To help reaching the Sustainable Development Goals, CGIAR must tap into Big Data. Within the programme on Climate Change for Agriculture and Food Security (CCAFS), researchers have already applied Big Data analytics to agricultural and weather records in Colombia, revealing how climate variation impacts rice yields. After defining its Open Data-Open Access strategy, CGIAR has launched an internal call for proposals for big data analytics platforms that will provide services to the Agri-Food system programmes and parners, and will interconnect the CGIAR data to other multi-disciplinary big data. The seminar will present the pespectives of the envisioned platforms.
Webinar@AIMS: Perspective on Big Data in the CGIAR
1. Big Data in
CGIAR
Elizabeth Arnaud1, Medha Devare2, Jacob
Van Etten1, Jawoo Khoo3, Andy Jarvis4
1 Bioversity International, Consortium
Office2, IFPRI3, CIAT4
4th February 2016, AIMS Webinar, FAO
2. What is CGIAR ?
An international organization that advances international
agricultural research for food security.
Photo CGIAR
15 Centers worldwide
40 years of research
Over 8,000 scientists,
researchers,
technicians, and staff
3. CGIAR Goals
1. Reduced poverty
2. Improved food and nutrition security for
health
3. Improved natural resource systems and
ecosystem services
Photo B. Wawa/CIMMYT
4. Big Data (as per IBM)
Volume, Velocity, Variety, Veracity
Identify patterns and gain new knowledge to answer
complex and unanswered questions
Entreprises remain agile in a rapid evolving environment
5. Strategic Collaborations
Existing high performance computing facilities and big data
analytical capabilities
Leverage this investment in capability and infrastructure
Strong partnerships across the Consortium and beyond it
Example of the Strategic Alliance between Google Maps and
FAO to make geospatial tracking and mapping products more
accessible
FAO's José Graziano da Silva and Google's Rebecca Moore celebrate
the partnership formalization at COP21 in Paris
6. Big Data for Agriculture
Massive Historical Agricultural Data in distributed
databases and repositories
New Highthrouput Data
Genomics/Genetics/metabolomics
Phenotyping
Geophysical through Remote sensing
Social and Economics
Citizen Sciences
Challenge: Secure Food and Health by addressing Yield
Gap and Climate Change
Photo: ICRISAT
7. 7 million samples of crop varieties and wild
relatives in genebanks worldwide
Using the emerging deluge of omics data to
unlock crop diversity on a massive scale
Biodiversity Informatics Platform
Currently 62 institutions
Initiative facilitated by
Global Crop Diversity Trust, Global Plant Council,
International Treaty for PGRFA,
CGIAR
http://www.divseek.org/
Mining the omics data as
collaborative effort IRRI
Illustration: Australian Phenomics
Facility
10. Remote sensing project across
CGIAR Research Programmes
Bill &Melinda Gates Foundation-supported remote sensing
project spans across the CGIAR
Linked to CGIAR Research Programs on Climate Change,
Agriculture and Food Security (CCAFS) and on Water, Land
and Ecosystems (WLE).
A complete lay of the land will be recorded by Use
Unmanned Aerial Vehicles (UAVs) to discern crop types,
any present diseases, and the effects of climate change.
Photo: CIP
12. To produce recommendations much more quickly on rice
varieties in Colombia
1. Access annual surveys and agronomic experiments from
commercial fields
2. Getting planting times for specific sites and seasonal forecasts
3. Pairing historical records with state-of-the-art seasonal
forecasts
4. Analyses with advanced algorithms from biology, robotics and
neurosciences
5. Search for weather patterns in previous years and checked which
varieties did best in those years
Analyzing large, uncontrolled,
real-world data for rice
13. Most productive rice varieties and planting times for specific
sites and seasonal forecasts Identified.
Recommendations could potentially boost yields by 1 to 3 tons
per hectare.
Scaling up the techniques from their pilot to Argentina,
Nicaragua, Peru and Uruguay
Analyzing large, uncontrolled,
real-world data for rice
The project won
UN Global Pulse’s Big Data Climate Challenge
CIAT
14. Crowdsourcing Farmers’
Preferences for varieties
Seeds for Needs projects
500 farmers per site will be given 3 blind varieties in small quantities to be
tested under their own conditions and feedback returned by phone
15. What Big Data can do for Livestock
Farmers ?
1 billion small-scale livestock keepers
Smalls-cale ‘mixed’ crop-and-livestock farmers
Livestock-based strategies for adapting to, and
mitigate, climate change -
Livestock feeding systems that both increase productivity
and reduce greenhouse gases.
From Kunming to Da Li, Yunnan Province, China (photo credit: ILRI/Stevie Mann).
Remote sensing/GIS for
Monitoring welfare of livestock
Grazing locations, distribution
Epidemiology/diseases distribution
Another Webinar !
16. Future of Big Data in CGIAR
CGIAR 2017-2022 Programmes and Platforms
17. Agri-food System Programmes
Big Data platforms will support 8 Agri-Food System
Programmes (AFSP)
1. Dryland Cereals and Legumes
2. Fish Forest and Agroforestry Landscapes
3. Livestock
4. Maize
5. Rice
6. Roots, Tubers and Bananas
7. Wheat
18. Global Integrative Programmes
1. Genebanks
2. Agriculture for Nutrition and Health program
(A4NH)
3. Water Land and Ecosystems (WLE)
4. Climate Change, Agriculture and Food
Security program (CCAFS)
5. Policies, Institutions and Markets (PIM)
19. CGIAR data & service platforms
1. Genebanks
2. Genetic Gain
Establishing CGIAR system genetic resources
capability
3. A platform on Big-Data, Information and
Knowledge
genetics & genomics, Agrobiodiversity data
spatial, biophysical, social and economics
Open Data-Open Access
20. IFPRI/CIAT led Proposal for
CGIAR Big Data Platform
‘Tools for Driving Interdisciplinary and Collaborative
Big Data Analytics ‘ - some elements
CGIAR Survey Platform Data
for data collected through mobile devices
from connected sensor network across trial sites
Provide ontologies/standards/tools for Data
Discovery & Analytics across data repositories
Proposed use cases for monitoring agriculture:
Scalable Satellite-based Crop Yield Mapper (SYCM):
Crop Water Productivity (CWP)
Remote Sensing for Agro-biodiversity Monitoring
Capacity development activities
21. More on the CGIAR Web Site
First Expression of Interest and Comments
http://www.cgiar.org/our-strategy/second-call-
for-cgiar-research-programs/crp-2nd-call-pre-
proposal-submissions/
Second call for full proposal
http://library.cgiar.org/bitstream/handle/10947/4
127/CGIAR-2ndCall-
GuidanceFullProposals_19Dec2015.pdf?seque
nce=1
22. Thank you
Medha Devare
Data and Knowledge Manager
Open Data-Open Access Strategy Coordinator
CGIAR Consortium Office
Jawoo Koo
HarvestChoice project Co-PI
IFPRI
Andy Jarvis
Director, Decision &Policy Analysis Area, CIAT
Flagship leader, CCAFS
Mali, E. Arnaud, Bioversity
Jacob van Etten
Theme Leader, Adaptation to Climate Change
Bioversity International
Editor's Notes
Enormous amounts of data generated with new technologies for measurement, collection and storage that defy conventional analysis techniques
A number of scientific organizations developed high performance computing facilities and big data analytical capabilities.
Leverage this investment in capability and infrastructure
Strong partnerships across the Consortium and beyond it
Example of the Strategic Alliance between Google Maps and FAO to make geospatial tracking and mapping products more accessible
Agrciutlre is comlex and requires access to multidsciplinary data sets and models.
710,000 accessions in CGIAR genebanks
offer the greatest, largely untapped opportunities.
and mobilize vast range of plant genetic variation to accelerate the rate of crop Improvement
The multispectral camera captures and measures light at visible and near-infrared wavelengths. That’s important because each plant variety has a small but measurable difference in the wavelength of light it reflects when in sunlight — a kind of distinctive “signature.” Measuring this spectral signature in field conditions in Africa can help researchers identify from the air whether a crop is sweetpotato, cassava or something else. It can also help them identify what variety of OFSP the crop is. - See more at: http://cipotato.org/press-room/blogs/cip-drone-study-over-sweetpotato-fields-of-east-africa-a-success/#sthash.JnM1vy72.dpuf?platform=hootsuite
The multispectral camera captures and measures light at visible and near-infrared wavelengths. That’s important because each plant variety has a small but measurable difference in the wavelength of light it reflects when in sunlight — a kind of distinctive “signature.” Measuring this spectral signature in field conditions in Africa can help researchers identify from the air whether a crop is sweetpotato, cassava or something else. It can also help them identify what variety of OFSP the crop is.
What’s more, this spectral signature may reveal whether individual sweetpotato plants are thriving and likely to produce many storage roots or whether they are stressed by drought, have a nutritional deficiency or are under attack by a virus or insect. Such changes can be detected in multispectral images before they can be seen in the visible spectrum, scientists say.
Getting spectral signatures with the drone is a key part of the CIP-led remote-sensing project — building what the researchers call a “spectral library” containing signatures for each variety of OFSP.
- See more at: http://cipotato.org/press-room/blogs/cip-drone-study-over-sweetpotato-fields-of-east-africa-a-success/#sthash.AbGklV4b.dpuf
Harvest results of annual surveys and agronomic experiments from commercial fields
Get planting times for specific sites and seasonal forecasts
Pairing historical records with state-of-the-art seasonal forecasts
Analyses with advanced algorithms from biology, robotics and neurosciences
Search for weather patterns in previous years and checked which varieties did best in those years
how agricultural biodiversity can help minimize the risks associated with climate change. The concept is simple – if farmers have better information and access to a wide range of varieties, they are more able to choose what best suits their conditions and cope with unpredictable weather.
Citizen science – Seeds4Needs run in 11 countries and the corwd sourcing started in Costa Rica, Kenya - 1500, Ethiopia 1500, India(15,000 farmers in India, Tanzania (1500) –
weather data are recorded every hour.
Parallel to the mother and baby trials, 500 farmers per site will be given 3 blind varieties in small quantities to be tested under their own conditions (the crowdsourcing approach) and will be asked to evaluate the material and provide feedback on their preferences, they will become citizen scientists. Data and
feedback will be collected by ERMCSD with the engagement of extension services after receiving appropriate training by Bioversity. The feedback will be collected using a simple questionnaire using mobile phones/tablets for immediate submissions to the data manager. This data will be linked to a global portal developed by Bioversity and CIAT to upscale the approach and will be analyzed using ClimMob software developed by Bioversity (van Etten, 2014).
‘I wonder whether wide-spread adoption of data collection and sharing by poultry farmers could have helped stop the spread of the avian flu ?’
(Todd Janzen, Janzen Ag Law, US)
To scaled up and out AFSP outputs to other countries and regions.
15 international research centers working together under the CRP structure, is well positioned to generate multidisciplinary, complementary big datasets and to demonstrate their complementary use in ongoing research programs. There are, however, major needs for modern approaches to data gathering, storage, and analysis across CGIAR
1. Supporting (Big) Data Generation and Management: To facilitate broad adoption of modern rapid, large-scale survey and crowd-sourcing tools for collecting new (big) data, efficiently managing, harmonized with other types of (small) data organized, aggregated, and digitized from existing sources.
2. Providing Tools for (Big) Data Discovery and Analytics: To facilitate the use of big data and analytics in agricultural research and development by providing tools for data exploration, visualization, delivery, and analytics and training.
3. Developing Case Studies for Monitoring Agriculture using (Big) Data and Analytics: To support the development of robust case studies on the use of big data analytics in multidisciplinary integrated modeling analysis frameworks and products with indicators to monitor the progress toward the Sustainable Development Goals (SDGs) and CGIAR’s Intermediate Development Outcomes (IDOs).
4. Organizing Activities for Capacity Development: To undertake a series of capacity development activities to ensure the long-term adoption of new techniques, especially with partnering organizations.
Scalable Satellite-based Crop Yield Mapper (SYCM): Stanford University developed the SYCM method, which uses crop model simulations to train statistical models and apply to satellite imagery within the Google Earth Engine platform in the cloud. By partnering with Google and Stanford University, the IFPRI-led Spatial Production Allocation Modeling platform will be further developed into a satellite-based crop mapping application, and to make such estimations consistent with agricultural statistics and surveys.
?Crop Water Productivity (CWP): IWMI has developed a body of research in the water productivity of different crops and over multiple scales, including tools and methods to assess this. By partnering with Google and UNESCO-IHE, this work will be further developed to provide a global, daily, ensemble ET product, which will be combined with SYCM to provide scalable assessments of crop water use from field to national and global levels.
?Remote Sensing for Agro-biodiversity Monitoring: Carnegie Airborne Observatory (CAO) at Carnegie Institution for Science uses LiDAR imagery to map the 3D structure of natural habitat. By partnering with CAO, we will be able to study and monitor the agrobiodiversity distribution at sentinel sites.