October 2009 SATrends Issue 99
     

Whats in a word
FAO and ICRISAT are developing Phase II of the AGROVOC agricultural ontology

The word “Ontology” describes the nature of being. In the area of Information Science the word means, the hierarchical structuring of knowledge about things by sub-categorizing them according to their essential qualities. In other words, an ontology is a collection of terms and relationships in a domain. But, lest we err on the side of oversimplification, the word ontology implies rich semantic relationships among terms and attributes, as well as strict rules about how to specify terms and relationships.

knitting words together for the AOS
Gudrun Johannsen of FAO (left) and Lavanya Kiran of ICRISAT knitting words together for the AOS.

An Ontology for Agriculture is currently being designed and developed at the Food and Agriculture Organization (FAO) of the UN by the Agriculture Information Management Standards group with the help of ICRISAT.

AGROVOC or the Agricultural Vocabulary is a structured controlled vocabulary designed to cover terminology in agriculture, fisheries, forestry, food and other related domains. It was earlier used for indexing and retrieving data in agricultural information systems. AGROVOC is available in 19 languages, including Hindi. With the advent of applications based on Web 2.0 concepts, the AGROVOC reach was realized in such a milieu, and thus began efforts to bring in more semantics into the controlled vocabulary. This multilingual thesaurus was revised and refined at ICRISAT in an earlier project.

The next phase of porting AGROVOC to an Agricultural Ontology Service (AOS) is now under way. The AGROVOC Concept Server (CS) is a concept-based system, which, compared to a traditional Knowledge Organization System, allows the representation of more semantics such as specific relationship between concepts and relationships between their multilingual lexicalization. It particularly allows for modeling localized differences in knowledge representation, ie, defining concepts via their language specific lexicalization and relations.

The Concept Server is freely accessible and facilitates collaborative editing. It serves as a pool of agricultural concepts and is a starting point for the development of specific domain ontologies, where multilingualism and localized representation of information are of prime importance. Many national level projects such as the AGROPEDIA (http://www.agropedia.net/) are being built using AGROVOC as the crux for knowledge inference and extraction.

The AGROVOC Concept Server Workbench (AOS CSWB) has been designed and developed in collaboration with Kasetsart University, Thailand. Ms Gudrun Johannsen a senior Information Specialist from FAO HQ was recently at ICRISAT for a period of 4 weeks to train the KMS team on the nuances in the AOS CSWB.

This advanced tool will facilitate the management of the multilingual ontology more easlily in a collaborative environment. Guest users are invited to test the AOS CSWB at http://naist.cpe.ku.ac.th/agrovoc

For more information contact: N.Lavanyakiran@cgiar.org or A.Sylvester@cgiar.org.

back to top Back to top


Boosting Fertilizer Use in South Africa
Private-public partnership helps to boost use of fertilizer in southern Africa

Fertilizers have been identified as the key to improve yields in places like the Limpopo Province, South Africa, where the soils are inherently low in fertility. However, farmers here do not use fertilizer because of affordability and availability issues. ICRISAT’s approach to this problem was to facilitate the creation of a public-private partnership between the Limpopo Department of Agriculture (LPDA); SASOL Nitro, a fertilizer manufacturer; and Progress Milling, a grain milling and trading firm with a network of rural retail outlets that sell fertilizer throughout the province.

The members of the partnership worked together to test the idea that selling smaller-than-usual size packs of fertilizer (10kg instead of 50kg) would help overcome the constraints.

For two years extension officers in the province along with ICRISAT and staff from the Limpopo Agricultural Strategic Team (LIMPAST) provided farmers in the area with information on micro-dosing, a strategy that allows farmers to maximize their returns from small investments in fertilizers. Then, in 2006/07, SASOL Nitro filled smaller bags with fertilizer. Progress Milling stocked the shelves in their outlets throughout the province with the new 10kg bags.

climate change
ICRISAT and LPDA demonstrate the benefits of micro-dosing to farmers in the Limpopo Province of South Africa.

Two household surveys carried out in the Limpopo Province show that farmers, especially those who had heard of micro-dosing but had limited previous experience with fertilizers, preferred the small packs as they were one-fifth of the investment required to buy a 50kg bag of fertilizer. The smaller bags were also easier and cheaper to transport home. Sales of fertilizer in the Progress Milling outlets have increased from 15 tons in 1999−2003 to 96 tons in 2005/06 and 140 tons in 2006/07. As a result of these efforts it appears that yields for those who used fertilizer increased by at least 25%.

While historically the public sector has catalyzed change in smallholder agriculture, this experiment shows that public−private partnerships can be instrumental in the promotion and adoption of new technologies. Farmers in the Limpopo Province are benefiting from the increased access to fertilizer whereas SASOL Nitro has seen an increase in the amount of fertilizer sold. The increased yields have meant a greater and more continuous supply of grain for Progress Milling. The public-private partnership has created an environment where all the players succeed in achieving mutual objectives.

For more information contact: i.minde@cgiar.org or j.dimes@cgiar.org.

back to top Back to top


Seeing is Believing – West Africa (SIBWA)
A Bill & Melinda Gates Foundation funded Quick-Win project

Every once in a while, land in the semi-arid tropics, especially in sub-Saharan Africa, dries up for extended periods of time. You remember your parents’ recollection of the great droughts. Over time you witnessed increased interest from non-farmers in your livelihoods, in your vulnerability to climate, and welcomed their suggestions.

ow fertility millet-cowpea field
Average furrow width, low fertility millet-cowpea field: 204 cm (Serkin Hawsa, Niger)

You’ve noticed some changes lately. Like, there is not much good land left to clear anymore. In some places there is not even any land available at all! That field over there, which used to yield bountiful harvests, now has patches of stress. More of your friends have gone to the city in search of seasonal jobs – or have permanently relocated. It’s as if land had shrunk and aged.

Remote sensing has an ever-growing wealth of information to monitor land processes and vegetation. From optical indices that intercept photosynthesis, to thermal sensors that sip out water stress, to microwaves that carve out canopy structure.

Remote sensing today is the unmatched source of spatial inputs to deal with the large, heterogeneous areas associated with agricultural landscapes. Now the big questions -- Could there be any reasonable justification that rural communities in Africa (or anywhere for that matter) be deprived of access to such key information. Critical metrics to help them inject elements of agricultural landscape design in their development plans? Meaningful maps to help them devise new development plans from scratch? How much longer will they have to wait? How much longer will they have to desert their homes and villages because science failed to provide one essential tool to slow down the exodus?

Very High Resolution (VHR) imagery has significant potential, not only to enrich farmer toolboxes and provide them with spatially explicit, quantitative information about their own farmland, but also as a mechanism to shed new light on fundamental drivers of change such as land scarcity, bringing rural communities to ponder constructively on practical steps towards sustainability and productivity.

SIBWA is one of the Bill & Melinda Gates funded “Quick win” projects (Phase 1), intended to realize the benefits of previously developed projects by targeting ready to go solutions, and to lay the groundwork for future projects implemented as part of Phase 2. SIBWA welcomes stakeholders aboard for a few months adventure in 6 agricultural smallholder communities of West Africa.

For more information contact: P.Traore@cgiar.org

back to top Back to top

ICRISAT-Patancheru, 502 324, Andhra Pradesh, India,
Tel +91 40 30713071, Fax +91 40 30713074,
icrisat@cgiar.org
, URL:www.icrisat.org