Main Menu
Development of a knowledge system for the selection of forages for farming systems in the tropics
Project ID
AS2/2001/029
Commissioned Organisation
CSIRO Sustainable Ecosystems, Australia
Project Leader
Dr Bruce Pengelly
Bruce.Pengelly@csiro.au
Phone:
07 3214 2348
Fax:
07 3214 2308
Project Budget
$837,721.00
Start Date
01/07/2002
Finish Date
30/06/2005
ACIAR Research Program Manager
Dr Bill Winter
Overview Objectives
The aim of this project was to develop a database that will help smallholder farmers in the tropics to select the best possible forage for their situation. The Selection of Forages for the Tropics (SoFT) database will allow smallholders to improve their productivity and thus benefit from an increasing demand for livestock products.
Project Background and Objectives
The demand for livestock products is expected to double by 2020. Smallholder farmers in developing countries, through more efficient and productive farming, will be able to benefit from this increase in demand. One way to achieve this is through planting well-adapted forages. However, it is often difficult for people in developing countries to identify and obtain the best possible forage. At present, much of the important information is fragmented, unpublished or published in media of limited circulation. A database is needed to capture the accumulated knowledge in a form that is applicable to all farming systems in the tropics.
Progress Reports (Year 1, 2, 3 etc)
Year 1:
Year 1 (01/07/2002-30/06/2003)
Objective 1. To develop a knowledge system for the identification of forages suitable for specified niches within smallholder farming systems
There have been three major tasks associated with this first project objective:
1) Selection of the parameters and states to be used in the selection tool and the information to be provided in the major SoFT output, the fact sheets associated with each species;
2) Design of the selection tool (acting upon dot point #1);
3) Acquisition of information from experienced agronomists and from the literature.
The attributes to be used for forage selection and the key elements (criteria and states for each criterion) of information to be recorded on database fact sheets were decided in a design workshop held in Bangkok in October 2002. That workshop was attended by the project management team plus forage agronomists from China and Thailand (a total of 12 agronomists) as well as Dr Peter Horne, CIAT, Laos. The key questions put to the workshop were:
Selection of the parameters and states to be used in the selection tool and the information to be provided in the major SoFT output, the fact sheets associated with each species.
The design of the selection tool (acting upon dot point #1)
Acquisition of information from experienced agronomists and from the literature.
The workshop provided excellent input into the design and development of the database. A critical outcome of the workshop was the decision that the final project product would feature static geographic information system (GIS) output rather than have a dynamic GIS facility. Any dynamic facility, while achievable, would have has very poor resolution because of the lack of detailed data inputs available. It was thought that there was a risk that potential users of the SoFT product would tend to apply the outcomes despite the poor resolution with potentially poor selection being made. Rather, it was agreed to include in the SoFT database outputs from current GIS research being conducted by Dr Michael Peters and Ms Rachel O'Brien at CIAT.
Following on from the Bangkok meeting, the design of fact sheets and selection criteria has been finalised and these designs incorporated into the LUCID selection tool. This selection tool is now well into advanced development as demonstrated by its use in the August 2003 workshops to demonstrate the forage data and selections that are now possible as a result of the information obtained from the early data acquisition activities. The data input tool and a 'fact sheet builder'(which enables easier compilation of fact sheets within the predetermined template) have been developed. Mr Arturo Franco (CIAT) investigated options for the World Wide Web version of the database in a visit to Brisbane, Australia in March 2002.
The major task of assembling data commenced in 2002-03 with workshops of experienced agronomists held in Brisbane and Addis Ababa in March and May respectively. By June 2002 a total of about 100 species had been reviewed at these workshops with data and expert comments recorded on adaptation and utilisation of each forage.
In addition to the data assembly from experienced agronomists at these workshops, major reviews of a range of published literature have commenced in Australia and Africa (ILRI) with the aim of producing a bibliography for each species in the database. This literature review is aimed at both the traditional scientific literature and less well known literature such as annual reports from various research projects.
Objective 2: To promote the system within the 'communities' who are using tropical forages.
The major activities under this objective have been undertaken in parallel with the regional workshops. More than 40 agronomists have already attended project workshops in the first year. The project leader also visited FAO and demonstrated the LUCID SoFT prototype in October 2002.
Objective 3: To develop a strategy for maintenance and updating the knowledge system.
Discussions are currently under way between FAO and the project team about new options for maintenance of the database. It is hoped that FAO will take a lead role in partnership with CIAT to maintain the database. Dr Stephen Reynolds, FAO, Rome, appears keen to collaborate and have the database linked to the FAO databases.
Year 2:
The development of a computer based selection tool and database is progressing. with the collaboration of a large number of forage experts from across the tropics and aims to deliver an expert system that will provide information on the management, use and variation between cultivars and elite accessions for over 120 major forage species for smallholder and large scale agricultural systems.
Develop a knowledge system for the identification of forages suitable for specified niches within smallholder farming systems
Nineteen attributes have been selected for use in the SoFT with between 2 and 8 states in each attribute. Selection criteria available in the SoFT database for selecting the most suitable forages for environments and uses
Climate/farming system attributes
Soil environment attributes
Plant attributes
Latitude x altitude
Soil pH
Plant family (legume or grass)
Rainfall (average annual)
Level of available soil Al/Mn
Life cycle
Length of dry season
Level of soil salinity
Growth form
Inundation
Soil drainage
Stem habit
Intended forage use
Soil texture
Cool season growth
Grazing pressure
Soil fertility
Frost tolerance (foliage damage)
Shade environment
SoFT selections are dependent upon the coding for each species in each attribute. This coding has been carried out by the Australian team members using their own experience in addition to the information from the workshops and literature reviews. The selection tool component of the database has been undergoing testing to ensure that selection outcomes are appropriate and that no obvious species are omitted.
Each fact sheet in the database will have about 5 photos. So far over 1000 high quality photos of forages have been sourced from forage scientists in Australia and overseas and have been placed in a database for filing and access. Photo subjects include juveniles, plants and forages being used.
Three workshops were held during the year at the of University of Hohenheim at Stuttgart Germany, in Cali, Colombia and in Hanoi, Vietnam, attended by more than 40 forage scientists. The workshop in Hanoi took place in conjunction with the annual meeting of the Forages for Smallholder Projects. Literature searches have continued and over 6,000 references have been assembled for the database.
Fact sheets have been written for about 70 of the most important fact sheets and a further 50 are expected to be completed before the database goes to press in March next year. These fact sheets are being written with a specific format and are of various lengths, depending primarily on the importance of the species.
Promote the system within the 'communities' who are using tropical forages
More than 70 agronomists have now attended project workshops across Asia, Africa, Central and South America, and Australia Negotiations are underway with FAO to have the database linked to their website and for FAO to be a distributor of the CD version. The project leader also plans to visit CABI in late 2004 to discuss the role of the database within their compendium system. Plans have also been made to have the database launched at the main congress session at the International Grassland Congress in Dublin in June 2005, and the web version of the project will be housed on the CIAT website, with links from the CSIRO Sustainable Ecosystems website.
Develop a strategy for maintenance and updating the knowledge system
Discussions with FAO and the project team about options for maintenance of the Database are continuing and it is proposed that FAO will take a leading role in partnership with CIAT to maintain the database. This arrangement will be crucial in ensuring that CIAT and others are equipped to deliver upgrades of the SoFT database.
Year 3:
Final Report covering this period has been received.
Project Outcomes
This project accessed the best available information for adaptation and use of 180 tropical forage species and their elite cultivars integrated in a single user-friendly database. The database, which includes a simple-to-use tool to assist in the selection of the best-bet species, is now freely available on the Internet (www.tropicalforages.info) and on CD.
The database has five main features:
information in fact sheets on the adaptation, uses and management of forage species, cultivars and elite accessions;
a selection tool built on LUCID that enables easy identification of best-bet species;
a bibliography of more than 6,000 references and abstracts on forage diversity, management and use which will enable users with poor library facilities to access summaries of some of the key literature;
global maps of climate adaptation for each species; and
a collection of photographs and images of species to help in their identification and use.
The database selection tool is an expert system based on the experiences of forage specialists who have worked for many years in tropical and subtropical regions of Africa, lower latitude USA, Central and South America, South and South-east Asia and Australia. Selections were made on the basis of 19 criteria. The project brought teams of experts together in workshops in Africa, Asia, South America, Central America, Europe and Australia over a two year period and had input from other forage specialists during the database development.
The principal outcome summarised information on tropical forage adaptation and use from expert knowledge, available literature and experiential sources made available in a readily accessible and consistent format. With availability on DVD and the Internet, the database allows researchers and advisors to select those forages most suitable for local conditions. Although the database was initially designed for use in developing countries, its content includes species adapted to farming systems in developed countries and is equally applicable in these regions. The database covers a wide range of forage uses and allows users to select among many different farming systems ranging from permanent to short term pastures, with applications in agroforestry, inter-row cropping, cut-and-carry, hedgerows, green manures and ground covers.
The database has been recognised as a valuable teaching tool for colleges and universities with feedback from many university staff from a wide range of countries attesting that the database and selection tool will have a major role in improving the way tropical forage science is taught.
The database was promoted during regional workshops, through partner organisations including CIAT and ILRI. The project team also conducted database demonstrations at the XX International Grassland Congress, held in Ireland in June 2005. Over 100 tropical forage agronomists trialled the selection tool and fact sheets. FAO became included in the project's partnership and agreement after discussions over the past 3 years. Involvement with FAO will further promote awareness of this database internationally.
The database resides on the CIAT Internet server and where it is maintained and updated. After a few months a need for information to be added to the fact sheets was identified and requests for additional funding have been made. CIAT is also exploring possibilities for the database to be translated into Spanish to encourage application in Latin America and the Caribbean.
Location
There are no project locations defined for this project.
