop-cuts in a farmer’s field. (Photo: CIMMYT)

A growing future for conservation agriculture

Over the last two decades, the development of environmentally sustainable and financially appealing farming technologies through conservation agriculture has become a key topic of agronomic research in South Asia.

“Conservation agriculture is based on three principles: minimum disturbance of the soil structure, cover crop and crop rotation, especially with legumes,” Paudel explains.

Leaving the soil undisturbed through zero-till farming increases water infiltration, holds soil moisture and helps to prevent topsoil erosion. Namely, zero-till farming has been identified as one of the most transformative innovations in conservation agriculture, showing the potential to improve farming communities’ ability to mitigate the challenges of climate change while also improving crop yields.

Still, the diffusion rate of zero-tillage has remained low. Right now, Paudel’s team is looking at a range of factors — such as farmers’ willingness to pay, actual demand for new technologies, intensification under input constraints, gender-disaggregated preferences and the scale-appropriateness of mechanization — to better understand the low adoption rates and to find a way to close the gap.

Can farm mechanization ease South Asia’s labor shortage?

In South Asia, understanding local contexts is crucial to streamlining farm mechanization. In recent years, many men have left their agricultural jobs in search of better opportunities in the Gulf countries and this recent phenomenon of labor out-migration has left women to take up more farming tasks.

“Women are responsible for taking care of the farm, household and raising their children,” says Paudel. “Since rural out-migration has increased, they have been burdened by the added responsibility of farm work and labor scarcity. This means that on-farm labor wages are rising, exacerbating the cost of production.”

The introduction of farm machinery, such as reapers and mini-tillers, can ease the physical and financial burden of the labor shortage. “Gender-responsive farm mechanization would not only save [women’s] time and efforts, but also empower them through skills enhancement and farm management,” says Paudel. However, he explains, measures must be taken to ensure that women actually feel comfortable adopting these technologies, which have traditionally been held in the male domain.

Gokul Paudel records the total above-ground biomass of maize and other maize yield attributes in a farmer’s field in Kanchunpur, Nepal. (Photo: Ashok Rai/CIMMYT)

From farm-tech to high-tech

Right now, amidst the global lockdown due to COVID-19, Paudel’s field activities are highly restricted. However, he is capitalizing on an opportunity to assess years’ worth of data on on-farm crop production practices, collected from across Bangladesh, India and Nepal.

“We are analyzing this data-set using novel approaches, like machine learning, to understand what drives productivity in farmers’ fields and what to prioritize, for our efforts and for the farmers,” he explains.

Although there are many different aspects of his work, from data collection and synthesis to analysis, Paudel’s favorite part of the job is when his team finds the right, long-lasting solution to farmers’ production-related problems.

“There’s a multidimensional aspect to it, but all of these solutions affect the farmer’s livelihood directly. Productivity is directly related to their food security, income and rural livelihoods.”

A changing landscape

About 160 km away from where he lives now, Paudel’s parents still own the farm he grew up on — though they no longer work on it themselves. They are proud to hear that his work has a direct impact on communities like theirs throughout the country.

“Every day, new problems are appearing due to climate change — problems of drought, flooding and disease outbreak. Though it’s not good news, it motivates me to continue the work that I’m doing,” says Paudel. “The most fascinating thing about working at CIMMYT is that we have a team of multidisciplinary scientists working together with the common goal of sustainably intensifying the agricultural systems in the developing world.”

Breaking Ground: Andrea Gardeazábal transforms data into meaningful information

Written by Mary Donovan on October 29, 2020. Posted in Features.

Andrea Gardeazábal has many titles — Monitor, Evaluation and Learning Manager, ICT for Agriculture — but the core of what she does is knowledge management. She merges monitoring, evaluation, accountability and learning (MEAL) with information communication technologies (ICT) to transform data into something meaningful.

A political scientist by training, Gardeazábal knows the power of data and statistics. As she began working on ICT-for-development projects in the field, she observed a lack of understanding of ICT and how the development sector could take advantage of these tools.

“I knew this was progressing very fast; that this was the future. Everyone was talking about ICT and the future with the internet of things, and social media was just getting started,” she said. So she asked herself, how could the development sector take advantage of these new technologies?

Gardeazábal was working on projects bringing computers to rural areas in Colombia, which did not have internet connection or electricity. The problem could not be solved simply with a machine. She wanted to understand how to use ICT for development in a meaningful way. This triggered an interest in MEAL, to understand how ICT benefits the development sector, or does not, and to reintegrate that information into project design and impact.

After working in ICT for civil participation, education and microfinance, she joined CIMMYT with the mission to understand ICT for agriculture. Now she merges ICT tools with MEAL, leading the design, development and operation of systems for data collection, data cleaning, data analysis and data visualization with the Integrated Development program’s projects in Colombia, Guatemala and Mexico.

Ensuring intended results

Monitoring, learning, accountability and evaluation is crucial to ensure CIMMYT delivers on its objectives. Monitoring means ensuring that operations in the field are happening as planned. Rather than waiting until the end of the project when the donor asks for a report, Gardeazábal’s team monitors operations in the field on a quarterly or yearly basis. The team, both in the field and at headquarters, uses this data to check that the project is achieving what was intended and make interventions or adjustments if necessary.

Evaluation looks at project results and evidence. The team collects evidence for every single data point that they have, and then evaluates that evidence for impact and results in the field. This data is not only related to yield increase, but includes sustainable production, capacity development, and adequate technology adaptation and adoption processes.

Accountability is transparency with funders, so that everyone involved in a project is accountable for the processes, decisions and impact. CIMMYT is able to show progress through a transparent relationship with funders.

Learning happens after the team collects information, produces results evaluations, and understands what was done well and where the process had to be redirected. This information can then inform design of new projects or project phases. “We use the data and analysis of each project to redesign or modify our plans for the next project or even what kinds of projects we want to conduct,” Gardeazábal said.

Andrea Gardeazábal merges ICT tools with monitoring, evaluation, accountability and learning to improve project design. (Photo: Francisco Alarcón/CIMMYT)

What ICT can offer

In the past, a MEAL team would collect data from a representative sample at the start of the project, then go back to the office and analyze that data. At the end of the project, the team would complete the same exercise, to see the difference from what they gathered at the beginning.

With ICT tools, researchers are able to gather and analyze robust data more quickly and can communicate efficiently with the beneficiaries of a project throughout its course. Artificial intelligence and machine learning algorithms can help in understanding large sets of data so that this information can strengthen and streamline the MEAL process and project impact.

“We don’t need to wait until the end of the project for the results in the field or to have a sense of what the farmers are saying and achieving. We have a lot of tools, from the ICT side, that help make monitoring and evaluation more efficient,” Gardeazábal explained.

An international award recognized some of these ICT tools earlier this year. Gardeazábal formed part of the winning team with members from the Alliance of Bioversity International and CIAT and the International Institute for Applied Systems Analysis (IIASA) working on groundbreaking data systems and tools that help over 150,000 farmers in Mexico.

The team tracked over 500 variables over different farming plots and analyzed them with geographic, weather and market data to help identify the best management practices for each plot. This information — including historic yield potential, local benchmarks, windows of opportunity, recommended agricultural practices and commodity price forecasting — is available to farmers through an app called AgroTutor (Android, iOS).

The importance of an enabling environment

However, Gardeazábal cautions against the idea that technology on its own is going to end poverty or increase food security.

“ICT is a vehicle for innovation in agriculture. Just having an app in the field is not enough to generate the change that we are actually looking for. You need an enabling environment, a network, engagement of the farmers and the buy-in of scientists to take advantage of ICT tools.”

From drones and satellite imagery to artificial intelligence, ICT tools can help CIMMYT carry out its mission by streamlining the data gathering and analytics processes.

However, this work is not done in isolation from the environment surrounding it. CIMMYT does not only work on increasing yields, but also manages resources and local networks in efficient ways. Teams must monitor data on air quality, water use and efficient information flows, analyze this data, and then return to the field with recommendations for the most sustainable production within integrated agri-food systems.

Creating impact through wider data sharing

Written by Nima Chodon on October 22, 2020. Posted in News.

The Bihar Convergence Platform for agriculture, a synergistic partnership to innovate and initiate targeted interventions that help farmers to have informed choices with proven scientific recommendations, has been consistently working to accelerate interventions and improve the lives and capacity of small and marginal farmers since its establishment in October 2019.

The Cereal Systems Initiative for South Asia (CSISA), in association with CABI and the Open Data Institute, hosted a six-day virtual interactive training in September for platform members on the theme “creating impact through wider data sharing.” The training aimed at strengthening technical expertise of the participants, creating an enabling environment to unlock the benefits of data sharing and developing space for participants to discuss, brainstorm and co-design initiatives to be implemented together by the platform in coming days.

The training ended with a common understanding about the challenges and constraints in agriculture because data is in silos. Furthermore, participants agreed on the need to look at the existing data with a broader lens to accelerate the pace of development in agriculture in the state. Participants expressed that sharing the data under set norms with standardized licensing could act as a catalyst to increase the benefits for smallholder farmers.

To constructively deal with the challenges in agriculture together, the platform members stressed the need to start analyzing existing data from a wider perspective and data sharing as the key for designing fact-based interventions for larger good and impact.

Platform members interact during virtual training. (Photo: Sugandha Munshi/CIMMYT)

The platform is chaired by the Vice Chancellor of Bihar Agriculture University, with key members from Bihar Rural Livelihood Promotion Society known as Jeevika, Bihar Agriculture University, Dr Rajendra Prasad Central Agriculture University, Agriculture Technology Application Resource Institute, ICAR-RCER, and the CSISA project, along with private groups like IFFCO, Bayer, and ITC.

Out of the many activities jointly implemented by the platform, the Data Ecosystem is the key arena where the platform works together in strengthening the impact of data and incorporating them in accelerating quality interventions for farmers.

This story was first published on the CSISA website.

International analytics award goes to CGIAR centers for sustainable agriculture tools

Written by Leslie Domínguez on May 14, 2020. Posted in News.

A researcher demonstrates the use of the AgroTutor app on a mobile phone in Mexico. (Photo: Francisco Alarcón/CIMMYT)

While technology and data analysis have revolutionized farming in high-income countries, farmers in low-income nations have often been left out. New tools developed by CGIAR centers are narrowing the digital divide for smallholder farmers. These tools are increasingly gaining international notice beyond the field of agriculture, as a recent international award attests.

The 2020 Innovative Applications in Analytics Award (IAAA) was awarded to the International Maize and Wheat Improvement Center (CIMMYT), the Alliance of Bioversity International and CIAT, and the International Institute for Applied Systems Analysis (IIASA). Their tools, which help farmers to increase productivity, use more sustainable practices and access markets, topped bids from international technology companies and universities with projects applied to various fields.

“The award from this collaboration highlights the strengths of bringing together our research centers under One CGIAR,” said Daniel Jiménez, an Alliance scientist who was part of the winning team. “It shows that we can collaborate and compete at the highest level when it comes to applying science to global challenges.”

CIMMYT, CIAT and the IIASA won with their ‘Integrated Analytics for Sustainable Agriculture in Latin America’ projects. Finalists included projects focused on a wide range of analytics tools used in pricing, optimizing gas transmission, taxi guidance and management of hospital discharges.

A farmer in Mexico uses the AgroTutor application in the field. (Photo: Francisco Alarcón/CIMMYT)

“The IAAA award emphasizes novelty and creativity in analytics applications along with real-world impact,” said Pallav Chhaochhria, Director of Equities Trading at Citigroup and co-chair of the IAAA selection committee, during the virtual award ceremony.

The winning submission recognizes groundbreaking data systems and tools by publicly funded researches and field technicians who advise more than 150,000 farmers who participate in MasAgro, CIMMYT’s bilateral collaboration project with Mexico for sustainable maize and wheat production.

The multi-disciplinary team developed a system to track over 500 variables during the growing cycle at each plot monitored. These variables are subsequently analyzed in combination with geographic, weather and market data obtained from open sources of information. Data analysts and researchers mine these databases to find correlations or patterns that help identify limiting factors and the best management practices for each plot.

The resulting analytics feed an application called AgroTutor — available on Android and iOS — which offers free information to farmers, including historic yield potential, local benchmarks, windows of opportunity, recommended agricultural practices and commodity price forecasting.

“Although sophisticated data analyses have been used in agriculture before, small and medium-sized farmers have seldom benefited from a combination of systems and tools that offer predictive and prescriptive site-specific analytics,” said Andrea Gardeazabal, Information and Communication Technology for Agriculture, Monitoring & Evaluation Manager for CIMMYT’s Integrated Development program. “MasAgro and AgroTutor have proven that resource-constrained farmers can benefit from high-end innovative analyses.”